HELEN GRAHAM

The Spanish
Republic at war
1936-1939
Cambridge
iosten**IA62ttf
 THE SPANISH REPUBLIC AT WAR
1936-1939

ft

\ . . by some distance the best book I have read in any language on

the Spanish Republic during the Civil War. The detailed - and
thoroughly analytical - narrative of the politics of the Republic
outshines anything that has been written before ... A very major
work by a mature historian writing at the height of her powers.’
Professor Paul Preston, London School of Economics and
Political Science

This is a new and comprehensive analysis of the forces of the Spanish

left - interpreted broadly - during the civil war of 1936-9, and the
first of its kind for more than thirty years.

The book argues two crucial propositions. First, that the wartime
responses (and limitations) of the Spanish left - republicans, social-
communists and anarcho-syndicalists - can be understood only
ists,

pre-war experiences, world views, organisational

in relation to their
structures and the wider Spanish context of acute uneven develop-
ment which had moulded their organisations over previous decades.
Second, that the overarching influence that shaped the evolution of
the Republic between 1936 and 1939 was the war itself: the book ex-
plores the complex, cumulative effects of a civil war fought under the
brutally destabilising conditions of an international arms embargo.

Helen graham is Reader

Spanish History, Royal Holloway,
in
University of London. Her book
Socialism and War: The Spanish
Socialist Party in Power and Crisis, 1936-1939 was published by

Cambridge University Press in 1991, and she has otherwise

published widely on the political, social and cultural history of Spain
in the 1930s and 1940s.
 Digitized by the Internet Archive
in 2016

https://archive.org/details/isbn_9780521459327
THE SPANISH REPUBLIC
AT WAR 1936—1939

HELEN GRAHAM

Mi Cambridge
'|j^ UNIVERSITY PRESS
PUBLISHED B“Y THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge, United Kingdom

CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, UK
40 West 20th Street, New York, ny 10011-4211, USA
477 Williamstown Road, Port Melbourne, vie 3207, Australia
Ruiz de Alarcon 13, 28014 Madrid, Spain
Dock House, The Waterfront, Cape Town 8001, South Africa
http:// www.cambridge.org

© Cambridge University Press 2002

This book is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 2002

Printed in the United Kingdom at the University Press, Cambridge

Typeface Baskerville Monotype 1 1 / 12 .5 pt System DTj^X 2g [tb]

A catalogue recordfor this book is availablefrom the British Library

isbn o 521 45314 3 hardback

isbn o 521 45932 x paperback
For Herbert Rutledge Southworth,
in memoriam 1908-1999
Paris, julio, tengo frio, mama, tengo frio. Estaba llorando Rosell
por Bonet, por Oviedo, por el fragil esqueleto del pajarillo de la
Libertad, por si mismo, y en la oscuridad crecia una besda cubica de
mandlbula poderosa y labios despectivos sobre un fondo de marchas
militares y gritos de rigor, rugidos invertebrados que expulsaban la
musica y la palabra.

Paris, July, I’m cold, Mamma, Pm so cold. Rosell was weeping for all
of them, for Bonet, for Oviedo, for the fragile frame of the tiny bird
that was Freedom, for himself, and out of the darkness there grew
a monstrous massive-jawed, sneering beast, against a background
of military marches and the obligatory shouting, incoherent roars
drowning out music and words.
Manuel Vazquez Montalban, El pianista (Barcelona: Seix Barral,
1985), PP- 270-1

You who will emerge from the flood

we have gone under
In which
Remember
When you speak of our failings
The dark time too
Which you have escaped
Bertolt Brecht, ‘To those born later’
 1

Contents

List ofplates page viii

List of maps x
Preface xi
Acknowledgements xiii

Introduction A fractured left: the impact of uneven

development (1898-1930) 1

1 The challenge of mass political mobilisation (1931-1936) 23

2 Against the state: military rebellion, political

fragmentation, popular resistance and repression

(18 July-4 September 1936) 79

3 Building the war effort, building the state for total war
(September 1936-February 1937) 13

4 Challenges to the centralising Republic: revolutionary

and liberal particularisms in Catalonia, Aragon and
the Basque Country 215

5 The Barcelona May days and their consequences

(February-August 1937) 254
6 Negrin’s war on three fronts 316

7 The collapse of the Republican home front 390

Glossary 426
Bibliography 434
Index 464

vii
 )

Plates

Between pp. 146 and 14 j

1 Francisco Largo Caballero (Ministerio de Educacion, Cultura y

Deporte, Archivo General de la Administracion)
2 (a) and (b) Madrid December 1936-January 1937 (Vera
front
Elkan Collection (HU71664 and HU 71662) Photographs
courtesy of the Imperial War Museum, London)
3 (a) Buenaventura Durruti: ‘Emulate the hero of the people’
(Ministerio de Educacion, Cultura y Deporte, Archivo General de
la Guerra Civil Espanola (Kati Horna collection))

3 (b) Wall posters (Abraham Lincoln Brigade Archives Collection,

Tamiment Library, New York University, Albert Harris Collection)

4 Home-produced armoured car c. January 1937 (Vera Elkan
10
Collection (HU 7 1512) Photograph courtesy of the Imperial War
Museum, London
5 Mijail Koltsov (Vera Elkan Collection (HU 71579) Photograph
courtesy of the Imperial War Museum, London)
6 Santiago Carrillo, Julian Zugazagoitia and Fernando Claudin
(FredCopeman Collection (HU 34724) Photograph courtesy of
the Imperial War Museum, London)
7 Catalan Pioneers youth group (Fred Copeman Collection (33003)
Photograph courtesy of the Imperial War Museum, London)
8 Parade in support of Popular Army (Fred Copeman Collection
(HU 33009) Photograph courtesy of the Imperial War Museum,
London)
9 Women’s factory labour (postcard) (Ministerio de Educacion,
Cultura y Deporte, Archivo General de la Guerra Civil Espanola)
Giral,Negrin and Azana visit the Madrid front (Ministerio de
Educacion, Cultura y Deporte, Archivo General de la
Administracion)

viii
 List of plates IX

11 Shattered houses after Barcelona air raid (Fred Copeman

Collection (HU 33 151) Photograph courtesy of the Imperial War
Museum, London)
12 Republican child refugees demonstrate co-educational principles
in action (Fred Copeman Collection (HU 33143) Photograph
courtesy of the Imperial War Museum, London)
13 Republican wall newspaper, 1938 (Fred Copeman Collection
(HU 33062) Photograph courtesy of the Imperial War Museum,
London)
14 Republican soldiers crossing the Ebro, 1938 (Fred Copeman
Collection (HU 331 17) Photograph courtesy of the Imperial War
Museum, London)
15 Wounded Republican soldier (Fred Copeman Collection
(HU 34628) Photograph courtesy of the Imperial War Museum,
London)
16 War memorial, Serra de Pandols, Catalunya (David Leach)
 Maps

1 The division of Spain 22 July 1936 page 107

2 The advance of the Army of Africa August-October 1936 1
14
3 Barcelona: urban development since 1850 431
4 The division of Spanish territory March 1937 432
5 The division of Spanish territory July 1938 433
 Preface

This book is concerned with the Second Spanish Republic during the

civil war of 1936-9 and the reasons for its defeat. Its central arguments
can be encapsulated in two crucial propositions. The first is that the
wartime responses (and limitations) of the Spanish left - republicans,
socialists, communists and anarcho-syndicalists - can only be under-

stood in relation to their pre-war experiences, worldviews, organisational

structures and the wider Spanish national context of acute uneven devel-
opment which had moulded their organisations over previous decades.
The second is that the overarching influence that shaped the evolution
of the Republic between 1936 and 1939 was the war itself.
It is remarkable the extent to which existing analyses have in practice

relegated the war to background noise or narrative filler between chunks

of political analysis that nevertheless remain largely dissociated from it.
But the war had a complex and cumulative impact on every aspect of
Republican polity and society. If we are to understand what happened
and why, then we have to recreate its texture. More particularly because
this was a civil war, and one fought under the devastating conditions

of Non-Intervention imposed by Britain and France. This meant virtu-

ally total international isolation and a de facto economic embargo that

placed the Republic - and only the Republic — at an enormous material

disadvantage throughout.
Coverage of Non-Intervention to date has concentrated on its interna-
tional diplomatic aspects and mainly on the first year of the war. But what
destroyed the Republic was the long-term impact of Non-Intervention
over nearly three gruelling years. It brought the daily erosion not only of
the Republic’s military capacity, but of its political legitimacy as well. For
economic embargo prevented the Republic from sustaining the social
and economic fabric of the home front and, in the end, from meeting
even the minimal requirements of its population in terms of food and
shelter. Vast too was the psychological cost of war under such conditions.

xi
The international political diplomacy that produced and sustained Non-
Intervention also repeatedly blocked all the Republic’s political exits,
making it impossible for it an end to the conflict in 1938.
to negotiate
In the last agonising months international mediation was still withheld,
even though it was the only course that might have reduced the risk
of massive violent reprisal against the defeated. In the end the Spanish
Republic collapsed inwards under the huge, intolerable pressures born
of the war. A war that others had forced the Republic to fight would end
by consuming it utterly.
 Acknowledgements

This book has been a long time in the making, and the volume of my debts
is commensurate with the time scale. Sir Raymond Carr offered unfailing
support as an academic referee. Along the way I enjoyed a Leverhulme
Research Fellowship, which permitted a sustained period of archival
research in Spain. Additional sabbatical leave from Royal Holloway al-
lowed me to extend this further. I am grateful here to the College, but
most especially to my departmental colleagues for covering teaching (and
innumerable chores) in my absence. Enrique Moradiellos and Susana
Botas offered me a warm welcome in Madrid, while Montserrat Delgado
Moreno has put a roof over my head there more times than I can re-
member. In London Penny Green and Bill Spence helped me through
the long haul. Historian friends and colleagues - in Britain and Spain,
Hispanist and otherwise - offered me intellectual sustenance, leads, in-
formation, advice, encouragement and kindness in amazing quantity
and variety. So my thanks here to: Michael Alpert, Julio Arostegui,
Richard Baxell, Jerry Blaney, Martin Blinkhorn, Kayvon Boyhan,
Hilary Canavan, Jim Carmody, Julian Casanova, Andrew Dowling,
Chris Ealham, Sheelagh Ellwood, Tim Fletcher, Jesus Garrido, Maria
Jesus Gonzalez-Hernandez, Liz Harvey, Gerald Howson, Joel Isaac,
Angela Jackson, Christoph Jahr, Tim Kirk, John Maher, Ricardo
Miralles, Enrique Moradiellos, Rudolf Muhs, Gerard Oram, Pilar
Ortuno, Hilari Raguer, Tim Rees, Nick Rider, Francisco Romero,
Ismael Saz, Angelo Smith, Sandra Souto, Dan Stone, Angel Vinas
and Mary Vincent. Santos Julia and Borja de Riquer gave me valu-
able archival orientation and helped track down recalcitrant refer-
ences. Sir Geoffrey Cox (News Chronicle correspondent in Madrid,
October—December 1936) provided morale-boosting encouragement.
My special thanks are due to Frank Schauff for generously sharing
with me extensive material from his research in the Soviet archives.
I am also grateful for the assistance of the Canada Blanch Centre
xiv Acknowledgements

for Contemporary Spanish Studies (LSE), whose library proved an

invaluable resource in the final preparation of the typescript. Tony
Kushner, Manuel Vazquez Montalban and Milton Wolff kindly agreed
words in my epigraphs. David Leach allowed me to
to the use of their
include his photograph of the Serra de Pandols memorial. The Republi-
can poster on the book jacket is from the collection ofjordi Carulla
and is reproduced with his kind permission. Bill Davies at Cambridge
University Press has shown exemplary patience during the long gesta-
tion of this book. Francisco Romero read and commented on numerous
draft chapters with his habitual incisiveness. Angela Cenarro has taught
me a great deal about the Spanish Civil War, as well as offering much
practical assistance. Paul Preston too has always been there to help. My
debt to him goes a long way back — for his constant support and quite
extraordinary generosity as a scholar over the many years of our friend-
ship. This book is dedicated to Herbert Southworth, miner, librarian,
bibliophile and pioneering historian of the Spanish Civil War — for his
passionate belief in the transformative power of forensic history and for
the monument to it which he has left us in his own work.
 INTRODUCTION

A fractured left: the impact of uneven development

(1898-1930)

The Spanish Civil War would begin with a military coup. Although
there had been a long history of military intervention in Spanish po-
litical life, the coup of 17-18 July 1936 constituted an old instrument

being used to a new end. It aimed to halt the process of mass political
democracy kick started by the effects of the First World War and the
Russian Revolution, and accelerated by the ensuing social, economic
and cultural changes of the 1920s and 1930s. In this sense, the military
coup against the democratic Second Republic in Spain was intended to
have the same function as the fascist take-overs that followed the com-
ing to power of Mussolini and Hitler in Italy and Germany. All these
European ‘civil wars’ (because civil wars can take many forms) had their
origins in the cumulative political, social and cultural anxieties provoked
by a process of rapid, uneven and accelerating modernisation (that is,
industrialisation and urbanisation) occurring across the continent. All
those who supported Spain’s military rebels in 1936 had in common a
fear of where change was leading - whether their fears were of material
or psychological loss (wealth, professional status, established social and
political hierarchies, religious or sexual (i.e. gendered) certainties) or a
1
mixture of these things.
That the military should function in 1936 as the ultimate guardian of
a certain kind of social and political order indicates not only the positive
fact of its own (at least relative) ideological cohesion but also the extent of
the fragmentation among other social and political groups. The histori-
calprotagonism of the Spanish army had its roots in the war-dominated
nineteenth century. But its enduring twentieth-century political protago-
nism was a consequence of the lack of any minimally coherent bourgeois
project for national development. The process of modern economic de-
velopment in Spain occurred late and very unevenly - even judging

1
For the wartime consequences of these fears, see chapter 2 below.

I
2 The Spanish Republic at war

by the standards across Europe as a whole. 2 As a result, Spain’s elites

were highly regionally fragmented. This was exacerbated by the events of
1898 when Spain lost the remnants of its old overseas empire (principally
Cuba and the Philippines). While elite groups and some sectors of Spain’s
middle classes were united in perceiving this as a political - even an ex-
istential - crisis, their responses to ‘the disaster’ were far from unified. 3

Most importandy, the economic consequences of imperial loss (in par-

ticular of protected markets) had galvanised the industrial plutocracy in
Catalonia, the most economically and sociologically advanced region of
Spain, to launch what would become a powerful middle-class movement
for regional autonomy. As a result, the first two decades of the twentieth
century saw a bitter and at times violent political struggle between the
‘old’ political centre in Madrid (representing the powerful landed elites)

and the aspiring plutocratic autonomists of industrial Barcelona over

the future direction of national economic policy. In essence this was a
dispute over who would pay for infrastructural modernisation. Not only
were Spain’s elites unable to agree here, but in their continuing disagree-
ment they would, by 1918, find themselves facing an actively mobilising
labour force.
At the start of the twentieth century, however, Spain was still a rural
sea out of which emerged a few urban, industrial ‘islands’. These were
confined predominantly to two areas. First, as we have seen, there was
Catalonia (especially the industrial belt of Barcelona) on the north-east
sea board which produced mainly textiles; and second, the north - the
Basque Country (Vizcaya) and Asturias - was an area of heavy industry
and mining. In a reduced number of other cities too (Madrid, Zaragoza,
Valencia, Seville) the development of small industrial sectors was also
gradually feeding urbanisation and the emergence of organised labour.
Uneven development and the consequent lack of an integrated national

2
J.Nadal, Elfracaso de la revolution industrial en Espana, 1814-1913 (Barcelona: Ariel, 1975); N. Sanchez
Albornoz (ed.), The Economic Modernisation of Spain 1830-1930 (New York: New York University
Press, 1987); G. Tortela, El desarrollo de la Espana contemporanea. Historia economica de los siglos XIXy XX
(Madrid: Alianza Universidad, 1994). A revisionist account in Leandro Prados de La Escosura,
De imperio a nation: crecimientoy atraso economico en Espana 17 80-1930 (Madrid: Alianza, 1988). A resume
in A. Shubert, A Social History ofModern Spain (London: Unwin Hyman, 1990), pp. 9-56.
3 S. Balfour, ‘The Loss of Empire, Regenerationism and the Forging of a Myth ofNational Identity’,

in H. Graham andj. Labanyi (eds.), Spanish Cultural Studies. An Introduction (Oxford: Oxford Univer-
sity Press, 1995), pp. 25-31; S. Balfour, The End ofthe Spanish Empire 1898-1923 (Oxford: Clarendon

Press, 1997), pp. 64-91. The frenetic debate over ‘regeneration’ derived from anxieties similar to
those that were provoking (ultimately social darwinist) discourses of ‘degeneration’ elsewhere in
Europe at that time. Cf. also M. Richards, A Time of Silence. Civil War and the Culture of Repression in
Franco’s Spain 1936-1943 (Cambridge: Cambridge University Press, 1998), pp. 47-66.
,
 A fractured left 3

market would in turn inhibit other sorts of exchange that might later have
mitigated or modified the consolidation of antagonistic social constituen-
cies and cultural perspectives or belief systems. None of these problems
and symptoms was specific to Spain, of course. Acute urban-rural divi-
sions along with emergent class tensions were the common by-products of
modernising change in early-twentieth-century Europe. But our knowl-
5
edge of hindsight, of the ‘hot civil war of 1936-9, inevitably leads us to
ask what, if anything, was particular about the Spanish experience.
Until the First World War there were relatively low levels of demo-
graphic mobility (from countryside to city) in Spain. The rural majority
was also highly atomised, living in villages and hamlets. Both these fac-
tors contributed to produce relative social stability This did not mean the
absence of social strife, but rather a situation where popular protest could
be easily contained within a given locality. And for Spain’s ruling elites -
5
composed of its ‘senior landowning partner in uneasy alliance with the
,

component - the primary* function of the state

5
‘junior urban-industrial
was to ensure this containment. While this could be ensured, and for
as long as there was no orchestrated political or social challenge to elite
hegemony, the fact of the elites’ own fragmentation/ regionalisation did
not matter. Popular rebellion could go on being considered by both elites
and the political authorities of the Restoration monarchy (1875—1923) as
purely a matter of public order.
Until the mid twentieth century, for most of the Spanish population it
was patterns of land ownership and agricultural exploitation that struc-
tured the political and social hierarchies they dwelt within and which
shaped their cultural worldviews. These structures and cultural per-
spectives varied enormously, however, between the north, centre and
south of Spain. The dominant form of land holding in the centre-south
(New Castile downwards) was the latifundio. These were vast estates,
run mainly by bailiffs in the absence of their aristocratic owners and
farmed by virtual slave armies of landless day labourers. In the north
and on the central tableland of Old Castile (the Meseta) the agrarian
norm was the peasant smallholder or tenant farmer. Individual peasants
and their families ranged from affluent to extremely poor depending
on a variety of factors — inheritance law, tenure arrangements and ge-
ography/relief (that is, land quality* and climate). But the fact of land
ownership itself tended to produce more conservative social attitudes.
In spite of the highly disparate levels of economic resource possessed by
these rural lower-middle-class sectors, there was often a sense that they
belonged to some commonality of landowners. This was more real than
r
4 The Spanish Republic at war

any sense of CQnneetion more sparse) urban equivalents

to their (rather
who increasingly had and aspirations. Moreover, within
different values
these rural worlds there were no integrative mechanisms operating that
might have fostered a sense of national belonging - such as markets, or
political/ cultural options like a functioning system of national primary
education or genuinely participatory forms of political representation.
Nor could military service fulfil this function as it did, to some extent, in
neighbouring France. The only wars in which Spain was involved (up to
the civil war of 1936-9 itself) were old and new colonial ones - in Cuba
and Morocco 4 respectively But these failed signally to stimulate bind-
ing or sustained patriotic sentiment among any significant proportion
of Spain’s middling classes. Cuba was the last gasp of a dying empire,
while Morocco was associated with the narrow economic interests of the
crown and some aristocratic and clerical sectors who owned iron mines
there. 5 In short, the colonial fervour exhibited elsewhere in Europe by
the emergent middle classes was not replicated in Spain. Most remained
indifferent, while the worker constituencies who bore the brunt of con-
scription were frankly hostile. 6 Spain’s middling classes remained highly
fragmented internally in a variety of economic, social and cultural ways -
and, needless to say, the peasantry felt quite distinct from landless ru-
ral labourers (whether far away or nearby) even though their levels of
subsistence may not always have differed gready.
Apart from land ownership, the major force of cohesion across social
classes in north and central Spain was religion. In the north and on
the Meseta the worldview of virtually the entirety of rural dwellers —
of whatever social extraction or status - was shaped by their Catholic
faith (although this was not necessarily a homogeneous entity). Popular

4 Under the terms of the Treaty of Cartagena (1907) the Great Powers had allotted Spain (which
already controlled the enclaves of Ceuta and Melilla) the task of policing northern Morocco.
5 Balfour, The End of the Spanish Empire 1898-1923', C. Serrano, Final del Imperio. Espaha 1893-1898
(Madrid: Siglo XXI, 1984). Morocco was a military protectorate - it was never setded by civilian
colonists. For the later political importance of Spain’s Moroccan policy, see P. La Porte, La atraccion
del Iman: el desastre deAnnualy sus repercusiones en la politica europea 1921 -23 (Madrid: Biblioteca Nueva,
2001). As La Porte indicates, it was only with the Annual catastrophe that the unpopular and
underfunded Moroccan campaign was brought to the political fore in Spain.
6
The best-known anti-militarist protest was the (also anti-clerical) Tragic Week ofJuly 1909: see
Joan Connelly Ullman, The Tragic Week: A Study ofAnticlericalism in Spain 1875 1912 (Cambridge, —
Mass.: Harvard University Press, 1968), esp. pp. 129-40; S. Balfour, The End of the Spanish Empire
1898-1923, pp. 1 4-3 1 and C. Serrano, Le Tour du Peuple. Crise nationale, mouvements populaires et
1

populisme en Espagne (1890-1910) (Madrid: Casa de Velazquez, 1987). On popular anti-militarism

see R. Nunez Florencio, Militarismoy antimilitarismo en Espaha 1888-1906 (Madrid:CSIC, 1990) and
C. Gil Andres, Echarse a la calle. Amotinados, huelguistasy revolucionarios. La Rioja 1890-1936 (Zaragoza:
Prensas Universitarias de Zaragoza, 2000).
 A fractured left 5

attitudes towards Catholicism and the Church constituted a major defin-

ing difference between north and south. In the centre/ north the Church
was popularly perceived as dignifying existence, providing humanity and
meaning in a ‘heartless world’. This perspective was sustained by the im-
portant pastoral role performed by local priests who forged close links
with their local communities. The Church also provided practical sup-
port - often in the form of rural credit banks - which offered a life-saving
resource to the small peasantry eking out a precarious and marginal
economic existence, perpetually threatened by crop failure and fearful
of falling prey to moneylenders .
7

By contrast, the very different daily experiences of the labouring

masses of centre-south Spain made them fiercely anti-clerical. There
Church and were perceived as pillar and perpetuator of an op-
priest
pressive landed order. Given the vastness of the latifundios and the fact that
monoculture was the norm in the rural centre-south, the landless labour-
ers were usually dependent on a single source of employment which, even
then, was only available for part of the year — at planting and harvesting
times. In the absence of any public welfare provision or other forms of
poor relief, this dependency created relations of social power which were
neo-feudal. The reality of power in rural Spain made meaningless the for-
mal existence of universal manhood suffrage since votes were implicidy at
the disposition of the local landowner (as sole employer), via the offices of
the local political boss or fixer [cacique), who in rural areas would quite of-
8
ten be the estate steward or bailiff . The latter’s close relationship with the
local priest symbolised for the rural workers of the south a microcosmic
expression of the Church’s legitimation-sanctification of the social and
political order which enslaved them .
9 This in turn explains the abiding

7 The residual influence of Carlist traditions was also important in northern Spain. On the later
consequences of this, see J. Ugarte Telleria, La nueva Covadonga insurgente. Origenes socialesy culturales
de la sublevacion de igj6 en Navarra y el Pals Vasco (Madrid: Biblioteca Nueva, 1998).
8
The cacique controlled the levers of power in a given locality - land, employment, taxation, judges,
magistrates etc. So in the north of Spain he might be a tax collector, lawyer or moneylender.
Monarchist Spain (1875-1923) had a two-party system whereby the dynastic parties (Liberals
and Conservatives) alternated in power (termed the turno). Each was guaranteed a ‘turn’ because
majorities in the Cortes (Spanish parliament) were manufactured by means of electoral fraud.
Candidates were imposed on constituencies according to previously agreed lists. The caciques'
major function was thus to deliver the vote and ensure that the agreed ‘result’ materialised. This
could be done either by means of a carrot (offering favours) or stick (coercion) depending on the
region and whose vote it was.
9 The rapprochement between the Church and the secular elites was the concomitant of a pro-
cess of distancing between Church and people that had its roots in the disentailment process,
W. J. Callahan, Church, Politics and Society in Spain 1750-1874 (Cambridge, Mass.: Harvard University
Press, 1984).
6 The Spanish Republic at war

anti-clerical element in the popular protests against this brutal clientelism

which were sparked from time to time by hunger and sheer desperation.
The lack of any substantial rural middle class in the south — beyond
those providing agricultural services for the latifundistas and thus bound
by ties of custom and obligation within the neo-feudal social structure -
made the nature of social oppression quite overt. At the same time,
however, the sporadic, ad hoc nature of rural rebellion in late-nineteenth-
and early-twentieth-century Spain and its atomised occurrence at the
levelof the village meant that it could be very easily quashed by the
efforts of the rural police force, the Civil Guard. This functioned virtu-
ally as a force of occupation in the countryside, thus constituting the third
element in the rural poor’s ‘unholy trinity’ (bailiff, priest and corporal of
the Civil Guard). The fact that the Civil Guard was used primarily (in-

deed almost exclusively) against the labouring classes reinforced in their

eyes the idea of the state as a uniquely repressive force - in so far as it

performed no countervailing ‘positive’ functions in areas such as public

education or (even minimal) social welfare provision.
Given the picture of rural Spain just indicated, it is unsurprising that
labour unions and worker organisations should have had little purchase
therein prior to the disruption of the old networks of power that occurred
with the coming of the Second Republic in 1931. The smallholding and
tenant farmers of the centre and north were made either impervious
or actively hostile to the left’s message through a combination of their
religious faith and the Catholic Church’s provision of material assistance.
But even in other areas of Spain where brutal social inequality was visible
and unmediated - for example along the lower south-eastern sea board
or, in particular, in the deep south of Andalusia and Extremadura - it

is still difficult to speak of a rural ‘proletariat’ as such prior to the First

World War. This is because it presupposes a level of articulation - in
terms of culture, consciousness and organisation - that scarcely existed
in the Spanish countryside before that time.
This did not mean, however, that the southern landless lacked political
beliefs of their own. These they had found in anarchist millenarianism
and During the second half of the nineteenth century and
direct action.
beyond, these offered the landless poor a far more functional - and in-
deed logical — response to extant power relations and the perceived
(im)possibility of incremental change than the state-reformism and grad-
ualist tactics of the Spanish socialists. Social reformism, implicidy predi-
cated on the bargaining power of labour, doubtless seemed as utopian or
incredible to the unskilled and powerless of the saturated southern labour
 A fractured left 7

market in the 1910s as millenarian anarchism now seems to some west-

ern historians of the late twentieth and twenty-first centuries. It w ould r

only be in the 1930s, with the promise of state-led agrarian reform, that a
socialist landworkers union would begin to make rapid headway among
5

the landless of Spain’s deep south.

But to speak of any area of the rural south as organisationally
5
‘anarchist prior to 193 1 can also lead to misconceptions. In Extremadura
and Andalusia there was certainly a deep-rooted popular culture of col-
lectivism thatchimed with anarchist ideas. The poor south was also a
milieu in which anarchist organisers circulated. But the level of stable
organisational implantation of what would become in 1910 the CNT
federation was - for a variety of practical, political and ideological
reasons - never very great before the relatively more conducive political
circumstances of the Second Republic. But by then, of course, the CNT
would be competing hard for southern recruits with the socialist-led
trade union, the UGT.
Beyond the CNT was, like the UGT, predominantly
rural south, the
an urban-based movement (The CNT was stronger on
until the 1930s.
the eastern seaboard, while the UGT had strongholds in the industrial its

north and in the geographical (and political) centre of Spain in Castile.)

This urban focus by the pattern of later civil wartime
is also indicated
collectivisation in Aragon or Valencia, for example, where the collectivist
movement in the villages of the agrarian hinterland w as
r
initiated for the
most part by anarchist cadres fanning out from the provincial towns or
10
cities. In the case of Aragon too, even the CNT’s urban base only really
developed in the years after the First World War. 11
In urban Spain, the First World War acted, as it did elsewhere in
Europe, as a catalyst and accelerator of social, economic, political and
cultural change. It contributed enormously to undermining an order
that had depended on the political demobilisation of the population.
Spain had not participated in the war. But ev en if ‘Spain did not enter
12 5
the war the w^ar entered Spain
. . . Because Spain was a neutral coun-
.

try able to sell to both sides, the war massiv ely stimulated its economy.
But while enormous profits were being made in war-related sectors,
a crisis of domestic production arose because the industrial base was

10
See chapter 4 below
11
J.Casanova. Anarquismoy revolution en la sotiedad rural aragonesa igj6—igj8 (Madrid: Siglo XXI.
-
PP- 32 9 .

12
F. Romero. ‘Spain and the First Worid War: The Structural Crisis of the Liberal Monarchy'.
European History Quarterly, 25 (1995), 532.
8 The Spanish Republic at war

insufficient to*meet normal domestic demand as well as the extraordi-

nary external demand. The result was massive internal inflation with
deteriorating living standards and shortages of food and staple com-
modities for many sectors of the working classes - especially those not
benefiting from the increased wages of war-related industry (although
such increases still did not match inflation). War-induced distortions in
the economy also produced severe sectoral unemployment that forced
thousands to migrate - mainly to Bilbao or Barcelona - in search of
work in flourishing war-related concerns. In response to the pressure of
working-class protest at the subsistence crisis, the UGT and the CNT
agreed a Labour Pact in July 1916 which was designed to pressure the
monarchist government into action against cost of living hikes.
The union rapprochement indicates the severity of the crisis
fact of this
because, for it to have occurred, the UGT
leadership had to overcome
its enormous aversion to any alliance with the CNT, which it associated
with confrontational movement (PSOE/UGT)
tactics. Spain’s socialist

had evolved from an urban base of artisans and skilled workers - typified
by Madrid itself, which, in spite of being the capital city, would retain
into the 1930s a predominantly ‘pre-industrial’ labour structure. 13 And
for all the UGT’s extension to the industrial north (among the iron and
steel foundry workers of the Basque Country and the coal miners of
Asturias), its organisational ethos remained extremely cautious. Its lead-
ership would never approach to labour
entirely cast off a certain guild
relations. Indeed, this was reinforced by the experience of the UGT-led
urban general strike of August 1917, launched as one segment of a demo-
cratic political challenge to the exclusivist order of the monarchy. But
when this bid failed, the socialists found themselves facing the onslaught
of the victors’ justice. The threat of confiscations (of union buildings,
property and printing presses etc.) and the organisational dislocation
that ensued traumatised the Madrid-based UGT leadership and sublim-
inally reinforced its commitment to a gradualist strategy of reform which
would avoid head-on challenges to the state or employers in the future.
As a result of 1917, then, the scene was set for subsequent dissension
in the UGT between this cautious veteran leadership and those whose
different experience of labour relations spoke to them of the need for
more direct, confrontational tactics. The most important example of this
would be the split between Madrid and the industrial north (Asturias
and Bilbao (Vizcaya)). The intransigence of northern industrialists - of
13
How this would change is a central theme of Santos Julia’s Madrid igji 1934. De la fiesta popular
a la lucha de closes (Madrid: Siglo XXI, 1984).
 A fractured left 9

foundry and mine owners (intensified by the post-First World War

slump in their fortunes) - and the angry expectations of the rank and file
were edging UGT leaders in the north towards backing more radical,
confrontational tactics. (These were precisely the kind of tactics being
espoused with renewed vigour by the CNT, which had drawn from the ex-
perience of repression in 1917 conclusions diametrically opposed to those
of the UGT leaders.) As a result, socialist union leaders in the north found
themselves in a stand-off with a disapproving UGT national executive in
Madrid. This union ideology and strategy would be a signif-
conflict over
icant factor in the emergence of the Spanish Communist Party in 1921 —
14
2. Its creation did not, however, resolve the internal debate over strategy

in the socialist movement. Indeed, the existence of a separate party (even

one as marginal as the pre-1936 Spanish Communist Party) created new
organisational jealousies and animosities that would make the internal
socialist debate even more fraught - as developments during the highly
charged years of the Second Republic demonstrate. The years following
1917 would see a spate of isolated and uncoordinated acts of rural rebel-
lion in Spain that are collectively known as the ‘bolshevik three years’
( trienio bolchevique). These rural outbursts helped stoke the fires of elite

anxiety about social disorder. But for all their increased frequency and
intensity, they were easily quelled by the Civil Guard. Rather it was in
‘minority’ urban Spain that the real challenge to the old order was
growing.
This process had germinated with the colonial disaster of 1898. This in-
jected new life
an urban middle-class republicanism that was highly
into
critical its dominant clerical and military influ-
of the old regime with
ences. In towns and cities, republican alliances began to erode the caciques'
control in the decade leading up to the First World War. In the war’s
wake came further accelerating demographic change, as workers moved
from countryside to town and city. Thus, just as the habitual mechanism
of political control, the influence of the cacique was foundering, so too the
,

old techniques for guaranteeing public order were looking increasingly

inadequate. Nowhere was this clearer than in Barcelona.
None of Spain’s other urban centres cast even a pale reflection of
the cultural, political and sociological complexity of Barcelona — Spain’s

14
G. Meaker, The Revolutionary Left in Spain ig 14-^23 (Stanford: Stanford University Press, 1974),
pp. 346-84, esp. pp. 369, 371. The emergent Communist Party would have its strongest base in
Bilbao, where for a time it controlled the Casa del Pueblo. On
the radical strategy of the Basque
socialists, see R. Miralles, ‘La gran huelga minera de 1890. Los origenes del movimiento obrero
en el Pais Vasco’, Historia Contemporanea, 3 (1990).
10 The Spanish Republic at war

main industrial conurbation and its only real metropolis. 15 Anti-state sen-
timent and hostility to the socio-economic order of capitalism were to be
found among organised workers and the poor elsewhere in Spain. But
nowhere else could radical left minorities draw on any comparable tradi-
tion or structure of proletarian/ popular political and cultural networks -
from the popular ateneus through alternative press and literary output, to
,

forms of alternative sociability. 16 The pumping heart of all of these was

the anarchist movement, whose creative energy offered a mirror image
of Catalonia’s unique bourgeois milieu of civic dynamism and its thriving
traditions of cultural autonomy. Anarchist culture in twentieth-century
Barcelona was also, of course, a political culture based on traditions
of community self-help and direct action (for example popular requisi-
tion against food shortages and neighbourhood resistance to evictions). 17
Barcelona’s anarchists were in semi-permanent mobilisation against the
old order. Their protest was anti-clerical - against the Church’s legitima-
tion of that order- and anti-militarist, not least because the old regime’s
18
colonial wars were fought predominandy by the poor.
But anti-clericalism was never the monopoly of Spain’s anarchists. 19
Between 1901 and 1909 in Barcelona it was to be the principal mobilising
mechanism of the Radical Republican Party, led by a shady demagogue
and political opportunist called Alejandro Lerroux. 20 The fact that for

15
Basque Vizcaya (focused on the city of Bilbao and the River Nervion) was
industrialisation in
also rapid. But there a ribbon of small industrial centres developed, with no comparable centre
to Barcelona. The political division between indigenous workers and migrant Spanish labour
was also far more extreme in the Basque Country than it was in Catalonia.
16
B. Hofmann et al., Anarquismo espanol y sus tradiciones culturales (Frankfurt-on-Main /Madrid:
Vervuert-Iberoamericana, 1995); P. Sola i Gussinyer, Educacio i Moviment Libertari a Catalunya
(igoi-iggg) (Barcelona: Edicions 62, 1980); S. Tavera, ‘La premsa anarco-sindicalista (1868-
193 1)’, Recerques, 8 (1977), pp. 85-102; L. Litvak, Musa Libertaria. Arte, literaturay vida cultural del anar-

quismo espanol (1880-igig) (Barcelona: Antoni Bosch, 1981); T. Kaplan, Red City, Blue Period. Social
Movements in Picasso’s Barcelona (1888-iggy) (Berkeley, Calif.: University of California Press, 1992).
17
C. Ealham, ‘Policing the Recession: Unemployment, Social Protest and Law-and-Order in
Barcelona 1930-1936’, unpublished Ph.D. thesis, University of London, 1995.
18
J. Romero Maura, ‘La Rosa delFuego’. Republicanosy anarquistas: lapolitica de los obreros barceloneses entre
el desastre colonialy la Semana Tragica i88g-igog (Barcelona: Grijalbo, 1989); X. Cuadrat, Socialismo

y anarquismo en Cataluna i8go-ign (Madrid: Revista del Trabajo, 1976); J. M. Huertas Claveria,
Obrers a Catalunya (Barcelona: Avene, 1982); P. Gabriel, ‘Sindicalismo en Cataluna 1888-1938’,
8 (1990) and ‘Poblacion obrera catalana’, Estudios de Historia Social, 32-3 (1985);
Historia Social,
A. Smith, ‘Anarchism, the General Strike and the Barcelona Labour Movement 1899-1914’,
European History Quarterly, 27 (1) (1997). See also A. Balcells, Trabajo industrialy organization obrera en
la Cataluna contemporanea igoo-igg6 (Barcelona: Ed. Laia, 1974) and Ullman, The Tragic Week. On
popular anti-militarism, see also n. 6 above.
19
On the social and cultural context of anti-clericalism, see F. Lannon, Privilege, Persecution and

Prophecy. The Catholic Church in Spain i8yy-igyg (Oxford: Clarendon Press, 1987), pp. 9-35.
20
J. Alvarez Junco, El emperador del Paralelo. Lerroux y la demagogia populista (Madrid: Alianza, 1990);

J. R. Mosher, The Birth of Mass Politics in Spain: Lerrouxismo in Barcelona igoi -igog (New York:
 1

A fractured left 1

a time he had considerable success in mobilising the Barcelona working

class indicatesone of the key specificities of the anarchist movement - its
refusal to engage in the sphere of organised (i.e. parliamentary or
municipal) politics. Given the absence from the Catalan frame of the
PSOE - whose scarcely dissembled centralism would also come to an-
tagonise such a strongly federalist region - this meant that the worker
constituencies of Barcelona were, effectively until the 1930s, deprived of
21
any real political voice for their interests.
After the end of the FirstWorld War, conflict was escalating in the
industrial heartland of Barcelona between the emergent influence of ur-
ban labour in the CNT, on the one hand, and employers’ opposition
to arbitration-based solutions on the other. This opposition stemmed in
part from a particular political mentality. But that was itself heavily in-
fluenced by longstanding economic factors. The Catalan textile industry
had always been ramshackle and undercapitalised. After the temporary
reprieve granted by the exceptional circumstances of the 1914—18 war, it
was plunged back into crisis because its products were not competitive
on the international market. Employer intransigence (which included
violent coercion) was an alternative to reduced profit margins. But it
also promoted the ascendancy of the exaltado or extreme anarchist wing
of the CNT over the more moderate syndicalists whose leaders were
among the first to fall to the bullets of assassins hired by Catalan in-
dustrialists and business magnates. The result was spiralling violence on
the streets of Barcelona between 1919 and 1923 as anarchist direct ac-
tion squads fought it out with the yellow unions Sindicatos Libres) and (

paramilitaries of shadowy provenance. With the connivance of General

Martinez Anido, civil governor of Barcelona from October 1920, Libres
gunmen were trained and armed in military barracks, and the notorious
Ley de Fugas operated on a scale that verged on industrial. (This meant
that CNT detainees were shot in cold blood, ostensibly ‘while trying to

Garland, 1991); a summary’ in F. J. Romero. Twentieth- Century Spain. Politics and Society i8g8-igg8
(Basingstoke: Macmillan, 1999), pp. 23, 25, 29. Lerroux’s shadiness was compounded by the fact
that he was funded by ‘Madrid’ in an attempt to block the progress of Catalan nationalism as a
political force.
21
It would be wrong to say, however, that the PSOE/UGT’s early failure in Catalonia was simply
the result of their centralist attitudes (after until the 1890s Catalonia was one of the UGT’s
all,

major bases). In the 1902 general strike the UGT missed the opportunity to lead a radicalised
labourmovement in Barcelona because - even then - its leaders shied away from radical tactics.
The resulting vacuum of leadership would be filled by syndicalists and anarchists, constituted
from 1910 as the CNT. The mood of fierce anti-clericalism, fanned by Lerroux’s demagogy,
would also alienate the PSOE. From here, PSOE/UGT centralism - which itself intensified as
a result of their lack of political purchase in Catalonia - became another grave source of tension
within the left. For the emergence of a Catalanist political left, see chapter 1.
12 The Spanish Republic at war

escape’.) Thousands of cenetistas were imprisoned or sent into internal

exile - whither they were dispatched in chains and on foot. In effect, a
low-intensity civil war was being fought. 22 It was a war not dissimilar in
its causes and objectives to that occurring contemporaneously in Italy,

where it would end with Mussolini and his Fascists being invited into
power by the old elites.
Back in July 1909 the Catalan bourgeoisie had given thanks to God for
delivery from the anti-militarist and anti-establishment rebellion known
as the Tragic Week. They had built the Church of the Tibidabo in
recognition of their salvation. 23 But in 1923 it was not prayers that the
Catalan patronal offered up, but a military invocation. When General
Miguel Primo de Rivera launched his coup in September 1923 he would
have the full backing of a Catalan establishment that had until then
claimed to be committed to the goal of regional autonomy within a new
constitutional settlement.
It was precisely the perceived threat of urban mobilisation in the wake
of the 1914-18 war, with revolutionary upheaval in Russia as a backdrop,
that brought Catalonia’s industrialists and business lobby to acquiesce in
the killing of constitutional politics in Spain. In its place they accepted
intervention by the ideologically ultra-centralist Spanish army because
it could restore order on the streets of urban Catalonia, thus securing
a form of social peace conducive to their own economic interests. The
army (or more precisely its officer corps) increasingly saw itself as the
patriotic guarantor of a certain conception of ‘Spain’ that was, according
to the officers’ lights, being repeatedly threatened - in 1909, 1917 and
subsequently - by internal enemies and in particular by regionalists and
organised labour. 24 Army officers were thus identifying as ‘anti-national’

22
F.Rey, Propietarios y patronos. La politica de las organizacion.es economicas en la Espana de la Restauracion,
igi4~2j (Madrid: Ministerio de Trabajo y Seguridad Social, 1992) and ‘Capitalismo Catalan y
golpe de Primo’, Hispania, 168 (1988); E. Gonzalez Calleja and F. Rey, El Mauser y el sufragio.
Orden publico, subversiony violencia politica en la crisis de la Restauracion, igij-31 (Madrid: CSIC, 1999):
also articles by Rey and A. Balcells in Estudios de Historia Social, nos. 42-3 (1987); L’Aveng, no. 192
(May 1995) is a special issue on pistolerismo; S. Bengoechea, Vuitanta-quatre dies de lock-out a Barcelona,
igig-20. Els precedents de la dictadura de Primo de Rivera (Barcelona: Curial, 1998) and Organitzacio
patronal i conflictivitat social a Catalunya. Tradicioy corporativisme entrefinals del segle i la dictadura de Primo
de Rivera (Barcelona: l’Abadia de Montserrat, 1994).
23 This week of anti-clerical looting and street fighting was triggered by the call-up of conscripts
for the Moroccan campaigns. Working-class Barcelona had been profoundly anti-militarist since
the return of malaria-ridden ‘walking dead’ from the Cuban war. The Barcelona bourgeoisie
took their cue from their French counterparts, who had constructed the Sacre Coeur after the
defeat of the Paris Commune of 1870.
24 The Spanish military’s searching for a new role after 1898 was also, of course, in part driven
by its desire to protect its own corporate interests: see G. Cardona, El poder militar en la Espana
 A fractured left 13

those groups most closely bound up with and symbolising the deep social,
political and cultural changes underway in Spain.
In spite of the Primo dictatorship’s legal restrictions on labour organ-
isation, however, the regime was not necessarily experienced by rural
and urban workers as more oppressive than the preceding Restoration
monarchist order under which associative freedoms and labour rights
had, in practice, been absent. Primo was able to stabilise his dictatorship
on a fairly low level of overt repression precisely because of the 1920s eco-
nomic boom with which he coincided. But precisely because of the boom
there was further large-scale population movement during the decade,
as impoverished agricultural labourers headed from the misery of the
rural south, with its quasi-feudal social relations of power, to the rela-
tively freer, but equally economically exploitative, urban environments
of Madrid and Barcelona - the latter the location of the international
exposition of 1929. Indeed, the dictatorship’s ambitious public works
projects — underwritten by a system of special loans whose interest re-
payments would so cripple the Republic’s reform project of the 1930s -
were themselves a stimulus to the demographic shift of the 1920s. In
the cities, the new arrivals worked m those sectors expanding under the
boom conditions - above all as unskilled labour on vast building sites and
public works projects constructing roads and the underground system.
With such a buoyant economic conjuncture Primo was able to in-
dulge his own paternalistic notions of government. He called upon the
UGT, which he saw as the ‘responsible’ face of labour, to collaborate
in the founding of a state system of arbitration boards. These the so-
cialist trade unionists would run as paid servants of the state from the

labour ministry. Such an offer fitted well with the veteran UGT leader-
ship’s own aspirations to partnership with the state (British labour was
very much the model here). As the cautious ugestista leaders saw it, fear-
ful of the ghosts of 1917, entrenchment in the state bureaucracy would

protect the organisational patrimony of the socialist movement. Their

decision to collaborate with the dictator would, however, arm a crucial
internal dispute with the parliamentary socialists who could not, under
any circumstances, accept that it was legitimate to be involved with an
anti-constitutional regime. But at the time this mattered far less to the
leaders of the UGT than the opportunity which they saw collaboration

contemporanea hasta la guerra civil XXI, 1983) and C. Boyd, Praetorian Politics in liberal
(Madrid: Siglo
Spain (Chapel Hill: University of North Carolina Press, 1979). Nor should we forget that the
corporate interests of some military sectors would be one of the motives impelling the July 1936
coup.
H The Spanish Republic at war

with Primo as affording them permanendy to overtake the CNT. In the

years leading up to the military coup of 1923 the anarcho-syndicalist fed-
eration had been outstripping the UGT
in recruitment and even making
inroads into socialist strongholds. 25
But with the dictatorship the CNT
had been declared illegal, as Primo saw this as the key to ending the
running war on the streets of Barcelona.
But for all the prominence of the anarchist exaltados in this war, it is vital
to remember that the CNT had other faces. Even if its Catalan branch
would dominate the CNT numerically until 1931, 26 the organisation was
a national one whose regional components each formulated a political
ethic and practice in the light of the material circumstances of their
particular memberships. In a country with such disparate levels of devel-
opment as Spain’s, this obviously also made for significant diversity within
the organisations of the left. As well as the southern millenarians and
Catalan and Aragonese street warriors, the CNT also included, for ex-
ample, a Northern (Asturian) federation in which skilled workers and ar-
tisans predominated. 27 It was the presence of this syndicalist component
that had reassured the UGT leaders sufficiently to make possible the his-
toric Labour Pact ofJuly 19 16.
28
A year later, a group of self-proclaimed
parliamentary syndicalists from Gijon even called for the formation of a
political party to represent the CNT’s interests. 29 Although this did not
prosper, the very fact that it was mooted demonstrates the clear political
affinities of some parts of the CNT with the UGT In the north, cooper-
ation between the two was also facilitated by their relative equilibrium
of strength in the region and the similarity of the industrial disputes they
were called upon to handle. At the same time, the fact that some cenetistas
saw the value of elaborating a parliamentary political strategy highlights
how the war-induced surge in labour mobilisation had crystallised im-
portant differences inside the CNT over ideology and practice - and
in particular over how to deal with the state. Indeed, in Catalonia too

25 B. Martin, The Agony ofModernisation. Labor and Industrialisation in Spain (Cornell University Press,
1990), p. 195 (for comparative figures).
26
In 1919 the CNT’s Catalan regional federation, with a membership of over 400,000, represented
half of the total CNTmembership for Spain, J. Peirats, La CNT en la revolucion espahola (Madrid:
Ruedo Iberico, 1971), vol. 1, pp. 27-8.During 1931-6 the Catalan federation would suffer a
decline in membership. But the fact that it retained its formidable mobilising power in the streets

and continued to provide the movement’s most prominent leaders meant that it would retain its
political dominance over the CNT nationally - see chapters 1, 4 and 5 below.
27 P. Radcliff, From Mobilisation to Civil War. The Politics of Polarisation in the Spanish City of Gijon igoo-

I937 (Cambridge: Cambridge University Press, 1996), passim.

28
P. Heywood, Marxism and the Failure of Organised Socialism in Spain (Cambridge: Cambridge Uni-

versity Press, 1990), pp. 41-2; A. Saborit, Julian Besteiro (Buenos Aires: Losada, 1967), pp. 87-8.
29 Radcliff, From Mobilisation to Civil War,
p. 128.
 ,

A fractured lefl 15

strong currents emerged in favour of modifying the CNT’s strategy in a

°
‘political’ direction and of reviewing its interaction with the UGT. 3

Between 1919 and 1923 syndicalist leaders (of whom the most famous
was Salvador Segul) who were fearful of the erosive effect that continued
violent direct action would have on the CNT tried to steer the influential
Catalan federation in the direction of organisational consolidation and
a more nuanced syndical strategy. 31 Although a majority of syndical-
ists still resisted the idea of parliamentary politics, these organisational
reforms were at root a response to the increasing complexity of indus-
trial and of industry’s interaction with the state. But Segui
organisation
and his supporters met the determined opposition of the majority of the
CNT - above all, in the Catalan and Andalusian federations. 32 Although
the CNT’s
1919 congress approved the conversion of its craft unions into
modern, vertical industrial ones (the Sindicatos Unicos), it rejected their
consolidation into national federations of industry. 33 But even progress
towards the Sindicatos Unicos was slow and uneven. 34 Primo’s collabora-
tion with the UGT ensured that the internal debate in the CNT did not
die. 35But the dictatorship’s criminalisation of the Confederation froze
the possibility of any organisational revision at the same time as it gave
radical, pro-direct action anarchists the upper hand once again in the
political argument over what the CNT was for and how it should be or-
ganised. (Just as the use of establishment violence against the CNT in the
post-First World War period had undermined the syndicalist reformers,
so too the fact that Segui was himself assassinated in 1923 boosted the
credibility of more radical currents in the organisation.) Nevertheless,
that the radicals felt it necessary to form a separate group, the Iberian
Anarchist Federation (FAI), in 1927 in order to defend anarchist ortho-
doxy within the CNT
indicates that the internal political differences
remained unresolved.

30 Their major focus was Angel Pestana and his supporters in the Solidaridad Obrera editorial group
(Joan Peiro et al.) as well as Salvador Segui, the Catalan CNT leader, C. Lorenzo, Los anarquistas
espanolesy el poder (Paris: Ruedo Iberico, 1972), pp. 44-5.
31
A. Pestana, Lo que aprendi en la vida (Madrid, 1972), Part 1, pp. 88-9, 100-1; Part 2, p. 87.
32
C. Lorenzo, Los anarquistas espanolesy el poder pp. 44-5.
,

33 Ibid.,
pp.36-7; Peirats, La CNT en la revolution espanola, vol. 1, p. 27; Radcliff, From Mobilisation to
War pp. 179-80.
Civil
34 The exception was in the already politically exceptional northern federation, Raddiff, From
to Civil War, pp. 179-80.
Mobilisation
35 Lorenzo, Los anarquistas espanoles, pp. 49—50; Pestana, Lo que aprendi en la vida. Part 1, pp. 102-4. As
Lorenzo indicates, some in the CNT, basing themselves on pure syndicalist ideas (i.e. ignoring
the forms of the state) saw the way forward for the Confederation in emulating the UGT’s role
in the regime’s labour tribunals (the comites paritarios).
1 6 The Spanish Republic at war

Despite th£ FAI’s injection of anarchist purism, the coming of the

Second Republic in 1931 would see the acceleration of these internal
CNT disputes over structure and strategy. (As we shall see in the next
chapter, they would produce the so-called treintista schism that split both
the CNT’s Catalan and Valencian regional federations.) The resurgence
of these debates probably had something to do with the numerical de-
cline of theCatalan federation in relation to the national organisation
as a whole. But more fundamentally (since the Catalan federation was
also affected), these growing internal divisions were part and parcel of
the larger process of social and economic change that had helped bring
about the Second Republic itself. Improved communications and the
beginnings of mass production, with more widely distributed printed po-
litical propaganda, were, by the 1920s, gradually transforming Spanish

workers - including the as yet unorganised - into a new political con-

stituency. With the Republic came a heightened sense of new possibilities
for the left that would see a further acceleration of popular political mo-
bilisation. In turn, some inside the CNT saw an enhanced opportunity
for the organisation’s partial political incorporation — at least sufficiently
to defend its members and social constituencies through the newly avail-
able channels of municipal politics and state labour agencies.
It has been a historiographical commonplace to contrast the solid

bureaucratic structures and practices of the UGT with the structural

inchoateness and direct action of the CNT. From this analysis, however,
we can conclude that to view the CNT and UGT as if they were structural
and ideological polar opposites is neither useful nor accurate. 36 Rather
each was a heterogeneous organisation containing divergent ideological
perspectives and labour practices moulded by different kinds of (still
predominantly urban) experience. Let us remember too that, just as
the arguments raged inside the CNT in the 1930s, so would they in the
socialist movement (PSOE/UGT) as the tensions that had provoked the
communist schism of 192 1 —2 resurfaced. In short, the UGT and the CNT
not infrequently resembled each other - and never more so than when
they were struggling to maintain organisational cohesion in the face of
and strategic dilemmas opened up by the Republic. It could
the political
scarcely have been otherwise given Spain’s acute regional disparities
economically, socially and culturally. If these factors had fragmented
what were relatively far more powerful elite groups (as clearly they had),
then how much more must uneven development have impacted on the

36 This observation should also be of use in the increasingly vexed debate over how to explain the
original pattern of anarchist and socialist organisational implantation in Spain.
 A fractured left
organisational forms and political goals of the relatively far less powerful
Spanish left. It is also important to remember that, apart from the brief
experience of the First Republic in 1873, the left - even broadly construed
to include all brands of republicanism - had not, prior to 1931, exercised
national political power in Spain in the modern period. 37
Only by understanding the multiple factors that had shaped the left -
or, more accurately, lefts - prior to 1931 can we hope to understand

properly what drove relations between them thereafter or what produced

the schisms and conflicts within each. Nor can we otherwise make sense
of what the Spanish left’s component parts said and did during the civil
war of 1936-9 - and, just as importantly, what they could not bring
themselves to think or do.
However, understanding all of this is made even more difficult by the
fact that after 193 1 the organisations of the Spanish left - republicans, so-
communists and anarcho-syndicalists - were also having to deal
cialists,

increasingly with other forms of political mobilisation, namely those af-

fecting Spain’s wide spectrum of middling classes. This process was also
kick started under the Primo dictatorship and by the 1930s it would sig-
nificantly have changed the political landscape, as these middling sectors
sought political representation for their interests. Moreover, progessive
sectors among the middling classes were joining not only specifically
republican parties but also a range of socialist and communist organisa-
tions - including the USC and the BOC in Catalonia and, from spring-
summer 38
1936, the united socialist-communist youth organisations.
This mobilisation of Spain’s middling classes was particularly evident
where powerful regional nationalist sentiment existed, and above all in
Catalonia. There the industrialist lobby’s defection to Primo would (by
1931) see the autonomy movement politically reconfigured under the
liberal-left leadership of the region’s urban professional middle classes,

but equally with support from middling rural sectors. Another key
facet here was the mobilisation of middle-class youth in the context
of a rising university population with equivalent rising expectations. 39
(The year 1927 would see the creation of the Federation of Univer-
sity Students (FUE) which was highly critical of the dictatorship.) This

unprecedented mobilisation of youth would also exercise a profound

37 Apart from 1873, the only other period since the French Revolution when the Spanish left could
be said to have had some power was during the constitutional triennium of 1820—3.
38 I refer to youth organisations in the plural because of the more complex situation in Catalonia:

see R. Casteras, LasJSUC ante la guerra (Barcelona: Nova Terra, 1977), pp. 113-30. For more on
this see chapter 1 below.
39 Ibid.,
p. 45.
1 8 The Spanish Republic at war

influence on *he political life of the Second Republic, especially during

the war.
More generally too the 1920s in Spain saw the acceleration of a pro-
cess of cultural modernisation which affected middling constituencies in
particular. The beginnings of mass production not only facilitated the
wider distribution of printed political propaganda, but also, along with
the radio, created a new leisure market which in turn consolidated a sense
of identity among the new urban professional and commercial middling
40
classes. Also crucial to this process of cultural change was the reception
of the Russian Revolution among these new middle classes. But this was
not primarily a question of ideology. 41 Spain’s new middle classes were
not aspiring marxist-leninists. They were, however, ‘aspiring’ in other
ways. Images of Soviet modernity in 1920s and 1930s Spain functioned
in lieu of domesticones (which were sparse). Their appeal tells us more
about the social aspirations of these urban groups than their political
ideas. The idea of the Soviet Union as an icon of modernity was some-
thing which particularly influenced lower-middle-class youth - that is,

the kind of young men and women who would join the United Socialist
and Communist Youth Organisation JSU) ( in droves during the second
half of 193 6. 42
Middle-class mobilisation in 1920s Spain also took the form of the bur-
geoning professional associations - of post office employees, rural doc-
tors, clerks, teachers and the like - who through their adaptation to new,

modern work methods came increasingly to have a sense of themselves

as performing a broader public service. 43 To these groups the constitu-
tion was important because it held out the hope of a non-arbitrary form
of power which could protect their own professional interests. Both as a
movement of opinion and in organisational terms, these groups would
be the most coherent oppositional sector of republicanism - indeed, this
was so precisely because theirs was a practical republicanism tied to
40 For example, commercial music and the cuple played a key role in stimulating radio in urban
middling-class homes: see S. Salaiin, ‘The Cuple Modernity : and Mass Culture’, in Graham and
Labanyi, Spanish Cultural Studies, p. 93.
41
R. Cruz,‘jLuzbel vuelve al mundo! Las imagenes de la Rusia y la accion colectiva en Espana’,
in R. Cruz and M. Perez Ledesma (eds.), Culturay movilizacion en la Espana contemporanea (Madrid:
Alianza, 1997).
42
Cf. ‘Russia, together with other expressions of modernity such as “aviation, the radio, telephone”
gave life great interest.’ E. Montero, ‘Reform Idealised: The Intellectual and Ideological Origins
of the Second Republic’, in Graham and Labanyi, Spanish Cultural Studies, p. 13 1.

43 Ibid.,
pp. 129—30. Increasingly conscious of their own professional specialisation, these sectors
put their ‘faith’ in technology and science - which was, of course, part of the perceived appeal

of the Soviet Union - and saw their own social advancement as integrally linked to ‘progress’
writ large.
 A fractured left 19

enlightened self-interest - and they played a crucial role in the collapse

of monarchy in 1931. Yet their potential usefulness thereafter was lost -
44

both as a social support base and as a source of new state personnel

for the modernising Republic. For republicanism in Spain was domi-
nated by academics and lawyers with an excessively legalistic view of
political change. Their conception of reform began and ended with the
mechanics of a top-down state project, and they thus failed to link up
organisationally with these professional groups. Given the undeniable
fact that Spain’s politically, socially and culturally
middling classes were
highly fragmented, and that their urban component was exiguous, 45 this
failure to integrate the professional associations in the project of Repub-
lican reform would seriously threaten its viability, as we shall see.
By the end of the 1920s as the boom ended and dictatorial debts
mounted, the Primo dictatorship entered terminal crisis. Its collapse
followed shortly after that of the peseta. But the agrarian and industrial
elites’ disaffection from the dictatorship remained essentially a political

one. For a start, members of the monarchist political class - men like
Niceto Alcala-Zamora, Miguel Maura and Jose Sanchez Guerra — were
hostile because the Primo dictatorship had destroyed the old two-party
system and thus rendered them politically obsolete at the same time as it

46
cut them by the state. More
off from the sources of patronage afforded
generally, there was mounting distaste among elite groups at the political
‘novelty’ of some of Primo’s policies — in particular his policies of class
co-option, such as the labour tribunals. The years of boom had blunted
their fears of labour unrest. Both agrarian and industrial elites held fast
to the naive belief that there could be a return in some shape or form
to the safely ‘traditional’ - i.e. demobilised and exclusivist - order of the
Restoration monarchy.
But existing levels of popular mobilisation had already made a return
to such an order quite unrealistic. This was manifest in the explosion
of anti-monarchist sentiment in the municipal elections of April 1931.
Although the caciques managed to retain control of rural constituencies,
urban Spain returned the ‘voto-verdad’ (authentic vote), revealing the
extent of popular disaffection with the existing order. On the basis of

44 E. Montero, ‘The Forging of the Second Spanish Republic: New Liberalism, the Republican
Movement and the Quest for Modernisation 1898—1931’, unpublished Ph. D. thesis, University
of London, 1989.
45 The classic (pre-193 1) republican experience of ‘pocket’ power - in a provincial city - is conveyed
well in Radcliff, From Mobilisation to Civil War.
46 Their disaffection also extended to the person of the monarch, Alfonso XIII, for his lengthy
support of Primo.
20 The Spanish Republic at war

this, a coalitren of the liberal-left, representing those social constituencies

excluded by the monarchist order, was able to press ahead and declare
the Republic. 47 But what had really created the political space for it was
not the strength or coherence of this coalition, but rather the political
disarray of Spain’s elites following the collapse of the dictatorship. In what
amounted almost to a fit of absence of mind, they found themselves faced
with the fait accompli of a Republic. 48
Some imagined that this might only amount to a change in the
still

form of the regime and one, indeed, whose novelty could even benefit
conservative Spain in the search for a means of relegitimising its own
power. (This is likely to have figured in the political calculations of erst-
while monarchist politicians-turned republicans, such as Alcala-Zamora,
later Republican president, who were extracted from the same social
world/families as the old elites.) But this ignored the fact that the col-
lapse of the monarchist system was due precisely to its inflexibility, to the
fact that it had proved incapable of assimilating new social constituen-
cies. Thus in 193 1 — in the context of a mounting international economic

crisis which would soon impact on Spain - elite groups faced increasingly

conscious and organised urban and rural working-class constituencies as

well as politicised republican sectors of the urban (and some rural) mid-
dle classes. Moreover, they were facing these constituencies in a situation
where political authority itself had, for the first time in the contempo-
rary period in Spain, passed to a coalition of forces that unambiguously
backed an agenda of social and economic reform. This coalition was now
also about to be in formal control of the very instruments of public order —
the police force and the army — to which Spain’s agrarian and industrial
elites had previously looked to avoid just such a political scenario.
But this state of affairs would prove less beneficial to the Spanish left
than might at first appear. In June 1931 progressive forces would take
control of national government in Madrid by dint of the authority in-
vested in them by the ballot box. But old Spain had retained most of
its economic power. (The Second Republic was not a socialist regime.)

Moreover, the elites still had tremendous social power in the localities.
The cacique system proper had been broken by Republican victory, but
the networks of power and influence that underpinned it had not. The
’

coming of the Republic would in fact institute a kind of ‘stand-off

47 This coalition was based on the Pact of San Sebastian, signed in August 1930 by republicans
and Catalan nationalists and soon supported by the PSOE. (In the Basque Country, the liberal
nationalist minority, ANV, participated, but the more influential PNV did not.)
48 S. Ben-Ami, The Origins of the Second Republic in Spain (Oxford: Oxford University Press, 1978).
 A fractured left 21

between new and old Spain. While the new exercised formal executive
political authority , the old still possessed formidable powerA 9
The scenes of jubilant Republican crowds in April 1931 appeared to
suggest the realisation of the historic dream of Spanish republicanism -
the unity of the awakened ‘people’ ( pueblo) carrying all before it. But the
experience of the subsequent five years would show that, ironically, the

necessity for unity was a lesson more easily assimilated by the political
right than by the left. Elite groups’ past political experience combined
with their sheer survival instinct saw them rapidly adapt to the new envi-
ronment. Conservative forces would use the new political instruments at
their disposal to block reform (for example by parliamentary obstruction)
as well as deploying their formidable economic resources and local power
bases to ensure that Republican laws remained inoperative even after
they reached the statute book in Madrid. This concerted obstruction, as
well as the political inexperience and strategic errors of the reformers, the
lack of money (above all at a time of economic depression) and, to some
extent, the instrinsic slowness of reform as a process - all these factors
opened up fissures within the ‘pueblo’ and on the political left that further
disabled change and advantaged conservative interests, as we shall see.
Nevertheless, one should not exaggerate the failure of change after 1931.
After all, those hostile to it would be sufficiendy frightened by the prospect
that the centre left’s electoral victory of February 1936 might herald the
re-emergence of a popular anti-hegemonic coalition bent upon reform,
forthem to resort to apocalyptic ‘solutions’ - namely a military coup.
What would happen after the launch of the coup had not been foreseen
by anyone. But looking back, we can see that the ‘afterwards’ of ‘hot’
civil war was also a means of resolving - albeit in a vasdy more violent

way — the uneasy balance or stand-off between class forces opened up in

1931, and, through that, the underlying question of which route Spain
would take to modernisation - the democratic or the dictatorial. 50

49 This was what the Austrian socialist thinker Otto Bauer acutely defined as a ‘transitional regime
of class equihbrium’. Otto Bauer, ‘Fascism’, in T. Bottomore and P. Goode (eds.) Austro Marxism
(Oxford, 1978) pp. 167-86. For a summary H. Graham and P. Preston (eds.) The Popular Front
in Europe (Basingstoke: Macmillan, 1987), p. 10. Cf. also S. Julia, ‘Manuel Azana: la razon, la

palabra y el poder’, in V
A. Serrano and J. M. San Luciano (eds.) Azana (Madrid: Edascal, 1980).
50 In the
1940s ‘Francoism’ itself would be constructed through the process by which the military
dictatorship mediated between the different sectors of capital in order to weld together a new
elite project - at the same time as it brokered power between the political ‘families’ and imposed
social discipline on the defeated. A. Cazorla, Las politicos de la victoria. La consolidacion del Nuevo
Estadofranquista 1938-1933) (Madrid: Marcial Pons, 2000); A. Cenarro, Cruzados y camisas azules.
(

Los 1936-1943 (Zaragoza: Prensas Universitarias de Zaragoza,

origenes del Jranquismo en Aragon,

1997); Richards, A Time of Silence.

'*

.
 CHAPTER I

The challenge of mass political mobilisation

(i93 1 -193 6)

1
jVivan los hombres que nos traen la ley!

The coming of the Second Republic saw the emergence of a governing

coalition of centre-left republican groups in alliance with the Spanish
Socialist Party (PSOE). Its reforming agenda was driven by a progres-
sive republican ideology borne by the liberal - but somewhat marginal -
sectors of Spain’s urban professional lower middle classes. Their numer-
ical slightness made the support of the PSOE crucial. The PSOE’s own
solidity as a support was provided by the electoral muscle of its near
million-strong trade union movement, the UGT.
The driving ethos of this 1931 coalition was to modernise Spain eco-
nomically, to initiate democratising reforms and to Europeanise the
country socially and culturally. These objectives were to be attained
via a series of legislative measures comprising agrarian, labour and
social reforms (including state provision of education). Land reform
(predominantly of the vast southern estates) was intended to create a
whose
large class of smallholding peasants in the style of France in 1789
acquisition of land would make them a permanent support base for
the regime. The Republic’s other key reform was of the military. This
had crucial political goals - namely to bring the institution fully un-
der civilian constitutional authority and, in time, to republicanise it. By
reducing the size of the notoriously ‘top-heavy’ officer corps it was also in-
tended to release much-needed funds to finance the rest of the planned
reform programme. Reform is always an expensive undertaking, but
the Spanish republicans were embarking upon it at a time when the
effects of European and world economic depression were just beginning
to be felt.

the men who bring us the rule of law!’ This was the greeting offered in one
1
‘Long live

Republican campaigners shortly before the declaration of the Second Republic, cited in
village to
E. Montero, ‘Reform Idealised’, in Graham and Labanyi, Spanish Cultural Studies, p. 129.

23
24 The Spanish Republic at war

‘Reformiitg’ the balance of socio-economic and political power in

Spain in this way was
perceived by the republicans as a means of de-
livering the classic goals of political liberalism. Increasing rural income
levels - and especially those of the landless proletariat of the south - was
intended too to create a larger domestic market in order to stimulate in-
dustrial growth. While redistribution would also fulfil social democratic
requirements of social equity per se, republicans also looked to it to create
a more inclusive and thus more stable society and polity in which to
pursue the national economic growth they sought.

THE POLITICAL FAILURE OF ‘HISTORIC’

REPUBLICANISM (1931 —3)
The republican agenda was without any doubt an extraordinarily am-
bitious and wide-ranging one. In part this reflected the stagnation of the
old regime and the long overdue need for basic modernising change to
bring Spain into line with its European neighbours. Conservative inter-
ests would lose no time in mobilising against change. But even more

ominous for the republicans would be the alienation over 193 1-3 of
social groups whose support was crucial to the viability of the reform
project - such as urban and rural labour and sectors of the provincial
middle classes. The reasons why this alienation would occur were com-
plex and the underlying problems were in considerable part connected
with the context of economic depression. But they were also the result
of strategic errors on the part of the republicans themselves.
The republicans had little sense of the need to build active political
alliances bottom up in society in order to ensure an adequate mass sup-
port base for the reforms they wished to make. This blind spot seems
ironic given both the accelerating political mobilisation underpinning the
Republic’s birth and the new context of representative democracy that
it had ushered in. We could explain it in terms of the republicans’ lack
of political experience. But while we should not minimise the impact of
this, or the obstacles faced by the reformers, this myopia is also indicative
of their particular understanding of politics. For the republicans, like the

conservatives who opposed them, belonged to an old political world that

was, at heart, uneasy with the idea of mass mobilisation. Spanish repub-
licanism was progressive in that it favoured certain structural reforms
to redistribute socio-economic power in Spain. But it was also conser-
vative in that modernising reform was envisaged as something to be
implemented top down by a political elite via the machinery of state.
 The challenge of mass political mobilisation 25

Indeed, for Spain’s republicans, ‘the Republic’ began and ended with
the state. Reform was perceived predominantly as an abstract, intellec-
tual problem — a view nowhere more clearly articulated than in the writ-
ings and parliamentary interventions of Spain’s pre-eminent republican
leader, Manuel Azana, prime minister of the liberal biennium of 1931 -3
2
and, from May 1936, president of the Republic. Strongly influenced
by regenerationist and Krausist thought, the republicans envisaged an
idealised state with extended and renovated powers. 3 But the actually
existing state in April 1931 was made up of institutions and personnel
inherited from the monarchist regime. There would be a significant con-
tinuity in the personnel working within the state bureaucracy —faute de
mieux since the incoming coalition simply did not have sufficient numbers
,

of experienced, politically conducive individuals at its disposal. 4 Again,

this was in part the inevitable consequence of the long exclusion of the left
(broadly construed) from power. But this lack of adequate personnel was
also compounded by the republicans’ blindspot around political mobil-
isation. As already mentioned, ‘historic republicanism’ - dominated by
lawyers, professors and educationalists - had previously failed to link up
with the relatively populous professional middle-class associations where
republican sentiment had developed significantly during the 1920s. By
failing to forge such links, the republican political class deprived itself
of much-needed technical and managerial expertise as well as losing
the opportunity to widen progressive republicanism’s popular support
base. 5 This would prove a costly failure. Political republicanism inside
the state machine lacked the necessary technical expertise to implement
and monitor the detail of reform on a daily basis. But nor, given the
republican conception of politics, did they necessarily understand why
this was important.
This conception also led the republicans to confuse the theoretical
political authority of government with real political power. The Spanish
republicans had a firm - if somewhat naive — belief in the power of the law

2
The idea of the Republic being synonymous with state action was most clearly articulated by
him: ‘ser republicano era solo una manera de entender el Estado
y las reglas del juego politico’,
J. Paniagua Fuentes, introduction to Azana’s Discursos parlamentarios (Madrid: CSIC, 1992). Also,
J. Marichal, El Intelectualy la politico, (Madrid: CSIC, 1990), p. 78.
3 E. Montero, ‘Reform Idealized: The Intellectual and Ideological Origins of the Second Republic’,
in Graham and Labanyi, Spanish Cultural Studies pp. 124—7.
,
Krausism was a strongly ethical school
of philosophy dominant among Spanish liberal reformers in the 1870s and 1880s, based on the
work of the post-kantian German philosopher Krause, a contemporary of Hegel.
4 On the difficulty of finding appropriate republican personnel, see M. Maura, Asi cayo Alfonso XIII
(Barcelona: Ariel, 1995), pp. 265-72.
5 See the introduction above.
26 The Spanish Republic at war

(
But while thdy had the authority to enact a new Constitution 6
juricidad ).
and bring legislation to parliament in Madrid, the task of implementing
these things would bring the reformers hard up against the reality of how
social and economic power in the localities of Spain (above all in majority
rural Spain) remained to a great extent in the hands of the old elites.
The images most associated with the Republic’s birth - of masses of
people in the streets, surging through the squares and open spaces of
the capital, clambering over public buildings and monuments, toppling
statues of the king - vividly depict the expectations raised by the new
regime among and economically disenfranchised sectors of the
socially
population, something which would further accelerate mass mobilisation
after 1931. But the republicans’ own political culture and experience did
not fit them to exploit its political potential. Indeed, they would soon
be responding (for example around issues of public order) in ways that
suggested a real fear of the uncontrollability of this process.
But this difficulty belonged not only to the republicans. Their coali-
tion partners, the Spanish Socialists (PSOE), were also in various ways
grounded in this statist, top-down understanding of political and social

change. Those socialists who identified primarily with the parliamen-

tary party rather than the union (UGT) shared much of the republicans’
elitist regenerationist ethos, while it was, paradoxically, in the socialists’

trade union wing that the republicans’ disquiet over mass mobilisation
would find its clearest echo. Influenced by their collaboration with the
Primo de Rivera dictatorship in the 1920s, the UGT’s veteran leaders
had envisaged the inheriting of state power in 193 1 as a means of squaring
a crucial circle. It could ensure expanding membership and influence for
the socialistmovement while maintaining a high degree of control, thus
not risking its organisational structures and patrimony - the traumatic
memories of 1917 had left an indelible mark. But when the UGT’s mem-
bership did begin rapidly to expand (and nowhere more than among
the rural south’s landless proletariat 7 ), then the union leadership’s atti-

tude became decidedly ambivalent. For the PSOE/UGT, like most other
European socialist movements of the time, had deeply ingrained views on
what constituted the ‘organisable’ working classes. Fears were expressed
about the likely effects of the mass influx of the politically uneducated
on the fabric of the organisation (its ‘historic profile’) and on its political

6
The Republican Constitution of October 1931 borrowed from previous radical republican
experiments (Mexico, 1917 and (especially) Weimar Germany, 1919).
7 The urban unskilled - for example on the building sites of Madrid - were another source of
anxiety.
 The challenge of mass political mobilisation 27

8
mission of reform. The socialist leaders seemed to have no idea about
what to do with the new members flooding in. Nor indeed were they ever
really utilised as a political constituency - apart, that is, from their sterile
deployment in the internecine war inside the socialist movement. 9
The challenge facing republicans and socialists was twofold. First, they
had to mobilise a viable support base for their own reforms. Second, they
had to develop strategies to defuse or counter anti-reform movements
of opinion that could foreseeably be mobilised against them within the
emergent system of mass parliamentary democracy.
But it proved impossible to mobilise an adequate support base. For
this to happen, republicans and socialists had to show that they could

convert aims into implemented policies. The proposed reforms com-

bined Azana’s ‘statist’ agenda with the social agenda of the PSOE. But
this was far too ambitious a programme to be realisable. Either of the

two agendas was, alone, guaranteed to provoke more opposition from

powerful elite groups than the government could deal with. Moreover,
both Azana and the socialist union leader, Largo Caballero, overesti-
mated the size of their electoral mandate for reform. Included within
the votes sustaining the government coalition there were probably quite
a proportion for Alejandro Lerroux’s Radical Party and others whose
commitment to a reforming agenda was, to say the least, ambiguous.
To make matters worse, Azana and the PSOE discounted the extent
to which their mandate for reform was dependent on the support of
the CNT’s social constituencies who were, thereby, left disenfranchised.
In sum, the internal tensions in the republican-socialist reform project
would prevent from ever mobilising a sufficient support base for itself.
it

Nor could it prevent the opposition from counter-mobilising.

Ironically, it was to be precisely those forces hostile to reform that
learned to adapt faster to the new political environment. The scale of
mass Catholic mobilisation between 1933 and 1936 was perhaps less
evidence of Spain’s ‘polarisation’ per se than it was of the liberal left’s
failure to achieve its own prior mass mobilisation, most crucially of
some of the lower-middle-class constituencies which then turned to the

Urgent calls for the mass political education of the new and prospective membership were made
at the PSOE’s 1932 congress; see J. M. Macarro Vera, ‘Causas de la radicalizacion socialista en la
II Republica’, Revista de Historia Contemporanea (Seville), i (Dec. 1982), p. 203. Similar fears would
resurface after the Popular Front electoral victory of February 1936.
9 P. Preston, The Coming
of the Spanish Civil War. Reform, Reaction and Revolution in the Second Republic
(henceforward CSCW), 2nd edn (London: Routledge, 1994); S. Julia, La izquierda del PSOE
(I935~I93 6) (Madrid: Siglo XXI, 1977) and H. Graham, Socialism and War. The Spanish Social-
ist Party in Power and Crisis, 1936—1939 (Cambridge: Cambridge University Press,
1991).
28 The Spanish Republic at war

populist rigfft, with fatal consequences for the Republican project. For if

ithad not been for mass mobilised conservative opinion, it would have
been difficult for the rebel officers to justify the - essentially Bonapartist -
coup ofJuly 1936, even in a country whose civil society was as relatively
underdeveloped as Spain’s. Although its leaders had been involved in
anti- Republican conspiracies since the regime’s beginning, it was only
the presence of mobilised civilian opinion that allowed the military rebels
10
to present their actions in 1936 as if they constituted a popular plebiscite.
The stakes in this battle of counter-mobilisation were made clear from
the start. From the formal declaration of the Republic in April 1931,

powerful sectors of the hierarchy of the Catholic Church in Spain in-

dicated their irreducible hostility to the political and cultural pluralism
Republican project. 11 In his pastoral letter of 7 May,
at the heart of the
Cardinal Segura, the Spanish Primate, offered a provocative homage
to the monarchy and in his collective letter ofJuly he publicly declared
the doctrine of popular sovereignty to be inimical to Catholic teaching.
With these declarations, the contest was now on to form the opinion
and achieve the support of ‘Catholic Spain’ (i.e. the provincial and rural
middle classes). But we should be clear that this was about constituting a
political force, not simply about giving voice to what already existed. For
many, maybe most, in the ecclesiastical hierarchy the issues were imme-
diately clear: the Republic was unacceptable per se precisely because it
was pluralist. But attitudes were much less clear-cut at the outset among
lay Catholics and many of the lesser/ ordinary clergy. It was only as a
result of the specific religious measures implemented by the Republic
that these sectors came to be politically and culturally alienated from it
more or less en masse.
In a country where religious loyalties and piety were as emotive and
powerful a mobilising force in some regions as anti-clericalism was (pre-
dominandy among working-class constituencies) in others, the Republic
simply could not afford to alienate the Catholic laity virtually in its en-

tirety. In these terms, the high-profile anti-clericalism of the republicans’

was a strategic error of considerable proportions. The
religious reforms
chamber elected in June 1931 to draft these reforms was driven by
vehement republican hostility to the Catholic Church as an institution.

10
Cf. the text of Franco’s Discurso del alzamiento — the radio broadcast made from Tetuan on
17 July 1936 justifying the rebellion in terms of the conspirators’ embodying the national will,
F. Diaz-Plaja, Laguerra de Espaha en sus documents (Barcelona: Plaza y Janes, 1969), pp. n-13.
11
Bishop Goma (later Primate of Spain) wrote on 15 April that ‘we have now entered into the
vortex of the storm’, quoted in Lannon, Privilege, Persecution and Prophecy, p. 179.
 The challenge of mass political mobilisation 29

In a mirror image of the right’s manichaeism, republicans proved en-

tirely unable to distinguish between state secularisation measures - such
as separation of Church and state, the provision of civil alter natives to
Catholic marriage and burial, or the provision of non-religious state ed-
ucation — and measures which infringed the democratic rights and sense
of identity of ordinary Catholics. In the latter category came legislating
for the exclusion from teaching of the religious orders or instituting a
plethora of municipal regulations which harassed Catholics daily: im-
positions on funeral processions, on the ringing of church bells, on the
outdoor celebration of patronal feasts. Sometimes wayside shrines were
removed, together with religious statues or plaques in village squares.
The implications of a secular state would no doubt have been strange
and initially unwelcome to many Catholics. But over time they would
have been assimilable. At any rate, they were not the stuff of which
counter-Republican mobilisation could have been made - unlike those
measures which directly interfered with the daily culture and identity of
Catholics and which were thus perceived as vindictive.
It may be that a further distinction needs making between republi-

can repression of popular Catholic culture (the cults around local \illage
saints, for example) and the question of educational policy. But even if

one were to make a case for the political importance of restricting the
teaching role of the religious orders, the fact remains that once again,
the republicans failed to implement their policy successfully. Although
the debarment of religious personnel was stipulated in the Constitution
of 1931, the specific legislation (the Law of Congregations) only reached
the statute books in May 1933 - barely five months before the disintegra-
tion of the republican-socialist coalition. In other words, little can have
happened before the coming to power of a centre-right government that
effectively froze the legislation. All the republicans had in fact achieved
was the creation of aggrieved constituencies that were, thus, ripe for mo-
bilisation by the enemies of reform. By the same token, the total removal
of state financial support for the Church provoked the alienation of the
lesser clergy - a sector whose initial position was one of guarded caution
but certainly not open hostility to the Republic.
One must also be wary of using late-twentieth-century conceptions of
civil rights (ethically compelling though these are) to assess republican

religious policy. While we may wish that the republicans had been more
liberal in this respect - thinking not least of the perennial philosophical-
political debate over means and ends — their illiberalism was of its
time. Moreover, they were also rather less illiberal and somewhat more
30 The Spanish Republic at war

concerned about constitutional rights (if not yet civil/human rights prop-
erly speaking) than their opponents. (Conservative Catholics were out-
raged that their beliefs and practices were being subjected to restraints.
But they themselves entertained no concept of civil and cultural rights
within the Spanish state for freethinkers or atheists.) In the last anal-
ysis, we have to remember that no aspect relating to the Church in

1 93 os Spain could be divorced from high politics. For many republi-

cans Catholic culture was, root and branch, a threat to the inculcation
of precisely the open, pluralist mentality needed to stabilise the demo-
cratic Republic in Spain. (We should remember too that the ecclesiastic
hierarchy was the most consistent and vociferous defender of the monar-
chy in the transitional period from the Primo dictatorship to Republic
(1929-31).) Moreover, there was also sometimes an important practical
dimension to the republicans’ measures: saving on the stipend to clergy,
for example, was one way of garnering scarce resources (even scarcer
because of the recession) to fund the programme of state school building.
However, perhaps the main point to grasp here for our purposes is that
the republicans saw their commitment to secularisation as a matter of
fulfilling certain ‘historic’ republican ideals or ideological principles. Just
as with agrarian reform or anti-militarism, it was perceived as another
‘cultural north’ and borne as a crucial ‘mark of identity’. But once again
the republicans had failed to think through the material consequences of
their policies in the new political environment. So the anti-clerical ten-
dencies of ‘historic’ republicanism armed a counter-movement without
having in place any strategy for dealing with it.

Catholic Action -the organisational hub of what would become the

mass Catholic party CEDA (strictly speaking a confederation of vari-
ous regionally based right-wing groups) — creatively elaborated a pro-
paganda line suggesting that the Church’s very existence was imperilled
by the atheistic material and spiritual depredations of the Republic. It
worked to good effect particularly among the intensely Catholic im-
poverished smallholders of central and north-central Spain. This pro-
cess of mobilisation was greatly facilitated by the fact that the Church
had a well-established social-organisational infrastructure (i.e. Catholic
Action’s own) embedded in the localities. The republicans had no compa-
rable structures on which to build. Moreover, the fact that their agrarian
reform measures tended to neglect the specific problems of smallholders
and tenant farmers also hugely facilitated the mobilisation of such
groups by the CEDA - a party that received massive subsidies from the
large southern landowners who stood to lose most from the Republic’s
 The challenge of mass political mobilisation 31

12
agrarian law of September 1932. Anti-clerical legislation alone did not
provoke the Spanish oligarchy’s campaign against the Republic, but in
bringing about practical unity on the right, it massively facilitated the
implementation of that campaign.
As a result, the provincial, commercial and rural smallholding classes
of the agrarian interior (above all of Castile and Leon) were definitively
conquered by resurgent conservatism. Via the CEDA or other conserva-
tive agrarian associations such as the CNCA (Confederacion Nacional
Catolico-Agraria 13 ) these sectors would effectively be recruited to the
political project of agrarian counter-reform. In the process, ‘Spanish’
nationalism itself was definitively appropriated not only as a force of
political conservatism (as had been clearly happening since the 1920s)
but now of populist conservatism.
Elsewhere in Spain, on the peripheries - both urban and rural — the
picture was less bleak for progressive republicanism. But here too polit-
ical and fragmentation were still the order of the day. Nowhere
tension
was this more evident than in relations between the Madrid government
and the highly Catholic and socially conservative Basque Country. The
Basque Nationalist Party (PNV), which was emerging as a significant
political force in the region, looked somewhat askance at ‘anti-clerical
Madrid’ while it was also concerned to keep the PSOE’s political in-
fluence at bay in Bilbao. 14 This was not only because the PSOE was a
socialist party, but also because it was a centralist one. In fact neither side

trusted the other. Although the republican-socialist coalition was open to

the possibility of a Basque autonomy statute, it wanted to ensure that the
devolved powers remained in the hands of Basque socialists and republi-
cans. Following the PNV’s — albeit relatively brief and abortive — alliance
with the Carlists in 1931, republicans and socialists saw it as representing
clerical conservatism and, especially given the anti-constitutional tenor
of some of the PNV-Carlist proposals, as far from politically trustworthy.
12
Preston, CSCW] E. Malefakis, Agrarian Reform and Peasant Revolution in Spain (New Haven/London:
Yale University Press, 1970); A. Bosch, ‘Nuevas perspectivas sobre la conflictividad rural en la
A
Segunda Republica’, Historia Contemporanea 9 (1993), 141 —66. recent case study of the south is in
,

E Cobo Romero, ‘El voto campesino contra la II Republica. La derechizacion de los pequenos
propietarios y arrendatarios agricolas jiennenses 1931-1936’, Historia Social 37 (2000), 119-42,

and Conflicto ruraly violencia politica (Jaen: Universidad de Jaen/Universidad de Granada, 1998).
13
The political organisation of Catholic smallholders in north and central Spain. Created in 1917,
it was a forerunner of the CEDA and provided the core of its mass base.
14
While the PNV was an influential force, the politically divided nature of the Basque region
(i.e. with the strong influence of the traditionalist right (Carlists)) meant that it was not hegemonic.

The PNV’s influence was predominant in the province of Vizcaya. But in its capital, Bilbao, the
PNV had to struggle against the PSOE. F. de Meer, El Partido Nacionalista Vasco ante la guerra de
Espana (1936-1937) (Baranain-Pamplona: EUNSA, 1992), p. 66.
32 The Spanish Republic at war

But the PNV, under a. new young leadership headed by Jose Antonio
Aguirre, espoused an open and pragmatic conservatism rather than the
closed integrist variety of Carlist Navarre. The PNV’s political leadership
was, moreover, significantly less conservative than its own lower-middle-

class support base - especially those parts of it located in the rural in-
terior. Accordingly, the pull of the Republican alliance would increase
for Aguirre and his party in proportion to their disaffection from the
integrist Catholic conservatism represented by the Carlists. (After it had
rapidly become an autonomy statute could not be used to bar
clear that
Republican secularisation and social reform policies from the Basque
Country, not only did the Carlists’ interest in a statute wane, but they
actively joined the monarchist right nationally in obstructing it.) Madrid
began to use the prospect of an autonomy statute as a ‘carrot’ to attract
the PNV into the Republican orbit. But the
between the twodistrust
meant that negotiations were inevitably slower and more complex than
those for the Catalan equivalent (promulgated in September 1932).
There was disagreement particularly about the extent of devolved finan-
cial powers and over who should control the police and army in the re-

gion. An accord had still not been reached when the centre-right came
to power in Madrid in November 1933 and the CEDA’s outright hostility
to autonomy blocked further progress. This hostility would result in the
PNV’s gradual, strategic rapprochement (though not entry) to what by
the end of 1935 would be a re-emergent republican-socialist coalition. In
this the efforts of the PSOE leader, Indalecio Prieto, were paramount. He

had close personal ties with the Basque Country and was determined to
strengthen the Republican coalition by bringing the PNV into its orbit.
Nevertheless, the basic republican thinking that social and educational
reforms would, in the medium term, contribute to stability and devel-
opment, allowing a new secular mentality to emerge as the basis for the
‘Republican nation’, remained problematic with regard to the Basque
Country, as it did in other ways. However, the fact that a formal commit-
ment to a Basque statute would feature in the electoral programme of
the centre-left coalition in February 1936 ensured that the PNV strategi-
cally accepted the programme, even though it did not join the coalition.
But the political and jurisdictional disputes that had constantly under-
lain the PNV’s tortuous path to a modus vivendi with the Republic during
1931-6 meant that the statute would still not have been promulgated
when the military rose in July 193 6. 15
Once again, Republican Madrid

15
Ibid., pp. 58, 67-72, 77.
 The challenge of mass political mobilisation 33

would use the statute (which it eventually ceded in October 1936) to tie

the industrial Basque Country to its warwould also

effort. Crucially, it

accept PXY leadership of the new pro\isional Basque wartime govern-

ment. But the fundamental disagreement over how much power it could
deploy would immediately erupt with full force. This was so not least
because the centrifugal impact of the military coup would by then al-
ready have conferred defacto ‘powers’ on the regions that far outstripped
anything the central Republican government had ever intended to
16
concede.
A similar jurisdictional dispute would also develop between the cen-
tral Republican government and Catalan nationalism - in spite of the
fact that 1 93 os Catalanism was clearly on the left and substantively in
agreement with the qualitative nature of republican reform, religious,
social and agrarian. Catalan nationalism of the centre-left had been a
fully subscribed Sebastian pact in 1930. 17 And when
member of the San
a coalition of political groups formed the Republican Left of Catalonia
ERC or Esquerra) in spring 1931. it had appeared the ideal interlocutor
for the liberal reformers of Madrid, led as it was by the urban professional
classes of Barcelona but with significant rural support in the region. In
short, Catalonia, as the most socially variegated area in Spain, had the
greatest potential for creating the counter-hegemonic alliance needed to
shore up the reforming Republic.
But the Esquerra 's relations with the republican-socialist government
of 1931 -3 w ere far from easy. In 1931 the Esquerra had initially declared
for an independent Catalonia in a federal Spain. But it had agreed to
forego this in return for Madrid ceding a statute of autonomy on generous
terms. But these terms, as the Esquerra saw it. never materialised. In spite
of the empathy over other sorts of structural reform and even though the
Madrid republicans recognised the Catalans' claims as licit in principle,
in the end their ingrained centralism was stronger. They sought to water
down the powers granted under the statute of 1932 and. even then,
delayed their transfer. That the Esquerra saw this as a promise broken
explains Catalonia's enthusiastic assumption, in the wake of the July
1936 military coup, of defacto pow ers which. Republican president Azana
18
w ould complain bitterly, lay beyond the statute.

1
See chapter 4 below for an analysis of wartime relations between the PXY and the central
Republican governments.
17
The 1920s had seen the political leadership of Catalanism pass from the conservative IJiga to
the centre-left; see the introduction above.
,8
For example, issuing currency and levying troops.
34 The Spanish Republic at war

Nevertheless, right from 1931 the Esquerra was a powerful political

force. Itdominated in Catalonia in a way that the PNV never did in the
Basque Country as a whole. The Esquerra’s success here can be gauged
by the fact that its identity would rapidly merge with that of the regional
government (Generalitat) ceded under the terms of the 1932 autonomy
statute. Moreover, in terms of agrarian reform, the Catalan republican
left would, between 1934 and 1936, fight and eventually win the right to
amend rural tenancies en masse. This had the effect of stabilising condi-
tions and increasing security for the small tenant farmers ( rabassaires who )

were the most numerous sector in the Catalan countryside and a source
of bedrock support for the Esquerra. The party fought this agrarian war
I9
first through legislative reform (the famous ley de contratos de cultivo and
)

then in the courts. Finally it was the ballot box, in February 1936, which
gave it victory. It would be this defeat that saw Catalonia’s (minority)
agrarian right — represented by the Institut Agrari de San Isidre - align
itself with its counterparts elsewhere in Spain and ultimately, in July 1936,
with the military rebels. 20
But although the existence of strong regional nationalisms problema-
tised the emergence of an overarching republican nationalism after 1931,
it is also true that the Catalan government’s dissatisfaction with Madrid

was much exacerbated by the fact of economic recession. With greater

budgetary resources the Generalitat could, for example, have funded its
own schools and thus ensured the dissemination of Catalan language and
culture. (Control of education remained beyond the autonomy statute
and, while Madrid recognised Catalan as an official language, it required
all school instruction to be undertaken in Spanish.) Financial stringency

would be responsible for more than the political disappointment of Cata-

lanists, however. For the commitment of the republicans - in Barcelona
as much as Madrid - to orthodox, deflationary liberal economics at a
time of international depression would make it impossible to provide a
credible level of social welfare relief for the urban and rural dispossessed.
Had it been possible to include them within a Republican ‘new deal’,
then urban and rural labour (or at least some sectors of it) could have
been mobilised as alternative support to compensate for the lack of a
sufficiently broad base among middling sectors. But the difficulties here
were enormous, as we shall see. It was the abiding distance between
the reforming Republic and its potential working-class support base that

19
The Law of Agricultural Contracts.
20
Conservative Catalan nationalism, in the shape of the Lliga, made some electoral gains around
1933-4, but not enough to unseat the centre-left coalition in Catalonia.
 The challenge of mass political mobilisation 35

made the price of its anti-clericalism too high. The combination of an

uncompromising religious settlement with monetarist economics would
deprive the republican-socialist coalition of any minimally sufficient
social support base.
After initial high hopes of Republican reform, worker disaffection
arrived quickly in metropolitan Spain - and most notably in the indus-
trial heartland of Barcelona. For many workers, their daily experience
was dominated by the absence of palliative reform (for the Republic had
21
promised )
it alongside the brutality of what appeared to be a them to
largely unreformed state apparatus in action. The underdeveloped
Spanish state had long been defined in terms of its security forces. Under
the old regime, the police and, in extremis the ,
army had functioned to
defend the established social order and elite economic interest in a highly
transparent fashion. While the coming of the Republic in principle
meant the chance to develop other, integrative, state functions — such
as education and welfare - in practice the options here were limited.
Even before the worst effects of international depression kicked in, the
Republic’s scope for enacting social and labour reform or for increasing
welfare spending was severely restricted. It was refused foreign loans
while at the same time it faced a highly unpromising national economic
situation, with a flight of capital as well as substantial debts inherited
from the dictatorship - especially in the form of the loans taken out to
fund Primo’s public works.
The international depression would also take its toll. The underde-
velopment of Spanish capitalism (and thus its lesser integration in the
international system) meant that the repercussions of the 1930s crisis
may have been relatively terms (for example there
less in macro-economic

was no sudden, new phenomenon of mass unemployment as there was

in Germany). But we should not make the mistake of assuming that the
impact of the crisis was therefore less severe for Spanish workers - who
included large numbers of economic migrants obliged to return home.
Moreover, Spain already had severe structural unemployment and highly
casualised and sweated labour patterns which, under the impact of reces-
sion, and in the absence of even the most rudimentary public welfare net,
pushed many sectors of the labouring classes to sub-subsistence levels.
But as the republican-socialist coalition was never conceived as a revolu-
tionary alliance, outright expropriation or other radically redistributive

21
Echoing the republicans’ own credo, there was a strong popular belief in the power of the letter

of the law. Cf. the epigraph to this chapter and n. i above.

36 The Spanish Republic at war

measures wejre not considered an option by either of the alliance’s com-

ponent parts.
Indeed, the only area in which the Spanish republicans were prepared
22
to depart from strict fiscal ‘rectitude’ was education. During 1931 they
took out special loans to underwrite a target of 27,000 new schoolrooms
(and teachers to staff them) in five years. In August 1931 a number of
‘teaching missions’ ( misiones pedagogical) were also established. In the form
of literacy classes, mobile libraries, travelling theatre exhibitions and civic
education, they brought ‘culture and politics’ to the villages of Spain.
While the project has attracted criticism because of its undeniably pater-
nalist overtones, it did reach people on a significant scale. 23 Indeed, the
subsequent conservative administration was worried enough to slash the
missions’ budget in 1934-5. But the missions’ main potential ‘public’ was
labour, and to construct this as a social support base for the progressive
Republic required rather more than the delivery of an abstract cultural
message - conservative fears notwithstanding. What it demanded was a
resolute, coherent and costed policy of practical social reform materially
to underpin republican ‘enlightenment’. 24 As one of the cultural mis-
sionaries memorably encapsulated the problem: [the rural poor] needed ‘

bread and medicine, and we had only songs and poems in our bags’. 25
But apart from education, the republicans never even saw fit to produce
26
a costing for the reform programme overall, within a formal budget.
On the other hand, from the start the republicans demonstrated their
strong line on law and order. A formative experience for urban labour
came with the Barcelona rent strike that erupted in the summer of
1931. The city and its surrounding industrial belt had a uniquely high

22
The ranks of political republicanism contained many teachers and educationalists who were
greatly influencedby the ideas of the Institute of Independent Education. They believed that
education was the key to modernising Spain.
23 M. de Puelles Benitez, ‘El sistema educativo republicano: un proyecto frustrado’, Historia
Contemporanea, 6 (1991), 159-71.
24 One can argue that a greater awareness of this existed in Spanish socialist ranks - cf. the
blueprint for social inclusion and nation building outlined in parliamentary socialist leader
Indalecio Prieto’s Cuenca speech (May 1936) demanding the ‘interior conquest’ of Spain. But the
same abstraction can also be adduced to criticise parliamentary socialist discourses of popular
mobilisation during the war; see chapter 4 below for a discussion of socialist premier Juan
Negrin’s wartime speeches.
25 C. Cobb in Graham and Labanyi, Spanish Cultural Studies pp. 136-7. The quotation, also cited
,

by Cobb, is the playwright Alejandro Casona’s.

26
J. M. Macarro Vera, ‘Social and Economic Policies of the
Spanish Left in Theory and in Practice’,
in M. S. Alexander and H. Graham (eds.), The French and Spanish Popular Fronts. Comparative
Perspectives (Cambridge: Cambridge University Press, 1989), pp. 171-84 and J. M. Macarro,

‘Economia y politica en el Frente Popular’ in Revista de Historia Contemporanea, 7 (1996), 129-50.

The PSOE’s Indalecio Prieto left the treasury for public works in autumn 1931.
 The challenge of mass political mobilisation 37

concentration of sweated factory labour and urban poor. 27 It was this

social context of the extreme impoverishment of unskilled labour, plus a

housing crisis within a deregulated housing market, which explain the
28
strike. It brought down the full force of Republican state discipline on
29 and
tenants and their leaders. They were subject to police harassment
‘preventive detention’. Their meetings were summarily banned, as were
worker newspapers that publicised strike-related matters. 30 Associative
rights were in practice being denied just as they had been under the
monarchy. That it was, on this occasion, the Republic’s newly formed
Assault Guards who were called in to supervise evictions merely rein-
forced the sense of continuity. The new Republican regime, fearful of
losing support among the urban (and especially the commercial) middle
measures in the name of the ‘authority principle’.
classes, justified these
It used the new Law for the Defence
of the Republic (passed in October
-
1931) to declare the strike an ‘illegal conspiracy’ against the regime
thus permitting the intensification of police action. 31
On numerous notorious occasions across Spain, Republican security
forces would clash fatally head-on with protesting workers: at Castil-
blanco (December 1931), in Arnedo (Logrono) and Llobregat (Barcelona
province), both in January 1932, and at Casas Viejas (Andalusia) in
January 1933. But beneath these high-profile incidents there lay a daily
experience of repression and exclusion.
In Barcelona especially, tensions increased as the republican authori-
ties found themselves unable to deliver promised welfare measures within
a severely restricted budget. (Nor of course did the establishment of an
autonomous Catalan government in 1932 make much difference here.)
The unemployed and others on the economic margins attempted to
27
For an excellent analysis of the material world of unskilled sweated labour which ‘made’ the
Barcelona working Ealham, ‘Policing the Recession’, chapter i, pp. 13-50.
class, see
28
ILO Spanish workers were the lowest paid in Europe - with the exception
figures indicate that
of the Portuguese. Yet the Spanish food-price index was higher than in recession-hit Germany,
where comparable wages were at least double. Ibid., pp. 142-3.
29 This was before the promulgation of the Catalan statute, so the Madrid government was still in
control of public order.
30 The old monarchist eviction law was revised, reducing the period at which eviction could be
enforced for rent arrears from 3 months to 8 days. Municipal byelaws were also pressed into
service to label flats ‘uninhabitable’ in order to remove striking tenants. The flats would then be
refilled with more pliable tenants, C. Ealham, ‘Frustrated Hopes. The 1931 Rent Strike and the
Republic’ (unpublished article), pp. 13-14. On the rent strike see also N. Rider, ‘The Practice of
Direct Action: The Barcelona Rent Strike’, in D. Goodway (ed.), For Anarchism. History, Theory,
Practice (London: Routledge, 1989), pp. 79-105.
31
As a a mass action, wound down by the end of 1931. But the failure to
result, the rent strike, as
resolveany of the desperate conditions that had sparked it meant that it would continue as a
sporadic action throughout the life of the Republic: Ealham, ‘Frustrated Hopes’, pp. 23-4.
38 The Spanish Republic at war

find their own solutions, for example by turning to itinerant street trade
or setting up ad hoc outlets (such as informal street stalls) selling cheap
food. But as these undercut established markets and shopkeepers, the
authorities, heeding complaints from the Barcelona Chamber of Com-
merce, sent in the Civil Guard (many of whose agents had served under
the monarchy) to arrest or dismantle the competition. 32 Pitched battles
regularly ensued in the working-class neighbourhoods of Barcelona be-
tween the police and the poor — both sellers and customers. There could
be no more graphic image of the social war waged between the
Republic and its dispossessed. 33 Republican law, once heralded as of-
fering these groups protection and redress, was, in the form of ‘public
order’, increasingly becoming a weapon against them. Moreover, the
interaction - calculated or otherwise - between public-order measures
and the Republic’s new labour legislation was systematically criminalis-
ing the most marginal groups of workers.
A key part of the labour legislation steered through under the auspices
of the republicans’ socialist coalition partners provided a national net-
work of committees (Jurados Mixtos) to settle labour disputes. But such
arbitration-based unionism, modelled on the practice of skilled sectors
of the UGT, was of little use to Spain’s army of unskilled, casualised
and easily replaceable industrial and agrarian labourers. Their lack of
bargaining power in the market place (above all in slump conditions)
made the direct action tactics spearheaded by the anarcho-syndicalist
CNT their only weapon. 34 But increasingly it was against these kinds of
labour strategies and their implementors that the Republic’s public-order
legislation was targeted. The Law for the Defence of the Republic was
in force throughout most of the period of republican-socialist govern-
The Law of Employer and Worker Associations (April 1932) 36
ment. 35
was used to much the same end. Militants were detained and union

32 The Republic created the Assault Guards as a new urban police force in 1931 and usually held
the Civil Guard in reserve for emergencies. But in Barcelona the Civil Guard was retained as
the normal policing body. The great continuity of police personnel (at least ‘in the ranks’ if not
in positions of responsibility) ensured the perpetuation of authoritarian ideas and a culture of
corruption. For policing in Barcelona and the failure of [Republican] professionalisation, see
Ealham, ‘Policing the Recession’, pp. 117-22.
33 Ibid.,
pp. 172, 192-3. This was a conflict which would continue unabated into the war years,
exacerbated by ever-increasing shortages, inflation and the black market. See chapter 5 below.
34 Although some CNTunions did accept the arbitration system - in spite of opposition from the
CNT’s national leadership. See the discussion of divisions in the CNT and CNT-UGT relations
later in this chapter.
35 In July 1933 it was replaced by the Public Order Act.
36 Ley de Asociaciones Profesionales, Patronales y Obreras; see J. Casanova, De la calle al Jrente El
.

anarcosindicalismo en Espana (1931-1939) (Barcelona: Critica, 1997), p. 55, p. 74, n. 13.

 1

The challenge of mass political mobilisation 39

premises closed. In the summer

of 1933 an anti- vagrancy law (Ley de
Vagosy Maleantes) was also introduced. 37 This permitted the detention
of those who could not prove that they had legal means of supporting
themselves. outlawed the financial collections on which the CNT’s
It also
organisation and strategies (especially spontaneous industrial action)
depended, and it threatened collectors with internment. 38
A test case under the Anti- Vagrancy Act was brought against a num-
ber of radical anarchist leaders from the FAI, including Buenaventura
Durruti and Francisco Ascaso - neither of whom was unemployed -
while they were on a speaking tour of Andalusia in 193 3. 39 Such legisla-
tive harassment more or less obliged anarchist and communist activists
-
above all in Barcelona - to operate in clandestinity, particularly when
they were organising the unemployed. Into this category came street

sellers and itinerant workers of all kinds who could also be detained
under the conditions of the anti-vagrancy law, in camps established for
the purpose. 40 It was not lost on those so treated that the much-vaunted
liberal freedom of association reached no further than skilled sectors of
the UGT. Republican law and order was effectively branding non-social
democratic constituencies of organised labour, plus anyone else forced
for reasons of survival to operate beyond liberal economic nostrums, as
41
‘enemies of the state’.

The republicans would seem to have believed that the mere existence
of ‘the Republic’ - or at least the dejure declaration of Republican liber-

ties —would serve to pacify economically and politically marginalised sec-

tors. Their constant evocation of ‘the people’ in parliamentary rhetoric

37 This was used as a kind of ‘judicial hoover’ (Ealham, ‘Policing the Recession’) to ‘regularise’ the
detention of all those being held at that time for reasons of perceived ‘undesirable’ behaviour or
lifestyle. The law was also retained after the re-election of the liberal left in the Popular Front
elections of February 1936.
38 The law specified financial collections by ‘clandestine’ organisations. This was a reference to the
CNT’s non-registration (prior to 1936) under the April 1932 law: Casanova, De la calle alfrente,
p. 141. The definition of ‘collections’ was also stretched to cover the meagre stipends paid to
CNT activists, thus allowing the latter to be targeted also.
39 Ealham, ‘Policing the Recession’, pp. 286-7.
40 Some camps were specially established, but a lack of government resources meant that existing
prisons (and in Barcelona prison ships) were used: Ealham, ‘Policing the Recession’, pp. 284-91.
Most camps were in Barcelona with some in the south. But statistical information and details
of camp regimes are sparse. A police report (Madrid interior ministry) refers to 107 individuals
being detained in Seville in September 1933 am grateful to Chris Ealham for this information
.

and for his help with the material in this section.

41
As discussed later in this chapter, Spanish socialist ethos (whether expressed by trade union
leaders or parliamentary socialists) also underwrote the notion of organisable (‘respectable’) and
unorganisable (disruptive/lumpen) elements of labour. One of the architects of the anti-vagrancy
law was the lawyer and leading parliamentary socialist Luis Jimenez de Asua.
40 The Spanish Republic at war

contrasts with their growing horror during 193 1-3 at the prospect of
actively mobilising masses whom they had not the least idea how to
cope with. As some of their middle-class constituencies expressed sim-
ilarly derived fears over the perceived growth of crime and disorder,
5
republican concern with the ‘neutrality of the street and public spaces
42
increased. These fears were probably a contributory factor in the re-
publicans’ failure to make effective public-order reform, and in partic-
ular, their signal failure to demilitarise the apparatus of public order.
Republican attitudes here, of course, also betray their inability to re-
linquish an elite conception of politics rooted in the ‘old demobilised
5
world of nineteenth-century liberalism. 43 But their heavy-handedness
in the area of law and order certainly accelerated urban worker disaffec-
tion. Meanwhile, protests sparked by economic hardship were sharp-

ened in the new political climate, which raised worker expectations

of redress. But budgetary constraints meant that the Republic simply
5
could not deliver the social ‘salvation demanded by mobilising worker
constituencies.
Nowhere was demands clearer
the redemptive dimension of worker
than in the rural deep south of Spain. For was there, of course, that it

the de facto power of the old elites remained most completely intact.
Such paternalistic social reforms as had been attempted under monarchy
and dictatorship were restricted to the urban arena, with the country-
side remaining off limits. There the Civil Guard had long been en-
meshed in clientelistic relations with the local landowning elites. As the
underlying relations of socio-economic power remained basically un-
changed across the political transition to a Republic, the Civil Guard
continued to be used as an instrument to prevent, or at least slow up,
the practical implementation of reform. 44 Slowness and obstruction af-
fected the agrarian reform badly, compounded as the problems already
were by technical, political and financial insufficiencies. By contrast,
the raft of labour reforms 45 initiated by the socialist labour minister

42 Casanova, De la calle alfrente , p. (inc. n. 12).

145
43 For further discussion of this, see the analysis below of republican politics on the morrow of the
February 1936 Popular Front elections.
44 The 1930s still saw the Civil Guard used in time-honoured ‘tradition’ to keep the starving off
uncultivated land and to prevent them poaching or scavenging for acorns and the like on estate
property
45 Most importantly, a minimum wage and maximum working day; requiring landowners to hire
through the labour exchanges and to contract workers on the basis of a strict rota; the role of
the state-chaired arbitration boards ( Jurados Mixtos) in both settling disputes and overseeing the
fulfilment of the new labour norms.
 The challenge of mass political mobilisation 41

Francisco Largo Caballero and overseen by the UGT’s cadres, did start
to make more of a material impact on both the political and socio-
economic balance of power in the rural centre-south. 46 But while the
rural elites understood the subverting potential of these norms (and
would mobilise accordingly), for the rural dispossessed - focused on
the icon of land reform - these changes seemed painfully slow and
piecemeal. 47
In terms of implemented policy, the socialist labour reforms probably
achieved more than did republican land reform. But given the struc-
tural ‘oversupply’ of unskilled labour generated by the latifundia system
of landholding, unemployment was a massive and endemic problem.
The scale of welfare relief needed was way beyond anything that could
have been contemplated within republican fiscal norms. The situation
for destitute landless peasants was worsened by the international de-
pression, which meant that state funds for public works relief were also
modest. Indeed, the exhaustion of the public works budget before the
particularly hard winter of 1932-3 explained much of the turmoil of that
year.
Land reform too was bound by similar financial constraints. Indeed,
these were exacerbated by one of the republicans’ fundamental political
principles - the inviolability of private property. For this meant that
they were committed to indemnifying owners (in government bonds)
at the fullmarket value of the land acquired. Resettlement could thus
occur only at a painfully slow rate. In such circumstances, no reform —
however politically subversive in the medium term - could have been
enough. 48 The implicit sense of urgency here was, of course, born of the
new political perceptions among the rural poor, of the belief that the
Second Republic would be synonymous with their own empowerment.
Their faith could not long outlast the interaction of elite obstruction and
republican limitations.
By mid 1933 the republican-socialist coalition was close to breaking
point under the dual strain of the alienation of the disenfranchised and

46 S.
Julia, ‘La experiencia del poder: la izquierda republicana 1931 -1933’, in N. Townson (ed.),

El Espana (1830-igyy) (Madrid: Alianza, 1994), pp. 183-4.

republicanismo en
47 This impatience needs also to be seen in a brutal social context where the most reactionary
of landowners and bailiffs taunted starving labourers to ‘corned Republica’ (literally ‘eat the
Republic’, or ‘let the Republic feed you’).
48 It is also unlikely that the republican model of settling plots of land on individuals would in any
case have been economically feasible in the medium term - given the inhospitable conditions
for agriculture in the south. But, once again, republican ideology had prevented the acceptance
of the PSOE/UGT’s preferred collectivist model of agrarian reform.
42 The Spanish Republic at war

the mounting elite offensive. When the split came in September the de-
cisive pressure was that applied to the republicans by an increasingly
vociferous business lobby. 49 This was determined to detach the Socialist
Party from government as the first step towards curtailing social re-
form. But the tensions between republicans and socialists and within
the socialist movement itself were also considerable by this stage. While
parliamentary socialists might have wished for greater republican pol-
and application over reform, it was the socialist union
icy consistency
leadership that most frustrated and exposed by the limits of republi-
felt

can economic strategy and political will. For these had exacerbated the
effects of the right’s obstruction of reform.
Throughout the life of the republican-socialist coalition government,
Spain’s employers’ federation had waged a high-profile and intensely
personal campaign against the labour minister, UGT general secre-
tary Largo Caballero. This was aimed primarily at ending the executive
power of the national labour arbitration committees. 50 At the same time,
the socialist union leaders were also increasingly fearful that the UGT’s
presence in government risked making seem responsible in the eyes of
it

its members improvements and land

for the slowness of social welfare
reform. In particular they were worried about the effect on frustrated
recent adherents to the socialist landworkers’ federation (FNTT), which
had expanded dramatically after April 1931 in the hope of government-
led reform. 51 These anxieties were massively heightened by the context
of violent clashes between urban and rural workers and Republican secu-
rity forces. The UGT’s leaders feared that their guilt by association with

a regime that not only neglected workers’ interests but also physically as-
saulted them would lead to a haemorrhage of UGT members to the rival
CNT. Given that the anarcho-syndicalist organisation had its own inter-
nal problems, with hindsight this was perhaps an exaggerated fear, but it
was nonetheless perceived as a real danger at the time. Nevertheless,
socialist departure from power still seemed ‘unthinkable’ — precisely

49 Socialist participation in government would split the Radical-Socialist Party, some of whose
components would in 1933 also spearhead the internal republican campaign against agrarian
and labour reform, N. Townson, The Crisis of Democracy in Spain. Centrist Politics under the Second
Republic iggi—igg6 (Brighton/Pordand: Sussex Academic Press, 2000), p. 170. Details of the
multi-pronged attack on the PSOE as a party of government are in given Macarro Vera, ‘Causas
de la radicalizacion socialista en la II Republica’, pp. 206-8. For the employers’ lobby, see
M. Cabrera, La patronal ante la Segunda Republica (Madrid, 1983).
50 S. Julia, ‘La experiencia del poder: la izquierda republicana, 1931-1933’, in Townson, El repub-

licanismo en Espaha, pp. 181-3.

51
Preston, CSCW, pp. 30, 31, 78.
 The challenge of mass political mobilisation 43

because no sector of the PSOE/UGT leadership had any political strat-

egy but gradualist reform. 52
As the tensions mounted inside republican ranks, the break with the
socialists came in September 1933. In the event, it was the socialists who

departed, under pressure from Largo Caballero. But his decision was
doubdess hastened by the fear that internal republican differences would
soon lead to the PSOE being asked to leave. With the left divided, and
a Republican electoral law in force which actively favoured coalitions,
the general elections of November 1933 saw the victory of a centre-right
alliance whose electoral backbone came from the CEDA. The disinte-
gration of the left coalition meant that the republicans were cut off not
only from the socialists’ electoral base but also from their organisational
resources, which were the nearest thing the left had to a national political
infrastructure. The lack of these assets would erode the political potential
of progressive republicanism, separated as it was now from government
resources, and with the mobilisation of a well-organised populist right
accelerating apace. For the socialists too, exclusion from power would
have no less seriously erosive effects. The trauma of departure would ac-
celerate and embitter divisions within the movement during 1933-5.
turn these would fatally undermine attempts to resurrect a republican-
socialist governmental alliance - although, in the end, this remained the

sine qua non of social reform in Spain.

THE INTERNAL FRAGMENTATION OF SPANISH

SOCIALISM (1933-6)

As conservative mass mobilisation consolidated around the CEDA, the

left’s social and political base became increasingly fragmented. While

this fragmenting process reached beyond the socialist movement to an-

archists, anarcho-syndicalists and communist groups, the most crucial

fissure on the left was that running through the PSOE/UGT. Having
blocked any resumption of republican-socialist alliance, Largo Caballero
increasingly identified himself in the public eye with those sectors of the
movement which claimed that the reformist option had now
socialist

demonstrably failed, that revolution was the inescapable conclusion and

52
This point is excellently made in Jose Manuel Macarro Vera’s subde and suggestive article
‘Causas de la radicalizacion socialista’. The author seems to assume, however, that evidence
of CNT crisis per se means that the UGT cannot have feared losing members to it. But this
assessment leaves out of account self-generating fears in a context of besieged social reform.
44 The Spanish Republic at war

that the PSOE’s organisation had to be ‘bolshevised’ in preparation for

its role as the revolutionary vanguard party. (In approximate terms, these
sectors constituted parts of the UGT, a majority of the socialist youth
leadership 53 and minority sectors in the PSOE.) But no such process of
bolshevisation occurred. Between 1934 and 1936 Largo spoke of a revolu-
tion that neither he nor anyone else in PSOE/UGT ranks was preparing.
Some radicals among the socialist youth believed in it, assisted in their
belief by the superficially theoretical discourse of left socialist intellectuals
such as Luis Araquistain and Carlos de Baraibar. Their discourses were
alsoused by Largo to bolster his own organisational position. Although
no conscious bad faith was involved on Largo’s part, this borrowing of a
revolutionary language was not enough to make him a revolutionary. His
revolutionism remained, throughout, purely the expression of a moral
preference that Spanish socialism should change. ‘That preference never
put down roots in the structure or practice of the organisation.’ 54 But it
provoked fear among conservative groups whose access to firepower was,
in the end, inevitably superior - in the shape of the army. Nor did Largo’s
supporters themselves make any contingency plans (such as the arming
of defensive militia) to counter the potential recourse of conservative
forces to ‘catastrophist’ solutions. When sectors of the officer corps rose
against the Republic in July 1936, the socialist left’s lack of any political
strategy, let alone a revolutionary one, would become fully apparent.
In order to explain this apparently contradictory behaviour, it is impor-
tant to disaggregate the political motives (conscious and unconscious) of
Largo Caballero and the socialist left’s trade union leadership from two
other components. 55 First, the context of increasing discontent and some
radicalisation of political intent among sectors of the socialist movement’s
grass roots, faced with the erosion of reform and working-class gains by a
conservative government. Second, the separate agenda of the radicalised
socialist youth leadership - which was not necessarily compatible with

that of the UGT’s veteran leadership. 56 In addition, socialist youth rad-

icalism almost certainly had an important sociological - as opposed to

53 But not necessarily of the youth membership as a whole: R. Vinas, La formation de las
socialist
XXI, 1978); H. Graham, ‘The Socialist
Juventudes Socialistas Unificadas (1934-1936) (Madrid: Siglo
Youth in theJSU: The Experience of Organisational Unity, 1936-8’, in M. Blinkhorn (ed.), Spain
in Conflict 1931 -1939 (London: Sage, 1986), pp. 83-102.
54 Macarro Vera, ‘Causas de la radicalizacion socialista’, p. 223.
55 For an analysis of the socialist left, see H. Graham, ‘The Eclipse of the Socialist Left: 1934-1937 \
in F. Lannon and P. Preston (eds.), Elites and Power in Twentieth-Century Spain. Essays in Honour of Sir
Raymond Carr (Oxford: Clarendon Press, 1990), pp. 127-51.
56 R. Vinyes, La Catalunya International. Elfrontpopulisme en I’exemple catala (Barcelona: Curial, 1983),

p. 256; Graham, Socialism and War, pp. 69-73.

 3

The challenge of mass political mobilisation 45

purely ideological - dimension. Although this is a relatively unexplored

area to date, the 1930s in Spain also saw the eruption of youth as a major
protagonist onto the political stage.
The and stronghold of Caballerismo remained, nev-
political centre
ertheless, the UGT executive - given both the numbers it controlled
(especially in the massively expanded FNTT) and Largo Caballero’s
immense personal prestige in the PSOE/UGT as a whole. It is signifi-
cant here that all Largo’s justifications for breaking the alliance with the
republicans hinged on the ‘wrongs’ of 193 1-3: on the employers’ cam-
paign against reform (and himself), on police violence against workers
and on the republican political class’s outrageous lack of solidarity (as
he saw it) with the PSOE in 1933. At no point did Largo ever justify
socialist departure from government in terms of an alternative revolu-

tionary strategy or any assessment of the future balance of political forces

in Spain. He did not do so because no such calculation existed on the
left. The UGT was far too valuable and hard won an organisation for its

substance and future to be risked in one revolutionary throw of the dice.

For Largo Caballero and the veteran union leadership, whose formative
experience had been that of the traumatic August 1917 general strike, the
threat of state repression never seemed far away — even when one held
ministerial office. For the UGT leaders, moreover, all strategy and policy
had to fulfil a central objective: the defence, nurture and augmentation
of the socialist organisation - identified, in a characteristically idiosyn-
cratic reading, with the deepest interests of the entire Spanish working
class.

But the organisational unity of the Spanish working class - however

envisaged - was a goal whose practical possibilities seemed to be erod-
ing precisely as the UGT’s political rhetoric of proletarian unity became
more insistent. The divisions and tensions both within and between the
socialist and anarcho-syndicalist leaderships and their respective move-

ments increased from April 1931 as the Republican environment stimu-

lated mass political mobilisation.
Largo Caballero’s objective in 1931 had been to use the leverage of
state power - in particular the labour arbitration boards - to exclude the
CNT from the frame. Unsurprisingly, this had further soured relations
with anarcho-syndicalist leaders already distrustful of Largo after his
collaboration with Primo in the 1920s. In fact, however, the UGT’s tactics
had rebounded. The slowness and obstruction of reform during 1931—
turned the UGT’s ministerial experience into a crisis of its own political
identity. In a sense this was inevitable since the UGT as well as the CNT
46 The Spanish Republic at war

contained April 1931 just as before) both radical and reformist

(after

practices of labour resistance. 57 Moreover, the was also acquiring a UGT

bigger following of unskilled, marginalised (often migrant) workers in the
1 93 os
- that is, precisely those constituencies worst hit by unemployment
and recession and for whom UGT arbitration-based unionism remained
inaccessible. 58
An absolutely fundamental fact to bear in mind here is that the grass
roots of organised labour - whether in the UGT or the CNT - did not
necessarily see parliamentary and direct action strategies 59 as mutually
exclusive, or existing in some kind of rigid ‘evolutionary’ order. Many
wanted access both simultaneously. Nor did ordinary workers neces-
to
sarily construct the UGT
and the CNT as binary opposites in ideological
terms. A certain amount of ‘mixing and matching’ went on. The richness
of anarchism’s political culture was undeniable, but the CNT’s unpre-
paredness to engage with the Republican political system - in particular
at the level of municipal politics - was a real weakness after 1931. It
was this, for example, that gave the Spanish Communist Party (PCE) its
opening in Seville in the early 1930s, where it began to build up a mem-
bership base. 60 (Just as the UGT’s ambivalence to radical tactics had
earlier given the Communist Party an entree in the industrial north.)
Moreover, it was probably this pragmatic and fluid political outlook
among the grass roots of worker constituencies that explains the under-
lying appeal of the re-emergent strategy of centre-left electoral alliance,
by spring 1936 known as the Popular Front. For many of those mobilised
in the socialist, communist and anarcho-syndicalist movements, the Pop-
61
ular Front was a parliamentary strategy, but it was also something more .

Precisely for this reason the PCE would have a fairly easy ride politically
when it made the transition from an exclusively worker front {/rente unico)
to support the Popular Front.
Largo Caballero and his colleagues in the UGT leadership were them-
selves allergic to any kind of direct action. But they also realised that the
feelings of their rank and file were not necessarily so clear-cut. The

57 See the discussion in the introduction above.

58 This is the situation which underlay, for example, many of the strikes in 1930s urban Madrid -
including those of 1936 (construction and hosteleria (cafes and catering)) when UGT strikers stayed
out on strike with their CNT counterparts, disobeying their own leaders’ calls for a return to
work. Julia, La izquierda del PSOE, pp. 253-64.
59 For example, spontaneous industrial action or community-based initiatives such as rent strikes,

housing occupations and campaigns against speculation (or food adulteration) by shopkeepers.
60
J.M. Macarro Vera, La utopia revolucionaria. Sevilla en la Segunda Republica (Seville: Monte de Piedad
y Caja de Ahorros de Sevilla, 1985), pp. 94-5.
61
See further discussion later in the chapter.
 The challenge of mass political mobilisation 47

revolutionary language that Largo began to adopt in his speeches, after

the socialists had exited from government in 1933, was (consciously or
otherwise) an attempt to contain radical currents in the socialist move-
ment without, supposedly, incurring any organisational risk. In fact the
risks would turn out to be great in more than one respect. For Largo

and his executive would call for a national revolutionary strike, but then
fail it. When the strike erupted, in October 1934, its
signally to lead
epicentreand leadership lay far from Madrid, in Asturias. As a result,
Largo and his executive committee would come to be enveloped by the
miners’ rising there. The judicial fall-out, when the UGT leaders were
brought to trial for rebellion against the state, activated all their worst
memories of 1917. But even after the trial Largo would still carry on em-
ploying a vehement revolutionary rhetoric - now because, after October,
he was also afraid of the UGT being outflanked on the left. But this was
a highly irresponsible course because Largo still had no organisational
blueprint for revolution. Moreover, the fact that the UGT was relatively
more centralised than the CNT meant that this attempt to square the
circle between radicalism and reformism would produce much greater
internal tensions and would do so much rapidly.
The CNT, as a looser structure, with regional federations that, his-
torically, had been virtually independent entities, had been able to live
with political difference more easily. But socialism’s governmental role
between 1931 and 1933 had, nevertheless, opened up the internal fault
lines in the anarcho-syndicalist organisation — for and against political
participation. Syndicalists such as Angel Pestana and Juan Peiro argued
that it was now time to abandon anti-parliamentarianism in order to
incorporate the CNT sufflciendy to allow the defence of its militants
and social constituencies through the newly available formal channels
of municipal politics and state labour agencies. 62 The new Republican
environment gave the CNT no choice but to compete in the political
arena with the PSOE/UGT. Failure to renovate libertarian ideology
and warned, would leave the CNT’s rank and file ex-
practice, they
posed and would ultimately erode its base as members were attracted
to the perceived benefits of its socialist rival as a union plugged into
state power. 63 In the opinion of the reformers, by refusing to sanction

62
CNT’s future industry minister in November 1936), had been part of the
Pestana, like Peiro (the
moderate group associated with Solidaridad Obrera that had acted as caretakers of the CNT in
the 1920s, opposing the (often exiled) radical anarchists of Solidarios/Nosotros.
63 For a forceful expression of this view, see Lorenzo, Los anarquistas espanolesy el poder, p. 202.
48 The Spanish Republic at war

the organisation’s political engagement, the purist anarchists were not

only assisting the UGT leadership in its bid to attract CNT members,
but were also damaging the welfare and rights of the anarcho-syndicalist
base. However, for the radical anarchists who, unlike large sectors of the
CNT’s own base, saw this question in binary terms, to propose political
engagement of any sort amounted to a betrayal of the purpose and value
of direct action and worker— state confrontations.
The reformers’ arguments would seem to find an echo in the member-
ship decline the CNT incurred during 1931-6. This would be steepest in
the key Catalan regional federation (CRT), which dropped from 300,000
members in 1931 to around 136,000 by May 1936. 64 This did not nec-
essarily impinge on the CNT’s considerable informal mobilising power
among radicalised and excluded social constituencies - particularly in
Barcelona — but it did denote some kind of organisational ‘crossroads’.
Indeed, this was clearly defined for it by the treintista schism 65 of 1931
when unions in Catalonia and Valencia representing skilled workers in
the most developed industrial sectors left the CNT. They judged their
position to be untenable, yet saw no means of winning the argument
66
over strategy against the radicals. Some of the most powerful of these
unions would never return. The fact that treintismo was so influential in
Catalonia makes the point very clearly that this profound division in
the CNT did not separate the organisation in the rest of Spain from the
Catalan ‘heartland’ but rather that it ran straight through that libertarian
heartland.
It is also worth noting that it was in areas where the treintistas were
strong in Catalonia that the 1930s also saw breakthroughs for quasi-
Catalanist political parties more left-inclined than the Esquerra. The
main example here was the Catalan communist formation, the Bloc

64 The CNT’s major regional federations — Catalonia, the Levante and Andalusia/ Extremadura -
allexperienced a decline during 1931-6. But the Catalan federation had dropped from its 1919
peak of 400,000 when it had represented half of the total CNT
membership for Spain: J. Peirats,
La CNT en la revolution espanola (3 vols., n.p.: Ruedo Iberico, 1971), vol. 1, pp. 27-8. For the Catalan
federation in the 1930s, see S. Tavera and E. Vega, ‘La afiliacio sindical a la CRT de Catalunya:
entre l’eufbria revolucionaria i l’ensulsiada confederal 1931-36’, in (various authors) Revolucio i

Socialisme, vol. 2 (Barcelona, 1990), pp. 343-63, esp. pp. 350, 353. For the south, see Macarro
Vera, La utopia revolucionaria, p. 62.
65 So-called because of its origins in the thirty signatories to a reformist CNT manifesto in August
1931: Peirats, La CNT en la revolution espanola, vol. 1, pp. 59-63.
66
They formed the Sindicatos de Oposicion (FSL). J. Brademas, Anarcosindicalismo y revolution en
Espana 1930-1937 (Barcelona: Ariel, 1974), pp. 76-7 91 -2 ,
, 1 17-2 1; E. Vega, El trentisme a Catalunya.
CNT (Barcelona: Curial, 1980) and Anarquistasy sindicalistas 1931-1936.
Divergencies ideoldgiques en la
La CNTy los Sindicatos de Oposicion en el Pais Valenciano (Valencia: Institucio ‘Alfons el Magnanim’,
1987), pp. 225-6.
 The challenge of mass political mobilisation 49

Obrer i Camperol (BOC 67 ) - which in September 1935 would merge

68
to become the core membership of the POUM. This coincidence of
treintistas with the BOC reinforces the point that the heart of the CNT’s
crisis was its continued rejection of parliamentary politics in a context

of accelerating mass political mobilisation. 69 This was also likely to have

been one reason why in a number of cities of historic CNT strength
(Valencia, Seville and, to a lesser extent, Zaragoza) UGT membership
was gradually rising during the period 1931 -6. 7 ° It was also in the UGT

that some of the treintista unions would eventually find a new home in
1936. For example, those in Sabadell and Manresa would join the Catalan
UGT in the political exhilaration and mass enthusiasm over worker
71
5
‘unity following the Popular Front elections of February 1936. Few
syndicalists would go as far as Pestana, who, in 1933, left the CNT to
72
form the Partido Sindicalista. But when he departed he left behind
an escalating debate over ideology and organisational forms that would
reverberate on and into the war period. Nor did this debate involve only
the CNT. It also cut across the FAI, the anarchist federation created back
in 1927.
The impulse to create the FAI had come from
concerned radicals
about the influence of syndicalist reformists in the CNT. But the image 73

of the FAI as a tight, cell-based organisation of the like-minded is some-

what misleading. The FAI was, by its own political definition, a loose
network of individual anarchist groups across Spain. 74 This form meant
that the FAI too reflected and transmitted the variations in anarcho-
syndicalist political perspectives across the regions of Spain. 75 Its groups
debated the same questions of ideological direction and organisational
hierarchy/ centralisation that the CNT did. And just as in the CNT, the
67
Workers and Peasants Bloc.
68
A. C. Durgan, B.O.C. igjo-igj6. El Bloque Obrero y Campesino (Barcelona: Laertes, 1996),
pp. 163-6. See also the analysis of left politics in Catalonia in 1931-6 below in this chapter.
69 The CNT had acted decisively to prevent the BOC’s entryism when it expelled BOC-led unions
in 1932. This was when the BOC created its own trade union, the FOUS.
70 Casanova, De la calle aljrente, p. 59 (n. 32); Macarro Vera, La utopia revolucionaria, pp. 48-51.
71
Balcells, Trabajo industrialy organizacion obrera en la Cataluna contemporanea igoo-igg6, pp. 156-7.
72
Peirowould also ‘withdraw’ defacto from the CNT when he saw the reformist route blocked. He
would only return after the military rising ofJuly 1936: Casanova, De la calle aljrente pp. 141-2
,

(n- 9 )-
73 S. Christie, We, the Anarchists! A Study of the Iberian Anarchist Federation (FAI) ig2y-igjy (Hastings:
The Meltzer Press/Jura Media, 2000), p. 21. See also the introduction above.
74 S. Tavera and E. Ucelay-Da Cal, ‘Grupos de afinidad, disciplina belica y periodismo libertario
t936-i938’, Historia Contemporanea 9 (1993), 167-8.
,

75 R. Fraser, Blood of Spain. The Experience of Civil War ig36-iggg (Harmondsworth: Penguin, 1981),
p. 548.
50 The Spanish Republic at war

opinions expressed in those debates varied. 76 The FAI certainly num-

bered radical anarchists among its members - including some powerful
and charismatic ones. But the FAI was not as politically homogeneous
as it is usually portrayed - particularly not beyond Catalonia. It is im-
portant to bear this in mind when considering the political evolution of
the FAI during the civil war.
The impact of the syndicalist reformers’ arguments on the CNT-
FAI as a whole during 1931—6 was diluted, however, by the way in
which other CNT-identified social constituencies were experiencing the
Republic ‘on the ground’. The fact of slow or stalled reform, of the exclu-
sion of unskilled and casualised labour from any ‘new deal’; then, after
November 1933, the centre-right administration’s rolling back of even
modest gains, all stymied the syndicalist reformers. As a result, organi-
sational reform proceeded at a pace that was almost imperceptible.
As agreed by CNT congresses in 1918 and 1919, the craft union struc-
ture had been converted into industrial unions - the sindicatos unicos ? 1
This meant bringing together in a single union (or branch) all the trades
operating within a given industry - for example, the building or textile
industry each constituted one industrial branch (among thirteen created
overall).
78
In 1931 CNT congress agreed the next stage of consolidat-
- something that had
ing these unions into national industrial federations
been successfully opposed by anarchist radicals back in 1919. But there
was little real progress here during 1931-6. Had the national federations
of industry been created, they could have overcome the CNT’s inver-
tebrateness. The CNT’s regional federations had always been virtually
autonomous entities. The national leadership was elected to represent,
but it had no power to enforce policy or organisational directives. In the
course of strikes in the 1930s the CNT did not even have the organisa-
tional means to communicate with its own constituent unions - which
seriously inhibited their efficacy. 79 In the absence of national federations,
itwas also difficult for the organisations of different industrial branches
to communicate with each other. Indeed, little efficient communication —
80 —
either top-down or laterally was possible beyond the regional level.

76 Tavera and Ucelay-Da Cal, ‘Grupos de afinidad, disciplina belica y periodismo libertario 1936-
l
93 %\ 73
P- T -

77 The Sants congress of the Catalan region (CRT) in 1918 and theCNT national congress in 1919.
78 On the CNT and industrial unionism see the summary in Fraser, Blood of Spain, p. 544 (n. 3).
79 The national executive could neither advise nor adequately gather information: Brademas,
Anarcosindicalismoy revolucion en Espana 1930-1937, p. 73.
80
The UGT was also defective here. The national executive in Madrid could, in normal cir-

cumstances, communicate relatively easily with all parts of the organisation. But in many areas
internal regional and provincial articulation simply did not exist. J. M. Macarro Vera discusses
 1

The challenge of mass political mobilisation 5

As a result, the CNT would have great difficulty coordinating modern

81
urban industrial strikes during the Republican years.
Although it has been little noted in the Anglo-American historiog-
raphy, a major ideological and organisational crisis was looming in the
CNT by 1936. In May, the CNT’s famous Zaragoza congress would
attempt to restore ‘unity’ by simply denying the cleavage. It endorsed
the anti-parliamentary line ‘by acclamation’, while also agreeing to the
reincorporation of some of the schismatic union federations that had
82
left after the But in reality the congress resolved nothing.
treintista split.

The between syndicalist pragmatists and purist anarchists over

conflict
the CNT-FAI’s future structure would still be pending when the military
rebelled in July 1936. After that point, the lack of national federations
of industry would seriously hinder the coordination of resistance. More-
over, it would also mean, ironically, that radical anarchists would have
no integrated organisational base from which to contest war policy with
their political opponents in the Popular Front.
But the consequences of internal dissension between the CNT’s lead-
ers or of conflict with their UGT
counterparts were not the only sources
of left fragmentation during 1931-6. There were also bitter and vio-
lent disputes between some sectors of the two memberships. This con-
flict frequently stemmed from the daily impact of recession in a soci-

ety where unemployment levels were high and few had access to any
kind of public relief. As recent work has shown, this also led to a lot
of new labour mobilisation at grass-roots level concerning specific lo-
cal grievances. 83 This situation was further embittered by the fact that
the CNT was also battling against the UGT’s bid to control the labour
market. Strike demands in 1932-3 - with over half of the strikes oc-
curring in the agricultural and building sectors - reflected this climate:
there should be no further reduction of the working day, a rota sys-
tem should be implemented to ration work, and a quota established to
ensure that the unemployed were hired. 84 There was virtual open war-
fare over control of the UGT-dominated labour arbitration machinery.

the Andalusian case, in S. Julia (ed.), El socialismo en las nacionalidades y regiones, Anales de Historia,
vol. 3 (Madrid: Fundacion Pablo Iglesias, 1988), pp. 105, 118. The great difficulty of lateral com-
munication between the different provincial and regional organisations of the UGT would be a
severe impediment in wartime.
81
Peirats, La CNT en la revoludon espanola, vol. 1, pp. 53-7.
82
CNT congresses tended to adopt decisions ‘unanimously through acclamation/display’. See
also Casanova, De la calle alfrente , pp. 63-4, 87 for a resonant description of the dimensions of
the CNT crisis.
83
Cf. Gil Andres, Echarse a la calle. Amotinados, huelguistasy revolucionarios.
84
Julia, ‘La experiencia del poder: la izquierda republicana, 1931-1933’, in Townson, El republi-

canismo en Espana, pp. 181, 185.

52 The Spanish Republic at war

The CNTVSeptember 1933 construction strike was primarily aimed at

gaining union recognition and equal bargaining rights by breaking the
UGT’s monopoly on hiring at new construction sites. In the agricul-
was constant conflict in 1931 -3 between socialist
tural sector too there
and anarcho-syndicalist unions over the functioning of the arbitration
boards. 85 Additionally tensions were exacerbated by the operation of
Republican public-order legislation, as we have seen. These antagonisms
had much in common with those that occurred between the SPD and the
KPD under the Weimar Republic. The KPD, like the CNT, increasingly
identified with and mobilised those politically and economically excluded
from the Republican order: the unskilled, the unemployed, migrants, the
marginal.
On the Barcelona waterfront, the battle between the socialist and
anarchist dockworkers’ unions for members erupted once again into vio-
86
lence. Members were important in their own right, but the particular
violence of the 1930s dispute (and resulting fatalities) derived from the
fact that work — as an increasingly scarce commodity — was allocated in
the Republic’s employment exchanges according to union size. There
were some other specific sources of tension too. For example, in the 1930s
the Catalan UGT became the new home for private security guards who
had previously belonged to the Sindicatos Libres, the yellow unions of the
192 os that had collapsed with the fall of the Primo dictatorship. Ugetistas (

in Catalonia were also sometimes used by employers to replace a striking

CNT workforce, thus earning themselves a reputation for scabbing.)
Elsewhere too there were other sorts of violent confrontation between the
rank and file of the two ‘sister’ unions. In Seville the intra-organisational
conflicts were made more fraught by the strong presence of communist
85 For the agricultural sector, see M. Perez Yruela, La campesina en la provincia de Cordoba
conflictividad

193 1 ~i93 6 (Madrid: Ministerio de Agricultura, 1979), pp. in, 119 ff., 228-9. For the construction
strike, see S. Julia, Madrid 1991 -1934 De lafiesta popular a la lucha de closes (Madrid: Siglo XXI, 1984),
.

pp. 229-58 and summary in S. Julia, ‘Economic Crisis, Social Conflict and the Popular Front:
Madrid 1931-6’, in P Preston (ed.), Revolution and War in Spain 1991 -1939 (London: Methuen, 1984)
pp. 137-58; S. Payne, Spain’s First Democracy.The Second Republic 1991-1996 (Madison: University
of Wisconsin Press, 1993), p. 144. For Catalonia, see A. Balcells, ‘El socialismo en Cataluna hasta
la guerra civil’, in S. Julia (ed.), El socialismo en las nacionalidades y regiones, Anales de Historia, vol. 3
(Madrid: Fundacion Pablo Iglesias, 1988), p. 34.
86
These membership wars went back to 1916, when the CNT made an attempt to take over La
Naval, a UGT dockers’ union: F. Romero, Spain 1914-1918. Between War and Revolution (London:
Routledge, 1999), p. 38. Then, during the Primo years, dockers in Barcelona, led by an ex-
cenetista, had joined the UGT: Balcells, ‘El socialismo en Cataluna hasta la guerra civil’, p. 33. See

also Gabriel Jackson, The Spanish Republic and the Civil War 1991 -1999 (Princeton, n.j.: Princeton
University Press, 1965), p. 285; C. Ealham, ‘Anarchism and Illegality in Barcelona 1931-37’,
Contemporary European History (July 1995); UGT executive committee minutes, Dec. 1936/Jan.
1937: see Graham, Socialism and War, p. 271, n. 51.
 The challenge of mass political mobilisation 53

unions in direct opposition to the CNT’s. 87 In Madrid new tensions

emerged as the CNT moved into what had traditionally been a UGT
stronghold.
The CNT was particularly strong among the large numbers of un-
skilled building labourers who, attracted by the boom in the industry in
the 1 92 os, had flocked to Madrid to work on the big construction sites.
Their situation had worsened as the industry, faced with the effects of
recession, began to cut back. With no security or bargaining power, con-
frontational tactics and strike action were increasingly the only options
available to these unskilled workers.
88
The UGT leadership, wary of the
increasing number of CNT-backed strikes which it saw as a threat to its

control in the capital, sought to reimpose the 1931 -3 arbitration machin-

ery of the Jurados Mixtos which it had run. 89 But this in fact only served
to exacerbate the inter-union conflict. By spring 1936 the situation was
explosive.
In April, the UGT’s fears of loss of control seemed to be confirmed
when, on the fifth anniversary of the Republic, many of its members,
ignoring the opposition of both socialist and communist leaderships,
CNT. The UGT press accused
joined the general strike declared by the
theCNT, tellingly, of fomenting ‘rebellion against the state while the 5

CNT’s responded by renewing its criticism of socialist collaborationism.

5
Largo Caballero was the ‘old corrupt collaborationist socialist plugging ,

into the state as he had in the 1920s during the Primo dictatorship and
now advising republican ministers on how
5
to ‘handle strikes in return
for favoured status and renewed state privileges for the UGT. The inter-
union conflict intensified with the Madrid construction strike that began
in June 1936 and was still unresolved at the time of the military uprising.
On 11 July the UGT (via its newspaper Claridad) called for a return to
w ork,
7
‘for the sake of the Popular Front, for the consolidation of victory
5
over the bosses ,
which was met by another scathing attack from the
CNT. More significantly a sizable percentage of the UGT base in the
capital (24 per cent) was also hostile to a return to work and sided with
the CNT. 9°

8/
Macarro Vera, La utopia revoludonaria, esp. pp. 214-42; 293-305; 313—43.
88
Julia, Madrid 1931 -1934, pp. 147-220.
89 The Jurados Mixtos in Madrid had broken down under grass-roots pressure in 1933.
90
Julia, La izquierda del PSOE, p. 260; yFeudo de la UGT
o capital confederal? La ultima huelga de la
construccion en el Madrid de la Republica’, Historia Contemporanea, 6 (1991), 207—20. The political
ramifications of the Madrid building strike would also reverberate during the war in the UGT
leadership’s reluctance to put Madrid building workers under military discipline (for example
54 The Spanish Republic at war

This other kind of war - into which the July military coup and ensuing
civil war erupted - brings home the extent to which the material effects

of economic crisis had driven a wedge between the social constituencies

of the left. The cumulative effect of such clashes in urban and rural
Republican Spain created a legacy of distrust that eroded the political
space for a substantive agreement between the and the CNT UGT
during the war. 91
But any feasible consolidation of forces to the left of the republican-
socialist alliancehad probably already been blocked for good by 1934
when, both the CNT and the UGT effectively inca-
in different ways,
pacitated the Workers’ Alliance (Alianza Obrera). This was originally
an initiative of the Catalan BOG, which wanted to use it as a launch
pad for activism outside Catalonia. 92 But precisely because the Alliance
was envisaged as an inter-organisational structure, it could, had it been
realised, have offered a real opportunity for the left to consolidate and to
coordinate its action. At the very least a functioning Workers’ Alliance
could have provided the crucial national coordination lacking on the left
in the spring and early summer of 1936 when the military conspiracy
was escalating.
However, the CNT refused point blank to participate in a ‘political’
initiative (although some treintista unions in Catalonia did ally them-
selves). Largo Caballero initially pledged socialist support. But in the
course of 1934 it became clear that he was only interested in the Work-
ers’ Alliance as a means of extending the influence of the PSOE/UGT

in areas where, traditionally, the socialist movement was weak — and

primarily in Catalonia itself. But other components of the Workers’ Al-

liance in Catalonia were wary of this - especially the Socialist Union of

Catalonia (USC). This had split from the PSOE in 1923 precisely because
of the party’s lack of sympathy for nationalist issues. Nor had relations
between the two improved under the Republic. The fact that, in the
April 1931 municipal elections in Barcelona, the PSOE had allied with
the notoriously anti-Catalanist Radical Party is indicative of its animos-
ity towards the USC. Then in 1933 the PSOE’s heavy-handed central-
ism killed off the USC’s attempt to agree a measure of organisational

when building trenches and fortifications in autumn 1936) for fear of alienating them. See chapter
3 below.
91
For examples of this ongoing conflict in wartime, see Graham, Socialism and War pp. 64
,
(n. 45),

82 (n. 51).
92 Democracy,^. 195-6.
Payne, Spain’s First
 The challenge of mass political mobilisation 55

devolution for Catalan socialism - even though this was in keeping with
the spirit of the 1932 autonomy statute. 93

The tensions within the socialist movement were not the only ones
operating against the Workers’ Alliance. Nor would the many under-
lying antagonisms at rank-and-file level that we have already discussed

bode well for its future. But in the event, the Alliance never reached the
starting line as a national entity. It was obstructed by the dead weight
of Caballerista bureaucratism. In part this was a manifestation of the
UGT’s ‘existential’ fear that any initiative that did not emanate from
itself was a threat to its control. In part too, however, it was a practical

response to the well-grounded fear that the small, struggling PCE saw
the Alliance as a way of attracting the socialist rank and file. But, what-
ever the case, without the near nationwide articulating capacity of the
PSOE/UGT, was simply not viable.
the Workers’ Alliance
Inside Catalonia too the Alliance would be beset by problems. It ini-
tially integrated virtually all of the forces on the Catalan political left: the

BOC, the USC, Rabassaires, Andreu Nin’s tiny Communist Left Party
(Izquierda Comunista) and (by December 1933) the small official Catalan
section of the PSOE as well as its (more important) UGT section. But the
Alliance’s two core components were, first and foremost, the BOC and
then the USC. The BOC, like the USC, had been born of a schism with
a centralist parent party, in its case the PCE, which was, like the PSOE,
unsympathetic to the political-cultural claims of nationalism. 94 But the
consequences of the two schisms had been quite distinct. The USC was
a small minority that had split off from a well-established parent socialist
party with a considerable national presence in Spain. The BOC, how-
ever, had effectively taken over the cadres of the Catalan federation of
the PCE which, when they ceded from the PCE at the end of the 1920s, 95

93 Balcells, ‘El socialismo en Cataluna hasta la guerra civil’, pp. 35-6. In 1932 the PSOE accepted
the Catalan statute as part of the general package of republican reforms. But this did not make it
‘pro-statute’. Both the main leaders of the PSOE in 1930s Spain - Prieto and Largo Caballero -
were hostile.
94 F. Morrow, Revolution and Counter-Revolution (1938; London: New Park Publications,
1976), p. 44;
E. Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil: un ensayo de
interpretation’, in Socialismo y guerra civil, Anales de Historia, vol. 2 (Madrid: Fundacion Pablo
Iglesias, 1987), p. 306.
95 The BOC was formed in November 1930 from the fusion of the Catalan federation of the
PCE (FCCB) with Jordi Arquer’s tiny dissident Partit Comunista Catala (CC). The organisa-
tional independence of the BOC was confirmed in 1931 with the expulsion from the PCE of
Joaquin Maurin, who would become the BOC’s main political leader and strategist. See Durgan,
B.O.C. iggo-igg 6 F. Bonamusa, El Bloc Obrer i Camperol (iggo-igj2) (Barcelona: Curial, 1974);
;
 ,

56 The Spanish Republic at war

constitutedAnost of the minuscule ‘parent’ Communist Party’s entire

national membership. In terms of political projects, the USC’s social-
ism was closer to a pragmatic social democratic credo and immersed in
a strong Catalanism, whereas the BOC leaders played down the nation-
alist question and offered a socialist programme expressed in radical,
96
workerist language.
But in spite of this difference of political emphasis, the BOC and the
USC had been organisational rivals since the beginning of the Repub-
lic since they were competing for members amongt the same sectors of
Catalan-speaking (and frequently Catalanist) urban lower-middle-class
professionals and skilled workers. 97 The smaller USC, caught between
the influence of the more powerful BOC and the CNT, had little choice
but to exist in the political orbit of the Esquerra, with which it collab-
orated in the Generalitat. 98 Through this the USC achieved an institu-
tional presence in Catalonia that it never could have aspired to on the
basis of its own flimsy electoral showing.
The USC saw its participation in the Workers’ Alliance in 1934 as
a means of bolstering both itself as a political entity and a reforming
agenda in Catalonia. The Alliance had brought together the political
left in the region, including the UGT’s Catalanwhich section, access to
the USC
had had vetoed only a year previously by the UGT’s Madrid
executive. Now the USC hoped that the Workers’ Alliance could provide
a valuable complement to its presence in the Generalitat. In other words,
the Alliance would serve as an important source of popular pressure to
maintain the rhythm of the Esquerra’s social and political reform in
parliament. In this USC aspiration we can glimpse something of the
underlying rationale of the Popular Front alliance, as it would emerge

J. Estruch, Historia del P.C.E. vol. i (Barcelona: El Viejo Topo, 1978), p. 60; Ucelay da Cal,

y comunistas en Cataluna durante la guerra civil’, p. 308.

‘Socialistas
96 Bonamusa, El Bloc Obrer i Camperol iggo-igj2] Durgan, B.O.C. iggo-igg6.
97 For the BOC’s urban base in Catalonia, see A. Durgan, ‘Trotsky, the POUM
and the Spanish
Revolution’, in Journal of Trotsky Studies, 2 (1994), 59 and also appendices 4-8 of Durgan, B.O.C.
I93°~I93 6- For the BOC’s rural support from 1931 see Durgan, B.O.C iggo-igg6, pp. 140-1,

x 47~54
— although the Esquerra trounced both the BOC/POUM and the USC in rural
Catalonia before the civil war.
98 Balcells, ‘El socialismo en Cataluna hasta la guerra civil’, p. 34; Durgan, B.O.C. igjo-igj6.
Extrapolating back from figures for the BOC/POUM, the USC and the Catalan federation of the
PSOE in July 1936 (for which see B. Bolloten, The Spanish Civil War. Revolution and Counterrevolution
(Hemel Hempstead: Harvester/ Wheatsheaf, 1991) (hereafter SCW), p. 397) it is likely that the
USC would have had at least three times as many members as the official PSOE section, while
the BOC was more than twice as big as the USC. In the pre-war period the Catalan UGT was
also the smallest of the UGT’s regional sections - although estimates vary quite considerably:
see Bolloten, SCW, p. 862, n. 21.
 The challenge of mass political mobilisation 57

later, in 1936, in Spain as a whole. (It is worth noting too that the USC
would maintain its own electoral alliance with the Esquerra throughout
the entire Republican period 1931-6, irrespective of the rupture of the
republican-socialist coalition in 1933 across the rest of Spain.)
But it was precisely the USC’s relations with Catalan republicanism,
and especially its governmental collaboration, that constituted the bone
of contention for the BOC. It saw Workers’ Alliance membership as
incompatible with participation in ‘bourgeois government’. This would
see both the Rabassaires and the USC depart. Thereafter, the Catalan
Workers’ Alliance struggled in vain against political marginalisation. The
USC understood that reform in Catalonia (and even more in the rest of
Spain) would always require -determined right and a
in the face of a
fragmented left - a broad coalition integrating both progressive middle-
class republicanism and the organisations of the political left. But this
crucial insight had not yet found its moment to be heard.

THE ASTURIAN OCTOBER (1934)

The one place that the Workers’ Alliance did achieve its goal of unity -
albeit in desperate and ultimately disastrous circumstances — was in
Asturias. The historical collaboration of the UGT and the CNT in the
north was crucial here. The region’s heterogeneous mix of small centres
of decentred (or ‘ribbon’) industrial production, smallholder agriculture,
port activity (centring on Gijon) and artisan production had given rise
to a great deal of overlap between the UGT and the CNT. Both unions
represented those engaged in heavy industry and in artisan and service
sectors. This created the basis for practical collaboration, especially as
neither union was sufficiently dominant in the region to exclude the
other. Pre-existing political sympathies were also bolstered during the
early Republican years by the Northern CNT leaders’ increasing hos-
tility to radical anarchist currents. This animosity crystallised with the
abortive general strike of December 1933 in Gijon, thus opening the way
for the forging of the Asturian Workers’ Alliance."
Paradoxically, given the circumstances of its birth, the Workers’ Al-

liance in Asturias soon found itself at the centre of growing political radi-
calisation as a result of economic tensions in the north. The recession had
brought the chronic crisis of the Asturian coalmining industry to a head.
Attempts to shore up the industry by savings that directly impinged on

99 Radcliff, From Mobilisation War

to Civil , p. 290.
58 The Spanish Republic at war

the livelihoods or safety of the workforce had brought gravely deteriorat-

ing labour relations. Once again, the UGT leadership (both its national
executive and northern mining federation (SMA)) found it-
that of its

self in the familiar position of trying to control a radicalising rank and

100
file. June 1934 had already seen the frustration of rural labour spill
out in a bitter and bloody FNTT strike. The spark that lit the fuse in
October was provided by the CEDA. Its ministers entered the Madrid
cabinet to claim, in a national climate of increasing protest and polarisa-
tion, the three most sensitive portfolios: Agriculture, Labour and Justice.
The Spanish left, mindful of the German and Austrian precedents, saw
this asfascism arriving by legal means. Its response was a general strike
throughout Spain. This largely failed, as did the Esquerra’s own, mainly
symbolic, protest from the Generalitat. But Asturias exploded into armed
rebellion - though many of those who took part saw what they were do-
ing as a defence of Republican social reform. Again, the line between
direct action and parliamentary action was seen as permeable - and not
only by the newly politically mobilised.
The miners held out for two weeks. But their towns were bombed by
the Spanish airforce, their coastal towns shelled by the navy and their
valleys finally overrun by the Spanish army. A harsh and extensive re-
pression ensued throughout Asturias in which General Franco, as defacto
head of the war ministry, deployed both Moroccan troops and the
Foreign Legion. Constitutional guarantees were suspended across Spain.
The impact on the left was catastrophic. Thirty thousand people were
imprisoned and many of them tortured. 101 Others went into exile -
among them, Indalecio Prieto, the intelligent and energetic leader of
the PSOE’s parliamentary wing. Party and union premises were closed
and the left’s press silenced. Socialist town councils were overthrown,
civil servants of liberal or left opinions were discriminated against, and

everywhere employers and management took the opportunity to dismiss

trade unionists and left activists en masse.
The damage wrought to the PSOE/UGT by the repression severely
traumatised its leaders - none more so than the veterans of the UGT, for

100
A. Shubert, The Road to Revolution in Spain. The Coal Miners ofAsturias 1860—1934 (U rbana/ Chicago:
University of Illinois Press, 1987), pp. 141-67 and see also pp. 121-40.
101
Ibid., p. 163. For repression and prisoner numbers in Asturias specifically, see Radcliff, From
Mobilisation to Civil War, p. 300 (n. 46). Between 3,000 and 4,000 received prison sentences of

one year or more. The amnesty issue eclipsed all others in the Popular Front electoral campaign
in Asturias, during which the CNT’s northern federation officially called upon their members
to vote: A. Shubert, A Reinterpretation of the Spanish Popular Front: The Case of Asturias’,
in Alexander and Graham, The French and Spanish Popular Fronts, pp. 222, 220.
 The challenge of mass political mobilisation 59

whom the socialist movement’s greatness was inscribed in its material

patrimony. In fact, the UGT’s Madrid-based executive had played no
part in the Asturian events, nor had they been directly involved in the
planning decisions taken by the Asturian Workers’ Alliance. Indeed,
given the total failure of the general strike in the Caballerista stronghold
of Madrid, some inside the socialist movement would soon be asking if the
noisy, self-proclaimed ‘socialist left’ had been actively involved anywhere.
The PSOE executive was, nevertheless, called legally to account for
the events. During his trial Largo Caballero, who was fearful of the
punitive confiscation of the PSOE/UGT’s assets, made his famous denial
of responsibility for the Asturian movement. This would offer the Spanish
Communist Party the opportunity it needed to come in from the political
cold. Even though its role in October had been marginal, the PCE eagerly
assumed responsibility for the events. 102 Initially the party projected it
as a symbol of proletarian unity in the drive to achieve a communist-led
‘united front’ of workers. But we can also see in the aftermath of October
the seeds of the PCE’s subsequent inter-class strategy of Popular Front:
for example in the party’s incorporation of progressive republicans in the
work of the PSOE’s prisoners’ aid committee. It was also after October
that the PCE’s Association of Anti-fascist Women (AMA) became active
103
in recruiting across the boundaries of social and economic class.
Until the shift to a Popular Front was endorsed in the summer of
1935 by the VII Congress of the Communist International (Comintern),
the PCE was still formally committed to the policy of a purely worker
front (frente unico). However, as we have just seen, by late 1934 the na-
tional context in Spain, as elsewhere in Europe, was in practice pushing
communist parties in the direction of what would soon become the new
policy. This is scarcely surprising, of course. The growth on the ground
of what in Spain was called ‘the climate of unity’ was everywhere a cru-
cial formative influence on the emergent Comintern policy of Popular

Front. 104
Prior to the end of 1934 such inroads as the very small Spanish Com-
munist Party had been able to make were the result of its being able to
appeal to a more radical mood than the PSOE felt able to: for example in
102
J. Diaz, 2 June 1935, the Monumental Cinema, Madrid, Tres ahos de lucha (3 vols., Barcelona:
Laia, 1978), vol. 1, pp. 42-3.
103
H. Graham, ‘Women and Social Change’, in Graham and Labanyi, Spanish Cultural Studies ,

pp. 108-9.
104
Alexander and Graham, The French and Spanish Popular Fronts, one of whose unifying themes is
that the Popular Front in Europe was a sociological phenomenon built from below as much as
‘directed from above’.
6o The Spanish Republic at war

Vizcaya and Asturias in the 1920s, as we have discussed in the introduc-

tion above. (It was no coincidence that the PCE union breakthrough in
Asturias happened at the same time as the UGT’s collaboration with the
Primo became an important PCE base when
dictatorship. 105 ) Seville too
the conclusion of Primo ’s extravagant works programmes in the 1920s
left a mass of unemployed construction workers to whom the UGT’s
106
traditional arbitration-based unionism had neither use nor appeal. In
the early 1930s in Seville the PCE was able to benefit from the CNT’s
refusal to engage politically with the Republic. Indeed, in the late 1920s
the PCE had acquired cadres and leaders (including future secretary
general Jose Diaz) from a number of former CNT unions in Seville. The
PCE also continued to benefit in Seville from the PSOE/UGT’s caution
and governmental compromise. This was the time when the PGE’s then
line of social fascism or ‘class against class’ appealed precisely because
it fitted pre-existing tensions and
conflicts between different working-
class constituencies which were generated or exacerbated by economic
depression. 107 As we have seen, the PSOE/UGT attempted to defend
itself from PCE (and CNT) encroachment with its time-honoured tactic
of combining a radical discourse with a moderate practice. But this was
less credible now that the PSOE was itself within the circle of political
power, rather than an outsider force as it had been through the years of
the Restoration monarchy.
The PCE’s continuing marginalisation in the 1920s and early 1930s
has always been attributed to the Comintern’s heavy-handed imposi-
tion of inappropriate and sectarian policies. This and
certainly occurred
had some devastating effects. For example, as already mentioned, wasit

largely responsible for the PCE losing the bulk of its membership when,
on the eve of the Second Republic, the Catalan federation walked out of
the party. But even apart from the Comintern effect, it is difficult to see
what political space (or audience) on the left could have been available
to the PCE prior to
1931. Its opportunity came precisely because of the
politicalconsequences of the Republic, the PSOE/UGT’s involvement
in it and the reluctance of both republicans and socialists to face up
to the challenge of mass political mobilisation in a society like Spain’s
with quite low levels of general and political education. Consciously or
otherwise, this was the challenge to which PCE secretary general Jose

105
The anarchist-communist miners’ union, SUM, was formed. But with the birth of the Republic
in 1931 the UGT’s Asturian miner’s union (SMA) regained prominence.
106
Preston, CSCW, p. 29; Macarro Vera, La utopia revolucionaria, passim.
107
Cf. the German case in Graham and Preston (eds.), The Popular Front in Europe, pp. 5-6, 21-3.
 The challenge of mass political mobilisation 61

Diaz was responding when in Madrid, in June 1935, he launched the

108
campaign for a ‘broad Popular Front’.

THE PGE AND THE POPULAR FRONT

The PCE’s great strength by 1935 derived from the fact that it was striving
for an would mobilise ‘the republican
inter-class political alliance that
nation’. This meant both middle-class sectors and the existing organised
labour movement, but it also meant reaching out to vast numbers among
the middling classes and workers who were as yet unaffiliated to any
political party or organisation. Of course, at one level this meant that the
PCE was simply applying the Comintern line. And certainly, given that
the party leadership was the ‘disciplined’ product of an earlier internal
purge in 1932 and also devoid of any notably independent spirits or
significant political theorists, it could not be argued that its adherence
to the Popular Front consciously went beyond an obedience to iron
party discipline and a very general belief in the political superiority
109
of the Communist There was also the additional and highly
Party.
problem of the personally overbearing Vittorio Codovilla, the
specific
Comintern delegate sent to Spain in 1932 to advise the party. 110 But,
at another level, these issues are irrelevant because Popular Front as
espoused by the Comintern and the PCE happened to be responding to
an underlying structural and conjunctural necessity of the Spanish left
(broadly understood) in the 1930s. First, there was the need to mobilise
the population to help build the Republican state and second, the need
to bring together the fragments of the counter-hegemonic project which
had split into its component parts as a result of the stresses of the early
Republican years of 193 1 -3. This cohesion was the essential precondition
of the left’s being minimally capable of mounting a successful challenge
to the right - at the level of state cadres and political process as well
as national symbols and values. 111 Nevertheless, the PCE was putting its
shoulder to the wheel here in the inevitable hope that it could also nourish
its own party by attracting sectors of the PSOE/UGT base. This would

108
Diaz, Tres ahos delucha, vol. i, pp. 35-64.
109
M. Azcarate, Derrotasy esperanzas. La Republica, la guerra civil y la resistencia (Barcelona: Tusquets
Editores, 1994), p. 209. Dolores Ibarruri and Antonio Mije, among others, were obliged to
perform 1932 in order to remain.
self-criticisms in
110
For more on Codovilla see chapter 3 below.
111
H. Graham, ‘Community, Nation and State in Republican Spain 1931-1938’, in C.
Mar-Molinero and A. Smith (eds.), Nationalism and the Nation in the Iberian Peninsula (Oxford:
Berg, 1996), pp. 133-47.
62 The Spanish Republic at war

produce increasing hostility and an accelerating organisational rivalry in

1935 between the communists and Largo Caballero’s wing of the socialist
movement. This was especially so because it was on the UGT that the
PCE would concentrate its campaign.

PARLIAMENTARY SOCIALISM I935~6 FROM !

FRAGMENTATION TO ‘UNITY’ - THE ROAD

TO THE POPULAR FRONT
Meanwhile, within the broader socialist movement, the debacle of the
Asturian October and the tremendous cost to the left of the ensuing
repression would rapidly conscript minds and hearts into action on two
issues. First, the reconstitution of the republican-socialist alliance: this
was crucial in the short term to win an election and release (via a political
amnesty) the thousands of prisoners taken after October. In the medium
term, it would ensure a return to a programme of progressive social and

economic reform by parliamentary means. Second, it was imperative

that the socialist left should not be allowed to disrupt this process —
especially not as the other conclusion to be drawn from ‘October’ was
that the party ‘left’ was in no meaningful sense revolutionary. 112 The
view growing apace in the parliamentary socialist wing of the PSOE
associated with Indalecio Prieto (but support for which also extended
into sectors of the UGT and socialist youth) was that the ‘left’ was both
ideologically bankrupt and organisationally disruptive. It had shown an
abysmal failure of leadership initiative in October 1934. Nor, in spite
of much radical rhetoric thereafter, had it produced a tangible policy
line of its own, much less one capable of building the power structures
necessary to orchestrate the anti-capitalist revolution whose imminence
it proclaimed. 113
For Largo and the UGT veterans, marooned without a policy since
their departure from government in 1933, a revolutionary discourse
seemed at some level to offer a ‘solution’ - the possibility of unblock-
ing rightist obstruction in government while also preventing any other
politicalgroup overtaking them on the left. To speak of revolution, more-
over, entailedno risk to the socialist organisation. In fact, the UGT
leaders would be proved wrong on all three counts. By 1935, the socialist

112
Graham, Socialism and War pp. 17-18.
,

113
The consequences of this assessment for the civil war period are analysed in depth in Graham,
Socialism and War. A resume in Graham, ‘The Eclipse of the Socialist Left 1934-1937 in Lannon
and Preston, Elites and Power in Twentieth- Century Spain pp. 127-51.
,
 The challenge of mass political mobilisation 63

left itself - never a coherent - was itself breaking up under the

force
competitive onslaught from the PCE. In the union and party, but above
all in the socialist youth, the attraction exerted by the Communist Party

would increase through the year.

Prieto was, meanwhile, doing his utmost to resurrect the republican-
socialist alliance. Notwithstanding his own ambivalent responses during
the Asturian episode, the parliamentary PSOE had seen it come like
an avalanche they were powerless to prevent and one whose wake had
set back the fragile beginnings of republican-socialist negotiation

(between Azana and Prieto) made in the course of 1934. Paradoxically,

however, October 1934 would assist the PSOE here in one crucial way.
The hysterical campaign launched by the rightist press against Manuel
Azana, as the sinister author of a bloody revolution, would pluck him
from the obscurity of the republicans’ ‘old-world politics’ to what one
might almost term modern political stardom. 114 After his release from
prison, Azana undertook across the spring and autumn of 1935 a mas-
sive publicity campaign to convince Spaniards, and particularly Spanish
workers, of the urgent need to rebuild an electoral agreement on the left.
Between May and October he toured major urban centres delivering
his discursos en campo abierto or open-air speeches. In Valencia he spoke
,

before more than 100,000 people; in Bilbao the crowd was even bigger.
On 15 October the campaign reached a crescendo at Comillas on the
outskirts of Madrid when nearly half a million people turned up to hear
Azana make his appeal for unity on the left. Nor, given the prevailing
political environment, was it a small matter for them to have come. In the
words of the British journalist Henry Buckley, who witnessed the event:

This meeting had not been widely advertised. It was frowned on by the author-

ities and in some Guard turned back convoys of trucks carrying

cases the Civil
spectators. All vehicles bringing people from afar were stopped some miles out-
side Madrid, thus causing endless confusion and forcing weary men and women
to trudge a long distance after a tiring ride. Admission was by payment. The
front seats cost twelve shillings and sixpence and the cheaper ones ten shillings
and half a crown. Standing room at the back cost sixpence. No one was forced
to go to that meeting. Presence there, in fact, was much more likely to bring the
displeasure of employer or landlord From the furthest points of Spain there
. . .

were groups who had travelled in some cases six hundred miles in rainy cold
weather in open motor lorries. 115

114
P. Preston, ‘The Creation of the Popular Front in Spain’, in Graham and Preston, The Popular
Front in Europe, pp. 97-8.
115
H. Buckley, Life and Death of the Spanish Republic (London: Hamish Hamilton, 1940), pp. 182-3.
64 The Spanish Republic at war

An icon of the larger political process supervening, Comillas resonates

still with the huge potential of that moment.
Prieto was able tremendous popular response to the discursos
to use the
en campo abierto to socialist left back to supporting a renewed
bring the
electoral alliance between the republicans and the PSOE/UGT. Largo
Caballero was not easily persuaded. But he could not gainsay - and
indeed was deeply impressed by - the strength of popular feeling, which,
he realised, also infused the socialist grass roots. And here October 1934
would be the single most important factor in worker mobilisation behind
the alliance whose electoral programme pledged an amnesty for all the
political prisoners taken in the aftermath of the Asturian rising.

The PCE also worked hard to persuade Largo of the folly of insisting on
a uniquely proletarian front - even sending in the veteran French union
leader,Jacques Duclos, to argue the case. However, the communists
were not the authors of Largo’s ‘conversion’: it is clear that a veteran of
Largo’s standing could, in the end, do the political arithmetic for himself.
Moreover, Largo - ever the wily veteran - also saw including the PCE as
a way of controlling its political pretensions by tying it (albeit indirectly)
into political responsibility for delivery of the electoral programme. Largo
therefore made his own participation in the electoral alliance conditional
on the PCE’s inclusion. Prieto was not keen, in spite of the PCE’s new
Popular Front mode. Azana and the republicans were even less happy.
But in the end it was agreed that the PSOE/UGT would represent
on the National Popular Front Committee all those forces to its left
6
included in the electoral pact." This made for a greater display of unity,
but it also meant that the PSOE would have to pay for this in hard
currency. For the allocation of the left’s total number of candidates was
taken not proportionally but from the PSOE/UGT’s own share. But
the biggest winners in this allocation of candidates were the republicans,
who were significantly over-represented in relation to their numerical
strength while the PSOE/UGT was under-represented. 117

THE POPULAR FRONT

As far as republican over-representation was concerned, however, this
was deliberate. Prieto understood, in a way that Largo Caballero did not,
that a purely worker front would simply not be strong enough, taking

signatories were: the PSOE, UGT, PCE, POUM, Izquierda Republicana (Azana), Union
116
The
Republicana (Martinez Barrio), Esquerra and Partido Sindicalista (Angel Pestana).
117
S. Julia, Origenes del Frente Popular en Espana 1934-1936 (Madrid: Siglo XXI, 1979), p. 145.
 The challenge of mass political mobilisation 65

the country as a whole. Given Spain’s social and economic profile, the
urban working class - even
if backed up by very newly incorporated

rural sectors - simply did not represent sufficiently broad strata to carry
the weight of structural political change alone. A liberal centre-left
political alliance like the Popular Front meant a guarantee of consti-
tutional norms and social reform via parliamentary legislation. But it

could not, by definition, be aprogramme of socialist reform. It could still,

nevertheless, suppose radical structural change - such as Prieto would
himself outline in May 1936 in his famous Cuenca speech, ‘La conquista
8
interior de Espana’." Of course, what this implicitly meant was that
Spanish workers were being asked to sacrifice some of their aspirations
in order to keep on board middling constituencies who might otherwise
be fished by the political right. But it had to be that or nothing. For
only an inter-class front like Popular Front could unite enough of the
-
fragmented ‘lefts’ defined as all those middling and worker constituen-
cies outside the power structures of old-regime Spain. Only together,
as a counter-hegemonic coalition, could they withstand the force of the
political elites whose power now linked up dangerously with that of
the GEDA’s mobilised popular conservative opinion. By 1936, then, the
left in the rest of Spain had arrived at the political position occupied

consistently since 1931 by a core of the Catalan left - republican and

socialist.

In Catalonia, both the Esquerra and the USC

had assumed the need
for a broad coalition in 1931 and had never abandoned it. One major
difference with the Popular Front in the rest of Spain, however, would be
that in Catalonia the PCE itself remained as marginal in 1935-6 as it had
been everywhere else prior to then. But if the PCE itself did not benefit in
Catalonia from the fall-out after October 1934, then specifically Catalan
communist and socialist formations would benefit. This did not happen
at the expense of the PSOE, clearly, but instead at the expense of the, until
then, politically unassailable Esquerra. Like the PSOE elsewhere, the
Esquerra had suffered heavy repression in the wake of October precisely
because of its organisational and institutional pre-eminence." 9

118
‘The interior conquest of Spain’. The allusion here was, evidently, to the need for domestic
reform in ‘post-imperial’ Spain. I. Prieto, Discursos fundamentals (Madrid: Ediciones Turner,
1
975)5 PP- 255-73.
119
It is worth insisting comparison between the PSOE and the Esquerra given the similarity
on this
of their semi-eclipse after 18 July 1936.The two parties suffered a similar process of internal
fracturing in spring and summer 1936 and both lost sustantial parts of their youth movements.
On this process in the PSOE, see Graham, Socialism and War, on the Esquerra there is a resume
in Ucelay da Cal, ‘Socialistas
y comunistas en Cataluna durante la guerra civil’, pp. 309-11.
 ,

66 The Spanish Republic at war

The Catalah left was influenced, as its Spanish counterparts were,

by the intensifying ‘climate of unity’ that had been growing since the
events of October 1934. Negotiations began in the spring of 1935 to try
to agree a basis for the organisational unification of the various socialist
and communist parties in the region. Partly this was stimulated by the
new opportunity to compete with the Esquerra. But the negotiations
were also a response (emotional as well as intellectual) to the now clearly
perceived need to overcome the structural fragmentation of pro-reform
forces — an awareness was infused with an even greater sense of
that
urgency by the international situation. Equally, the discussions between
the (socialist) Second International and (communist) Third International
as well as the latter’s new Popular Front policy increased the sense of
propitiousness.
One singular feature of this process in Catalonia, however, was the
remarkable fluidity of the potential alliances within the political (i.e.
120
socialist and communist) left. (It should be borne in mind too that a
good number of the leaders of the political left in Catalonia had also
begun in the CNT.) It even seemed possible that the proposed merger of
socialists and communists might include both the BOC and the marginal

Catalan section of the PCE (PCC) - something that, had it come to

fruition, would have gone some considerable way to erasing the ‘family’
schism of the late 1920s. But the Catalan communists insisted on the
exclusion from any putative merger of Andreu Nin’s Communist Left
because of his links with Trotsky.
At the core of the unity talks in Catalonia were the biggest groups,
the BOC and the USC. Although they had been rivals since 1931, a
merger would have produced a new player well on its way to holding
its own as a partner in alliance with the Esquerra. But the outcome of

the Workers’ Alliance in 1934 did not augur well. And indeed, in the
end, the attempt would fail because of basic disagreements between the
two leaders over the political objectives and the geographical range of
any new party. While the USC’s Joan Comorera was as enthusiastic a
supporter of a broad Popular Front- type alliance in 1935-6 as he had
been in 1931, Joaquin Maurin of the BOC wanted this commitment to
be strictly time limited to the election - that is, as a tactic to secure a
political amnesty. And while the USC saw its desired role very clearly
in terms ofprotagonising Catalan politics, Maurin wanted to play down
120
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, pp. 298-9, 305—7
308, 318—19 and Balcells, Trabajo industrial y organizacion obrera en la Cataluna contemporanea 1900—
1936, p. 143.
 The challenge of mass political mobilisation 67

the BOC’s Catalanism and (again as in 1934) to launch the new party
on a political career throughout Spain.
This would see the BOG ally instead with Nin’s Communist Left to
form the POUM in September 1935 — in spite of Trotsky’s own objec-
tions, vehemently made to Nin. But although Nin’s group was politically
sympathetic to Maurin’s radical political and pan-Spanish aims and even
keener to minimise the new party’s Catalanism, it could add little by way
of an extra-Catalan base on account of its own minuscule size. It had
members in Madrid and Valencia, but so few that they counted as a grou-
puscule rather than a political party.
121
Many BOC members were un-
easy about the merger. But most stayed in the new party because they felt
that Maurin’s political intelligence and strategic skill offered a sufficient
guarantee for the future. A minority, however, left the BOC in protest at
the merger and joined the Catalan Communists (PCC) instead. 122 The
animosity between those who stayed and those who left was consider-
able, and this would feed into the Catalan left’s war of political position
after 18 July 1936. Nevertheless, the merger did make the POUM by the
end of 1935 the largest socialist party in Catalonia by some margin. 123
Its core strength, however, derived from the Catalan - and Catalanist -

members of the BOC in the region’s other urban centres (for example
Lleida, Girona and Tarragona). This might have been its strength. But
it would in fact become a weakness during the war. In Maurin’s absence

(he was caught in the rebel zone) the contradictions would intensify be-
tween the POUM’s own base — rooted in the Catalanist lower-middle
classes - and the leadership’s quasi anti-Popular Frontist strategy. 124

THE EMERGENCE OF THE PSUG

The failure to agree terms for an alliance with the BOC would send the
USC in search of other potential allies to bolster its political position.
Comorera, for all that he was as moderate a social democrat as the USC
contained, was of course influenced by the general sense of urgency on
the left over the need to achieve strength through unity. But we should

250-300 strong, while the BOC had several thousand members: V Alba,
121
Nin’s group was around
Dos revolutionaries: Joaquin Maurin, Andreu Min (Madrid: Seminarios y Ediciones, 1975), p. 389. See
also P. Pages, El movimiento trotskista en Espana (1930-1935) (Barcelona: Ediciones Peninsula, 1975).
122
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 311.
123
Balcells, Trabajo industrial y organization obrera en la Cataluna contemporanea 1900-1936, p. 146. On
the eve of the war the POUM had an estimated 6,000 members, though by December 1936 it

claimed 30,000, F. Morrow, Revolution and Counter-Revolution in Spain (1938; London: New Park
Publications, 1976), p. 45; Bolloten, SCW, p. 405.
124
See chapter 4 below.
68 The Spanish Republic at war

also remember Comorera would agree to en route to

that everything
the creation of a single socialistand communist party in Catalonia (the
PSUC, formed in Barcelona in the summer of 1936) was driven by the
USC’s basic goal ever since it had properly begun to operate as a political
party in 193 1 -2 : that it shouldform the ruling nucleus ofa strong Catalanist social
125
democratic party which would eventually displace the republican Esquerra . The
PSUC offered a form of socialist unity in Catalonia that would allow the
USC to lead. The larger POUM was excluded. At the same time, the
merger in the PSUC of the various other fragments of the socialist and
communist left in Catalonia would also provide crucial ‘ballast’ which
would allow the new party to break out of the Esquerra’s orbit and
hold its own politically. The exclusion of the POUM also seemed to
remove the threat of Catalan socialist unity becoming subordinate to a
nationwide political strategy, such as the POUM had wanted and which,
if successful, would once again have relegated the USC to a subordinate

role - this time not to the PSOE, but to the POUM itself. Nor would
the PSUC be subordinate to the PCE, for it was envisaged not as a
section of the PCE, but as a separate and independent party (and as
such the PSUC would be accepted by the Comintern 126 ). PSUC unity
permitted the reinforcement of the Catalan left against the Esquerra
(to calm memories of 193 1-3) but without breaking the Popular Front, as
the POUM (and indeed the CNT) argued for, but to which the USC, as
moderate social democrats, were absolutely opposed.
The prospect of PSUC unity also, crucially, freed the USC from the
irksome influence of the PSOE. The intransigence of the highly central-
ist PSOE, which had resisted any concession to the Catalanism of the
USC since its emergence in the 1920s, is a crucial factor in explaining
the tactical preparedness of Comorera to do business with the Catalan
Communists and the Comintern. In this respect it is highly significant
that the decision taken in July 1935 to propose Comintern affiliation to

the next USC

congress was made by the USC executive without any
prior talks having taken place with their Catalan Communist (PCC)
counterparts. Indeed, in summer 1935 no apparatus yet existed for li-

aison of any kind between the USC and PCC. 127 Nor did the fact that
freedom from the anti-Catalanist PSOE came at the price of Comintern

125
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 305. Moreover,
that Comorera did precisely this is evident in Comintern wartime criticisms of him. P. Togliatti,
Escritos sobre la guerra de Espana (Barcelona: Critica, 1980), pp. 247—8.
126 127
See chapter 4 below. Balcells, ‘El socialismo en Cataluna hasta la guerra civil’, p. 38.
 The challenge of mass political mobilisation 69

membership seem particularly risky. After all, as part of the new Popular
Front line the Comintern had made its own a discourse that was highly
sympathetic to nationalist claims. No doubt the Comintern’s delegate to
Catalonia, the Hungarian Erno Gero, who was involved in December
1935 in discussions to bring the USC and the PCC together, would have
missed no opportunity to stress this. For the USC, as a party of the as-
piring lower-middle classes, the image of the Soviet Union as an agent
of social modernisation is also important in appreciating the appeal of
128
the new Not least, the rank-and-file enthusiasm for unity on
policy.
the which would reach a crescendo after the Popular Front election
left,

victory of February 1936, meant that the PSUC merger was an option
that enjoyed significant popular support. In January 1936, the PCC would
join in the unification discussions between the USC and the tiny Partit
I2
Catala Proletari (PCP). 9
This pre-history of the PSUC resolves the conundrum of Como-
rera’s apparently contradictory attitudes both to unification itself and to
130
Comintern affiliation. In an ideal world Comorera would have pre-
ferred a Catalan social democratic party independent of ties with either
the Spanish Socialist or Communist Parties, let alone with their respec-
tive Internationals. But Comorera viewed everything
in the real world,
strategically. The PSUC unification process was, anyway, gathering mo-

mentum across the spring and summer of 1936 as the constituent parties
held meetings to ratify the unification. But the final, joint unification

congress was only scheduled for late August, and the issue of the PSUC’s
relationship to the Comintern remained in a very real sense ‘in the
air’ — even if were being made by the USC. 131 Como-
positive noises
rera himself was probably less enthusiastic about Comintern affiliation
than his favourable public pronouncements in March and June 1936
indicate. At the same time, however, he would certainly have felt far
better disposed towards the Popular Frontist Catalan Communists and
their International than towards the Catalan Socialists. After all, it was

128
For more on this see the discussion in chapter 3 below.
129
An offshoot of the radical nationalist Estat Catala, thePCP (founded 1932) was of almost
no importance in itself. But by the end of 1934 it had come to control the powerful
political
Barcelona shopworkers’ association, the CADCI. Ucelay da Cal, ‘Socialistas y comunistas en
Cataluna durante la guerra civil’, p. 309.
130
A resume of these is in Bolloten, SCW, pp. 397-8 (inc. nn. 10-13).
131
From autumn 1935 through to spring 1936, USC leaders opposed to the new policy were
displaced: Balcells, Trabajo industrial y organizacion obrera en la Cataluna contemporanea igoo-igj6,
p. 153; ‘El socialismo en Cataluna hasta la guerra civil’, p. 38.
70 The Spanish Republic at war

the Caballerista-led Catalan PSOE federation in Barcelona which in

May 1936 sealed a long tradition of obstructionism by obliging Como-
rera to renounce USC support of the republicans in the Catalan gov-
ernment (and thus resign from his ministerial post in the Generalitat)
as the condition of PSOE entry to the PSUC unification discussions. 132
Committed as Comorera was to supporting the Catalan government, he
had to accept the Caballeristas’ ultimatum since, for his PSUC strategy
to work, he had to take the Catalan PSOE with him into the new party.
The Madrid PSOE would try to block this because of the Comintern
connection. But this time, in spring 1936, unlike in 1933, the Catalan fed-
eration of the PSOE and, more importantly, the Catalan (which UGT
theUSC’s own union, the UGSOC, had already joined) would break
away and back the PSUC. 133
In Catalonia as elsewhere in Spain, the Popular Front elections
of February 1936 were stoking popular enthusiasm for political unity
on the left. In May 1936 the POUM’s own trade union, the FOUS,
with approximately 6,000 members, also agreed to join the Catalan
UGT. 134 The latter was also strengthened by the influx of unions in
Sabadell and Manresa that brought around 14,000 and 3,300 mem-
bers apiece. Their entry to the Catalan UGT
was also highly symbolic.
Sabadell had been a key part of the CNT’s support base in Catalonia
over the previous half-century. Manresa too was one of the treintista bases
that found a home in the CNT. Of the approximately forty thousand
members whose unions had left the CNT in 1931 and 1932, nearly half
would end up in the UGT. 135 In Catalonia too, the youth movements (of
both nationalism and the political left) were being increasingly drawn to
the vibrant, activist politics of the Popular Front. 136 The sphere of influ-
ence for what would soon be the PSUC
was growing along with these
currents. But the nascent party’s golden opportunity to compete with the
Esquerra would only come a few months later when the military coup
ofJuly completely transformed the political terrain, by breaking the link
between electoral strength /institutional authority on the one hand and
political power on the other. 137

132
Balcells, ‘El socialismoen Cataluna hasta la guerra civil’, p. 39.
133
For more on the PSUC and its wartime membership, see chapter 4 below.
134 Bolloten, SC W, pp. 407-8. The extremely optimistic (not to say unrealistic) intention was to
radicalise the UGT in preparation for merger with the CNT.
135
Figures from Balcells, Trabajo industrialy organizacion obrera en la Cataluna contemporanea igoo-igg 6 ,

pp. 156, 158-9.

136
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, pp. 310, 31 1.
137
See chapter 4 below.
 1

The challenge of mass political mobilisation 7

THE ‘hot’ SPRING OF 1936

In the rest of Spain too the Popular Front elections had seen an explosion
of popular political energy that showed no signs of abating. Everywhere
the left’s electoral victory had unleashed demands for the rapid rein-
statement of workers who had been
sacked after October 1934. There
were calls too for urgent practical measures against unemployment and
for the acceleration of social reform. In Asturias, local Popular Front
committees - some newly created after the elections - showed their po-
138
tential for becoming independent vehicles to press workers’ demands.
The CNT was also vehemendy critical of the restrictions on the political
amnesty passed by the Popular Front on the morrow of its victory. The
amnesty covered those convicted for political motives in the aftermath of
October 1934. But it excluded those the CNT called ‘social prisoners’ - a
category that included many CNT and FAI members detained under
the anti-vagrancy legislation or in the insurrectionary attempts in the
early years of the Republic (193 1-2). This popular political pressure -
in the streets, in public spaces, in town halls - ran counter to the prefer-
ence of both republicans and the PSOE to ‘contain’ the Popular Front
within parliamentary channels and to assert the authority of central
government. 139 The PCE for its part sought to maintain extra-
parliamentary popular political mobilisation wherever it could. But, by
the same token, its preference was for legal forms of protest. 140 But as
we have noted before, many ordinary Spanish workers saw no necessary
incompatibility between parliamentary action and direct action of more
radical means of bringing change. Those who had joined the
hue as a
FNTT in the hope of state-led reform did not necessarily see this as
debarring them from taking part in land seizures in March 1936.

138
Shubert, ‘A Reinterpretation of the Spanish Popular Front: The Case of Asturias’, in Alexander
and Graham, The French and Spanish Popular Fronts pp. 213-25; Radcliff, From Mobilisation to
,

Civil War pp. 301-3 (on the ‘two Republics’ and the conflict of symbols between 14 April and
,

1 May celebrations in 1936 Gijon). The continued existence of local Popular Front Committees

in Asturias after the February 1936 elections would provide a basis for local organisation after
the 18 July coup.
139
Prieto himself commented that the Popular Front as a mass movement had no reason to exist
once the elections were over:J. Vidarte, Todosfuimos culpables (2 vols, Barcelona: Grijalbo, 1978),
S.

vol. 1, p. 99. On liberal republican anxieties and obsession with ‘the neutrality of the street’, see
S. Julia, Manuel Azaha. Una biografia politico (Madrid: Alianza, 1990), pp. 459-69; Casanova, De la
cable alfrente , p. 145 (n. 12); H. Graham, ‘Spain 1936. Resistance and Revolution: The Flaws in
and A. McElligott (eds.) Opposing Fascism. Community, Authority and Resistance
the Front’, in T. Kirk
in Cambridge University Press, 1999), p. 64.
Europe (Cambridge:
140
Cf. the similar moderating influence exerted by the PCF during the June strikes in France:
D. A. L. Levy, ‘The French Popular Front 1936-37’, in Graham and Preston, The Popular Front
in Europe, p. 69.
72 The Spanish Republic at war

For the republicans in government after February, the fear of such pop-
ular action was out of all proportion to its extent and, moreover, to what
141
it signified. The republicans certainly feared itmore than they did the
military conspiracies against the Republic that had been rumoured ever
since the right’s defeat at the polls. These anxieties ruled out any pos-
sibility of Azana’s government addressing the still pending issue of the
demilitarisation of public order. In this crucial question, as in so many
others, the republicans would, once again, reap the worst of all worlds.
In 193 1 they had inflamed the feelings of many officers by pursuing with
vehement rhetoric the ‘responsibilities’ campaign against those alleged
to have committed crimes or to have been corrupt or incompetent under
the old regime. 142 But in reality few officers had been called to account.
So the republicans had alienated a powerful and dangerous sector in the
officer class, yet in practice failed to ‘disarm’ them. 143 The July rebellion
would be triggered by the military declaring a state of war - something
which (under the terms of the Republic’s own 1933 public order Act)
still fell within their competence. 144 Had the republicans demilitarised
public order, not only would it have prevented the conspirators from
using legality as a cloak for their actions, but it would also have served
as an important consolidation of civilian constitutional authority on the
symbolic plane. For this too was an important component in the consti-

tutional ‘education’ of the officer corps. In sum, the failure to demilitarise

public order stands as the example, par excellence , of the lethal nature of
Republican idealism.
To make matters worse, the government that issued from the February
1936 electoral victory was not in fact a ‘Popular Front’ government at all,
but an entirely republican one. Largo Caballero had made his support of
the electoral pact conditional on the assurance that the PSOE would not
participate in government thereafter. The Caballeristas — remembering
the experience of 1931—3 - argued instead for the formation of an ex-
clusively workers’ alliance. Simultaneously, they propounded the view —
largely constructed by Largo’s political lieutenant, the party theorist Luis

141
The republicans inherited the ‘state of alarm’ from the caretaker administration in charge
before the February Popular Front elections and they maintained it virtually uninterrupted
until the military coup in July 1936. M. Ballbe, Orden publico y militarismo en la Espana constitucional

(i8i2-ig8j) (Madrid: Alianza, 1983), p. 387.

142
Preston, Franco. A Biography (London: HarperCollins, 1993), pp. 76-7; C. P. Boyd,
‘“Responsibilities” and the Second Republic, 1931-6’ in M. Blinkhorn (ed.), Spain in Conflict
*93 * ~*939 (London: Sage, 1986), pp. 14-39.
143 Ballbe, Orden publico y militarismo en la Espana constitucional, pp. 391, 393.
144 Ibid.,
pp. 13, 336.
 The challenge of mass political mobilisation 73

Araquistain, that the republicans should be allowed to ‘exhaust’ them-

selves in government and then ‘power’ would drop like a ripe fruit into
the hands of the socialists, who would then fulfil their historic mission
and inherit the state. But this puerile ‘theory’ completely ignored the
collateral effects of republican erosion: most particularly, that the weak
all-republican cabinet was entirely incapable of dealing with the over-
riding challenge facing the Republic in the spring of 1936 - the military
conspiracy escalating in the garrisons. Indeed, given the shape of repub-
lican demons, to many among them this would not even appear to be
the overriding danger.
The parliamentary PSOE and Indalecio Prieto in particular (who had
an impressive personal intelligence network) were far less sanguine about
the situation in the garrisons. In May 1936, in an attempt to bolster
the government by recreating the coalition of 1931— 3, Prieto sought
to replace Manuel Azana as prime minister in a two-pronged strategy
which saw Azana appointed as Republican president. But the strategy
failed because of opposition from Largo Caballero and his supporters,
who threatened to split the PSOE if Prieto went into the cabinet. Party
colleagues, including Prieto’s close friend Juan Negrin, tried to persuade
him to go ahead anyway and call Largo’s bluff. But whether for fear
of the damage that might be done to the PSOE or because he was
not prepared to risk his own political reputation, Prieto refused. 145 As a
result the Republic found itself in the worst of all possible positions. The
government continued without the PSOE but also deprived of Azana’s
stewardship. Azana had agreed to be proposed as president only on
condition that Prieto take over from him in the government. So Prieto
was ill advised to have pursued his strategy at all if he was not prepared
to brave the worst of the Caballeristas’ threats. There can, of course,
be no guarantee that with Prieto as premier the Republic could have
defused the military time bomb. But the memory of his renunciation
in May 1936, and what he saw as its consequences, would return to
haunt Prieto during the war. For whatever the difficulty of attempting to
prevent the rising, it paled into insignificance beside the enormous task
of defending the Republic once the military rebellion was a fact. In the
cupola of the PSOE, the realisation that a golden opportunity had been
lost would weigh like a millstone after 18 July 1936. Ultimately it was

the politics of May 1936 that would drive the actions of Prieto and his

145
Graham, Socialism and War, pp. 35-6, 100. See P. Preston, Comrades (London: HarperCollins,
1999) f° r both Prieto’s and Azana’s point of view, pp. 224-6, 260-1.
74 The Spanish Republic at war

PSOE colleagues when thfey forced the famous cabinet crisis a year later
in May 1937 which saw Largo Caballero resign as prime minister. 146 But
even this could not neutralise the corrosive effect of the memory of May
1936 among leading socialists. The resulting paralysis of political will in
the party hierarchy would be one important factor contributing to the
deadly isolation ofJuan Negrin as Republican prime minister after May
1937, aswe shall see.
By May 1936 the socialist movement was effectively split into two, even
though Prieto had chosen not to enter the government. 147 But although
this made it appear as if the Caballerista wing was in control, in fact it was
148
riven by internal contradictions and growing dissension. Its crisis was
both organisational and political. Politically, it had blocked Prieto’s strat-

egy, but it was difficult to discern what specific alternative it proposed.

Certainly there is no evidence that Largo took any steps towards the
practical implementation of a nationwide inter-organisational alliance
such as the Alianza Obrera had had the potential to be in 1934. 149
In fact a seismic shift was taking place which would dramatically
landscape on the left. Crucial to this shift was the
alter the political
PCE. In Popular Front mode, it found itself to have more and more
common ground with and the parliamentary PSOE. The PCE’s
Prieto
great advantage over the socialists, however, was that it was campaigning
wholeheartedly for an inter-class alliance, while the socialist movement
was divided. But there was something else too. The PCE was the only
political force in Spain arguing for an inter-class alliance based on mass
mobilisation - necessary both as a resource to allow state building and
also as the essential political legitimator of that state-building project
itself. This crucial unifying and modernising dynamic at the heart of the
PCE’s Popular Frontism was encapsulated in the term ‘pueblo laborioso’
(‘the productive nation’) which the party coined and used from February
50
93 6
-’
i

146 147
See chapter 5 below. Graham, Socialism and War pp. 27-8. ,

148
Graham, ‘The Eclipse of the Socialist Left 1934-1937’, in Lannon and Preston, Elites and Power
in Twentieth-Century Spain, p. 134.
149 In September 1936 Largo Caballero would throw his weight behind the reconstitution of a
left-liberal coalition government of precisely the sort he had consistently blocked between 1934

and 1936. See chapter 3 below.

150
This built on Jose Diaz’s call for a ‘concentracion popular antifascista’ at the 2 June 1935 Madrid

meeting, Tres ahos de lucha, vol. 1, pp. 47, 51, 61. Cf. the May 1935 Comintern manifesto (signed by
Diaz, Andre Marty and Palmiro Togliatti), which referred to ‘the socialist, communist, anarchist
and syndicalist toilers of Spain’, and also to ‘the toilers of Catalonia, the Basque Country and
 g

The challenge of mass political mobilisation 75

There was an aura of political vibrancy adhering to the PCE by the

spring of 1936 — even though it was still quite a small political party. 151
It had made some headway in attracting support in the UGT since

1935. But it was in April 1936 that the PCE achieved its most startling
successwhen the national leadership of the fifty thousand-strong socialist
youth federation (FJS) agreed to merge with its much smaller communist
counterpart. 152 In part this was the logical conclusion of the radicalisation
of some sectors of the socialist youth organisation. But politics here has to
be understood as rather more than a matter of abstract marxist-leninism.
Many young Spaniards who were politically active on the left in spring
1936 simply saw the PCE as a more exciting and attractive option than
the socialists, whose views on the proper role of a youth movement had
changed litde since the days of the founding fathers. It was thus the PCE
that channelled the youth breakthrough to organised politics in 1930s
Spain.
Although Largo Caballero had not been consulted about the youth
merger, he spoke in its favour at the massive celebratory meeting held

in the Ventas bullring, Madrid, on 5 April. Given the sheer numerical

superiority of the FJS, Largo no doubt envisaged the unification as lit-
tle more than the reabsorption of an errant minority by the ‘historic’

socialist movement. He also thought that it would give the socialist left a

boost in its bid to thwart Prieto. 153 But quite the opposite would occur.
For although Largo did not yet know it, the IJS leaders were in the

Morocco’. The PCF also referred to ‘an alliance of all toiling people’, Humanite, 5 June 1935.
There was still a certain ambiguity (indeed tension) around whether ‘toilers’ defined primarily
the working class or whether it also included middling-class sectors. This obliquely reflected
the disagreements that still Comintern over the advisability of moving from a
existed inside the
homogeneously worker or united front to the inter-class Popular Front: E. H. Carr, The Twilight
of the Comintern (London: Macmillan, 1982), pp. 148-50, 317.
151
The PCE probably had in the region of 40, 000 members, although it would claim over 100,000,
Bolloten, SCW, p. 831; Guerra y revolucion en Espaha igj 6-3 (4 vols, Moscow: Editorial Progreso,
196&-77), vol. 1, p. 87. For some useful data on the comparative wartime growth of the PSOE,
the PCE and the UGT, see M. Ortiz Heras, Violencia politico en la II Republica (Albacete ig3C—iggg)
(Madrid: Siglo XXI, 1996), p. 109, n. 43. This corroborates the speed of the PCE’s growth,
compared to a much more modest increase for the PSOE. But it was the UGT that grew fastest
of all.
152
The Communist Youth (UJC) had around 3,000 members. For the youth unification see
Graham, Socialism and War, pp. 29-33 an d ‘The Socialist Youth in the JSU’, in
19, 22,
M. Blinkhorn (ed.), Spain in Conflict iggi-igj6 (London: Sage, 1986), pp. 83-102, Vinas, La
formacion de las Juventudes Socialistas Unificadas (igj4~igj6). There was a precedent for this. In
1920 the Socialist Youth had departed the PSOE to form Spain’s very first Communist Party:
153
Both things are clear from Largo’s speech, reproduced in various newspapers, including the
PCE’s Mundo Obrero, 6 April 1936.
76 The Spanish Republic at war
'* •

process of shifting their affiliation to the PC E. 154 In the process they

would bind the new united youth organisation to PCE and Comintern
discipline. In effect, the whole of the FJS would be lost to the PSOE.
But this would only become clear several months later, and the story of
how it happened is inextricably bound up with the exceptional wartime
conditions of autumn 1936, and in particular the siege of Madrid. 155
We have no way of knowing if the merger decision would have been
ratified by FJS congress had normal peacetime life gone on. Certainly
there was opposition inside the FfS to their own executive’s decision.
Many socialist youth sections simply refused to obey the April instruc-
tions. But the military coup exploded before any FfS congress could be

held. The exceptional conditions of war would thus end by binding the
majority of the socialist youth to membership of a united youth organ-
isation, theJSU, which the PCE saw as the flagship of its Popular Front
policy.
But increasingly prepared though Prieto’s socialists and the com-
munists were to reinvigorate the Republic’s parliamentary project, the
military rebellion took them almost equally unawares. The PCE had
been training a militia, the MAOC (the Milicias Antifascistas Obreras
y Campesinas), in the Casa de Campo scrubland adjacent to Madrid.
But, like the party, it was fairly small. The Caballeristas were no better
off, however. Their verbal revolutionism had never extended to the
preparation of armed action. Indeed, their implicit definition of the
revolution as a quasi-mystical event meant that it required no practical
preparation (and thus involved no organisational risk for the UGT). In a
sense, it was this mystical conception of how the revolution would occur
(as a spontaneous rising should the military dare to oppose the popular
will) which ‘legitimised’ the singular inaction of the socialist left/UGT
national leadership - with deadly results. All militia organisation and
training in the tense summer of 1936 was done ad hoc on the basis of local
party and union cells. (The same was obviously true for the CNT.) There

was no national coordination, even within the socialist movement, and,

of course, none between the organisations of the left such as a Worker’s
Alliance might have provided. Absorbed by its own political balancing

154 The FJS leaders informed Largo of this only in the autumn of 1936 — in December, according to
one protagonist: see F. Claudin, Santiago Carrillo. Cronica de un secretario general (Barcelona: Planeta,
1983), P- 45- This inevitably caused a definitive break between Largo and the socialist youth
leaders. Santiago Carrillo himself always dated his decision to join the to the siege of PCE
Madrid in November 1936, but it seems likely that this was the culmination of a process that
had begun back in the spring of 1936.
155
See chapter 3 below.
 The challenge of mass political mobilisation 77

act and mesmerised by republican crisis, the socialist left, with its eyes
on statepower, was caught virtually defenceless by the rebellion. For all
156
of the Spanish left, 1
8 July 1936 was, in every sense, a cataclysm.

CONCLUSION
Looking back, we can say that the reforming project of 1931 was eroded
by the effects of economic crisis. Republican commitment to laissez-
faire economics in conditions of depression made it impossible for the
republican-socialist coalition to implement the kind of welfarist social
and economic reform that could have integrated urban and rural labour
(or substantial segments of it) in the Republic. The Republic could not
broaden its support base to the left while it remained committed to
orthodox liberal economics. The poor, unemployed and unskilled would
look instead to those who articulated a radical critique of the existing
order that direcdy addressed their plight.
Second Spanish Republic, like that
But, ultimately, the stability of the
of its Weimar counterpart of the 1920s in Germany, was never seriously
threatened by those to its left. It could always control such disaffection
through the use of the police. Quite a different matter was the destabil-
ising impact of the right’s political mobilisation of a mass conservative
anti-reform movement. Liberal republicans’ doctrinaire anti-clericalism
had facilitated the mobilisation of Catholics across categories of socio-
economic class. But the sequence of events examined here has also sug-
gested that such a mobilisation might have been impeded if republicans
and socialists had themselves been able to elaborate a strategy of mass
mobilisation that would have allowed them to compete more successfully
with the right for control of the middle ground. But, ironically for the left,

republicans and socialists belonged to an old political world that feared

rather than sought popular mobilisation. The failure to understand the
need for this was to cost the Republic dear. The popular, or counter-
hegemonic, alliance existing in potential in 1931 had fragmented. The
Popular Front victory in the February 1936 elections offered the possi-
bility of reversing that process to some extent. But political opponents

were about to make sure that there would be no time for this. It was the
existence of mass mobilised conservative opinion that made the military
rebellion of July ‘viable’. But it was the action of the coup itself that finally
ripped apart the liberal project in Spain. For it removed from the

156
Cf. M. Alpert, El ejercito republicano en la guerra civil, 2nd edn (Madrid: Siglo XXI, 1989), p. 18.
78 The Spanish Republic at war

Republican orbit a substantial sector of the social groups supporting

forms of economic liberalism. 157 The wartime Republic still bore within
it and political project of historically inclusive liberalism. But
the cultural
thiswould now have to be carried forward from a weaker social base. For
not all the constituencies opposing the coup subscribed to that project.

157
I have two things in mind. First, that some middling groups were lost outright because the
military rising succeeded (for example Granadan vega). But there is a second and more
in the
important point here - that the coup’s success also closed off further political evolutions. It seems
of a military backlash in 1936 was going to allow the republicans to justify
likely that the threat
tougher public-order policies against the left. Likewise, the republicans’ parliamentary pro-
gramme of social and economic reform could also have justified tougher government measures
against worker direct action, such as land seizures. (The Second Republic was not a socialist
Republic - whatever the military rebels claimed.) This scenario might, in time, have permitted
republicans, and even the PSOE, to make inroads into provincial middle-class support for the
Radicals and CEDA. All this is counterfactual, of course. But the crucial point is not: the coup’s
success did more than simply freeze existing political choices - it wiped out other potentials for
good.
 CHAPTER 2

Against the state: military rebellion, political

fragmentation, popular resistance and repression

(i 8 July—4 September igj 6)

Show me the words that will reorder the world.

1

2
On 18 July 1936 the military rebellion that had erupted the previous
day in Melilla, Spanish North Africa, spread to garrisons on the main-
land. 3 In symbolic terms it revealed the serious limits of Republican
control over the state. In practical terms, in shattering both army and
police 4 command structures the rebellion deprived the liberal republican
government of the coercive force it needed to exercise centralised con-
trol of resistance measures. Without unified, coherent security forces -

which in the 1930s remained the defining institution of the central state
in Spain - the government’s authority collapsed. The capital city of
Madrid became, for a time, just another ‘island’ of conflict. Everywhere
they could, the and unions declared a general strike as the
left’s parties
first stage of mobilisation against the rebel military. (In Madrid this hap-

pened predominantly under PSOE and UGT direction, in Barcelona

under the CNT’s.) Proletarian protagonism obviously owed a great deal
to workers’ awareness that they had most to lose should the military
rebellion succeed - an awareness that had been heightened over time
by a string of bloody working-class defeats in 1920s and 1930s Europe

(London: Royal National Theatre/Nick Hern Books, 1994),

1
Tony Kushner, Angels in America
Part 2: Perestroika, Act I, Scene 1, pp. 1-2.
2
The ultra-reactionary rebel cabaldrew its strength from the fact that it was backed by a junior
whose career prospects/professional aspirations had been curtailed by Republican
officer class
budgetary restrictions. The bulk of senior army officers remained loyal to the Republic. See
Cardona, El poder militar en la Espana contemporanea hasta la guerra civil and for Franco, Preston,
Franco, pp. 69-143.
3 On 18 July the rebels extended their control in Spanish North Africa to Ceuta, Tetuan and
Larache.
4 Here the Republic reaped the consequences of failure to demilitarise public order,
fatal its

M. Ballbe, Orden publico y militarismo en la Espana pp. 391, 393-5. If Civil and
constitucional,

Assault Guards had been under civilian control, then their role could have been decisive in enough
places to defeat it instantly. Sometimes Civil Guards were reluctant to oppose the Republic as the
constituted authority. But their passivity was all the rebels needed.

79
8o The Spanish Republic at war

(Italy 1922,Germany 19^3, Austria 1934) as well as by the military repres-

sion following the Asturian miners’ rising in northern Spain in October
I934- 5
In Madrid at party and union headquarters members signed up to
form militia forces. And some sympathetic Republican officers who were
in a position to do so issued arms to them directly, in spite of republican
government threats to have officers who did so shot. 6 The quantities of
arms which could be provided in such ways, however, were too small to
offer a serious or sustained defence against rebel military firepower. The
left consequently called for the systematic arming of party and union
cadres.
The centre-left republicans who constituted the Madrid government
under the weak premiership of Santiago Casares Quiroga 7 were hor-
rified by the thought that in order to defend liberal Republican legal-

ity they might have to arm the very proletarian cadres whose political

agenda they feared and whose mobilisation they had, consequently, re-
8
sisted since February 193 6. Whether for this reason, or for others, the
government had consistendy underrated the threat posed by conspira-
tors in the military. 9 (This very much reflected the views of Azana 10 - for
Casares and the other ministers were the president’s intimates.) Casares,
with extremely little information to go on as a result of the fragmenta-
tion of communications consequent on the coup, played down the scale
and significance of the garrison revolts in the desperate hope that they
might indeed turn out be a minor affair." In lieu of material defence
to
measures, a petrified Casares issued a number of decrees dissolving those

5 See chapter i above.

6
J. Zugazagoitia, Guerray vicisitudes de los espanoles (Buenos Aires, 1940; 3rd edn, Barcelona: Critica,
1977), P- 58; Vidarte, Todos fuimos culpables, vol. 1, p. 238.
7 Casares led the small Galician republican group, Organizacion Regional Gallega Autonoma
(ORGA). (He was appointed prime minister when Azana became president of the Republic in
May 1936.) Zugazagoitia, Guerray vicisitudes, pp. 39-46.
8
See chapter 1 above.
9 There is the famous story of Casares’ outburst against Prieto’s ‘menopausal’ tendencies when
he tried to warn the premier against the military threat, I. Prieto, Convulsiones de Espana. Pequenos
detalles de Mexico: Ediciones Oasis 1967-9), vol. 1, p. 163; Zugazagoitia,
grandes sucesos (3 vols.,
Guerray vicisitudes , by Ignacio Hidalgo de Cisneros, Casares’
pp. 39-41. For other vain attempts
aide-de-camp, see C. de la Mora, In Place of Splendour (New York, Harcourt, Brace & Co., 1939),
pp. 209—10, 216—20. (Casares was both premier and war minister.) See also S. Carrillo, Memorias
(Madrid: Planeta, 1993; 6th edn, 1994), p. 168 and Ballbe, Orden publico y militarismo en la Espana
constitucional, pp. 389-90.
10
Juan Marichal suggests that even Azana could not quite construct for himself the military threat
in terms as vivid as those in which he posed the danger of an armed proletariat: see introduction
to M. Azana, Obras completas (4 vols., Madrid: Ediciones Giner, 1990), vol. 3, p. xxxii.
11
A. Cordon, Trayectoria (Paris: Coleccion Ebro, 1971), p. 224.
 Against the state 8i

military units involved in the rising and relieving troops of their duty of
and the latter of their commands. Early in the
allegiance to rebel officers
evening of the 18th, the government broadcast claims on Madrid radio
12
that the rebellion had been extinguished everywhere in Spain. It was as

if Casares were hoping — against all rationality - that such constitutional

measures would be sufficient to alter the course of events, since he was
not prepared to arm the unions.
The coup itself had in fact precipitated a full-scale crisis for historic
republicanism as swathes of its natural constituents - smallholders and
tenant farmers, traders, shopkeepers and small entrepreneurs - were
definitively lost to it in rebel-conquered territory. Moreover, in the confu-
sion of territorial dislocation, many middle-class republican functionaries
and elected officials prevaricated or ‘went missing’, fleeing their public
responsibilities. The judiciary collapsed. 13 From the ministries down to
municipal and village councils, the state was ceasing to function either
through a physical absence of personnel or through the vacillation of
those still present. In the provinces, civil governors ‘awaited instructions
from Madrid’, but neither the Madrid government nor the leadership of
any other political party or organisation loyal to it had adequate means of
acquiring an overview of what was happening across Spain. The Madrid
government, for all it was in permanent emergency session, 14 had few
resources, and to discover how the political geography on the penin-
sula was evolving it was reduced to a process of piecemeal canvassing
(by telephone or telegraph). A response of Arriba Espana!’ (‘Long live
‘i

Spain!’) inspired greater alarm since it signified rebel control. But re-
publicans everywhere - leadership and cadres - feared for their lives
and were uncertain where their political allegiance should lie in such
a fluid situation. They were caught between their fear of the military’s
visceral anti-republicanism (in spite of the rebels’ initial professions of
‘good faith’) and their awareness that the party and union militia pro-
viding the Republic’s emergency defence represented a threat to their
own preferred forms of social and political order.
In an attempt to heal the breach within republicanism, while also
avoiding both the shedding of blood and the arming of the workers, in
the early hours of the morning of 19 July Republican president Manuel

12
Thereby also alerting much of the population to the alarming fact that the military rising had
been nationwide.
13
Fraser, Blood of Spain, p. 178.
14
First in the war ministry (located in the Palacio Real (del Oriente) - called the Palacio Nacional
during the Republican period) and then in the interior ministry in the Puerta del Sol.
 .

82 The Spanish Republic at war

Azana appointed a new prime minister to head an emergency adminis-

He was Diego Martinez Barrio, leader of Union Republicana,
tration. 15
president of the Cortes and the man who most symbolised republican-
ism’s social conservatism, pragmatism and compromise. His brief was
to conduct telephone negotiations with the rebels’ leader in the north,
General Emilio Mola, 16 in order to achieve a truce. But within some
three hours it was clear that the republicans’ gamble had failed. The
rebels were not interested in negotiating, 17 and the net result of Martinez
Barrio’s attempt was only to lose republicanism its last shreds of credibil-
ity with the proletarian cadres who were facing down the military rebels

in the streets of Madrid and in the Guadarrama sierra to the north of the
capital. Martinez Barrio’s efforts were seen as temporisation, which, in
casting doubt on his supporters’ commitment to resisting the coup, only
accelerated the final eclipse of republicanism. By midday on 19 July he
had no choice but to resign. 18
Time was ever more of the essence. The republicans’ indecision -
epitomised by Casares’ refusal to arm - had
the workers’ organisations
already caused lives to be lost. President Azana - not exempt from the
epidemic of procrastination himself 19 - appointed the same day an-
other all-republican cabinet under the premiership of chemistry profes-
sor Jose Giral, Azana’s personal friend and close associate in Izquierda
20
Republicana (Left Republicans). Giral’s tacit brief was to arm the mili-
tia in order finally to quell the garrison revolts. It was hard for Azana to
find a republican politician prepared to accept leadership responsibility
in this crisis situation, something which only emphasises the depth of the
crisis which had struck republicanism. Giral was motivated in great part

15
Although the suggestion of a ‘moderate’ (i.e. in fact a fairly conservative) republican cabinet
to negotiate with the rebels came from the conservative republican lawyer and leader of the
small National Republican Party, Felipe Sanchez Roman. For the members of Martinez Barrio’s
cabinet, seej. M. Gomez Ortiz, Los gobiernos republicanos. Espaha 1936-1939 (Barcelona: Bruguera,
z
977 ),4° (originally listed in the Gaceta de Madrid, 19 July 1936).
P-
16
Mola was the ‘director’ of the military conspiracy. The intended general-in-chief of the rising,
Sanjurjo, was killed when the plane bringing him from Lisbon to Burgos crashed soon after
take-off on 20 July. H. Thomas, The Spanish Civil War (Harmondsworth: Penguin, 1977), p. 254.
17
Mola explained this by reference to the irreconcilable values/objectives of the two sides. But
obviously for the rebel chiefs there was also a personal imperative. For even if an amnesty had
been negotiated as part of the truce, the leaders had effectively ‘burned their boats’ as far as
professional advancement under a Republican regime was concerned.
18
Azana, Obras completas, vol. 4, pp. 714—16; D. Martinez Barrio, Memorias (Barcelona: Planeta,
i
983), pp. 356-68; Zugazagoitia, Guerra y vicisitudes, pp. 63-5.
19
Prieto expressed his exasperation at Azana’s ‘abcesos de vacilacion’, A. Velez, Informaciones
(Madrid), 10 November 1977 (Velez was the pseudonym of Jose Maria Aguirre, Largo
Caballero’s secretary).
20
For a list of cabinet members see Bolloten, SCW, pp. 46-7
 Against the state 83

by personal loyalty to Azana. But his decision is also indicative of the re-
luctant acceptance by some republicans that, if the military would not ne-
gotiate, then the Republic’s survival depended on the emergency popular
defence forces with whom a modus vivendi would thus have to be reached.
Given the dislocation of army and police, these extremely eclectic
local resistance movements provided important reinforcement for loyal
elements in the security forces. Together they were all that stood between

the Republic and defeat at rebel hands. Giral immediately petitioned the
French Popular Front government for arms on 19 July. He simultaneously
decreed the arming of party and union militia as well as the reopening of
workers’ centres and union headquarters closed by Casares.
21
CNT mil-
itants were also released from Madrid’s Carcel Modelo (Model Prison).
Provincial civil governments were instructed by phone to distribute
arms. (Although the instructions often came too late or else were simply
22
not implemented. ) But we should not make the mistake of assuming
that Giral controlled these forces his government was prepared to arm.
Indeed, at this stage, the Madrid authorities were barely in control of
what was happening in the capital itself.
There the militias’ desperate search for arms and food supplies, es-
pecially in the early weeks, accelerated the process of state dislocation
which the military rebellion had detonated. Both the Spanish Socialist
and Communist Party leaderships intervened to support the quartermas-
ter officers attempting to impose some sort of limits and discipline on
the often excessive and always uncoordinated militia demands. 23 (Such
was the mistrust of the professional military in the wake of the coup
that when the small emergency staff of officers keeping things running
at the War Ministry refused to meet militia requests, they instandy laid
themselves open to the accusation that they were crypto-rebels.)
The Spanish Socialist Party (PSOE), which along with its trade union
(UGT) was the most prominent organisational force in the capital at the
start of the conflict, formed the Motorizada militia, which also fulfilled a

public-order function in the city. Most of its membership was drawn from
the minority of the Madrid socialist youth which had not left to join the
united socialist-communist youth organisation (JSU) in April 193 6. 24

21
On i4july 1936 the government had closed monarchist, Carlist and anarchist centres, J. Lozano,
La Segunda Republica. Imageries, cronologiay documentos (Barcelona: Ediciones Acervo, 1973), p. 224.
22
Cordon, Trayectoria, p. 248.
J. Martin Blazquez, I Helped to Build an Army (London: Seeker & Warburg, 1939), pp. 125!!
23

24
Zugazagoitia, Guerray vicisitudes, pp. 58-9, 130; Vidarte, Todosfiiimos culpables, vol. 1 p. 59; Graham,
,

Socialism and War , p. 30. The PSOE and its youth militia replicated this public-order function in
several cities: see, for example, Ortiz Heras, Violencia politica en la II Republica, p. 106.
84 The Spanish Republic at war

The JSU itself, rapidly expanding from among previously unaffili-

ated young people, organised a number of militias - Octubre, Largo
Caballero, Joven Guardia and Pasionaria. The Spanish Communist
Party (PCE), which was also rapidly recruiting new members in Madrid
and the central zone from among unaffiliated youth and the ranks of
the professional military, had its own militia, the Madrid-based MAOC
(Milicias Antifascistas Obreras y Campesinas). Formed back in 1933,
the MAOC also had a youthful
profile. Since acquiring an effectively

February 1936 Popular Front elections, it had

legal existence after the
functioned mainly to protect party meetings and premises. But it had
also trained in the Casa de Campo with instructors who were often serv-
ing army officers. The MAOC
would later become the nucleus of the
PCE’s Fifth Regiment (Quinto Regimiento), whose function in the initial
training and shaping of the new Republican army was to be so crucial. 25
But not only had the military coup fragmented the army. By inducing
the collapse of Republican government at every level it also massively
facilitated the upsurge in popular political violence which followed that
26
collapse. This sudden explosion was primed by rage at what was seen
as the rebels’ attempt to put the clock back to old-regime order by force,
after their failure by electoral means. Although the intensity of this post-
coup popular political violence varied across Republican territory, it was
everywhere instigated by urban workers and landless labourers, who di-
rected it overwhelmingly at the sources and bearers of the ‘old power’ —
whether material (by destroying property records and land registries 27 )
or human (the assassination or brutalisation of priests, Civil Guards,
25 On the MAOC, seej. Modesto, Soy del Quinto Regimiento (Barcelona: Editorial Laia, 1978), pp. 47,
49-50, 61-2 and E. Lister, Memorias de un luchador (Madrid: G. del Toro, 1977), pp. 66-7; Alpert,
El ejercito republicano, p. 18. See also chapter 3 below.
26
See Ortiz Heras, 99-100; J. D. Simeon Riera, Entre la rebelioy
Violencia politico, en la II Republica, pp.

la tradicio (Lima durante La Republica y la Guerra 1931 -1939) (Valencia: Diputacio de Valencia,
Civil.

I 272-6 for a suggestive, if

993), PP- problematic, discussion of the ‘pre-modern’ dimension of
this popular violence. Sometimes, however, the means of redress were other: see the case of
rural labouring women in Pozoblanco (Cordoba) who demanded that the ‘senoritas de derechas’
(‘conservative ladies’) should be required to participate in the olive harvesting, G. Garcia de
Consuegra Munoz, A. Lopez Lopez and L Lopez Lopez, La represion en Pozoblanco (Cordoba:
Francisco Baena, Editor, 1989), p. 74. As several authors note, paseos were often carried out
by those from the surrounding area but not the same village as the victim(s). This could be
attributed to the practical security offered by relative anonymity. It was also the pattern of pre-
war anti-clerical violence (see Thomas, Spanish Civil War, p. 54) and probably owed something
to community tabus. Ortiz Heras, Violencia politico en la II Republica, p. 105; Simeon Riera, Entre la
rebelioy la tradicio, p. 208; C. Castilla del Pino, Preterito imperfecto. Autobiografia (Barcelona: Tusquets
Editores, 1997), pp. 186-8; Fraser, Blood of Spain, pp. 353, 358, 362.
27
A. Bosch-Sanchez, Ugetistas y libertarios. Guerra civil y revolucion en el Pais Valenciano, 1936-1939
(Valencia: Institucio Alfons el Magnanim, 1983), p. 32; Simeon Riera, Entre la rebelioy la tradicio,

pp. 205, 273 (n. 72).

 Against the state $5

28
and shopkeepers associated with speculative pric-
police, estate bailiffs,
ing and other exploitative practices 29 ). There is a clear link between
post-coup popular violence and pre-war conflicts: for example over the
blocking of land or labour reform legislation in certain localities or over
worker dismissals after the general strikes of 1934 or over conflicts (again,
over the implementation of social and labour reforms) in the aftermath
of the February 1936 Popular Front elections. In the early months of
the conflict, acts of terror perpetratedby the population at large would
also be triggered by the news of mass shootings and other atrocities
in rebel territory, as well as by the direct experience of enemy air at-
tack, which saw assaults on imprisoned conservatives in a number of
places. 30
In the end, however, such acts of terror cannot be explained solely by
reference to the conscious decision of individual (or collective) perpetra-
tors. Violence as a popular response is always shaped by the dominant
culture: 31 those who died embodied in the eyes of those who killed the
privilege and property of a closed social, economic, political and juridical
order that had daily done violence to them by excluding their most ba-
sic needs from its purview. It was the symbolic centrality of the Catholic

Church as an institution to this exclusion that explains the notorious anti-

clerical terror. The collapse of authority caused by the
dimension of the
militarycoup ushered in a wave of killing of religious personnel unprece-
dented in the long and complex history of anti-clericalism in Spain. 32

28
Although latifundistas were habitually absentee landlords, landowners are listed among those
killed in the south, Garda de Consuegra Munoz, Lopez Lopez and Lopez Lopez, La represion en
Pozoblanco, p. 69 unpaginated appendix: ‘Victimas de la represion republicana’).
(see also
29
The inclusion of shopkeepersmay seem an anomaly in an analysis of attitudes to the state.
However, as discussed in chapter 4 in relation to Barcelona (but not exclusively to it), the Chamber
of Commerce frequendy called upon local government authorities to deploy police to ‘resolve’
disputes between shop owners and those engaged in alternative food procurement and sale for
poor and marginal sectors of the population. For similar sorts of direct action, see Radcliff, From
to Civil War pp. 249-304.
Mobilisation ,

30 Garcia de Consuegra Munoz, Lopez Lopez and Lopez Lopez, La represion en Pozoblanco , pp. 55,
60; G. Cox, The Defence ofMadrid (London: Victor Gollancz, 1937), p. 183; Ortiz Heras, Violencia
politica en la II Republica, pp. 106-8; G. Jackson, The Spanish Republic and the Civil War, p. 343;

I. Gibson, Paracuellos como jue (Barcelona: Argos Vergara, 1983), pp. 178-9.

31
For one formulation of this, see ‘Del terror y la violencia’, in F. Savater, Para la anarquiay otros
enfrentamientos (Barcelona: Orbis, 1984), p. 68.
32
On anti-clericalism in pre-war Spain, see especially: Ullman, The Tragic Week: also
;
Romero
‘ >
Maura, La rosa del Fuegci \
Alvarez Junco, El emperador del Paralelo. Lerrouxy la demagogia populista\
D. Castro Alfin, ‘Cultura, politica y cultura politica en la violencia anticlerical’, in R. Cruz
and M. Perez-Ledesma (eds.), Cultura y movilizacion en la Espaha contemporanea (Madrid: Alianza
Editorial, 1997), pp. 69-97 1 J- de la Cueva Merino, ‘El anticlericalismo en la Segunda Republica y
la guerra civil’, in E. la Parra and M. Suarez (eds.), El anticlericalismo espahol contemporaneo (Madrid:
Biblioteca Nueva, 1998).
86 The Spanish Republic at war

But we still have a relatively undifferentiated picture of wartime anti-

clerical violence. The existing historical bibliography tends to chroni-
cle rather than analyse .
33 Anthropological studies can open up fruitful
avenues for the historian. But too great a concentration on the symbolic
plane can also obscure the quite specific renegotiations of political power
which anti-clerical violence signified in particular situations 34 For these .

reasons we are in need of thorough local studies informed by interdisci-

plinary theoretical perspectives .
35 It is also important to remember that
not all forms of popular terror against Catholics during the civil war were
necessarily anti-clerical in origin. For example, the ‘pillars’ of Catholic
associational life belonged to the local economic elites or
also frequently
were leaders of the sometimes both). 36 The killing of
political right (or
religious personnel was for many contemporary observers - in Spain and
beyond — the most symbolically and ethically charged of all the forms of
violence perpetrated during the civil war. For some commentators this
remains the case today — even though its dimensions are dwarfed by
other forms of killing carried out during the war (in both zones) against
secular social constituencies .
37
Whatever one’s assessment here, the fact
remains that retaliatory popular terror - anti-clerical and otherwise -
happened because it was perceived as offering the prospect of tabula rasa :

a satisfyingly instantaneous dissolution of political oppression as well as

38
reparation for accumulated social hurts .

33 The standard works on anti-clerical killings during the civil war are Mgr Antonio Montero,
Historia de la persecution religiosa en BAC, 1961) and V Carcel Orti,
Espana igg 6—igjg (Madrid:
La persecution religiosa en Espana durante la Segunda Republica (igji ~l33g) (Madrid: Ediciones Rialp,
1990); see alsoj. Sanchez, The Spanish Civil War as a Religious Tragedy (Notre Dame, Ind.: University
of Notre Dame Press, 1987).
34 Some work of B. Lincoln, ‘Revolutionary Exhumations in Spain, July
useful exceptions in the
and R. Maddox, ‘Revolutionary Anti-
1936’, Comparative Studies in Society and History , 27 (2) (1985)
Clericalism and Hegemonic Processes in an Andalusian Town. August 1936’, American Ethnobgist,
22 (1) (1995), pp. 125-43.
35 It would be useful if empirical work on a region like Aragon - such as the study currently under
way on Barbastro (Huesca province) - could address the question of the qualitative differences
in wartime anti-clericalism. (Aragon saw no church burning in 1931, and ‘traditional’ forms
of anti-clerical demonstration had not included this previously.) Yet churches in Aragon (and
Catalonia) were major foci of wartime anti-clerical violence in 1936.
36 Casanova, Caspe 1336-1338. transformaciones sociales durante la guerra civil
J. Conflictos politicos y
(Zaragoza: Institution Fernando Catolico, 1984), p. 45.
el

J. M. Sole i Sabate
37 As the most important local studies make clear: see, for example, and
J. Villaroya i Font, La repressio a la reraguarda de Catalunya (1336-1333) (2 vols.,
Barcelona: Abadia de
Montserrat, 1989-90); F. Moreno Gomez, La guerra civil en Cordoba 1336-1333 (Madrid: Editorial
Alpuerto, 1985). For an overview, see Santos Julia et al. (eds.), Victimas de la guerra civil (Madrid:
Ediciones Temas de Hoy, 1999), pp. 117 —57 (which also makes clear that the myth of the mass
killing/ rape of nuns was precisely that, pp. 140, 152-3).
38 ‘People were killed for pointless things - for example, because someone sang in church or was
a bellringer.’ But, as local historian J. D. Simeon Riera comments on this oral testimony, ‘The
motives might seem pointless to our way of thinking now, but for those who did the killing, the
 Against the state 87

Popular violence — urban and rural — occurred in the aftermath of

a military rebellion which had disarticulated all of the Republic's pre-
existing political and state structures. Therefore, it is not particularly
useful to attach to it — as many historians and other commentators per-
sist in doing - the prefixes of political mov ements or organisations: thus,

for example, its frequent labelling as 'anarchist terror - as if it were some-

7

how the 'product of CNT leadership directives. 39 In some places, local

7

libertarian cadres identified with and channelled popular expressions of

anger. 40 But the picture which emerges from local studies is clear: what
happened in villages and neighbourhoods was very’ often not under the
control even of the local committee, let alone of any political authority
beyond that. 41 The uncertain correspondence between the left's political
organisations and coerciv e action in the aftermath of the coup is epito-
7
mised by the v exed term 'the uncontrollables incontrolables). As the war (

went on, so its meanings would shift and expand. 42 But, originally, it was
coined to describe the expropriationary terror and assassination imple-
mented by anonymous groups or militia forces which often claimed, or
had imputed to them, libertarian credentials.
The CNT faced an especial difficulty in refuting such accusations be-
cause libertarian organisational forms had always been quite loose. The
FAI, in particular, had always consisted of tiny activist groups operating
independently of all organisational controls - a situation that had already-
created friction inside the CNT before the war. After 18 July 1936 the
boundaries of almost all the Republic's political organisations became
more porous, and the CNT's more than most. Nor was there anything
to prev ent individuals or groups engaged in robbery or extortion from
simply* using libertarian symbols as a cov er for their activities.

Church singer and the bellringer were part of a world that had to be annihilated’. Enin la rebelio

y la tradicio , p. 273, n. 73. Such lay religious functions would no doubt also have invested those
community members with a social power resented by others.
39 There is a particularly exaggerated example of this in T. MitchelL Betrayal of the Innocents
( Philadelphia: University of Pennsyhania Press. 1998 . pp. 86-8. This is part of a larger problem

with certain accounts of 1930s Spain which see those wiio opted for the direct political action
automata - on even more crudely, the dupes’ -
7
(often associated with the CNT) as ideological
of radical political leaders in the CNT An example of this is in Payne. Spams First Democracy.
40
Simeon Riera, Entre la rebelio j la tradicio pp. 203, 205, n. 37.
.

4
J.
M.
Sabin. Prisiony musrte en la Espaha de la postguerra Barcelona: Anaya-Mario Muchnik. 1996).
p. 16; Casanova, Caspe 1936—1938, p. 46. By the same token local committees w ere also usually
laws unto themselves: there is a Yalencian example in Simeon Riera. Entre la rebelio 1 la tradicio.

P- i94
;
42
Thus it was used to denounce more and in tra -organisational political strife in
specifically inter-
the Republican zone, although again facilitated by w eaknesses and gaps in state power - see
chapter 5 below. The 'uncontrollables was a term also used - loosely - to designate the activities
7

of fifth columnists by the later stages of the w ar.

88 The Spanish Republic at war

Conversely, a simplistic construction of ‘the prisoner’ as always and

everywhere a fully fledged comrade-in-arms was still retained by some
anarchist sectors - particularly in the FAI. In the aftermath of the 18
July coup, in areas where they were powerful enough so to do, liber-
tarians released entire prison populations onto the streets in a de facto
extension of the Popular Front’s February 1936 amnesty for political
Many of these prisoners also joined the libertarian militia.
prisoners. 43
For example, the del Rosal column in Cuenca and the Valencia-based
Iron Column (Columna de Hierro) both recruited (the latter heavily)
from the ex-inmates of the San Miguel de los Reyes prison. The Iron
Column would become notorious throughout the Valencia and Teruel
areas during the latter half of 1936 for ad hoc expropriation, targeting both
smallholders and the urban commercial middling sectors and, increas-
ingly, for running battles with reconstructing local police forces. 44 Iron

Column activities caused serious political friction within the libertarian

organisation in Valencia and probably reinforced support for militarisa-
tion therein. 45 Even more threatening for the Republic as a whole was
the growing alienation of the rural and urban middling classes of the
Valencia region, who inevitably interpreted the Iron Column’s activities
as pure brigandage, thus deepening the social fissures exposed by rebel-
lion. The pressure on the CNT’s national leadership to curb its militants
would increase as the escalating needs of wartime mobilisation put a
premium on social unity. This would be one of the factors influencing
the leadership’s institution of mechanisms of centralised control in the
CNT in the course of 1937. 46
Nevertheless, even where the FAI is concerned, one must be careful
not to identify it too readily with all the myriad forms of anonymous
violence occurring in Republican territory after the collapse of the state.

Lumpen activity occurred in many other permutations, while, conversely,

the FAI was more than simply a sponge for political or other forms of
desperado. 47 But whether or not those who inflicted death, terror and
other forms of coercion had, or claimed, organisational affiliations to
the CNT-FAI or other left political entities, the notion that they were
somehow being ‘directed’ to carry out atrocities by specific national

43 Ealham, ‘Policing the Recession’, p. 412.

44 Bolloten, SCW, pp. 333-42. On del Rosal, see Zugazagoitia, Guerra y vicisitudes, p. 180; I. Prieto,
De mi vida, vol. 1 (Mexico: Ediciones El Sitio, 1965), pp. 324-5.
45 Casanova, Anarquismoy revolucion, p. 113. The CNT’s Juan Peiro publicly denounced the activities

of uncontrollable elements: F. Jellinek, The Civil War in Spain (London: Victor Gollancz, 1938),

p. 441.
46 See chapter 5 below. 47 Jellinek, The Civil War in Spain pp. 331—2, 340—1.
,
 Against the state 89

or regional political leaderships bears scant relationship to how things

occurred on the ground. Indeed, the entire mode of explanation which
seeks to show how ‘orders were given’ seems to borrow rather too heavily
from the conspiratorial mind-set (and publications) of the contemporary
right, which went on to produce the {post hoc and justificatory) Francoist

Causa General. 48 The forms of social and political violence which in-
fested the Republican zone in the aftermath of the military coup were
far too complex and chaotic to have been generated by conspiracy.
The history of post-coup popular violence in rural areas - anti-clerical,
revolutionary and otherwise — may at first sight appear easier than its
urban counterpart to elucidate, simply because the relation of power
between the protagonists, as well as the socio-economic and political
tensions in village communities, was relatively less complex and existed
on a smaller scale. But beyond a certain point, this ‘transparency’ is
deceptive. As Julian Casanova indicates in the case of Aragon, the his-
torical evidence with which we operate is so fragmentary and ambiguous
that it is difficult to ascertain how specifically local and personal feuding
connected up with structural political conflict. 49 Nevertheless, in urban
Spain the greater anonymity and fragmentation of life as well as the sheer
concentration of population certainly make it more difficult to establish

where anti-clerical and anti-capitalist motives ended and the settling of

other - less overtly political - scores began.
In Madrid, as elsewhere, one of the most terror-inducing forms of
popular violence was the paseo, 50 or execution at the margins of the judi-
cial process, carried out by militia patrols acting on their own authority.
A brutal form of settling political and class accounts, th epaseo was at root
a product of the deep social cleavages exposed by the military rebellion.
But the anonymous, nocturnal form of the paseo the ultimate unaccount- ,

ability of the process, also made it a perfect cover for settling all man-
ner of personal scores and for motives of sheer material acquisitiveness.
Moreover, whether the motive was revolutionary justice, crypto-rebel
provocation, material advantage or some mixture of these, the paseos

48 A nationwide investigation into ‘red wartime crimes’, this effectively constituted the victorious
Francoist state’s lawsuit against the defeated.
49 Casanova, Anarquismoy revolution, pp. 253, 254, 258. The same can be said of PSOE wartime

factionalism.Correspondence from local organisations to the national executive (in the Archivo
Historico de Moscu, FPI) at times give a sense of how local disputes were being reclothed in the
lexicon of factionalism.
50
one which always ended in death. While the paseos were not exclusively
Literally ‘a stroll’, but
an urban phenomenon, they were a form whose ‘potential’ was obviously greater in cities and
towns.
90 The Spanish Republic at war
*4 ,

soon intensified the already high levels of social insecurity and distress
from which no sector of the Republican population was immune 51 .

In these first weeks of the conflict when the initial shape of the rebel and
Republican zones was being defined by force of arms, virtually the only
articulation between the areas that would, from September, gradually
be built up into the territory of the Republican state came from an ad hoc
communications network provided largely by the transport unions of the
UGT 52 Information thus relayed to the
. executive’s headquarters UGT
in Madrid’s Calle Fuencarral was passed on to the Giral cabinet, which
was also being propped up by Indalecio Prieto and the Socialist Party
executive committee in permanent/ emergency advisorial session. Both
Prieto and his fellow executive member Juan Negrin (later Republican
finance minister and premier) also ran significant personal risks nightly by
participating in informal patrols to curb the wave oipaseosin the capital .
53

They were morally repelled by the arbitrariness of the paseo 54 Indeed, a .

determination to end such abuses was what drove many socialists in their
efforts to restore Republican state power.
However, neither the UGT nor the PSOE - as the other component(s)
of the pre-war Popular Front axis - had themselves managed to avoid
the negative effects of the centrifugal blast detonated by the military
rebellion. This removed the control that the national leaderships of both
union and party exercised over their respective organisations (already
fragmented by the effects of the internal dispute), rendering them for a
time as ‘federal’ as the anarcho-syndicalist CNT had in reality always
been. Nor would this fragmentation prove easily or rapidly reparable. In
various areas of Republican territory, local socialist sections unilaterally
opted to merge either themselves or their local party and union newspa-
pers with their PCE counterparts (where these existed ).
55 In Malaga the
socialists would consider merging with the CNT, complaining bitterly of

51
On the repercussions for worker constituencies, see Ortiz Heras, Violencia politico, en la II Republican

p. 1 00 . A related factor increasing working-class insecurity was the increased occurrence ofviolent
internecine labour conflict, discussed below.
la UGT de Espana igoi-iggg (2 vols., Barcelona: Grijalbo, 1977),
52 Amaro del Rosal, Historia de
vol. 2, pp. 493-6; F. Largo Caballero, ‘La UGT
y la guerra’, speech October 1937 (Valencia,
T
937), P- H; F. Largo Caballero, Mis recuerdos, 2nd edn (Mexico DF: Ediciones Unidas, 1976),
p. 166; R. Llopis, ‘Las etapas de la victoria’, Spartacus, October 1937, p. 4.
53 Marcelino Pascua in S. Alvarez, Juan Negrin. Personalidad historica , vol. 2 (Madrid: Ediciones de
la Torre, 1994), p. 280; (and for Valencia) M. Anso, To fui ministro de Negrin (Barcelona: Planeta,
j
976), pp- 165-6.
54 The fate of a potential victim of repression could be very fluid. In a street situation, it depended
often on whether someone else defended or attacked when the person was first accused. See M.
Ortiz Heras, Violencia politico en la II Republica, p. 105; Zugazagoitia, Guerra y vicisitudes, pp. 78-82.
55 Graham, Socialism and War pp. 75, 269 (n. 25).
,
 Against the state 9

being ‘abandoned by their

5
own - a reaction which,
national leadership
again, derived from the fracturing of communication and organisational
structures as well as from the acute shortage of arms (about which the
socialist Madrid could do litde ). 56 At other times such ini-
leadership in
tiatives of syndical unity grew out of the solidarity forged in the heat of
5
militia defence, which was then transmuted into local ‘unity committees.
Conversely, the fragmentation of state and union power also allowed
already acute tensions/ conflicts between some sectors of the CNT and
UGT rank and files to break into outright violence. In Barcelona, Aragon
and Cartagena, property was ransacked and deaths and disappearances
of ugetistas reported 57 This intra-union antagonism was probably worst
.

in Barcelona, where the clashes - especially between the dockworkers

5

unions - produced fatalities 58 .

What we can deduce from this contradictory simultaneity of inter-

union conflict and collaboration is that the military rebellion - for all the
singularity of the threat it posed to the Republic — did not erase the pre-
war dynamic of intra-left relations with all their tensions, hostilities and
contradictions. Indeed, what many accounts of the period tend to ignore
is that while the coup fractured organisational structures, it left intact

memories of conflict and deeper-rooted patterns of collective political

behaviour and social identity. The underlying picture in Republican ter-
ritory was thus complex and contradictory. The wartime political unity
around which the left’s entire discourse was constructed would, from the
start, be up against serious obstacles. The fragmentation of the PSOE

and UGT, moreover, combined with the virtual eclipse of republicanism,

meant the effective dislocation of the Popular Front alliance.
5
In conditions of such unprecedented government crisis, the ‘islands
of local or regional resistance were strengthened in their particular-
ism. Local resistance to the military rising in the north, north-east,
centre, Valencia region (comprising Valencia, Castellon and Alicante)
and south was largely orchestrated by those parties and union organisa-
tions of the left59 in situ, but, temporarily at least, was unconnected to any

56 Letter
to the PSOE national executive, 15 October 1936, in AH-23-16 (FPI); UGT
executive
minutes, 9 December 1936 (the UGT
executive had learned of the proposed merger only inci-
dentally via the Malaga press), Graham, Socialism and War p. 185; Vidarte, Todos fuimos culpables,
,

vol. 2, pp. 649-50.

57 Graham, Socialism and War pp. 64, 82, 86-7.
,

58
For the pre-history of CNT-UGT conflict in Barcelona see chapter 1 above. For an analysis of
the post-18 July situation in Barcelona/Catalonia, see chapters 4 and 5 below.
59 The (Republican) Basque Country - i.e. predominandy industrial Vizcaya - is the exception
here: the conservative Basque Nationalist party (PNV) retained political control. See chapters
4 and 5 below.
92 The Spanish Republic at war
•
central (or even regional) leadership: anarchists and anarcho-syndicalists
in Aragon; anarcho-syndicalists, anarchists and Gatalanist communists
(POUM) CNT and the UGT in the Valencia region;
in Catalonia; the
the PSOE and the UGT CNT and the PCE) in
(backed up by the
Madrid; anarchists, communist and UGT cadres in Malaga, UGT and
CNT cadres injaen and Badajoz. These forces organised both the initial
popular resistance and the committee structures which supported and
supplied it.

The immediate key to Republican survival, however, lay in Barcelona

and Madrid. In the former, the rising failed rapidly in the face of worker -
and predominantly CNT — mobilisation, seconded by loyal Civil and
Assault Guards. The municipal police were also loyal, being headed by
Colonel Frederic Escofet, who, along with Major Perez Farras, 60 had
led the Catalan Guard in defence of the Generalitat in October 1934.
The Catalan republican left reaped the dividend for having vetted the
police service after its electoral victory in February 1936 to ensure that
only those loyal to the Generalitat remained in positions of power. While
Catalan premier Luis Companys had held out against calls to arm the
for fear of libertarian strength -
-
61
workers on the evening of 18 July
the CNT had managed to storm several depots and some sympathetic
officers had, as elsewhere, allowed them access to the arsenals. Thus
resourced, they went out to meet the disparate rebel columns and picked
them one by one, before they could converge to consolidate their
off,

strength in the city centre. General Goded, arriving from Majorca to

take control of the rising, was instead taken prisoner. By the evening of
19 July only two barracks held out: San Andres on the outskirts of the
city and the Atarazanas near the port. Both would be stormed by CNT
militia and Catalan security forces (Assault and Civil Guards), which
thus ensured the complete suffocation of the rising in Spain’s most radical
and cosmopolitan city. 62 Goded’s briefly worded recognition of defeat, in
which he appealed to his followers to lay down their arms, was broadcast
across Republican territory and gave the morale of Loyalist defenders
a tremendous boost. Barcelona was a vital victory for the government,
60
Perez Farras would act as the anarchist leader Buenaventura Durruti’s military adviser in Aragon:
Thomas, Spanish Civil War, p. 316.
61
He had similarly refused to arm the workers’ organisations during the events of October 1934
in Barcelona.
62
Accounts of the military rising and worker resistance in Barcelona can be found in J. Perez Salas,
Guerra en Espaha (1936-1939) (Mexico, 1947); F. Escofet, Al servei de Catalunya i la Republica (Paris,
r
973); Jellinek, The Civil War in Spain D. Abad de Santillan, Por que perdimos la guerra (Buenos
;

Aires, 1940) and F. Borkenau, The Spanish Cockpit (1937; London, 1986). There is a summary in
Thomas, Spanish Civil War, pp. 232-6.
 Against the state 93

for it ensured that all of Catalonia would remain loyal. But that victory
would be the beginning of a long and intense struggle for political power
between radical libertarians and the affluent urban and rural middle
classes who constituted ‘liberal Catalonia’.
In Madrid, meanwhile, the 18th had seen armed workers surround the
Montana barracks where General Fanjul was awaiting reinforcements
from Getafe, Cuatro Vientos and Carabanchel. But the rebellion had in
fact already been put down in these places. Fanjul was thus isolated and,
without reinforcements, knew that he had insufficient forces to take the
city centre. Early on 19 July the order came through from the Giral cab-
inet to arm the worker militias. Lorries were sent from the government
arsenals to the headquarters of both the UGT and the CNT in the capi-
tal. But the bolts for the vast majority of the 65,000 rifles thus distributed

were still inside the Montana barracks, where those in charge refused
to relinquish them in spite of orders from the war ministry to this effect.
The militia siege of 19 July thus gave way to full and ultimately successful
assault the following day, during which the bolts were duly acquired - if
at a significant cost. For in spite of support from the Republican airforce
which bombed the barracks, the death toll among the assaulting forces,
mown down by insurgent machine guns mounted on the barracks
windows, was extremely high. It was also high among the officers within
(although Fanjul was himself taken prisoner). Some committed suicide
while many more fell victim to militia anger - heightened as they came
under fire on their approach to the barracks to accept the surrender
which the white flag posted by the rebels ostensibly signified.
Once Madrid was safe for the Republic, the militia forces set off north-
wards to the Guadarrama sierra in order to stem the rebels’ military ad-
vance on the capital. Mola’s advance had been impeded by the necessary
dissipation of the northern troops - some to be sent to San Sebastian and
others to Aragon. The mixed columns advancing on Madrid (soldiers
plus Carlist Requetes and Falangists) were halted at the Somosierra pass
in the Guadarrama and at the Alto del Leon to the north-west. They
were handicapped by a lack of arms and ammunition but also by stiff
militia resistance where they had expected a ‘walk-over’.
Militia action was an important component in the failure of the re-
bellion throughout most of populous, urban Spain and its hinterlands.
Nevertheless, there were enough examples of urban labour movements
being defeated in the July days for us to be wary of claiming that the
militia alone were sufficient to guarantee Republican survival in the face
of the garrison revolts. Madrid and Barcelona were very specific in terms
94 The Spanish Republic at war

of the sheer scale of proletarian organisation, and even there the militia’s
resolve was reinforced by support from professional army officers loyal to
the Republic and - most importantly of all - by the regular police forces
(Assault Guards, Civil Guards and (in Barcelona) the Catalan govern-
ment’s own police, the Mozos de Escuadra). 63 In the cases where the
rebels took control in July of cities in which the left was strong - most
notably in Seville (the most revolutionary city in Andalusia and the key
to control of the region), Zaragoza (Aragon) and Oviedo (Asturias) -
these tended to be victories achieved because the working-class forces
lacked coordination or were surprised or outmanoeuvred in some way
Nevertheless, no city in Spain was held for the Republic without the
assistance of at least some part of the police. 64
In Valencia, the ambivalent attitude of the garrison’s military com-
mander prolonged the political limbo until the last day of the month.
Distrusting his intentions, the workers’ organisations had declared a gen-
eral strike 19 July and surrounded the garrisoned troops. In an at-
on
tempt to break the stalemate and avoid bloodshed, on 21 July Giral’s
government appointed a delegate body for the whole of the Valencia
region which was to take power in Valencia (capital) in the government’s
name. 65 But the central state’s writ had ceased to run in Valencia. In the
streets Falangists killed workers and workers burned churches. The CNT,

63 In Barcelona the (loyal) military commander of the Catalan region (General Llano de la
Encomienda) did not have the military forces to defeat the rebels. The fact that he could count on
the police was absolutely crucial: Barcelona had 3,000 Civil Guards, 3,200 Assault Guards and
300 Mozos de Escuadra, a total of 6,500 men against 2,000 military rebels. For the police role in
Barcelona, see Escofet, Al servei de Catalunya i la Republica and V Guarner, L’aixecament militar i la
guerra civil a Catalunya (Barcelona, 1980) - Escofet was councillor for public order in the Catalan
government and Guarner his jefe de servicios.
64 According to R. Salas Larrazabal, just over half of the Civil Guard remained with the Republic,
along with 60 per cent of the Carabineros (customs police) and 70 per cent of the Assault Guards.
In Guadalajara there was Civil Guard support for the militia and at Jaen the corps remained
loyal to the Republic. In Malaga the militia had the Assault Guards on their side. In Zaragoza,
there was no such support: Casanova, Anarquismo y revolucion and J. Cifuentes Chueca and
P. Maluenda Pons, El asalto a la Republica. Los ongenes del franquismo en Zaragoza (1936-1939)

(Zaragoza: Institucion ‘Fernando el Catolico’, 1995). (For an important critique of the conse-
quences of the Republic’s failure after 1931 to demilitarise the police, see M. Ballbe, Orden publico
y militarismo en la Espaha constitucional, pp. 317-96, especially pp. 394-5.) For General Aranda’s
deceit in Oviedo (he proclaimed his loyalty to the Republic, sent the Asturian miners off to
liberate Madrid, and then pronounced for the rising), see summary in Thomas, Spanish Civil War,
p. 236. Oviedo was besieged by Republican forces (miners’ militia predominantly) until October

1936, when rebel troops would succeed in breaking through: see the collective works La guerra en
Asturias (Madrid, 1979) and Historia General de Asturias (Gijon, 1984), vol. 9; Fraser, Blood of Spain,
pp. 250-4.
65 Bosch Sanchez, Ugetistasy libertarios, p. 18. Decree of 22 July (in the Gaceta de Madrid) stipulated

that its jurisdiction should cover the provinces of Valencia, Alicante, Castellon, Cuenca, Albacete
and Murcia.
 Against the state 95

which trusted neither the military nor the Madrid government, had al-
ready taken the lead in establishing a joint union (CNT-UGT) executive
committee which demanded to be armed against potential rebellion.
(In its eyes the delegate body was discredited from the outset because it
was headed by Martinez Barrio, who had just attempted to treat with
the military conspirators.) Then on 25 July a party of Civil Guards, sent
along with worker forces to help take Teruel, turned their guns on the
66
militia en route and passed to the rebels. This provoked a rebellion in
the barracks, and a number of pro-Republican soldiers fled with arms.
This in turn permitted an assault by the workers’ militia, assisted by loyal
Civil Guards, on 3 1 July. The assault extinguished the military threat, but
it also signalled the end for the tenuous hold of Madrid’s delegate body,

which was now eclipsed by the unions’ executive committee (Comite

Ejecutivo Popular).
Elsewhere, however, the military rebellion succeeded rapidly and rel-

atively easily within the first two days (18-19 July) - most notably in the
conservative rural Spain of the north down to the centre, where it had a
significant measure of civilian support extending to the popular classes.
While the Basque industrial heartland of Bilbao (Vizcaya) was held for
the Republic (if soon territorially isolated from it), the northern Carlist
strongholds of Navarre and Alava as well as virtually the whole of Old
Castile/Leon, with all its major centres (Burgos, Valladolid, Zamora,
Salamanca all the way down to Caceres in Extremadura), plus the
Canary Islands were in rebel hands. 67 By 22 July Galicia in the north-west
corner would also be almost entirely rebel-controlled in spite of desperate
resistance in the left’s urban bases, most notably in the ports of Vigo and
La Coruna. Nevertheless, by the end ofJuly 1936 the rebels had in fact
failed to take control of more than a third of Spain’s national territory.
In the other two thirds of Spain power was, as we have seen, almost
everywhere intensely fragmented. The dominant view of this situation —
whether in memoirs or subsequent historical analysis - has tended to
be negative. But this assumes an exclusively ‘top-down’ perspective on
the committee phenomenon and one, moreover, strongly influenced by
a retrospective appreciation of the escalating military threat faced by the
Republic. If we view the committees ‘from below’, however - from the
contemporary perspectives of rank-and-file participants at the moment

66
Jackson, The Spanish Republic and Civil War, p. 265; Bosch Sanchez, Ugetistasy libertarios p. 19.
,

67
Of all the Balearic Islands (Majorca, Minorca, Ibiza and Formentera) only Minorca was held
Republic (until its surrender in 1939) - although Ibiza and Formentera were briefy retaken
for the
by Republican forces in August 1936 and held until late in the year.
 )

96 The Spanish Republic at war

they were mobilised by *the left’s local leadership cadres - then we have
to understand that ‘resistance’ to the rebels was spurred predominantly
by the possibility of direct action to transform the local environment,
the lived unit of experience or patria chica - be it village, town or urban
neighbourhood — by means of gaining control over decisions affecting
daily life.

Everywhere the coup had been quelled - aside from the Republican
Basque Country - there was a mushrooming of locally oriented solutions
to the organisation of everyday life: from transport, communications and
water supply to the cooperativisation of food supplies, workshops, news-
papers, restaurants and barber shops. Money was frequently abolished
and a system of coupons or vales to cover basic needs was instituted by
individual village or urban committees - and, particularly in the latter
case, these were often issued by many different committees simultane-
68
ously. In some areas agriculture, industry and commerce were partly
collectivised.
But although committee and militia were widely established forms
of organisation and all owed their initial existence to the powerful cen-
trifugal charge of the military coup, this does not mean that they were
qualitatively similar beyond superficial aspects of nomenclature. What
committee and was strongly inflected by the regionally
militia ‘meant’
diverse historical experiences and political cultures to be found among
Spain’s proletarian constituencies. While the committee phenomenon
was everywhere particularist, 69 few sectors of Spain’s working classes
(leadership cadres included) possessed the kind of ideological collateral
which allowed them to think of the committees as the building blocks of
a new order. 70 Within these variables, the direction/ potential of commit-
tees and militia was also significantly shaped by the rapidity with which
the imperatives of military defence impinged upon them.
Madrid saw union and neighbourhood committees formed as well
as collectives and cooperatives overseen by worker committees both
in the municipal sector and to some extent in private industry, com-
merce and the service sector. But this occurred on a significantly smaller
scale than in Barcelona and it was driven more by practical imperatives

68
R. Abella,.Z<2 vida cotidiana durante la guerra civil. La Espana republicana (Barcelona: Planeta, 1975),

pp. 17, 88.

69 Cf. the refusal of collectives to pay state taxes, ‘which only support good-for-nothings gandules
(

and police harassment’, Abella, La vida cotidiana durante la guerra civil, p. 89.
70 Nor indeed was localism necessarily radical simply because it excluded the notion of the state:
see M. Vilanova, ‘L’Escala y Beuda: dos formas de propiedad y de lucha social durante la guerra
civil’, Historiay fuente oral 3, ‘Esas Guerras’ (1990), 39-66.
,
 Against the state 97

than by popular ideological and cultural preferences. 71 The two cities

were the only Spanish ones to have a population exceeding i million
in 1930. But although Madrid was an important administrative centre,
it was not yet a significant industrial one. Moreover, the difference be-

tween Madrid’s 40,000 building workers and 25,000 metalworkers and

Catalonia’s 200,000 textile workers, 70,000 metalworkers and 70,000
building workers was not just one of scale; it also represented a step
change in the range and potentialities of labour culture. 72 In Madrid the
close proximity of rebel columns (initially to the north and later from
the south) concentrated attention on the need to guarantee the coordi-
nated economic/ military supply and transport services essential to the
emergency defence operation. Thus an estimated 30 per cent of Madrid’s
productive industry was brought relatively rapidly under military or gov-
ernment control 73 — with union cadres serving as the instruments of this
process. The UGT, as the dominant labour union in the Madrid area, had

always been far less interested than the CNT in syndical economic con-
trol. But, in practice, the Madrid CNT also came to accept government

control as necessary in the circumstances. CNT members worked along-

side those of the UGT in war production - even if some tensions remained
between the respective leaderships over questions of organisational pre-
rogatives. 74 In the Madrid area the CNT’s militia forces were also less
inclined to dispute the need for centralised organisation and discipline. 75
In Asturias too, in the isolated (non-Basque) north, the pressures of the
two-pronged rebel military advance on the city of Oviedo, besieged by
Republican forces, concentrated minds on issues of military defence. 76
This would go some way towards facilitating the reconstruction of cen-
tralised political power within the region. But equally significant here was
the special political and syndical culture of the north. 77
While both the CNT and the UGT in the north had remained open to
the use of direct action into the 1930s, they were also far less uniformly
hostile to the idea of parliamentary and other political strategies for

71
The expropriation of small businesses or industries was very rare. Although a notice might
announce ‘here one works collectively’, it usually referred to a newly brokered profit-sharing
scheme, not workers’ control: Thomas, Spanish Civil War, p. 293.
72
Fraser, Blood of Spain, p. 295. In Madrid industry meant small workshops, while in Barcelona
these existed alongside large-scale factories.
73 S. Payne, The Spanish Revolution (London: Weidenfeld & Nicolson, 1970), pp. 236-7.
74 The Madrid FAI was less reconciled to government control, however.
75 See chapter 3 below.
76
Fraser, Blood of Spain, pp. 240-1. There is a nicely observed sense of the isolation of Asturias in
Thomas, Spanish Civil War, p. 310; on the north see alsojellinek, The Civil War in Spain, pp. 407-16.
77 See, the introduction and chapter 1 above.
98 The Spanish Republic at war

achieving social reform. This underlying sympathy between the two had
been the key to their successful pre-war collaboration in the Asturian
Workers’ Alliance. Although this had failed everywhere else in Spain
as the intra-left unity initiative it was intended to be, in the north it
had provided the organisational matrix for the famous armed rising of
October 1934. 78 (Less spectacularly, but even more crucially, the experi-
ence of organising emergency supply and defence functions during the
rising would provide a blueprint for collaboration and survival in July

1936.) The impassioned amnesty campaign to free the thousands im-

prisoned after the failure of October 1934 was a key process in forging
closer CNT-UGT leadership collaboration in the north. It reinforced a
common understanding among socialists and anarchists there of the im-
portance of a left politics of pragmatism geared to intermediate goals. In
the process, a certain ideological heterogeneity was also given its head,
which then came to be reflected in the solutions given to homefront
organisation during the war. 79
There was, thus, a reasonably cooperative relationship between the
CNT’s war committee, based in its stronghold, the port city of Gijon,
and the socialist-led Popular Front committee in Sama de Langreo, in
the heart of the mining belt. CNT leaders were as concerned as their
socialist counterparts to counteract the hyper-fragmentation of power
symbolised by the seemingly endless replication of local neighbourhood
committees. The same CNT leaders were also instrumental in ensuring
80

that Bank of Spain funds in Gijon were handed over to the Popular
Front committee, thus effectively securing them for government use.
Nor did the CNT’s war committee oversee any systematic purge of the
police force in Gijon, still less establish parallel security forces on a par
with the patrol committees being formed at the same time in urban
Catalonia.
Although there was conflict in Gijon over the anarchists’ desire to ex-
propriate small traders and shopkeepers, in other respects the CNT and
UGT concurred. The CNT accepted the Sama de Langreo Popular
81

Front’s veto on any interference with the property or individual commer-

cial rights of rural smallholders whose farms constituted the predomi-
nant form of agriculture in the region. Thus agrarian collectivisation

78 See chapter above. 79 Radcliff, From Mobilisation Civil War, pp.

i to 305-7.
80
Fraser, Blood of Spain, pp. 240-1.
81
For the long history of anarchist mobilisation around issues of popular consumption in Gijon,
see Radcliff, From Mobilisation to Civil War, passim. But even on this issue the northern lead- CNT
ership was remarkably conciliatory: Fraser, Blood of Spain, p. 244; Jellinek, The Civil War in Spain,

pp. 407-16.
 ,

Against the state 99

was never seriously on the agenda in Asturias - even though such rural
82

constituencies had significandy less economic and political power than

did their Catalan or Levantine counterparts. (Not least because Asturias
had no urban middling any significance outside rebel-held
classes of
Oviedo.) The regional CNT and FAI also accepted the UGT’s line on
pre-existing industrial property rights in Asturias. Workers’ control com-
mittees existed - UGT-led in the coal mines and CNT-directed in the
steel works. But their function was largely one of monitoring produc-

tion - itself left largely in the hands of such politically reliable foremen
and engineers as had not fled or who could be procured from elsewhere.
Given the UGT’s dominance in the mines, the production commit-
tees were overseen almost from the beginning by a central government
representative. 83
Economic innovation and cooperativist/committee activity in
Asturias was heavily concentrated on organising the practical needs of
urban life and, in particular, the supply of food and essential services
(including education). As well as being eminently pragmatic, this also re-
sponded to a deeply ingrained community-based idea of politics that was
further reinforced by the isolation of the Republican north in 1936—7. 84
Although private distribution systems continued, the Asturian Popular
Front was collectively responsible for underwriting the consumption of
poorer sectors of the population. 85 It was thus an attempt to deliver -
86
within the ever-increasing constraints of the war - what had been the
Asturian Front’s political agenda between the February 1936 elections
and the military rising: namely the ‘reclaiming of the Republic’ for worker
constituencies and for social reform, by means of extra-parliamentary
(but legal) mass political mobilisation.
From the above analysis is thus becomes clear why CNT leaders in
the north were far less resistant than their counterparts elsewhere to the
dissolution of their own war committee. While defence imperatives obvi-
ously figured in the calculation, they had no serious ideological qualms.
The northern CNT joined the new provincial Popular Front committee,

82
A few large landowners who were pro-rebel were expropriated in November 1936, but this was
the exception rather than the norm: Fraser, Blood of Spain p. 243 (n. 1).
83
Thomas, Spanish Civil War, p. 310.
84 See Radcliff, From Mobilisation to Civil War, pp. 279, 307. 85 Ibid.,
pp. 306-7.
86
Food supply posed a challenge not least because from early September the Republican north
was isolated from both France and the rest of the Republican zone except by sea. There were
droll remarks about cats being afraid of Gijon’s inhabitants (Thomas, Spanish Civil War, p. 311).
The food problem, though acute, was eased by the relative lack of population pressure and then
by evacuation in 1937.
100 The Spanish Republic at war

which, though socialistrled, moved to Gijon in September 1936. Nor,

some organisational conflicts with its other members, did
in spite of
the CNT oppose what was effectively the ‘bottom-up’ reconstruction
of government power. New town councils were set up in September—
October 1936 - the earliest this process occurred anywhere in Republican
territory. 87
A not dissimilar picture would also emerge in August and the following
88
months in the Valencia region. There the Popular Executive commit-
tee allowed the multiform committees which had mushroomed during
the emergency defence to be replaced by a reconstructing Republican
government authority. Thus by September 1936 all three provinces in the
region (Valencia, Castellon and Alicante) would have re-established local
Popular Front authorities through which the Republican government
gradually recuperated its power. 89 For although the Valencian CNT was
influential, its regional federation was the heartland territory of treintismo ,

a reformist syndicalism far less hostile to the state’s political power than
its radical libertarian counterparts in adjacent Catalonia. 90 This political
outlook also explains the extent of the Valencian CNT’s own efforts to
articulate the atomised rural collectivisation carried out by local com-
mittees which had fragmented the region’s economy to such an extent
that it was jeopardising basic processes of distribution and supply crucial
to Republican defence. 91 The CNT also faced the thorny issue of ad
hoc/ indiscriminate ‘requisition’, a collateral effect of the emergency de-
fence which, as elsewhere, was threatening to alienate middling social
constituencies. 92
For in Valencia the CNT had to recognise the sociologically divided
nature of the region. While it was strong in urban areas — and espe-
cially in the port of Valencia — its rural base was somewhat weaker. 93

87
As part of this September the Madrid government appointed the Asturian Popular
process, in
Front leader, veteran Belarmino Tomas, as governor of Asturias and Leon; in November
socialist

the Popular Front committee was renamed the Council of Asturias and in December the Consejo
Interprovincial de Asturias y Leon.
88
I use ‘Valencia region’ here to denominate the provinces of Valencia, Castellon and Alicante
(known collectively as the Pals Valenciano or Levan te).
89 Bosch Sanchez, Ugestistasy libertarios p. 41.,

90 Ibid., p. 22 citing the Valencian CNT on ‘the committee plague’. For a discussion of treintismo
see chapter 1 above.
91
Bosch-Sanchez, Ugestistasy libertarios, pp. 19, 22-3, 38-9.
92 Note the acute comment by syndicalist leader J. Lopez from Fragua Social (Valencia: CNT press)
on ‘una masa permanentemente sublevada que pedla vlveres’ (a perpetual mass revolt demanding
foodstuffs) in Bosch Sanchez, Ugetistasy libertarios, p. 19 (inc. n. 6); see also p. 22. There were also
serious clashes over requisition between peasants and militia in Aragon: see chapter 4 below.
93 Bosch Sanchez, Ugetistasy libertarios, pp. 45-6.
 ,

Against the state IOI

Moreover, in the Valencian countryside important rural constituencies

of smallholding and tenant farmers were bastions of social and often
political conservatism .
94
In the latter case this was inflected by strong
regionalist sentiments. But though these distanced Valencia’s middling
classes from the ultra-centralist ideals underpinning the military rising 95 ,

these social constituencies were, nevertheless, hostile to the collectivism

96
associated with the proletarian defenders of the July days In fact, agri- .

cultural collectivisation in Valencia, though not negligible, would still

remain marginal to the economy of the region as a whole But fears of .

97

what might happen remained prevalent among tenants and smallhold-

ers, fuelled by the knowledge of incidents where GNT cadres on the

ground had exercised coercion. Thus mutual distrust and sharp social
divisions between individualists and collectivists, as well as the increas-
ing influence of Republican state agencies, would set the stage for future
confrontations in the Valencian countryside — in spite of the regional
CNT’s overall pragmatic commitment to the voluntary principle.
In the rural zones of neighbouring Catalonia the CNT-FAI leader-
ship similarly acknowledged the rights and property of the populous class
of rural smallholder and tenant farmers. What made Catalonia differ-

ent, urban dimension - in particular Barcelona (Spain’s

however, was its

only metropolis) and its surrounding industrial belt. There a unique

configuration of industrial production and urban life had, over decades,
produced a rich and complex set of popular and proletarian cultures
unparalleled anywhere else in Spain. Among some sectors of Barcelona

94 Ibid., 47-8; Thomas, Spanish

pp. 23, Civil War, p. 305; Jackson, The Spanish Republic and the Civil

War, pp. 246, 264.

95 For the fate of Luis Lucia, leader of the Valencian Derecha Regional (Regional Right) (which was
a constituent part of the mass Catholic Party (CEDA) prior to the war), see P. Preston, Comrades
(London: HarperCollins, 1999), p. 326; Thomas, Spanish Civil War, p. 242 (and n. 1).
96 ‘Speaking to them
about collectivisation was like speaking to them in Greek’, as one local CNT
leader observed, Bosch Sanchez, Ugetistasy libertarios, p. 48.
97 Rural expropriation and agrarian collectivisation occurred most notably in Aragon and in the
Republican centre-south (New Castile downwards). (For Aragon, see discussion and note be-
low.) There was also a considerable amount in the provinces of Toledo and Guadalajara: see
Bosch Sanchez, Ugetistasy libertarios, p. 372 for useful comparative statistics and also J. Casanova
(ed.) El sueho igualitario. Campesinadoy colectivizaciones en la Espaha republicana 1936-1939 (Zaragoza:

Institucion Fernando el Catolico, 1988), passim. Onjaen see L. Garrido Gonzalez, Colectividades
agrarias en Andalucia: Jaen (1931-1939) (Madrid: Siglo XXI, 1979). There is substantial individual
memoir material on collectivisation both by Spaniards and non-Spaniards. This is of variable
quality, but with the proviso that such material always offers a ‘worm’s-eye view’ it also provides
invaluable insights into the passionate commitment and hope - also a condition of history -
which fuelled collectivisation as a cultural as well as a social and economic endeavour. For a
brief, useful summary of the historiographical debate see J. Casanova, ‘Anarchism, Revolution
and Civil War in Spain: The Challenge of Social History’, International Review of Social History 37
(! 992 ), 39 8-4 ° 4 -
102 The Spanish Republic at war

labour it had produced a much more explicit and focused hostil-

also
ity to the political, social and economic status quo. These perceptions
were also mediated by libertarian ideas, thus consolidating a collective
awareness of integral dispossession rather than just of poverty or political
inequality. The failed military coup, in provoking the collapse of state
institutions, thus bequeathed a real legacy of revolutionary social con-
flict. The emergency defence of 18—19 July offered workers the chance
to go further. They created a network of popular committees, collec-
tivised industry in significant quantities 98 and implemented a range of
cooperative experiments in the organisation of social life. In neighbour-
ing Valencia industrial workers resisted the coup with a general strike
but then used their leverage far more conventionally to try to gain im-
proved pay and conditions from their employers. 99 Set against this, the
specificity of urban Catalonia becomes clear. Moreover, the experiment
stretched beyond into the adjacent rural zone of eastern Aragon. Catalan
100
anarchists carried collectivisation to its villages. Thus, under the juris-
diction of libertarian Barcelona, Aragon became the agrarian hinterland
101
of Catalonia’s urban revolution.
In the rural south of Spain too, land was frequently collectivised by
villagers in the days after the coup attempt. 102 The UGT’s agrarian fed-
eration, the FNTT, was involved along with the CNT. In contrast to the
process in Aragon, collectivisation in the south had deep roots in the
pre-war period, driven by ideological opposition to the social, economic
and political effects of the latifundia. The fact that the land expropri-
ated belonged almost entirely to (often absentee) pro-rebel elites made
collectivisation a less immediately fraught political issue for the Repub-
lican authorities than was the case elsewhere, as did the involvement of
the UGT, which was much more government-friendly than the CNT’s

98 Most industrial and commercial collectivisationoccurred in urban Catalonia, with a concen-

tration in the Barcelona area; see chapter 4 below.
99 Bosch Sanchez, Ugetistasy libertarios, p. 29.
100
Casanova, Anarquismo y revolution, pp. 36, 218; W. L. Bernecker, Colectividades y revolution social.
El anarquismo en la guerra civil espanola 1936-1939 (Barcelona: Critica, 1982), pp. 251—2 (German
original, Anarchismus und Biirgerkrieg gfr Geschichte der Sozialen Revolution in Spanien 1936-1939
(Hamburg, 1978).
101
For more on Aragon, see chapters 4 and 5 below. For more on collectivisation therein, see
Casanova, Anarquismo y revolution and Anarchism and revolution in the Spanish civil war: the
case of Aragon’, European History Quarterly, 17 (1987); J. Casanova, El sueno igualitario\ G. Kelsey,
Anarchosyndicalism, Libertarian Communism and the State: The CNT in Zaragoza and Aragon (Amsterdam,

1991); P. Broue and E. Temime, The Revolution and the Civil War in Spain (London: Faber & Faber,
1972); important oral testimonies in Fraser’s Blood of Spain.
102
Garrido Gonzalez, Colectividades agrarias en Andalutia: Jaen (1931 -1939), pp. 27-37; Bosch
Sanchez, Ugetistasy libertarios, p. 372.
 Against the state 103

103
southern federations . Inevitably, however, given the chaotic aftermath
of the coup, the overall effect was one of serious fragmentation. Even in
the province ofJaen, where the FNTTs influence was predominant, col-
lectivisationwas occurring beyond the control of union leaderships, and
village committees were often hostile (as they also were in other regions) to
any interference from outside 104 On the other hand, there were \illages
.

where the coup seemed to induce paralysis and once committees had
105
established themselves, they did very litde Given that the immediately .

pressing need was for the collection of the harv est which was essential
106
to the supply of urban Republican zones both scenarios presented ,

problems. A significant proportion of the collectives in the south-west

would be felled rapidly in August and September by the rebels' military
advance. But those that remained, in the Republican south-east and in
Jaen, would soon be the object of concerted government efforts to impose
107
state regulation and control .

From this brief survey of developments across Republican territory it

is clear that the local forms of social and economic reorganisation born
out of the emergency defence were everywhere highly atomised. This
reflected the tendency of most people to identify exclusively with their vil-
lage or neighbourhood of origin patiia chica) - a normal consequence in
(

an underdeveloped state which lacks an entirely unified, interdependent

economy. Committees and emergency defence did not everywhere pro-
duce collectivisation, however. And where it did occur, urban or rural, it
108
too was fragmented, a heterogeneous and uneven mix of forms It was, .

moreover, frequently beyond the control of even those regional UGT and
CNT leaderships struggling to articulate it. In Aragon, as we have seen,
agrarian collectivisation was impelled by urban-based cadres from neigh-
bouring Catalonia. And virtually everywhere it was developed, apart
from in the Republican south, the mov ement lacked a strong pre-war
109
collectivist tradition Heterogeneous and highly ‘invertebrate’, most of
.

103
The FNTT represented the interests of the landless in search of land. How ever, the UGT’s
involvement also points to another of opinions which had historically fed into the debate
set
around southern collectivisation: those of specialists (economists and agronomists) w ho saw it 7

as a means of rationalising southern agriculture as part of a plan to stimulate and modernise

the Spanish economy.
104
Thomas, Spanish Civil liar. pp. 305-6; Abella, La vida cotidiana durante la guerra civil, p. 89.
105
Jaen examples in Thomas, Spanish Civil War, pp. 306-7.
106
Garrido Gonzalez, Colectividades agrarias en Andalucia: Jaen (1931-1939), p. 57.
107
For a discussion of this, see chapters 4 and 5 below 7
.

108
Bosch-Sanchez, Ugetistas y Ubertarios pp. 31-9; Fraser, Blood of Spain,
, p. 232 (n. 3): Thomas.
Spanish Civil War, pp. 306^-7.
109
For a rare exception in the Valencia region, see Bosch Sanchez, Ugetistas v Ubertarios . p. 36.
104 The Spanish Republic at war

these local initiatives were only made possible by the paralysis of the state,
which, for a time, also paralysed the social opposition to collectivism.
Even in the one possible exception to this picture of contingency -
urban Catalonia — the fact that revolutionary forms of social and eco-
nomic organisation had the opportunity to develop derived in significant
measure from the region’s considerable distance from the front line of
military defence in 1936. Barcelona was the Republican city farthest from
the military front. The fact that were impelled
its collectivist initiatives
by radical libertarians certainly did not help infuse them with the need
for any greater economic or political articulation. But it was also more
generally true that outside Madrid and the central zone there was still no
sense of the need for centralisation. This was because there was still no
real consciousness of ‘the war’. Thus, while in some areas of rebel Spain
the first waves of conscription were being organised within three weeks
of the coup, in this same period in Republican territory we cannot refer
in any meaningful sense to a single ‘Republican’ war effort, still less to
a single goal. The energy, enthusiastic improvisation and heterogeneity
of the emergency defence were initially its strength, but they would soon
come to symbolise its underlying weakness.
What would ‘bring the state back
in’ to the Republican equation was

the encroaching experience of the war in the south. The Republicans

would soon be confronting much more than a series of ill-coordinated
and (until then) only very partially successful garrison revolts. Early
on 19 July, as Franco himself arrived in Tetuan, a 200-strong contin-
gent of troops from North Africa (indigenous troops regulares com- ( )

manded by career officers (Africanistas)) landed at the mainland port

of Cadiz in the far south-west. The workers’ movement there had de-
clared a general strike, offering fierce resistance to the African veteran
and Carlist sympathiser General Jose Enrique Varela and his supporters
110
in situ. The arrival of the Moroccan troops not only assured Varela’s
victory at Cadiz, but also underwrote the rebel order in the whole area
of Cadiz-Algeciras-La Linea. This African contingent - supplemented
by a few other relatively small-scale transfers from Morocco 111 - would
also participate in the repression of popular resistance in Seville.

110
Cadiz was apparently called ‘Rusia Chica’ (‘Litde Russia’) by the right because of the strength
of socialist support there. On the rising in Cadiz, see A. Garrachon Cuesta, De Africa a Cadiz y de
Cadiz a la Esparia Imperial (Cadiz: n.p., 1938); F. Espinosa Maestre, La justicia de Queipo. (Violencia
selectivay terror fascista en la II Division en 1936) Sevilla, Huelva, Cadiz, Cordoba, Malaga y Badajoz

(Seville: Centro Andaluz del Libro, 2000), pp. 57-72.

111
Preston, Franco p. 152;
,
M. Alpert, Laguerra civil espano la en el mar (Madrid: Siglo XXI, 1987), p. 86.
 Against the state io5

More extensive troop transportations were supposed to follow since

the rebels intended to use the Army of Africa (including Spain’s Foreign
Legion - the tercios) as their shock troops to extinguish popular resistance,
thus guaranteeing their control. But the insurgents’ plans were thwarted
by the rebellion of the navy crews (on the main fleet of warships sailing
112
south towards Algeciras) against their pro-rebel commanders. The
Straits were effectively blocked." 3 This meant that the rebels’ southern
advance was also blocked as they could not now get their crucial pro-
fessional forces to the mainland. At the same time, the levels of popular
resistance the rebels had met meant that their bid for power had been
aborted in many areas, and Mola’s northern advance, also hindered by a
lack of munition, was halted at the Guadarrama sierra outside Madrid.
It was clear to the insurgent leaders that as the balance of power stood,
a rebel victory seemed a remote possibility.
It was at this point on 19 July that Franco, coordinating the southern
campaign from Tetuan, called for assistance from Italy and, in view of
Mussolini’s initial refusal to supply transport planes, a few days later on
22 July Franco petitioned Hitler directly. The provision of aircraft by
Hitler" 4 (and simultaneously by Mussolini, who had revised his opinion
by 27 July" 5 ) to fly the Army of Africa to the mainland effectively gave
which to turn a foundering coup into a war. By
the rebels the forces with
the end ofJuly there was an air ferry of troops from Morocco to Seville
which in ten days saw 10,000 troops transferred. (By 5 August there
would also be troop ships crossing the Straits under Italian air cover.
The Republican navy could do little to stop this - causing significant
Loyalist demoralisation - since its ships were debarred from refuelling

112
D. Sueiro, Lajlota es roja (Barcelona: Argos Vergara, 1983). There had also been a similar sailors’
rebellion in the south-eastern port of Cartagena. J. Martinez Leal, Republica y guerra civil en
Cartagena (1931-1939) (Murcia: Universidad de Murcia, 1993), pp. 169-85.
113
In fact a trickle of African troops continued to come across the Straits from Tetuan in small
transport planes: Preston, Franco p. 154. But this
,
was neither substantial nor rapid enough to
provide the insurgents with a mainland army.
114
Ibid., pp. 156-62. Hitler sent(in what was known as ‘Operation Magic Fire’) thirty Junker

JU-52 transport Franco had approached Hitler directly after initial requests direct to
aircraft.
the German Foreign Office had been rebuffed; see P. Preston, ‘Mussolini’s Spanish Adventure:
From Limited Risk to War’, in P. Preston and A. Mackenzie (eds.), The Republic Besieged: Civil
War in Spain 1936-1939 (Edinburgh: Edinburgh University Press, 1996), p. 21.
115
For a variety of reasons - mainly to do with his understanding of Britain’s pro-rebel position
and an awareness that France and (initially) the Soviet Union would not aid the Republic -
Mussolini reversed his initial decision. He did this in ignorance of Hitler’s decision to assist
Franco, even though the Duce’s decision was taken at more or less the same time as Hitler’s:
Preston, ‘Mussolini’s Spanish Adventure: From Limited Risk to War’, in Preston and Mackenzie,
The Republic Besieged.
 o6 The Spanish Republic at war

or using the port facilities at Gibraltar by the British authorities 116 there
(as they were also debarred from Tangier in spite of its free port status 117
and further harassed by the presence of German warships patrolling the
Moroccan coasts). The Germans also sent some Heinkel fighters and
volunteer pilots and mechanics from the Luftwaffe. Within a week of
petitioning, the rebels were thus receiving regular supplies of armaments
and ammunition from both Nazi Germany and Fascist Italy. Between
the end of July and October 1936 868 flights were to carry nearly 14,000
men plus artillery and 500 tons of equipment to mainland Spain. 118
Courtesy of their fascist suppliers, the insurgents were now escalating
their offensive against the Republic. They would strike with all the force of
superior firepower and technological advance that their foreign backers
could provide. It was to be the agrarian south, inevitably, which first felt

the full blast of the escalation.

Much of the south had been held by the Republic in the initial phase
of the rebellion (see map
1). Indeed, a significant portion would never

be conquered by the insurgents: the south-east coast, from

militarily
Alicante down through Cartagena to Almeria (rebel conspiracies were
defused or resisted in all three 119 and the interior down to and including
)

much of Jaen province (where no rising occurred) 120 remained Republi-

can until the surrender at the end of March 1939.
The highly conservative city of Granada, 93 kilometres south ofJaen,
to the military rebels on 20 July, although fierce residual resistance
fell

continued for several days in the working-class district of the Albaicin,

which was bombed and shelled into submission. A ferocious repres-
sion followed. The military authorities gave the Falangist death squads
free rein to liquidate the left extra-judicially. The paseos on which the
116
The Franco to make a formal request that Repub-
British naval authorities actively invited
lican shipsbe excluded both from the port facilities and from Gibraltar’s territorial waters.
E. Moradiellos, ‘The Gende General: The Official Perception of General Franco during the
Spanish Civil War’ and P. Preston, ‘Mussolini’s Spanish Adventure: From Limited Risk to War’,
both in Preston and Mackenzie, The Republic Besieged p. 4 and p. 38 respectively.
,

117
The pro-Franco sympathies of Italy’s Minister Plenipotentiary in Tangier, De Rossi, who was
also Chairman of the Control Committee administering the port, ensured the exclusion of
Republican ships from Tangier: Preston, ‘Mussolini’s Spanish Adventure: From Limited Risk
to War’, in Preston and Mackenzie, The Republic Besieged, p. 31.
118
Preston, Franco, pp. 161-2.
119
M. Ors Montenegro, La represion de guerray posguerra en Alicante (1936-1945) (Alicante: Instituto
de Culturajuan Gil-Albert, 1995); V Ramos, La guerra civil (1936-1939) en la provincia de Alicante

(3 vols., Alicante: Biblioteca Alicantina, 1974), vol. 1, pp. 85-111. Martinez Leal, Republicay
guerra civil en Cartagena, pp. 169-85; R. Quirosa-Cheyrouze y Munoz, Politicay guerra civil en

Almeria (Almeria: Cajal, 1986), pp. 113-20; Thomas, Spanish Civil War, pp. 242, 251.
120
F. Cobo Romero, La guerra civil
y la represion franquista en la provincia de Jam (1936-1950) (Jaen:
Diputacion de Jaen, 1993).
1936

July

22

Spain

of

division

The

Map
io8 The Spanish Republic at war

Falangistas took their victims - who ranged from the poet Federico
Garcia Lorca through liberal professionals to labour activists, rank-and-
workers and Popular Front supporters of any kind - ended in ex-
file

ecution against the cemetery wall. 121 The scenes were such that they
drove the caretaker insane. Thousands met their deaths in the paseos
5
as the rebels directly removed
opponents while also ensur-
their ‘red
ing the submission of recalcitrant sectors of the population through this
non-exemplary, generalised terror. 122
Much of the south-west had been - the
initially held for the Republic
-
major exceptions being the isolated enclaves of Seville General Queipo
de Llano’s fief - and Cordoba. 123 (Though in the south, as elsewhere,
the insurgent military had to liquidate many army officers — includ-
ing high-ranking ones - who refused to rebel. 124 But once the highly )

trained and well-armed Army of Africa had reached the mainland en

masse there was little pro-Republican sectors of the population could do.
,

Although they were numerous, these civilian resisters had no military

training or experience and scant arms. (In fact, to speak of the southern
defenders of the Republic as ‘militia’ is rather misleading.) The African
Army troops swept out from Seville (capital) on a campaign of wide-
scale repression in the province. (The precedent here had been set in

121
I. Gibson, The Assassination of Federico Garda Lorca (London: W. H. Allen, 1979); R. Gil Bracero,
Granada: jaque a la Republica (Granada: Caja General de Ahorros de Granada, 1998).
122
All the rebels’ political opponents were described as ‘red’. But the term was also applied in-
discriminately to entire social constituencies - predominantly to the urban and rural working
classes, but also to Republican-identified intellectual and liberal professional sectors, including
regional nationalists in Catalonia and the Basque Country. In the post-war period ‘red’ came
to mean whomever means of removing either their lives
the rebel victors chose so to label as a
or their civil rights.
123
For the military rising and ensuing repression in Seville (capital), see A. Bahamonde y Sanchez
de Castro, Un aho con Queipo: memorias de un nacionalista (Barcelona: Ediciones Espanolas, n.d.
17938]), pp. 23-7 ;J. de Ramon Laca, Bajo la ferula de Queipo: como fue gobernada Andalucia (Seville:
Imprenta del Diario Fe, 1939), pp. 18-20; A. Braojos Garrido, L. Alvarez Rey and F. Espinosa
Maestre, Sevilla 36: sublevacion fascistay represion (Seville: Munoz Moya y Montraveta, 1990), esp.
pp. 2 1 1— 21; J. Ortiz Villalba, Sevilla 1936: del golpe militar a la guerra civil (Cordoba: Diputacion
Provincial de Sevilla, 1998); Espinosa Maestre, Lajusticia de Queipo pp. 73-117; I. Gibson, Queipo
,

de Llano: Sevilla, verano de 1936 (Barcelona: Grijalbo, 1986), pp. 80-92. There is also N. Salas,
Sevilla fue la clave: republica, alzamiento, guerra civil (1931 -1939) (2 vols., Seville: Castillejo, 1992),
vol. pp. 281-363, vol. 2, pp. 409-91 - although its hagiographical aspect in regard to Queipo
1,

de Llano leads the author to make some dubious assertions (see Espinosa Maestre, La justicia de
Queipo, pp. 46-7, 56-7, 95, 319 for a critique of Salas’ work). For Cordoba, see below.
124
J. Vila Izquierdo, Extremadura: la guerra az>z7 (Badajoz: Universitas Editorial, 1983), p. 57. ‘Conspir-
ators were, in the main, africanistasremoved from active commands in late February 1936: they
had in most cases to get rid of Azanista garrison-commanders before “pronouncing” against the
government’, R. H. Robinson, The Origins of Franco’s Spain (Newton Abbott: David and Charles,
1970), P- 376, n. 3; Cordon, Trayectoria p. 224; Thomas, Spanish Civil War, pp. 250, 266 (for a list
,

of the six generals and an admiral shot by the rebels).

 Against the state 109

Asturias in October 1934, when, at Franco’s initiative, they had been

used violently to repress the workers’ rebellion in the northern mining
125 and land
cuenca. )
The Republic’s agrarian reform was thus reversed
and power handed back to the latifundistas (owners of the large estates),
who often rode along with the army to reclaim their lands manu militari 126 .

Rural labourers were killed where they stood, the ‘joke’ being they had got
their ‘land reform’ at last - in
form of their burial plot. 127 In pueblos
the
across the rebel-held south there was systematic brutality, torture, shav-
ing and rape of women and mass public killings (of both militia fighters
128
and civilians - male and female) in the aftermath of conquest. Where
there was a particularly strong radical or collectivist tradition or where
there had been land occupations or militancy in the spring/ summer 1936
or after the rural landworkers’ strike ofJune 1934 or as a consequence
(though this more rarely) of the October 1934 revolt, the apoplectic
rage of a feudally minded ruling elite saw villages wiped off the map by
129
repression. (And when, at the end of the war, the repression was ex-
tended and institutionalised by the triumphant rebel forces throughout
125
As the war minister’s special adviser, Franco had effectively been in charge of the Asturian
repression: Preston, Franco p. 103. ,

126
Salas, SevillaJiie la clave, vol. 2; ‘El comienzo: la “liberacion” de Lora del Rio (1936)’, in Cuadernos
de Ruedo Iberico (Paris, 1975), pp. 46-8. For the case of Arahal (Seville), see Fraser, Blood of Spain ,

p. 158 (n. 1); Salas, Sevillafue la clave, vol. 2, pp. 623, 650-1; and Carmen Munoz, ‘Masacre fascista
en Arahal (Sevilla)’, Interviu, 91 (9-15 Feb. 1978), pp. 38-41 (of interest for its oral testimonies;
the figures cannot be corroborated). Latifundistas rode with the army columns across the south:
for Cordoba, see L. Collins and D. Lapierre, Or I’ll Dress You in Mourning (London: Weidenfeld &
Nicolson, 1968), pp. 85, 93-7. For the repression in Cordoba, see F. Moreno Gomez, La guerra
civil en Cordoba (1936-1939) (Madrid: Alpuerto, 1985), passim but esp. pp. 284-325; Espinosa

Maestre, La justicia de Queipo, pp. 119—24.

127
Zugazagoitia, Guerray vicisitudes, p. 84; F. Moreno Gomez, ‘La represion en la Espana campesina’.
in J. L. Garcia Delgado (ed.), El primerfranquismo: Espana durante la segunda guerra mundial (Madrid:
Siglo XX, 1989), p. 192.
‘Prelude to World War: A Witness from Spain’, Foreign Affairs, 21(1) (October
128
J. T. WTiitaker,
1942), 104-7 and {same author) We Cannot Escape History (New' York: Macmillan, 1943), pp. 111 —
14; E. Taylor, Assignment in Hell’, in F. C. Hanighen, Nothing but Danger (New York: NTC,
I
939), pp. 68-73; Buckley, Life and Death of the Spanish Republic, p. 235; M. Koltsov; Diario de la

(Madrid: Akal, 1978), pp. 96-7; Moreno Gomez, La guerra civil en Cordoba, p. 265
guerra espahola
(for women’s deaths see pp. 80, 86, 373); A. Braojos Garrido etal. Sevilla 36, p. 244. For incitement

to rape,cf. Gibson, Queipo de Llano, pp. 160-1, 431 and G. Brenan, Personal Record 1920-1970

(London, 1974), p. 297; Castilla del Pino, Preterito imperfecto, p. 196. For more on rape/other
punishment of republican women and its significance, see discussion and notes below. See also
below for the repression in Extremadura.
129
Moreno Gomez, La guerra civil en Cordoba (e.g., case of Palma del Rio, pp. 377-82 and continuing
into autumn in conquered territory - the case of Fuenteovejuna, pp. 438-43) and F. Moreno
Gomez, ‘La represion en la Espana campesina’, in Garcia Delgado, El primerfranquismo, p. 19 1;
F. Espinosa Maestre, La guerra civil Diputacion Provincial de Huelva, 1996).
en Huelva (Huelv a:
(Although isolated from the rest of Republican Spain by the rising at
Seville, Huelva, with its
‘red’ miners, had initially been kept for the Republic only to fall to the rebels after a delayed
rising by the Civil Guard. The large number of huidos (fugitives often operating as guerrillas)
iio The Spanish Republic at war

the south irfthe form of highly summary legal proceedings, rural work-
ers would be found guilty in mass ‘trials’ and executed - without any
apparent intended irony - for the crime of military rebellion.) The colo-
nial mentality permeating the rebels’ southern campaign is more then
amply demonstrated by Franco’s letter to Mola of n August. In the
context of explaining that the conquest of Madrid remained the mili-
tary priority, Franco stressed the need to annihilate all resistance in the
‘occupied zones’, especially in Andalusia. 130 As Army of Africa troops
under Varela swept south-eastwards to connect up Seville with the other
rebel enclaves of Cordoba and Granada during August and September
131
1 93 6, this process of removing the ‘dangerous element’ (‘elemento

peligroso’) continued. 132

But the main thrust of the advance from the south was in rebels’
- the capital, Madrid. They saw it as
the direction of the greatest prize
the hub of Republican resistance whose conquest would win them the
war. Franco, having landed in Seville on 2 August, directed the Army
of Africa’s troops. The bulk of these, under the overall command of
Lieutenant-Colonel Juan de Yagiie, a veteran of the Moroccan wars and
the most influential military supporter of the Falange, had begun the
march up towards the capital. Taking village after village as they went,
the columns left a trail of carnage and terror in their wake. 133 Ultimately
there was nothing the towns’ hastily organised defenders — still less an

who became the premier target of rebel repression was a result of the virtual impossibility
of escape, surrounded as Huelva was by hostile territory.) For later repression in Malaga and
(later again) in Jaen, see E. Barranquero Texeira, Malaga entre la guerra y la posguerra (Malaga:
Arguval, 1994), pp. 199-228 and Cobo Romero, La guerra civily la represion jranquista en la provincia

dejabv Richards, A Time of Silence pp. 130-1.

.
; ,

130
Preston, Franco p. 165. ,
Note weapons in
also Franco’s frequent requests to Italy for chemical
1 936-7 which reflected his earlier experiences in North Africa, A.Vinas, Franco, Hitlery el estallido
,

de la guerra civil (Madrid: Alianza, 2001), pp. 29-112, esp. pp. 109—12.
131
L. M. de Lojendio, Operadones militares de la guerra de Espaha 1936-1939 (Barcelona, 1940), p. 108;
J. M. Martinez Bande, La campana de Andaluda (Madrid, 1969), pp. 73ff. There was a failed

Republican attempt to reconquer Cordoba on 20 August (under Miaja): see Thomas, Spanish
Civil War, pp. 380-1, 490, 493, 494. Moreno Gomez suggests that Miaja’s military strategy was
deeply flawed, La guerra dml en Cordoba, pp. 368-9.
132
Moreno Gomez, La guerra civil en Cordoba, pp. 438, 463—4 (where he cites radio broadcasts by

a Franciscan priest on the necessary cleansing of red elements (‘it is imperative that we uproot
and destroy the poisonous and degenerate seed of marxism from the soil of the fatherland. Exile
is not enough: we have to obliterate it’). The result of this is described in Collins and Lapierre,
Or I’ll Dress You in Mourning, pp. 93-9.
133
Thomas, who remarks (quoting the Portuguese press) on the severe
Spanish Ciml War, pp. 373-4,
levelof repression - a thousand deaths at Almendralejo (including those of 100 women) on
6 August. Both Gerald Brenan and Herbert Southworth comment on the initial openness of
reporting of these southern massacres in the Portuguese press, Brenan, The Spanish Labyrinth
(1943; Cambridge: Cambridge University Press, 1990), p. 322 and H. R. Southworth, Elmito de
la cruzada de Franco, 2nd edn (Barcelona: Plaza & Janes, 1986), p. 218.
 Against the state II

atomised rural labour force - could do to protect Republican land and

labour reforms, or their cherished collectives, pitted as they were in open
country against lorry-loads of seasoned troops,artillery and German and

Italian airbombardments. They would fight desperately as long as they

had the cover of buildings or trees. But the defenders were not trained
in elementary ground movements or even in the care and reloading
of their weapons. Moreover, as reports of the atrocities committed by
Yagiie’s troops mounted, even the rumoured threat of being outflanked
was enough to send them fleeing, abandoning their weapons as they ran.
A vast army of refugees fled before Yagiie’s army northwards.
On August the Army of Africa reached Merida, an old Roman
io
town near Caceres (most of which was rebel-held from early on 134 ).
Merida was an important communications centre between Seville and
rebel-friendly Portugal. Its defenders went out to engage the oncoming
rebel troops outside the town, fighting ferociously in the battle for the
River Guadiana. This was the first serious opposition the Africanistas
had encountered. But the resisters could not hold them and they broke
through to the town. 135
Shortly afterwards, initial contact was made
with General Mola’s forces. The two halves of rebel Spain were thus
joined into what would come to be called ‘the National zone’ (‘la Espana
nacional’). In Merida, meanwhile, the executions began with those of
the entire defence committee. 136
Yagiie then turned west to capture the frontier town of Badajoz, the
capital of Extremadura. On 14 August his forces reached the outskirts
of the walled where the garrison commander was in charge of a
city,

small nucleus of soldiers and several thousand inexperienced civilian

resisters 137 - many armed only with scythes and hunting shotguns. The

inequality of the contest was increased by the fact that the defenders had
to put down a Ci\il-Guard mutiny which undermined their material,
energy and confidence just as they had to confront the besieging troops.

134
For the rising and repression see J. Chaves Palacios, La represion en la provincia de Caceres durante la

guerra civil (1936-1939) (Caceres: Universidad de Extremadura, 1995).

135
On the battle for Merida, seej. Chaves Palacios, La guerra civil en Extremadura (Merida: Junta de
Extremadura, 1997).
136
This included two women, Rita Aznar and Anita Lopez: Vila Izquierdo, Extremadura: la guerra
civil p. 46; Lopez is also mentioned by Hugh Thomas (apparendy also citing the Portuguese
,

press), Spanish Civil War p. 373 and by Victor Chamorro, Historia de Extremadura (6 vols., Madrid:
,

\lctor Chamorro, n.d. [1985]), vol. 5. Shirley Mangini’s more recent enquiries confirm Lopez’s
existence (and profession - a pharmacist) but no more. S. Mangini, Memories ofResistance. Women’s
Voices from the Spanish Civil liar (New
Haven/London: Yale University Press, 1995), p. 75.
137
Estimates for the number of militiamen range from 8,000 (Thomas, Spanish Civil War, p. 373)
to 3,000. Later estimates suggest that the original 3,000 from Badajoz grew by around another
2,000 as refugees fled the rebels, Vila Izquierdo, Extremadura: la guerra civil, p. 50.
 ,

1 12 The Spanish Republic at war

Nevertheless, the resistance was solid. A built-up city was a much harder
target than untrained fighters scattered around villages or in open coun-
try. It would take two assaults by artillery and bombs before Yague’s
shock -tercios could breach the city walls. Once they succeeded, however,
a savage repression ensued. was chaotic, indiscriminate
Initially, there
slaughter and looting in the streets by tercios and regulares (enraged among
other things because this first experience of solid resistance had caused
serious casualties amongst their own ranks). Later, the more system-
atic repression began. Falangist patrols stopped workers in the street to
check if they had fought to defend the town. They would rip back their
shirts to see if their shoulders bore the give-away bruising of rifle re-

The defenders were herded into the bullring- turned concentration

coil.

camp and machine-gunned in batches. After the first night the blood
ran ‘palm-deep’ according to the witnesses interviewed by American
whose famous report on the Badajoz massacre cata-
journalist Jay Allen,
pulted the Spanish war into newspaper headlines throughout Europe
138
and America . The shooting at Badajoz would continue for weeks (and
the provincial repression for months after). 139 No less an authority than
Yagtie himself would soon confirm the witnesses’ accounts of repression
when, interviewed by another American journalist, John T. Whitaker
(who accompanied him for most of the march on Madrid), he made
his - now famous - reply: ‘Of course we shot them. What do you ex-
pect? Was I supposed to take four thousand reds with me as my column
advanced racing against time? Was I supposed to turn them loose in
my rear and let them make Badajoz red again ?’ 140 Bodies were left for

138
blood was supposed to be palm-deep on the far side of the lane. I don’t
‘After the first night the
doubt it. men - there were women, too - were mowed down there in some
Eighteen hundred
twelve hours. There is more blood than you would think in 1,800 bodies.’ Jay Allen, report in
the Chicago Tribune 30 August 1936 - although the massacre was first reported by two French
journalists and by a Portuguese reporter, Mario Neves. The latter’s (censored) report appeared
on 17 August 1936 in the Diario de Lisboa. In 1982 Neves finally returned to Badajoz, where he was
Granada Television series The Spanish Civil War. Vila Izquierdo, Extremadura:
interviewed for the
la guerra civil, memoir is in La matanza de Badajoz (Badajoz: Editorial Regional
pp. 54-8. Neves’
de Extremadura, 1986) (Portuguese original, A chacina de Badajoz (Lisbon, 1985)). The volume
also contains the text of his original newspaper reports. According to Jay Allen’s original Chicago
Tribune article, war booty - gold watches and jewelry from the dead citizens of Badajoz - went
on sale in Portugal at bargain prices.
139 Contemporary journalist accounts refer to approximately 2,000 people killed in the initial
mass executions in the bullring. Recent area studies of the repression estimate that some 5,000
people were killed in Badajoz province. Chaves Palacio, La guerra civil en Extremadura F. Espinosa;

Maestre, Lajusticia de Queipo, pp. 161-87; J. Casanova, in Julia et al., Victimas de la guerra, pp. 77
and 194 for the repression in Badajoz during 1937-8, in the context of the ongoing war effort
and shifting position of military fronts.
140
M. Neves, La Matanza de Badajoz (Badajoz, 1986), pp. 13, 43-5, 50-1. Se also Jay Allen, ‘Blood
Flows in Badajoz’, in M. Acier (ed.), From Spanish Trenches: Recent Lettersfrom Spain (London: The
 Against the state ii3

141
days in the streets to terrorise the population and then heaped to-
gether in the cemetery and burned without burial rites. Simultaneously,
the ‘liberation’ of territory by the rebels would be celebrated by the re-

opening of churches, by masses and baptisms and other public religious

ceremonial.
The fall of Badajoz sealed the Republic off from Portugal while it

gave the rebels unrestricted access to the frontier with the power that
had been their first international ally. 142 From the beginning, Oliveira
Salazar had permitted the rebels to use Portuguese territory to link their
northern and southern zones. (The Portuguese police also repeatedly
returned refugees to certain death. 143 ) Indeed, access to Portuguese help
had been an important factor in Franco’s decision to forego the more di-
from Seville to Madrid across the Sierra Morena via Cordoba.
rect route
This had also wrong-footed the Republicans, who concentrated their
exiguous military defensive forces in the region (under General Miaja)
on the Madrid-Cordoba line. Now, with all the south-west coast from
Cadiz to Huelva and the entire land border with Portugal beyond un-
der rebel control, Yagiie’s forces continued from Badajoz up the roads
north-eastwards towards Madrid (see map 2).

The columns split for a time to cover roughly parallel routes. The
first route took one of the three columns through Trujillo to Navalmora
de la Mata
(occupied on 23 August). To the east lay the valley of the
River Tagus, which offered no serious natural obstacles. The collectives
formed after the March 1936 land occupations were easy targets on
whose members hard deaths were inflicted. Massacre at the hands of
the Army of Africa was, once again, the brutal lesson meted out to
those who had dared to challenge the socio-economic status quo. On the
route taken southwards through the Guadalupe mountains, however, the
remaining two Moroccan columns had a more difficult time. Here they
were met by Republican government troops from Madrid under General
Jose Riquelme. In Medellin part of one column came close to destruction

Cresset* Press, 1937), pp. 3-8; Whitaker, ‘Prelude’, pp. 104-6; J. J. Calleja, Tagiie: un corazon al
rojo (Barcelona, 1963), pp. 99-109.
141
This was a constant feature as the rebel columns moved up through the south towards Madrid,
although international press agencies usually censored this kind of detail. Various journalists’
testimonies about Santa Olalla and Talavera de la Reina are cited in H. R. Southworth, Guernica!
Guernica! A Study of Journalism, Diplomacy, Propaganda and History (Berkeley, Calif.: California
University Press, 1977), pp. 53, 420-1 (n. 69). See also Espinosa Maestre, La justicia de Queipo,
pp. 189-204.
142
Portugal’s unfettered support was taken by Mussolini to indicate Britain’s essentially pro-rebel
sympathies, which, in turn, spurred his own intervention: Preston, ‘Mussolini’s Spanish Adven-
ture: From Limited Risk to War’, p. 38.
143
These included the socialist deputy for Badajoz, Nicolas de Pablo.
i4 The Spanish Republic at war

Map 2 The advance of the Army of Africa August-October 1936

 Against the state i *5

at the hands of the Republican air squadron, organised by the French

writer Andre Malraux, in its first serious engagement. While it could
not challenge the faster Italian fighter planes that gave the rebels local
144
control of the air, the Republic’s overriding military weakness at this
stage remained its inexperienced militia.
Untrained in elementary ground movements, the militia were con-
stantly being outmanoeuvred by the tercios and regulares and forced to
retreat. In the harsh conditions of the barren Tagus valley on the ap-
proach to Talavera, the vulnerability of their fighters meant that retreat
was the only option for the Republican commanders. While the volun-
teers themselves still seemed to believe that their (undoubted) bravery
would find its own recompense, the government simply could not afford
to risk all their men in a general engagement. Unpreparedness and
lack of training forced constant Republican retreat all the way back
to Talavera itself. There, defensive positions were established to the
fore of some 10,000 volunteer fighters inside Talavera. But at dawn on

3 September the Moroccan columns surrounded the town and, having

taken the aerodrome and railway station on the outskirts, assaulted the
centre, overwhelming its defenders in street fighting. In a bare month the
rebels had advanced almost 500 kilometres. And now the last important
town between the rebels and Madrid had fallen.
The August defeats and ensuing repressions continued into Septem-
ber — terrible and seemingly inexorable. There was litde or nothing
unarmed or poorly armed and untrained workers and landless peas-
ants could do against the highly disciplined and well-equipped insurgent
forces. Their bravery was epic, but on all the evidence it was failing. Even
occasions of prolonged resistance were not that plentiful, and when they
occurred were usually based on natural obstacles or the advantage of
urban terrain. 145 But the rebels had already launched a war which could
not be won militarily unless the Republicans could meet and hold them
in pitched battle. Moreover, on each occasion that resistance was broken,

144
There were some Republican air raids in the early weeks of the war against rebel-held towns,
including Granada, Cordoba, Segovia and Valladolid. These resembled First World War raids
in terms of intensity and casualties caused. It was, of course, German and Italian firepower that
enabled the rebels to achieve levels of destruction on a par with those seen in the Second World
War.
145 ‘Militia units were able to put up sporadic resistance in some places by dint of the energetic effort
so, time and again this failed to prevent the flattening of the resistance
of their leaders, but, even
and ensuing disorderly retreat - notwithstanding which there could be many instances of militia
bravery during the fighting itself.’ This was the opinion of Vicente Rojo, future chief of staff of
the Republican Army, in V Rojo, Asifue la defensa de Madrid (Madrid: Comunidad de Madrid,
1987), p. 60.
1 1 6 The Spanish Republic at war

the, quite literally, terrible price paid had an ever more devastating (and
146
unaffordable) impact on Republican morale .

Terror had always been a key weapon in the Army of Africa’s ar-
moury In the aftermath of the military rising it was first seen in evidence
in the bloody repression of the working-class neighbourhoods of Seville
(capital). In one of these, Triana, Army of Africa soldiers 147 rounded up

all the men they found, knifing many of them to death in the streets .
148

Now in the ‘long-haul’ war in the south, mass terror was being deployed
to facilitate a rapid military advance on Madrid and to ‘pacify’ the con-
quered nothing might jeopardise
territory, consolidating rebel control so
that advance. But the rebel commanders’ deployment of mass terror was
about more than short-term tactics. While Badajoz was undoubtedly
a message aimed specifically at those in Madrid contemplating resis-
tance 149 in targeting specific social sectors en masse — whether or not they
,

were active combatants — the rebels were in fact redefining ‘the enemy’
as an entire social class - the proletariat produced by modernisation and
perceived (by the rebels and their elite civilian backers) as ‘out of control’.
What occurred in the killing fields of the south was highly visible be-
cause of the presence of numerous foreign war reporters. The scale of
the immediate repression also marked out the south. But it is important
to remember Dantesque repression was simultaneously
that a similarly
being enacted everywhere in rebel-held territory Moreover, it was happen-

ing in places controlled by the rebels from the outset: where there was no
objective military threat, no significant political resistance and no Army
of Africa — in short, where one would be hard-pressed to find a ‘war-
150
situation’ at all . Nor is it feasible to argue that much of the initial

146
For the cumulative psychological effect of this on the militia’s capacity to contribute to the
defence of Madrid, see Cox, Defence ofMadrid, p. 70.
147 These soldiers belonged to a small advance contingent flown across in a Fokker from Morocco.
148
Some women were also taken prisoner. There are eye-witness accounts in Braojos Garrido et al.,
Sevilla 36, pp. 2 1 1 -2 1 (La toma de los barrios populares). A
laconic reference to the ‘pacification’
of Triana is in the account of rebel journalist M. Sanchez del Arco, El sur de Espaha en la reconquista
de Madrid (Seville: Editorial Sevillana, 1937), pp. 31-2.
149 Through both the refugees and the French press, news of the massacre soon spread throughout
the whole Republican zone: Cordon, Trayectoria, p. 256.
150
It is regularly adduced that the southern repression was uniquely attributable to a strategy of

war elaborated by the rebels in view of their own numerical exiguity in relation to the bulk of the
civilian population. Zamora is a good example of repression where the rebels were in control
from the outset: see P. Fidalgo, A Young Mother in Franco’s Prisons (London: United Editorial, 1939),
passim and R. Sender Barayon, A Death in Zamora (Albuquerque: University of New Mexico Press,
1989), PP- to6ff. In western (rebel-held) Aragon too there was a vicious repression out of all
proportion to the resistance offered: see J. Casanova et al., El pasado oculto. Fascismo y violencia
en Aragon (1936-1938) (Madrid: Siglo XXI, 1992); Cifuentes Chueca and Maluenda Pons, El
asalto a la Republica. Los ongenes delfranquismo en Zaragoza (1936-1939), pp. 44-83 and A. Cenarro,
 7

Against the state 1

1

5
violence stemmed, as in the Republican zone, from ‘uncontrollable
groups. For nowhere in the rebel zone was there a collapse of public
order. Falangists and other vigilante volunteers of the right could at any
time have been disciplined by the military authorities who underwrote
public order from the beginning. Not only did this not happen, but mil-
itary and civilian-instigated repression existed in a complementary re-
lationship. The military authorities were, thus, sanctioning widespread
151
terror throughout rebel territory. Indeed, the director of the coup,
General Mola, had both envisaged and announced the need for just
such an extensive application: ‘we have to terrorise, we have to show we
are in control by rapidly and ruthlessly eliminating all those who do not
think as we do 152 In both northern and southern Spain the terror had a
5
.

common political-strategic dimension: violent repression functioned as

public spectacle and threat, as a means of liquidating opponents which
would also enforce orthodoxy among those left alive, thus increasing the
level of social control exercised by the military authorities. 153 If anything
distinguished the south from the north, it was a quantitative, not a qual-
itative factor. In the rural south the social structure of the population
5
meant that there were far more ‘enemies to be dealt with. The presence
of a populous, professional killing machine in the Army of Africa facil-
5
itated the ‘solution ,
although Falangists and other right-wing civilian
volunteers played their part too. In the northern half of Spain the mili-
tary was short of personnel, which explains why, from the outset, Carlist
requetesand other right-wing volunteers played a more prominent role in
the killing process. It may also be possible to show that the rebel forces
doing the killing in northern parts of Spain had a different cultural per-
ception of their victims from those in the south. But they still saw killing 7

them as the solution. 154

Elfin de la esperanza: fascismoy guerra civil en la provincia de Teruel (1936-1939) (Teruel: Diputacion
Provincial de Teruel, 1996), pp. 67-91.
151
Cenarro, El fin de la esperanza, pp. 73—5; C. Garcia Garda, Aproximacion al estudio de la
represion franquista en Asturias: “paseos” y ejecuciones en Oviedo (1936—1952)', El Basilisco, 2
6
( ) (1990), 76
.

152
This principle was established in Mola’s series of confidential instructions initiated on 25 April
1936: G. Cabanellas, La guerra de hs mil dias (2 vols., Buenos Aires: Grijalbo, 1973), vol. 1,
pp. 304—5; F. Bertran Giiell, Preparation y desarrollo del alzamimto (Valladolid: Libreria Santaren,
1939), PP- 119-24. See also text of Mola’s ‘Bando de declaration del estado de guerra, 19 July
1936, in E. Mola, Obras completas (Valladolid: Libreria Santaren, 1940), pp. 1173—76.
153
Such as the sudden increase in church attendance among the urban liberal middle classes:
Castilla del Pino, Preterito imperfecta, p. 212 and c£ C. Barral, Anns de penitencia (1975; Barcelona:
Tusquets, 1990), pp. 75-6.
134
Whether there was a quantitatively greater rebel repression in the south overall (in terms of
the percentage of the total population of Extremadura and Andalusia) we cannot yet know
 , ,

1 1 8 The Spanish Republic at war

Also significant is the manner in which the ‘enemy’ so often met his or
her death at rebel hands: the mass public executions (sometimes with
the victims roped together) followed by the exhibition of corpses in the
streets for days, the mass burning of bodies 155 the quasi auto-da-fe of ,

a socialist deputy in the Plaza Mayor of Salamanca or the fact that

executions in the centre/ north of the rebel zone often took place on
established saints’ and feast days 156 Both physical and psychological
.

torture were also habitually inflicted on prisoners, and they were also
publicly humiliated - especially the women prisoners 157 All these forms .

of violence (in which I include the humiliation) were functioning as rituals

through which social and political control could be re-enacted. Thus the

definitively. One would need to be able to compare each and every Spanish province from 18
July 1936 up to and including at least three years of ‘post-war’ (i.e. post-1939) repression. At
the moment (March 2002) any conclusion is provisional, as only half of Spain’s provinces have
been researched (and some, to date, only studied in part): see Julia et al., Vktimas de la guerra
pp. 407-12 for an explanation of the state of current research and tables of findings to date.
The known figures for the south - and especially Badajoz - are very high. But as yet we lack
any figures for Galicia or for Castilian provinces such as Guadalajara and Cuenca, where the
indications are of heavy repression.
155
The exhibition and burning of corpses took place most infamously at Badajoz, but both occurred
in other places (north and south), and everywhere mass graves were to be found. Sender Bar ayon,
A Death in Zamora (bodies at the roadside, p. 137; common graves, pp. 149-50, 155, 162, 163);
Garcia Garcia, ‘Aproximacion al estudio de la represion franquista en Asturias’, pp. 69-82
(includes information on different kinds of paseo, on common graves, bodies on the streets and
razzias (searches) of working-class areas).
156
The Salamanca incident (involving the death of Andres y Manso in the first days after the rising)
is described in a letter from Julio Alvarez del Vayo (16 April 1937) in the Archivo de Barcelona
(AB), Azana’s correspondence RE 135 (carpeta 11 (5)) and L. Gonzalez Egido, Agonizar en
Salamanca. Unamuno, julio-diciembre 1936 (Madrid: Alianza, 1986), p. 82 Also the similar execution
.

of a socialist leader in Calatayud was reported in Heraldo de Aragon (Zaragoza) and then by El
Socialista (Madrid): R. Abella, La vida cotidiana durante la guerra civil p. 38. There are many accounts
reporting the auto-da-fe atmosphere which permeated rebel Spain: see A. Ruiz Vilaplana, Doy
fe. . .un ano de actuacion en la Espaha nacionalista (Paris: Editions Imprimerie Cooperative Etoile,
1938); various authors (inc. Flory), Galice sous la botte de Franco (Paris, 1938); A. Bahamonde, Un
ano con Queipo (for Andalusia). For the south see also Moreno Gomez, La guerra civil en Cordoba,

p. 287. Such spectacles continued well into the war, for example the ‘terror and fiesta’ in the
main square (called the Plaza del Torico) of Teruel (Aragon), Cenarro, Elfin de la esperanza, p. 75
and Garcia Garcia, ‘Aproximacion al estudio de la represion franquista en Asturias’, pp. 74-5.
On public executions during religious festivals, see J. de Iturralde, El catolicismo y la cruzada de
Franco (Bayonne, 1955), pp. 88-9, 93 (the festival of the Virgen del Sagrario in Pamplona);
Altaffaylla Kultur Taldea, Navarra 1336. De la esperanza al terror (2 vols., Tafalla: author-editor,

1992) and Fraser, Blood of Spain, p. 165.

157
See Fidalgo, A Young Mother and p. 24 for a variation on the ley de
in Franco’s Prisons, passim
jugas - for which see also Munoz, ‘Masacre en Arahal (Sevilla)’, p. 38; Garcia Garcia,
fascista
‘Aproximacion al estudio de la represion franquista en Asturias’, pp. 74-5, 82. For ritual public
humiliation of Republican prisoners, see also Espinosa Maestre, La justicia de Queipo, pp. 209,
217-22; S. Ellwood, ‘Spanish Newsreels 1943-1975: The Image of the Franco Regime’, Historical
Journal of Film, Radio and Television, 7 (3) (1987), 230-1; Braojos Garrido et al., Sevilla 36, p. 244;
Castilla del Pino, Preterito imperfecto, p. 205. On the rape and other punishment of women, see
discussion below.
 ,

Against the state i l

9

violence was also a form of exorcism of the underlying fear of loss of

control which was the subconscious linkage uniting the military rebels
158
with their various groups of civilian supporters .

All those against whom violence was done belonged to groups whose
accelerating political mobilisation (channelled by the Republic) consti-

tuted a threat to the ultra-hierarchical, monolithic national order on

which Spain’s past (and future) national ‘virtue’ and greatness were
perceiv ed as being founded. (Local studies of the repression demon-
strate quite clearly that those targeted the length and breadth of rebel
Spain were precisely those on whom the Republic’s reforming legisla-
tion had conferred and political rights for the first time in their
social
lives. )
No area of Spain was exempt from this threat, and thus no
159

160
area could be exempt from the ‘cleansing’ repression The dialectic .

of fire and sword, the necessary suffering of the ‘heretic’ is reminiscent

7

161
of Counter-reformation v alues and norms But in the interior land- .

scape of rebel leaders and many of their followers these melded with
other more ‘modern’ discourses of disease and racial impurity in which
the Republicans’ ‘marxist barbarism’ was explained as a lethal virus, the
germ of ‘anti-nation’ which if not ‘cleansed’ out to the last trace, would
contaminate the healthy body of ‘Spain’. Disease equalled disorder and,
more significantly, vice versa 162 Likewise, the widespread complicity of
.

158
Also suggestive here - if indirectly so - are the anxiously repeated references juxtaposing
Africanista terror and the rebel army’s restitution of ‘tradition’ in Sanchez del Arco, El sur de
Espana en la reconquista de Madrid: see, for example, p. 164.
159 While unskilled urban and rural labourers bore the brunt of the repression in the south, in other
zones of distinct social composition there was still a notable level of repression falling on various
kinds of leaseholding peasantry who had disputed the terms of their leases under the impetus
of Republican leglisation: Cenarro, Elfin de la esperanza pp. 88-9; J. Casanova, ‘Guerra civil
^Lucha de clases? El dificil ejercicio de reconstruir el pasado’, Historia Social, 20 (1994), 135-50.
160
The lexicon of cleansing and ‘public hygiene’ is also to be found in the Republican zone,
although not in the discourse of the state/ political leaders of the Republic: J. Casanova, in Julia
et al., Victimas de la guerra civil, p. 70.
161
The expectation that the Republican population should suffer is repeatedly pronounced by
rebel cadres and supporters of all kinds. For example, the intelligence officer in Burgos who in
1937 in response to his Quaker interiocutors’ concerns about how there was significandy greater
material deprivation being endured by the Republican population commented that ‘just as soon
as we can get things cleaned up in the North then there’ll be more suffering there too’: Dan
West, ‘Needy Spain’, Reports from the field, vol. 2 (report authored Feb. 1938), FSC/R/Sp/4.
For an analysis of the political functions of ‘suffering’ and ‘penitence’, see Richards, A Time of
Silence.
162
See, for example, thecomments of two rebel press officers: first. Captain Rosales (on the taint
to Spain'sbloodstream which had come through the industrial cities of the coast), cited in
M. Richards, A Time of Silence, p. 62 and those of the Africanista Captain Gonzalo de Aguilera
(Conde de Alba de Yeltes), in Whitaker, ‘Prelude to World War’, pp. 107-8 (Whitaker judged his
social and political ideas to be typical of ‘scores and hundreds of others on the Franco side’).

(Whitaker also discusses Aguilera in We Cannot Escape History, pp. 108-9); see a*so Charles Foltz,
120 The Spanish Republic at war

priests in the denunciation, killing and torture of those deemed oppo-

nents has to be understood as an active element within this ideological
framework rather than a response to popular anti-clerical violence in
Republican territory. 163
In view of these discourses and where they led, it is probably too restric-
tive to categorise only the war in the south as colonial. Franco’s specific

reference to Andalusia as an ‘occupied zone’ has tended to highlight the

south. Certainly there was a reciprocal influence between latifundismo’s
feudal values and the colonial experience of the Africanistas (an inter-
change intensified by the contiguity of Africa and Andalusia). But above
and beyond this, the entire rebel project was constructed as a colonial
enterprise in which the target was Spain as a whole. 164
In the decades after the loss of the remnants of Spain’s empire in 1898,
the military elite developed an ideological identity for themselves as the
defenders of the unity, hierarchy and (thus) the cultural and political

Jr (who also considered Aguilera’s views typical of the landowning elite’s), The Masquerade in Spain
(Boston: Houghton Mifflin, 1948), p. 116; P. Kemp, Mine Were of Trouble (London: Cassell, 1957),
p. 50; Neves, La matanza de Badajoz, p. 60. Further testimonies in Southworth, Guernica! Guernica!,
pp. 50-3. Also P. Preston, ‘Slaves, Sewers and the Nationalist Uprising’ (unpublished essay on
Aguilera). Consider also the self-revealing anecdote that some among the rebel political class
believed that the Republicans were plotting to infect them with epidemic disease: Southworth,
Guernica! Guernica!, pp. 463-4 (n. 43). Likewise, General Mola’s later outburst (in April 1937,
to the Condor Legion’s General Sperrle) about the need to wipe out the industrial centres if
Spain was to be healed, in A. Vinas, Guerra, dinero, dictadura. Ayudafascistay autarquia en la Espaha
There were also tests conducted in 1938-9 on
de Franco (Barcelona: Critica, 1984), pp. 102-3.
Spanish Republican and International Brigade prisoners by a military psychiatrist, Antonio
Vallejo Nagera, in search of the ‘bio-psychic roots of Marxism’, A. Reig Tapia, Ideologla e historia:

sobre la represion franquistay la guerra civil (Madrid: Akal, 1986), p. 28; Richards, A Time of Silence,
pp. 57-8.
163
Kemp, Mine Moreno Gomez, La guerra civil en Cordoba, pp. 463-4;
Were of Trouble, pp. 76, 80;
Mario Neves, the Portuguese journalist who reported the
Fraser, Blood of Spain, pp. 116, 166;
Badajoz massacre, recalled similar comments by a priest: see Neves’ interview in The Spanish
Civil War (Granada Television, 1982), episode 2 (‘Revolution, Counter-Revolution and Terror’).

A similar picture obtained in the north: see Fidalgo, A Young Mother in Franco’s Prisons, pp. 14-15;
Sender Barayon, A Death in famora, p. 163; A. Cenarro, Cruzados y camisas azules. Los origenes del

franquismo en Aragon igg6-igqy (Zaragoza: Prensas Universitarias de Zaragoza, 1997), pp. 203-4
and (same author) Elfin de la esperanza, pp. 89-90; Garcia Garcia, ‘Aproximacion al estudio de
la represion franquista en Asturias’, p. 78. For continuing violence in the post-war period, see
M. Torrent Garcia, yQue me dice Usted de los presos? (Alcala de Henares: Talleres Penitenciarios,
1942) - Torrent’s behaviour is discussed by H. Southworth in P. Preston (ed.) Spain in Crisis
(Hassocks: Harvester Press, 1976), p. 272; and in El mito de la cruzada de Franco, pp. 320-1;
F. Arrabal, Carta al General Franco (Paris, 1972), p. 159; Preston, Franco, pp. 323-4.
164
The conquering rebel armies would enter northern cities in the same manner as those in the
south, in the spirit of razzia (a term of nineteenth-century North African derivation meaning a
raid or hostile incursion for the purposes of conquest and plunder), J. A. Sacaluga La resistencia
,

igjy-ig62 (Madrid, 1986), pp. 5-6. See also the description by a pro-rebel
socialista en Asturias

priest (Father Alejandro Martinez) of the army’s entry to Gijon (Asturias) in October 1937, when
they ‘sacked it as if it were a foreign city’: ‘it was as if a “certain species” had to be liquidated’,
Fraser, Blood of Spain, pp. 424-5.
 1

Against the state 12

homogeneity of Spain. 165 In the process, they effectively internalised

the empire: first, in the sense that metropolitan Spain itself became the

empire (according to the monarchical constitution, the colonies had,

anyway, been provinces of Spain), and second, in that many within the
166
military elite interpreted their defence of ‘Spain’ as an imperial duty. By
the 1 93 os this defence was directed against both disintegrative regional
nationalisms and the working class as the bearers of political change and
cultural difference.
The cultural difference which sparked the most vehement, pathologi-
cal loathing and, in turn, some of the most extreme cases of ritual, violent
humiliation was digression from rigid gender norms. Women who took
part in the armed resistance
(
milicianas and others identified as ‘red’ were
)

shot alongside men. Many were also raped before they were killed. More
frequently, defeat was branded upon ‘red’ civilian women by means of
shaving their heads and administering doses of castor oil - with the
inevitable consequences (although sometimes these punishments were
167
also the prelude to murder). The evident misogyny of rebel elites was
rooted in the same fear of losing control: hence the consistent displays of

165
For the impact of imperial decline and re/ degenerationist thought on the development of the
Spanish right’s ideology into the civil war period, see M. Richards, ‘Civil War, Violence and
the Construction of Francoism’, in Preston and MacKenzie, The Republic Besieged, pp. 197-239
and Richards, A Time of Silence.
166
For the concept of internal colonialism where rival nationalisms are developed by antagonistic
groups within a state, see M. Hechter inj. G. Kellas (ed.), The Politics ofNationalism (Basingstoke:
Macmillan, 1991).
167
There is a significant amount of women’s memoir material which attests to these consistentforms
women. See picture of women of various ages with shaved heads (Montilla,
of rebel attack on
Cordoba) in Moreno Gomez, La guerra civil en Cordoba, p. 93. Contemporary right-wing ac-
counts also make reference to shaving, for example journalist Cecil Gerahty’s The Road to
Madrid (London: Hutchinson, 1937), p. 95. References also in Collins and Lapierre, Or Til Dress
You in Mourning, p. 97 ;
Garcia Garcia, ‘Aproximacion al estudio de la represion franquista en
79-80; Fraser, Blood of Spain, p. 272. There are also reports of atrocities against
Asturias’, pp.
women (who subsequently became refugees) in Quaker correspondence, FSC/R/Sp/i file
1 (Barcelona 1936-7). Such punishments continued to be meted out by victorious Falangists
in village and town in the immediate post-war period. See also Y. Ripa, ‘La tonte purifica-
trice des republicaines pendant la guerre civile espagnole’, Identites feminines et violences politiques

(i936~i946). Les cahiers de ITnstitut d’Histoire du temps present, 31 (Oct. 1995), 39-51* Women po-
litical prisoners also continued to suffer rape during police interrogations. T. Cuevas, Carcel de
Mujeres (Barcelona, 1985) and translated as Prison of Women: Testimonies of War and Resistance in
Spain (Albany, n.y.: State University of New York Press, 1998); Giuliana di Febo, Resistenciay
movimiento de mujeres igg6-igy6 (Barcelona: Icaria, 1979), pp. 18-20, 88, 107; F. Romeu Alfaro,
El silencio roto: mujeres contra elfranquismo (n.p.: n.p., 1994), p. 40; C. Garcia, Las carceles de Soledad
Real. Una vida (Madrid: Ediciones Alfaguara, 1982), p. 97 mentions a gaol rape of which there
are many reports; cf. Fidalgo, A Young Mother in Franco’s Prisons, p. 22; J. Dona, Desde la nochey

la niebla. Mujeres en (Madrid, 1978; 2nd edn, 1993), p. 17 1; Richards, A Time

la carceles franquistas

of Silence, pp. 52, 55, 64. For women’s undocumented disappearances, see Cenarro, Elfin de la
esperanza, pp. 80, 89-90.
 1

122 The Spanish Republic at war

m %

pathological hatred manifested towards even the memory of the milicianas

168
in the post-war years .

women in general were obsessively reduced to their sexuality by

‘Red’
rebel commentators who thus projected their own fears onto the Repub-
lican enemy 169 . In the same vein, note the myth of rape vouchers sup-
posedly distributed as payment to the Republican militias. In fact, it was
Africanista officers themselves who came closest to realising this when
they paid Moroccan troops in war booty that included access to ‘red’
women. The contradictory layers of rebel pathology also led to peculiar
philosophical debates between officers over whether such women were
still not fundamentally ‘Spanish’ (i.e. ‘white’ in a different sense) which
made handing them over to the regulares a questionable action 170 Even .

within their own terms of reference this seems a pointless debate, since
Republican women were regularly subjected to severe physical abuse
171
(including rape) by assorted ‘Spanish’ sectors within the rebel camp .

All kinds of were imprisoned - from teenagers to seventy-

women
year-old grandmothers. Many were imprisoned with their babies, some-
times newborn. Conditions were infrahuman, and not infrequently the

168
Hate-filled commentaries accompanying pictures of milicianas were published in the post-war
Francoist press, including in Seccion Femenina women’s magazines.
169
For the extraordinary case of General Queipo de Llano’s sexual psychopathology - as manifest
in his Seville radio broadcasts - see Brenan, Personal Record, p. 297; I. Gibson’s Queipo de Llano
reconstructs the text of these broadcasts. On war, sexuality and loss of control, see also the
thought-provoking analysis in J. Labanyi, ‘Women, Asian Hordes and the Threat to the Self
Studies, 73 (1996), 377-87 - see esp.
1

in Gimenez Caballero’s Genio de Esparto Bulletin of Hispanic

,

p.382 (for his hysterical denunciation of Madrid as a whore and Medusa from the pulpit of
Salamanca Cathedral) and p. 385 (for Labanyi’s fruitful incorporation of Theweleit’s concept
of war as the ‘ultimate permissible “controlled explosion” of the self’).
170
See Whitaker, ‘Prelude to World War’, pp. 106-7 an d Taylor, ‘Assignment in Hell’, p. 61 for an
example of racist theories among southern landowners and rebel officers. At the same time, both
they and priests waxed lyrical about the cleansing services offered by the African troops, their
underlying racism buried here beneath the image of these troops as part of the larger imperial
enterprise embodied in the ‘Crusade’, A. Lunn, Spanish Rehearsal (London: Hutchinson, 1937),
p. 66;Sanchez del Arco, El sur de Espaha en la reconquista de Madrid, pp. 95, 165, 205, 248 (‘at the
hour of liberation [of the Alcazar of Toledo] women of Castile received from African hands a
bread as white as Communion bread [the war] was a Mudejar enterprise against the Asiatic
. .
.

hordes’ (p. 205).

17
War rape is a complex issue for which a growing theoretical and comparative bibliography exists.
But neither the Spanish Civil War nor post-war repression have yet been subject to such analysis
(although see Alberto Reig Tapia’s comments in Ideologla e Historia, p. 145 and Espinosa Maestre,
Lajusticia de Queipo, pp. 249-55). My purpose here, however, is purely to locate the subject in the
broad context of rebel pathologies. There was no comparable phenomenon of consistent, mass
physical abuse of women in Republican territory as contemporary pro-rebel commentaries
note: see, for example, Sanchez del Arco, El sur de Espaha en la reconquista de Madrid, p. 55 and
H. R. Knickerbocker, The Siege of the Alcazar (London: Hutchinson, 1937), p. 86. Subsequent
research also bears this out, including in respect of nuns: S. Julia et al., Victimas de la guerra,

pp. 140, 152-3. (Few female religious were killed, and their sexual intimidation was rare.)
 Against the state i 23

children died. Indeed, this seems to have been part of the punishment for
their gender transgression: one prison remarked that ‘red’ women offical

had forfeited their right to nourish their young 172 Nursing mothers were .

shot along with the grandmothers and teenagers. For the child survivors
the price of nourishment (via Falangist social welfare) involved what one
witness (herself a prisoner) described as ‘moral suffering: obliging or-
phans to sing the songs of the murderers of their father; to wear the
uniform of those who have executed him; and to curse the dead and
blaspheme his memory 173 ’.

In the end, what the military, Falangists and other rightist volunteers

did to Republican men and women responded to something other than

tactical necessity in a military conflict. The startling uniformity of the
degradation and objectification inflicted upon Republican prisoners -
and in particular the remarkable need of their captors to break not only
their bodies but also their minds before killing them (and, even where they
were not killed, to leave them, as it were, psychologically ‘reconfigured’

by their experience of prison/ repression) - was servicing the underlying

rebel project: to (re)build a homogeneous/ monolithic and hierarchised
society .
174
The rebels’ objective and the insistent desire to reach the inner
life of the ‘enemy’ correspond very closely to what, in another national
context, we would term fascist 175 It
. has been powerfully argued that
fascism is a kind of ‘colonialism come home’: in both there is a need
to subjugate - if not annihilate - the threatening ‘other ’. 176 Yet the
‘other’ — or outgroup - is also crucial to the process of national reorder-
177
ing . For the rebel (and later Francoist) project of national reordering,
became what the Jews were to that other,
the Spanish working classes
more notoriously renowned Volksgemeinschaft. ‘Biological’ racism is,

172
Fidalgo, A Young Mother in Franco’s Prisons, p. 28. 173
Ibid., p. 31.
174
For a strikingly similar attempt to refix the identities of political prisoners by coercing them
into signing declarations of ‘repentance’, see the case of post-civil war Greece in the essay by
Polymeris Voglis in M. Mazower (ed.), After the War was Over. Reconstructing, the Family, Nation and
ig4j~ig6o (Princeton, n.j.: Princeton University Press, 2000), p. 77.
State in Greece
175
Cf. H. Arendt, The Origins of Totalitarianism (New York, 1966), p. 245.
176
For fascism as ‘colonialism come home’ in relation to the ideas of both Hannah Arendt and
postcolonial theorists Fanon and Cesaire (whose phrase it is), see D. Stone, ‘Ontology or
Bureaucracy? Hannah Arendt’s Early Interpretations of the Holocaust’, European Judaism, 32
(2) n-25. (Although Arendt is herself referring to the common territorial expansionism of
(1999),
fascism and colonialism.) On the Spanish right’s internalisation of the empire, see H. Graham,
‘War, Modernity and Reform: The Premiership of Juan Negrin’, in Preston and Mackenzie,
The Republic Besieged, p. 178 (n. 40) and ‘Popular Culture in the Years of Hunger’, in Graham
and Labanyi, Spanish Cultural Studies, p. 238.
177
On the instrumentality of German antisemitism and the organisational function of outgroups,
see Arendt, The Origins of Totalitarianism, pp. 238, 226, 241 (respectively).
124 The Spanish Republic at war
• ,

after all, ultimately explicable as a cultural category: thus one would

always expect the specificities of a given national history and ‘tradi-
tion/ myth’ to construct both the outgroups and the arguments which jus-
tify their exclusion/ annihilation. Thus the crucial instrument of Franco’s

Volksgemeinschaft was the Castilian Catholic Church - both as an or-

ganisation and in terms of a particular construction of Catholicism whose
political function was to subjugate and exclude. 178 The project would re-
main confined within Spain. But no less than the German variant, it
required the brutal eradication of ‘disorder’ and the subjugation of dif-
ference and change. Like Nazism (and, in a substantially different socio-
economic context, Stalinism) the Spanish rebels used ‘massacres ... for
the purpose of establishing a circumscribed, rational political commu-
nity’. 179 On this reading, then, Mola’s war directives encapsulate not a
strategy but a pathology.
As Yagiie’s forces conquered Talavera de la Reina on 3 September, the
needs of an escalating war situation were forcing themselves acutely on
socialist, communist (and to a lesser extent republican) political leaders
in Madrid. They were learning a hard and crucial lesson - paid for
in blood by the thousands of militiamen and women who fought and
died in the south. The Republic could not now win unless it was able
to meet the rebels in pitched battle. meant - sooner or later -
And that
having to confront the cumulative material and technological expertise of
their Axis backers - including the most sophisticated military-industrial
180
complex of the day, the Nazi state gearing itself up for war.
The Republic, as the legitimately elected government of Spain, had
straightaway (19 July) attempted to secure war material for itself from
the western democracies. But initial promises from the French Popu-
lar Front government were withdrawn in the face of British disapproval

178
Catholic organisation and discourse provided the mediating device whereby the regime in-
tegrated and nationalised the bulk of the Spanish population: see A. Botti, Cielo y dinero. El
Espaha (1881-igyj) (Madrid: Alianza Universitaria, 1992).
nacionalcatolicismo en
179
Arendt, The Origins of Totalitarianism, p. 186. The repression carried out by the rebels in Spain
both during and after the civil war was vasdy more violent and extreme than anything which
occurred in Italy under Mussolini. We have known this empirically for some time. Yet few
analytical conclusions have yet been drawn. The first Spanish work to deal with the civil war
repression as a whole (as opposed to empirical local studies which limit themselves to quantifying
deaths), S. Julia et al., Victimas de la guerra civil, still implicidy treats rebel repression and popular
violence in Republican territory as if they meant the same thing. For the opposing view, see
A. Reig Tapia, Ideologla e Historia and (a brief encapsulation in) Cenarro, El fin de la esperanza,
pp. 68-9.
180
For the Republican (defenders’ and civilians’) culture shock faced by modern warfare (planes
and bombs), see Cox, Defence ofMadrid, pp. 15, 30, 32.
 Against the state !25

and opposition from the bulk of the French Radicals whom the social-
istpremier Leon Blum feared would jeopardise his domestic reform
programme if he insisted on official French military aid to the Spanish
181
Republic. In Britain, unease at the potentially destabilising impact of
conflict inSpain crossed into overtly pro-rebel attitudes in some quar-
ters of the policy-making establishment. (The Spanish Republic was
perceived as less capable than the rebels of guaranteeing capital and
property - not least in respect of significant British investment in Spain.
The fact that it was precisely the had
act of military rebellion itself which
provoked the violence and disorder that so shocked British diplomats and
political leaders did not, however, seem to register in these circles.) British
fears and hostilities obliged Blum, also acutely aware of France’s vulner-
able defences, to sponsor Britain’s preferred policy of ‘Non-Intervention’
By 24 August the diplomatic consensus for this was achieved
in Spain.
when Nazi Germany agreed to join (primarily to lock France into the
agreement and thus tie Blum’s hands). As the Axis powers were in
effect already intervening with impunity, what the policy amounted
to was an arms embargo imposed in practice exclusively against the
182
Republic.
As a result of this, the Republican leaderships were reduced in August
and September to scrambling for arms piecemeal through ad hoc pur-
chasing agents. For many reasons this was both a hideously expensive
and highly inefficient process - not least because the plethora of Spanish
buying agents, dispatched by the many Republican committees and thus
acting independently of each other, ended up competing to purchase
183
scarce material thus driving the already high prices higher still. All
this was, of course, a direct result of the organisational dislocation and
fractured communication channels caused by the rebellion. Yet given the
Republic’s near total international isolation, its ability to withstand the
Axis-backed insurgents - indeed, its very survival — depended absolutely
on coordinating its efforts and on maximising its internal resources. Nor

181
G. Howson, Arms for Spain (London: John Murray, 1998), pp. 21-6. The Radical Pierre Cot,
who was Blum’s air minister, supported the Republican cause.
182
The Non-Intervention committee set up to administer the agreement was inaugurated in
London on 9 September. Representatives were present from every European country' except
Spain itself, Portugal (which had formally agreed to Non-Intervention under great pressure from
its and Switzerland, which also accepted the policy but declined to compromise its
ally Britain)

neutrality by attending the committee’s discussions: M. Alpert, A New International History of the
Spanish Civil War (Basingstoke: Macmillan, 1994), pp. 40-61.
183
For more on the acute problems of procuring war material, see chapter 3 below.
126 The Spanish Republic at war

in the periocl from 1 8 July to early September did there appear to be

much prospect of foreign assistance for the Republic from any other
quarters. Not only were Britain and France tied into Non-Intervention,
but so too seemingly was the Soviet Union. Once it became clear that the
offer of French military aid was in great danger of stalling, prime minister
Giral hastily dispatched a desperate request to the Soviet Union. 184 But
although Stalin was prepared to send a few military advisers, he hesi-
tated before the Republican request for military hardware - fearful of the
destabilising potential of an escalating conflict in Spain, given the vul-
185
nerability of the USSR’s own frontiers. Faced with these hesitations,
and rebel advance, the Republic’s own initiative was clearly going to be
imperative. A single, overarching political structure - Republic-wide -
was the sine qua non for planning or implementing a single, coordinated
war effort. This was evident to most among the Madrid leadership -
anarcho-syndicalists included. But it was probably best learned by Juan
Negrin, the socialist deputy soon to become the Republic’s new finance
minister.
Meanwhile, the war was bearing down on Madrid from the south. It
was physically manifest in the tide of refugees who poured in having fled
before the rebel army. To the north at Guadarrama lay the rebels’ other
forces under Mola. On 23 August the military airfield at Getafe, on the
city’s perimeter, was bombed, and on the 25th that of Cuatro Vientos -

even closer. On 27-8 August the population of Madrid suffered their

first air raids - indeed, the first of their kind to occur anywhere in

Europe. Their occurrence led to the formation of house committees in

each residential block to implement basic civil defence measures (such as

184
The Republic’s emissary, the distinguished legal historian and former education minister
Fernando de los Rios, arrived in Paris on 23 July. Given the climate of embarrassment and
fear he encountered there in the parliamentary Socialist Party and the waves being made by
the conservative press, it cannot have taken him very long to realise how precariously placed
was the whole question of Republican aid: Howson, Armsfor Spain pp. 24-5. Giral’s request of
,

25 July to the Soviet government is reproduced in R. Radosh, M. R. Habek and G. Sevostianov

(eds.), Spain Betrayed. The Soviet Union in the Spanish Civil War (New Haven/London: Yale University

Press, 2001), p. 21.

185
F. Schauff, ‘Hitler and Stalin in the Spanish Civil War 1936-1939’, unpublished research paper
(presented at the Annual Conference of the Society for Historians ofAmerican Foreign Relations,
Toronto, 24 June 2000), p. 8. According to Schauff, ‘informally and during Stalin’s holidays it
was agreed that the Soviet Union would send instructors to Spain. At the beginning of August,
two were selected and a group of about a dozen Red Army officers were sent to Spain later
that month.’ But at the start of September, Litvinov, the People’s Commissar of Foreign Affairs,
informed the Soviet ambassador to Madrid that there would be no arms sent to Spain for the
moment. See also D. Smyth, ‘Soviet Policy Towards Republican Spain: 1936-1939’ pp. 92-3
and P. Preston, ‘Mussolini’s Spanish Adventure: From Limited Risk to War’, pp. 39-40 both in
Preston and Mackenzie, The Republic Besieged.
 Against the state 127

listening for sirens — the signal to go down to the cellars — or attempting to

ensure a black-out). Gradually, as the socialist and communist leaderships
in Madrid organised civil defence mobilisation, the war began to enter
popular consciousness. It was now an only too tangible reality - perceived

directly through personal experience and in the news/ rumours of inces-

sant defeatsborne by the refugees. On 21 September Yagiie’s troops
would take Santa Olalla, mounting a public execution of 600 militia-
men in the main street of the town: ‘they were unloaded and herded
together. They had the listless, exhausted, beaten look of troops who
5186
can no longer stand out against the pounding of German bombs.
The conviction was growing among the Madrid-based leaders of the
5
Republic that something was needed: the ‘apocalypse had to be or-
else
187
ganised. The defeats in the south and the aerial bombardments were
a constant reminder of the need for military preparation and popular
mobilisation, in both of which the Spanish Communist Party especially
would soon show itself to excel.
The scale of the challenge facing the Republic also made changes
in government essential. By September it was clear that Giral’s all-

republican cabinet could not command the political respect or support

of a broad enough spectrum of Republican Spain. Its credibility was

tainted by popular memories of earlier republican attempts to treat with

the rebels. Its authority was further undermined by the string of mili-
tary defeats and by public-order disasters. The loss on 3 September of
Talavera, to the south-west of Madrid, and of Irun on the border with
France at the western end of the Pyrenees, intensified the Republic’s
physical vulnerability. The inability of Irun’s defenders to hold on
munition-starved and bombed by Italian planes - meant that the rebels
acquired some thousand square miles of fertile farm land, densely popu-
lated, with many important factories. It was also a major strategic defeat,
cutting off the Republican Basque Country, Santander and Asturias from
France and the rest of the Republic. 188 Worker confidence in the repub-
lican leadership was reduced to virtually nil. But as great, if for different
reasons, were the anxieties to which middling social constituencies within

186
In the words of American journalist John T. Whitaker, ‘Prelude to World War’, Foreign Affairs,

21 (1-4) (Oct. 1942-July 1943), 105-6.

187
Andre Malraux’s famous metaphor in his novel UEspoir (Paris, 1938); see also Cox, Defence of
Madrid, p. 28.
188
Previously Spanish militia had been able to pass through France from the Catalan frontier to
the Basque side. For their part, the rebels could now travel by rail all the way from Hendaye
(France) to Cadiz on the south-west coast: J. M. Martinez Bande, La guerra en el norte (Madrid:
San Martin, 1969), pp. 91-2.
128 The Spanish Republic at war

itsterritory were still prey. ‘Bottom-up’ violence continued in the form

of paseos and sacas (the removal from gaol and assassination of impris-
oned rightists). On 23 August, as news of the mass killings at Badajoz
began to circulate in Madrid and as the bombers closed in on the capital,
there occurred a notorious loss of control in the Carcel Modelo (Model
Prison). In unclear circumstances a fire broke out inside, attracting a
hostile crowd to the prison. After panicked staff had fled, and before the
socialist militia, the Motorizada, could arrive on the scene, the building
was assaulted by the crowd and some 70 political prisoners (out of ap-
proximately 3,000 held) - including numerous prominent conservatives
and right wingers - were shot, more or less by ‘popular acclamation’. 189
As a result, the government immediately moved to establish a system of
popular courts (the Tribunales Populares) in an attempt to forestall any
recurrence of such devastation. 190
Republican-zone violence was thrice negative. First, it was prevent-
ing the minimal inter-class consensus necessary to create the coherent
structures essential for the successful prosecution ofan escalating war.
Second, in that was perpetrated by the least powerful sectors of society,
it

it remained very uneven and incomplete - in contrast to what was oc-

curring in rebel Spain - and thus left social and political opponents
in place, later to demand counter-measures (or to contribute to ero-
sive fifth-column activities). Third, and not least, the paseos and sacas
meant by alienating the liberal
that the Republic risked political defeat
capitalist democracies - something which a good many CNT leaders
understood just as well as did their republican, socialist and communist
counterparts.
The detonation of the military coup in July had blasted open the
fault lines running deep in Spanish society. It destroyed the fragile anti-
oligarchical Popular Front alliance of urban and rural workers with
middle-class sectors and thus precipitated a state crisis of unprecedented
proportions. Almost everywhere within the atomised regions and local-
191
ities (apart from the Republican Basque Country )
the military rebel-
lion had unleashed conflict between antagonistic social constituencies.
The coup had failed on its own terms - in that it did not achieve com-
plete political and territorial control of Spain. However, the level of

189
Zugazagoitia, Guerray vicisitudes, pp. 128-30; Borkenau, The Spanish Cockpit pp. 125-6; Fraser,
,

Blood of Spain pp. 175-6, Thomas, Spanish Civil War, p. 404. The Model Prison incident was the
,

most notorious, but both before and after there were also other sacas from both the Modelo and
Madrid’s other prisons: Porlier, Ventas and San Anton.
190
See chapter 3 below for discussion of Republican public-order policy.
191
See chapter 3 below.
 Against the state i 29

fragmentation thereby provoked had pushed the Republic far back from
a reasonable starting point for fighting the kind of war rapidly being
imposed on it. In retrospect, therefore, it is possible to see the coup as an
inadvertent strategic success. This was especially the case given that the
rebels - by the use of terror and with an authoritarian Catholic lowest
common denominator - would have a much easier time uniting their
fragments.
The challenge facing the veteran socialist union leader Francisco
Largo Caballero, appointed to replace Giral as prime minister on
4 September 1936, was, first and foremost, to legitimise Republican gov-
ernment in the eyes of the proletarian forces who had led the emergency
defence. This would permit the rebuilding of military and civilian state
structures capable of erecting an adequate front-line military defence,
while also articulating and mobilising the Republic’s internal resources
against the rebels and their fascist backers, and mending political fences
with the western democracies in whose eyes the Republic had been di-
minished by its loss of public, political control in the aftermath of the
July days. The support of Britain and France (or at least their genuine
was a basic condition of Republican victory. For unless the
neutrality)
embargo underlying Non-Intervention was lifted, it would be impossi-
ble, in the medium term, for the Republic properly to undertake the
enormous task of militaryand state reconstruction necessary to wage
the full-scale war now looming. In the mean time, while embargo was
a reality, it was imperative that the Republicans cement the home
front in order to ensure aswide as possible a social mobilisation of the
Republican population behind the war effort.
Such a mobilisation, in turn, however, required the establishing of a
new modus vivendi between the proletariat in arms and the Republic’s mid-
dling classes. Both represented fragments of social constituencies. The
loss of large sections of the landless proletariat in the rebel-conquered
and Oviedo,
south-west, plus radical urban centres like Seville, Zaragoza
meant the amputation of much of the social base which might have driven
a more radical wartime political agenda; while the middling-class sec-
tors still with the Republic were those of the historically separatist north

and confederal north-east (Vizcaya and Catalonia). The challenge for

the Republican government would therefore be how to inst J war
consciousness and, linked to that, an idea of the ‘necessary state’ in the
differing social constituencies which made up its base.
The opposition of some working-class sectors, but also strongly region-
alist middle-class constituencies, to the central state-building enterprise
130 The Spanish Republic at war

would lead to bitter, energy-diverting social and political conflicts in the

Republican zone which impeded and undermined the war effort. But
they could scarcely be avoided. A Republic that was in the process of
reconstructing (post-coup) its democratic forms could not, without an-
nihilating its own raison d’etre , solve this dilemma in the fashion of the
rebels: by the mass physical liquidation of the enemy or the blanket use
of terror as a weapon of social control.
 CHAPTER 3

Building the war effort, building the state for total war
(September igg 6-February igjy)

You who will emerge from the flood

In which we have gone under 1

On 4 September 1936 Republican president Manuel Azana formally

confirmed the appointment of a new government to replace Giral’s.
The new prime minister was the veteran socialist union leader Francisco
Largo Caballero, who also took charge of the war portfolio. His support-
ers had been arguing for some time that he was the only figure capable
of bringing the whole of organised labour, including the CNT, on board
the war effort. Prieto and his supporters in the PSOE also recognised
the truth of this and told Azana so — in spite of their own profound
political differenceswith Largo. The new cabinet comprised twelve mini-
sters. There were four republicans, including one from the Catalan
Esquerra. (Basque nationalist (PNV) representation would come later
2
in the month when Madrid agreed to a Basque autonomy statute. )
But the three Madrid-based republican representatives, including ex-
premier Giral himself as minister without portfolio, were very much on
the cabinet’s margins. Its core was provided by the Socialist Party, which
had six ministers. In addition, there were two from the Communist Party
(PCE) in agriculture and education.
Largo’s value as prime minister was chiefly a symbolic one: his lead-
ership gave an impression of political unity under left leadership which
was designed to pacify the radical mood of the Republic’s proletarian
defenders. Workers saw Largo as a guarantee of their interests and, in
particular, of no compromise with the military conspirators. 3 A cabinet

1
‘To those born later’, Bertolt Brecht.
2
Largo originally (5 Sept.) designated the PNV’s leader, Jose Antonio Aguirre, as his minister for
public works. There then followed negotiations over the ‘price’ of Basque participation. After
the central government had agreed to promulgate the Basque statute forthwith, Manuel de Irujo
joined the cabinet as minister without portfolio, Gaceta de Madrid, 26 Sept. 1936.
3 Graham, Socialism and War, pp. 55-9.
 .

132 The Spanish Republic at war

led byhim was thus the bnly viable way of rescuing the concept of gov-
ernment, which had suffered greatly as a result of the military rebellion
and subsequent republican temporising with the rebels. 4 Largo’s sym-
bolic importance thus overrode the considerable reluctance of President
Azana, the republicans and the parliamentary Socialist Party to accede to
left socialist leadership of the new cabinet. 5 They would all have preferred
a republican-led cabinet reinforced by PSOE participation - something
the Soviet leadership and for the same reason -
had also actively sought,
the more favourable impression such a government line-up would have
on western government/ policy-making opinion internationally. 6 But this
sort of coalition was vetoed by Largo at the start of September, just as
he had vetoed a republican-socialist government back in May. Largo’s
insistence had also carried the day over the inclusion of the Communist
Party in the cabinet. Stalin had opposed this, but Largo made his own
acceptance of the premiership conditional on it7 - fearful that, other-
wise, the PCE would gain political credit at socialist expense from the
freedom of opposition.
Largo was widely perceived among working-class constituencies as
a symbol and guarantee of left unity and proletarian ascendancy. But
it was in fact Largo’s underlying reformism which made it ‘safe’ for

Popular Frontist socialists and republicans to accept his appointment in

September 1936. 8 If nothing else, the experience of 1934—6 had made it
clear to Indalecio Prieto, his close friend and colleague Juan Negrin and
the rest of their, like-minded, collaborators in the parliamentary PSOE
that the socialist had no viable ideas or strategies of its own. (On
left

hearing that Largo was to form a government, Negrin remarked that it

was the victory of October 1934 and worse than if the rebels had taken
Getafe (the military airfield near Madrid). But this was certainly not
an expression of trepidation about any radical policy potential. 9 Negrin
was alluding instead to the likely international ramifications of such a

4 Koltsov, Diario de laguerra espanola,

p. 62.
5 Azana had no real input in Largo’s appointment except for formal endorsement. On parliamen-
and preferences, see Graham, Socialism and War, pp. 56-7
tary socialist rationale
6
Azana’s continuity in the presidency was a crucial source of legitimacy vis-a-vis international
observers. For the Soviet perspective, see A. Elorza and M. Bizcarrondo, Queridos camaradas. La
Internacional Comunistay Espana 1919-1939 (Barcelona: Planeta, 1999), pp. 308-12.
7 Bolloten, SCW, p. 12 1; Elorza and Bizcarrondo, Queridos camaradas, pp. 312-13.
8
On the socialist left see Graham, Socialism and War and ‘The eclipse of the socialist left 1934-1937 ’

in Lannon and Preston, Elites and Power in Twentieth- Century Spain and Julia, La izquierda del PSOE.
9 Zugazagoitia, Guerray vicisitudes, pp. 154—5; M. Anso, Tofui ministro de Negrin, p. 15 1; see also Pascua’s
comments in Alvarez Juan Negrin (Documentos), pp. 278, 281. The socialist left’s ideological sterility
,

and strategic bankruptcy were nowhere clearer than in the procedural techniques through which
they chose to continue waging war inside the Socialist Party while the Republic itself was fighting
for its life: Graham, Socialism and War, passim.
 Building the war effort !33

government line-up as well as to the party left’s political mediocrity and

strategic bankruptcy.) This deep and savage division inside the socialist
movement, however, meant that Largo’s cabinet was badly fragmented.
The three members of the UGT-identified left socialist contingent (with
the premiership /war, interior and foreign ministry portfolios) scarcely
communicated with the three parliamentary socialist ministers (Prieto
in the navy/ air ministry, Negrin in the treasury and the veteran socialist
Anastasio de Gracia, who was in charge of industry and commerce).
Such atomisation at the heart of the cabinet might have mattered
less if the left socialists had themselves formed a strong, coherent nu-

cleus for government. But this was far from the case. In terms of hard
policies, developments during 1934—6 had revealed that the socialist left
was in practice indistinguishable from the rest of the PSOE. Largo, a
veteran, reformist union leader, probably saw his own arrival at the pre-
miership as setting the seal on his camp’s organisational victory in the
internecine socialist conflict. With a few notable exceptions, the socialist
left derived from and identified with the trade union rather than the so-

cialist parliamentary party, and there was a certain tendency among the

left’s members - Largo included - to perceive political office in terms of

the institutional advantage which could accrue to the UGT. Largo was
nevertheless well-meaning, strong on patriotic duty and very concerned
to acquit himself honourably as prime minister. But he had litde grasp
of the magnitude of the task awaiting his government as the military
rebellion escalated to full-scale war in the south, and he would prove a
disaster as war minister.
Indeed, the PSOE split itself revealed just how litde real political ex-
pertise the party left could call upon. Largo’s choice of the ex-Radical
socialist Angel Galarza (only recendy affiliated to the PSOE) as his inte-
10
rior minister underlines this weakness. Galarza was widely perceived
as something of a political opportunist. Nor, in the event, would he
prove a successful or efficient interior minister. 11 In all, Galarza’s ap-
pointment did litde to reinforce left socialist coherence or credibility in
the cabinet. This was, moreover, all the more significant as the left’s third
ministerial appointee to the foreign ministry, Julio Alvarez del Vayo, was
moving away from his close personal allegiance to Largo towards the
Spanish Communist Party. It is likely that Largo’s political adviser, Luis

10
The alternative was the landworkers’ union (FNTT) leader Ricardo Zabalza: del Rosal, Historia
de la UGT, 526-7. Julian Zugazagoitia in Guerra y vidsitudes, pp. 152-3 tries to make
vol. 2, pp.
the best of it - but the implication even here is that Largo did not have a lot of options to choose
from.
11
Graham, Socialism and War, pp. 63-4.
i34 The Spanish Republic at war

Araquistain,Avas the left’s preferred candidate for the foreign ministry

portfolio. appointment was vetoed by President Azana, who re-
But his
garded Araquistain as the brains behind the May 1936 cabinet fiasco
(when Largo had blocked Prieto’s bid for the premiership - with deadly
results). After the war, Largo’s political-military secretary, Jose Maria
Aguirre, would claim that Araquistain had been vetoed as a result of
12
Soviet pressure over armaments. But the Soviet Union had not yet
decided on intervention and, irrespective of the Soviet opinion of
Araquistain, both republicans and parliamentary socialists were already
vehemently opposed to him. 13
Parliamentary socialists and republicans both sought a smaller, more
homogeneous Popular Front cabinet staffed by themselves, with token
Catalan and Basque representation and supported by the Communist
Party from outside the government. Such a line-up would have been
better able to address the urgent problems of military reorganisation and
state infrastructural rebuildingwhich were looming large by September.
Internal party political tensions would still have existed to some extent,
of course, but such a government’s political expertise and international
perspective and connections would have allowed it to compute more
rapidly the policy implications for the Republic of fascist intervention
and the war in the south. But such a Popular Front government would
not have had the confidence of the proletarian forces at the forefront
of the July defence - and most particularly the CNT’s cadres, whose
support was absolutely crucial to war mobilisation. Largo’s appointment
was, therefore, inescapable.

12
Aguirre in an undated (but post- 1945) note to Araquistain’s son (Araquistain correspondence,
legajoyi, no. 22a(AHN) and again in a series of articles in Informaciones 8-10 Nov. 1977, published
under the pseudonym A. Velez. Aguirre claims Largo told him (i.e. on 3 Sept. 1936) that del Vayo
was ‘a Soviet agent’. Bolloten cites this testimony without comment, SCW, pp. 123-4. But the
language suggests a suspicious amount of prejudice of hindsight. (Indeed, only a few pages earlier
(p. 120) Bolloten himself refers to the still functioning relationship between Largo and del Vayo

in September 1936 and elsewhere in his book gives a highly critical assessment of Aguirre, whom
Bolloten knew personally during the war, as a ‘young upstart . . . arrogant and inexperienced’
which must also cast doubt on his testimony, SCW , p. 353.) Azana’s memoirs make no reference
to the episode.
13
Stalin only finally made up his mind to send the Republic war material in mid September, Elorza
and Bizcarrondo, Queridos camaradas, pp. 314, 317, 322-4; Howson, Arms for Spain , pp. 124-15;
F. Schauff, ‘Hitler and Stalin in the Spanish Civil War’, unpublished research paper (presented
at theAnnual Conference of the Society for Historians of American Foreign Relations, Toronto,
24 June 2000), p. 8. While Araquistain’s left socialist political trajectory scarcely recommended
him to a Soviet leadership concerned to assuage France and Britain, the harsh view presented by
the Comintern representative to Spain, Vittorio Codovilla, must be understood in the context
of an attempt to deflect Comintern secretariat criticism from himself: Elorza and Bizcarrondo,
Queridos camaradas , pp. 315-16, 319-20.
 Building the war effort *35

Bringing the CNT

government was the biggest political chal-
into
lenge facing Largo Caballero’s government. The prime minister himself
was from the outset concerned to ‘implicate’ the CNT in governmental
responsibilities in order to defuse their potential criticisms of cabinet de-
cisions. Largo was anxious, as ever, about the UGT’s ‘credit’ /‘credibility’
and fearful of the CNT stealing a march over the socialist union now
that its leadership was fulfilling a governmental function. But at root,
the imperative of CNT inclusion in the Republican cabinet, even more
than Largo’s own appointment,
derived from the need to relegitimise
and consolidate Republican government and state functions so badly
eroded by the military rebellion. However, this internal political impera-
tive conflicted greatly with the political requirements of the international
situation. The - should anar-
overriding fear within the Popular Front
chists be included in the government - was of confirming the hostility
of the western democracies whose material support (and, first and fore-
most, the lifting of the Non-Intervention arms embargo) remained their
primary/immediate goal. Although not blind to these image problems,
the PCE was more flexible. The Comintern was clear about the need to
rein in the CNT in Barcelona, but it also appreciated that CNT support
would be crucial to the effective war mobilisation of Republican society 14 .

(Its references to the need to separate ‘good’ anarcho-syndicalists from

the ‘lumpen’ and ‘low lifes’ represent an initial attempt to skirt the polit-
ical dilemma posed by the CNT. 15 ) But neither parliamentary socialists

nor republicans were comfortable with the idea of anarcho-syndicalist

16
representation, and President Azana was vehemently opposed .

Indicative of this fear and unease was the fact that Largo initially of-
fered the CNT’s national committee just a single ministry in September.
Unsurprisingly, this was rapidly rejected by a majority of its regional
federations 17 as beneath the CNT’s (political) dignity and due, given its
size/ strength and the key role it had played in the emergency defence

14
The PCE reinforced with favourable comments all the signs of reformist ascendancy in the CNT:
for example the public recognition within the CNT of the need for centralisation to serve the
war effort, Guerra y revolucion en Espana, vol. 2, pp. 52-4.
15
Cf. Andre Marty’s report (10 Oct. 1936), Elorza and Bizcarrondo, Queridos camaradas, p. 327 and
cf. p. 340 and Guerra
y revolucion vol. 2, pp. 10-11, which insists on
,
the same manichean division
within the CNT.
16
M. Azana, Apuntes de memoriay cartas ,
de Enrique de Rivas (Valencia: Pre-Textos, 1990), p. 26;
ed.
S. Julia, ‘Presidente por ultima vez: la crisis de mayo de 1937’, in A. Alted, A. Egido
Azana en
and M. F. Mancebo (eds.), Manuel Azana: Pensamientoy accion (Madrid: Alianza Universidad, 1996),
p. 246.
17
Plenary session of regional federations held 3 September 1936, C. Lorenzo, Los anarquistas espanoles
y el poder 1868-iggg (Paris: Ruedo Iberico, 1972), p. 180.
136 The Spanish Republic at war

of the Republic. The safne national plenum which rejected the single
ministerial portfolio did, however, accept the principle of GNT participa-
tion in government. The fact that the text of the CNT resolution made
such participation conditional on the prior restructuring of government
and state along syndical lines should not deflect our attention from the
crucial importance of this decision, which was taken - albeit as a result
of much heated debate - only three and a half months after the apparent
victory of the purist anarchist currents at the Zaragoza congress in May
i93 6 -

Throughout September 1936 the GNT - in the form of successive

plenary meetings of its regional federations - continued to argue with a
lone voice that the Republican government should be reconstituted in the
form of a National Defence Council. The Popular Front groups would all
be represented on this, but majority control would be split between five
GNT and five UGT members. The CNT’s proposal met with steely resis-
tance, however, from all these other groups, including the UGT. Then,
on 24 September, the CNT accepted the dissolution of the Catalan
Anti-fascist Militia Committee - until then the most important source of
political authority in the region - whereupon the CNT’s representatives
18
entered the Catalan regional government effectively as ministers. The
precedent was thus set and on 18 October the delegates at the CNT’s
national plenum of regional federations finally agreed to allow partici-
pation in the central Republican government tout court something which ,

the CNT’s ultra-pragmatic general secretary, Horacio Prieto, had been

advocating since the formation of Largo’s government itself. 19
But even for the pragmatic syndicalists in the CNT, Largo needed
to offer more substantial representation in the cabinet for them to feel
able to justify participation before their own membership. In the ensuing
negotiation process Largo revealed himself as the most resolute cham-
pion of a conventional, bourgeois government structure. In this he was
in part undoubtedly concerned to calm British and French fears over
the Republic’s perceived radicalism. Alongside this one can also detect
the veteran UGT fighter concerned not to give any advantage to the
CNT. Most significantly, however, in his championing of the Popular
Front government mode, Largo revealed that the socialist left - for all
its radical rhetoric between 1934 and 1936 - was as comfortably an-

chored as the parliamentary wing of the PSOE in the ‘old liberal’ world

18
For events in Catalonia after i8July 1936 and an analysis of the CNT’s wartime political evolution,
see chapters 4 and 5 below.
19
Lorenzo, Los anarquistas espanolesy el poder, p. 185. (Lorenzo was Horacio Prieto’s son.)
 Building the war effort 137

20
of bourgeois political hierarchy and limited popular mobilisation. The
outbreak of war, in bringing Largo Caballero to power, fully exposed the
contradictions underlying the socialist left.

The UGT leadership had all along continued to share with its parlia-
mentary PSOE counterpart a very singular view of the socialist move-
ment’s ‘manifest destiny’, according to which it would inherit the state.
This implicit statism meant that the union’s national executive had been
as profoundly traumatised by the events of July-August 1936 as its party
counterparts. For the rebellion’s disintegrative impact took on the
its toll

UGT’s organisational hierarchy as much

on party political
as and state
structures. Local union committees acted autonomously even when they
claimed a UGT denomination. Neither Largo Caballero nor the rest of
the union leadership had any intention of supporting such localist initia-
tives, which significandy undermined their own control of the UGT. It is

in this context that one must understand the leadership’s energetic and
persistent calls from the start for the nationalisation of industry. Certainly
there were cogent reasons for this in terms of articulating an integrated
national war effort. But an equally crucial agenda here was ensuring the
rearticulation of the UGT itself and the recovery of leadership control.
21

The entire historical experience of the UGT, and in particular the in-
creasing identification with state power in the 1920s and 1930s, informed
the decision of Largo Caballero to pursue both aspects of this agenda
(i.e. the concentration/ rationalisation of war production and the Spanish

socialist movement’s consolidation of political power) from government

after September 1936 - rather than by making an alliance with the CNT.
But although Largo Caballero instinctively favoured a moderate po-
litical course, his premiership was marked by a certain sense of policy

limbo. For Largo was reluctant to undertake any ov ert or vigorous policy
for which he might be criticised by one or other popular consistency, and
especially where his action could be construed as endorsing the ‘old’ hi-
erarchies of power against the workers. The impact of this contradiction
would weigh heavily from the start on the Republic’s most imperative
task — the rapid reconstruction of its defensiv e military capacity.
The impact of the rebel coup had left Largo’s predecessor, Giral, with
no option but to enact the formal dissolution of the army. The degree
of organisational disruption caused by the rebellion varied from area to
area - the centre (Madrid) zone being relatively less affected. But nowhere
in Republican territory was it possible to deploy army units in the rapid

20 21
See chapter i above. Graham, Socialism and War, p. 185.
138 The Spanish Republic at war
.

and reasonably coherent way the rebels could. 22 Even in Madrid, when
the Republican government was using regular troops, these constituted
the remains of units dislocated by their commanders’ departure which
were then lumped together with militia forces and often put under the
command of any available lieutenant, sub-lieutenant alferez or sergeant, (
;)

even though they were often strangers to the men they were required to
direct. 23 And nowhere did Republican militarised forces exceed 2,000 at
the outset. Moreover, the massive distrust of the officer class per se which
the rebellion had caused everywhere among the Republic’s proletarian
defenders made a symbolic break with the old army structures cru-
cial to successful military mobilisation thereafter. Giral had appointed
a committee headed by Captain Eleuterio Diaz Tendero, 24 formerly
prominent in the progressive republican army organisation the UMRA
(Union Militar Republicana y Antifascista). This committee investigated
the political reliability of all those officers who remained at the Republic’s
disposal, removing those whose loyalty was materially in doubt. 25 But
the underlying conservativism of the Spanish officer corps would clearly
create an enormous - and in many ways irresolvable - problem for the
Republic at war in its attempts to ascertain military loyalty.
Even before Largo came to power, it had been realised by Giral and
others that any successful attempt at military rearticulation had to start
from the basis of the militia. Initially — on 3 August — Giral’s government
had decreed the creation ofa volunteer army on the French revolutionary
model - something both more conducive to the republicans’ ideological
framework and comforting to their sense of order than the perilously un-
26
predictable, multi-form militia units. But the militia - dynamised by the
sheer centrifugal force unleashed by the rebellion and buoyed up by the
popular anti-militarism it had magnified - simply could not be sidelined.
The protagonism of the militia thus required other measures designed
to lay the basis of centralised control. This process began with the decree

22
See Cordon, Trayectoria, p. 250 for some idea of the huge extent of the fragmentation the coup
caused in the officer corps.
23 Even in the Madrid military region it has been calculated that the forces remaining to the
republicans were 70 officers and 1,313 NCOs and soldiers out of a previous base of more than
828 officers and 10,425 NCOs and men. This was just not enough to allow the conventional
restructuring of an armed force. Alpert, El ejercito republicano p. 30;
,
Azana, Obras completas, vol. 3,

P-487 •

24 Cordon, Trayectoria, pp. 235-6; Alpert, El ejercito republicano, pp. 17, 367.
25 In the Civil Guard a high proportion of its - perhaps in excess of 40 per cent - were
officers
removed by the Republican authorities after 18 July 1936 - in spite of the corps’s initial role in the
suppression of the rebellion. Alpert, El ejercito republicano, p. 25 (and see pp. 20-32 for an estimate
of the military forces available to both sides).
26
Cordon, Trayectoria, p. 249.
 6

Building the war effort *39

of 8 August creating a new Inspectorate of Militias covering the central

zone, the south and the Valencian area. (The centrifugality unleashed
by the coup meant that both Catalonia/ Aragon and the north were still
beyond the reach of Madrid’s authority) The Inspectorate’s brief was to
end irregular or arbitrary militia requisitioning. By the end of the month
its operations had grown to such an extent that it had entirely superseded

the original volunteer battalion initiative, whereupon the government

formally recognised the militia as the building blocks of the new army. 27
By 1 August the Inspectorate had decreed conditions of service for
those militias which agreed to be regulated by it (food and io pesetas a
day for each miliciano). At one level this was welcome to militia members
because it rationalised and thus (in theory at least) speeded up what
was a notoriously haphazard system of payment and provisioning. 28
But standardisation was effectively the first stage in the process of
militarisation because it required the presence in each militia of an army
quartermaster’s (Intendencia) official to ensure the proper utilisation of
public funds. 29 The connection between payment and control was thus
established, and by the end of the month the government was already
threatening to withhold militia wages in those cases where the requisite
membership lists had not been forwarded to the authorities.
This militia system, already in the course of transformation, was thus
inherited by Largo with the premiership in September. In theory (and
rhetorically)he entirely approved the principle of accelerating the cre-
ation of a new Republican army (Ejercito Popular). The series of devas-
tating militia defeats in the south made the military case unanswerable.
Moreover, the initiative also offered a powerful means of reinstilling the
centralised political order he instinctively sought in the Republican zone.
In practice, however, there was often rather more ambivalence in Largo’s
manner of proceeding. He knew that endorsing the policies required to
achieve militarisation would inevitably bring him into head-on conflict
with some sectors of the organised working class, including even some
socialist sectors, with unavoidably negative consequences for his personal
reputation within the labour movement and the left as a whole. This is

27
Decree of 28 August, Gaceta de la Republica.
28
‘The miliciano can never be sure when he’s going to be paid, nor does he know who is paying
him - the (military) unit he’s in, his union, political party or the War Ministry.’ C. Contreras
(Quinto Regimiento), in Milicia Popular, 6 Sept. 1936, cited in Alpert, El ejercito republicano, p. 39.
29
M. Azana, Apuntes de memoriay Cartas, ed. de Enrique Rivas (Valencia: Pre-Textos, 1990), p. 68
refers to the mayhem in the early days of a huge excess of rations paid in excess of active
militiamen; see also Dieguez’s report to the Madrid Defence Council on 12 Dec. 1936,
I.

J. Arostegui andj. A. Martinez, La Junta de Defensa de Madrid (Madrid: Comunidad de Madrid,

1984), pp. 341-2.
 9

140 The Spanish Republic at war

no doubt one reason why Largo undermined the fundamental rationale

of militarisation established by his general staff by tacitly allowing the
CNT to form some homogeneously anarchist military units. 30 And if
this happened on the centre front, then clearly the chances of his taking

a resolute stand in the near future on the matter of the military incor-
poration of Catalonia and the Aragon front under central government
control were distinctly remote. Largo’s reluctance to take unpopular ac-
tion - although probably largely unconscious -- derived no doubt in part
from the same prickly sense of amour-propre that underlay his infamous
dogmatism. At root, Largo did not really understand either fast enough
or fully enough the imperatives of Republican militarisation. His lack of
ease with the idea also stemmed from the fact that he had not grasped
that what was being advocated - above all by example in the PCE’s
Fifth Regiment - was light years away from any attempt to resurrect an
ancien regime army. 31 This also underlay his reluctance to put his weight
behind an effective programme of military conscription. A decree mobil-
ising all able-bodied men aged between twenty and forty-five passed on
32
29 October was simply not enforced. Similarly, Largo was not prepared
to gainsay the bureaucracies of the UGT and CNT building federations,
which refused to exert influence on their members to build trenches after
working hours. 33 In overall terms, Largo’s resolve was inhibited by the
enormous weight of the contradiction between the grass-roots radical-
ism whose symbolic guarantor he had become and the moderate course
which his political preferences and past practice dictated.
Throughout the second half of September the PCE’s leaders, and in
particular Antonio Mije, one of the Republic’s four deputy commissar
generals, did their utmost to impress upon Largo the desperate need to
accelerate militarisation and civil defence in Madrid. Mije also suggested
a four- or five-member war council to overcome the problems inherent
in the unwieldy and lumbering (although still politically necessary) form
of the Popular Front cabinet. But the prime minister rejected all these
proposals out of hand, adducing practical difficulties, even though the
military case the communists made was unanswerable. Largo undoubt-
edly felt overwhelmed by the desperate circumstances, but it was a culpa-
bly obtuse response which he gave to Mije’s observation that the way to
stop militia fleeing under attack was not to shoot the offenders but to dig
30 Bolloten, SCW, pp. 331-2.
31
Azana, ‘Cuaderno de la Pobleta’, Obras completas, vol. 4, p. 862.
32 Bolloten, SCW,
pp. 346-7.
33 Jackson, The Spanish Republic and the Civil War , p. 312; cf. D. Ibarruri, En la lucha. Palabrasy hechos
1 93 6-c93 (Moscow: Editorial Progreso, 1968), pp. 59-65; Azana, Obras completas, vol. 4, p. 819;
Preston, Comrades, p. 292.
 1

Building the war effort 1 4

proper trenches: according to Largo, Spaniards were too proud to hide

in the ground. 34 After the fall of Toledo on the 28th, Mije pointed out
to Largo that the same would happen in Madrid unless they speeded up
fortificationwork. He proposed a joint war ministry for which Largo and
Prieto would be responsible. For obvious reasons of a personal and (in
the circumstances) petty nature - namely his enmity with Prieto - Largo
rejected the suggestion instandy. By October, Mije - for all his jovial
Andalusian personality - was frustrated and exhausted. It had become
clear that anything emanating from the PCE leaders would instantly
be rejected by Largo for reasons of personal or organisational jealousy.
For this reason, in October the Soviet ambassador, Rosenberg, took the
brunt of trying to persuade Largo of the need to accelerate the defence
of Madrid - something which indicates just how crucial Stalin deemed
the survival of the Republic to be as an advance front to defend Soviet
frontiers against Nazi aggression. In time, the tensions between Largo
and the PCE would be replicated between premier and ambassador. We
shall return to these in due course, but it is important to emphasise at
the outset that their origin lay, not in arguments over the unification of the socialist

and communist parties but in the clash over the speed of implementation
and quality of Republican defence policy and, thus, in Largo’s running
of the war ministry.
To Largo’s suspicious narrow-mindedness, we also have to add the
massively debilitating consequences of his extremely bureaucratic ap-
proach. To be sure, the Spanish left as a whole was politically inexperi-
enced by dint of its lengthy exclusion from power. But Largo’s particular
inability to evaluate the bigger picture, or to discriminate between the
issues on which he really needed to focus as wartime premier and those
more trivial matters which could be delegated to others, seriously in-
hibited the Republic’s recovery capacity. Some sense of being out of his
depth seemed to convert itself into an exaggerated doggedness (if Largo
didn’t understand the point, everything stopped in its tracks until he had)
which only magnified the effects of his inefficiency. 35 This already diffi-
cult situation was made worse by certain other understandable physical

34 Elorza and Bizcarrondo, Queridos camaradas, Andre Marty’s Comintern report of

p. 325, citing
10 Oct. 1936. This is echoed in other memoir examples see Thomas, Spanish Civil
literature; for
War p. 432. Koltsov’s diary entries for
,
September/ October abound with scathing references to
Largo’s imbecility in matters military - see the heavy irony in the entry for 30 Sept. 1936 (points
8 and 9).
35 Cordon, 258-61 (especially p.
Trayectoria , pp. 261); Koltsov, Diario de la guerra espahola, p. 62 records
Prieto’s withering insights. Koltsov was beside himself over Largo’s inefficiency and procrasti-
nation (p. 19 1). In its very precision and lack of stridency, Cordon’s assessment is particularly
damning; also Gabriel Jackson’s comments on cabinet meetings, The Spanish Republic and the Civil
War, p. 366.
142 The Spanish Republic at war

limitations. At sixty-seven years of age, and in less than robust health,

Largo apparently saw no reason why the larger crisis should cause him to
modify his night-time routine and regularly retired between eight at night
and eight in the morning - during which time he was rigidly incomunicado.
While it was quite reasonable for Largo to bar individual militia leaders’
direct access to his person (indeed, the Militia Inspectorate forbade this),
as war minister, he needed to be more flexible with officers of the general
staff.

These deficiencies of prime ministerial leadership as well as the gen-

eral cabinet fragmentation made work of military and state-
the crucial
administrative reconstruction more difficult. It also meant that in both
areas crucial reconstruction work was being carried out initially in iso-
lated pockets, both by individuals and small groups, usually without much
specific government guidance or support 36 .

As we have seen, the Militia Inspectorate provided a key base from

which to begin centralising control on the Madrid front. More than any
other body, it did most every day, both practically and psychologically, to
articulate a vast array of disparate militia forces formed in the July days,
shaping them into the Republic’s first battalions. It was the Inspectorate
which did much to free up the war ministry from the incessant and
(inevitably) unstructured demands of individual militia leaders 37 The .

Inspectorate itself was reorganised on 20 October, when it passed directly

under the control of the military chief of operations on the central front,
being renamed the Comandancia Militar de Milicias. Through both its
control of militia payment 38 and through the dissemination of a constant
stream of circulars and instructions it instilled the need for discipline,
endlessly arguing/reiterating the point that the militias were now no
longer autonomous forces but part of a larger defensive enterprise. An
idea of the Gomandancia’s efficacy can be gauged from the fact that by
the end of the month some of these battalions were being successfully
incorporated into the Mixed Brigades (Brigadas Mixtas), the integrated,
autonomous units being established on the central front as the basis of
the new Republican army under construction 39 .

36 Elorza and Bizcarrondo, Queridos camaradas, p. 326.

37 Cordon, Trayectoria, pp. 241, 258, 260. Cf. J. Lopez in Bosch Sanchez, Ugetistasy libertarios, p. 19.
38 Within the Commandancia’s territory such payment ceased to be transmitted via parties or
unions from around 21 October 1936.
39 Thus, a Mixed Brigade would incorporate cavalry, heavy and light artillery, a signal corps and
sappers, as well as infantry troops, in order to make the brigade a flexible and independent unit.
For more on the Mixed Brigades, see Alpert, El ejercito republica.no, pp. 76-80. For the reorganised
Comandancia de Milicias and its achievements under its director, Servando Marenco, see Alpert,
 Building the war effort 143

Largo’s appointment, and that of Colonel Jose Asensio 40 (whom Largo

promoted to General) as his under-secretary at the war ministry, also
marked a process of evolution towards the construction of a central gen-
eral staff, the lack of which was making it impossible for the Republic
even to begin to develop an overarching strategic plan. Yet prioritisation,
not least of scarce resources, was essential if it was to sustain a war effort
against the better-supplied and equipped rebels. The July rebellion had
obviously left the pre-war general staff, as part of the state, in chaos. There
was a desperate shortage of capable military staff, yet, given the circum-
stances, very few of the available officers were trusted. The government
relied on a small group of fervently republican officers - Hernandez
Sarabia, Casado, Menendez, Estrada, Fontan, Fe and Cordon — to be-
gin the enormous task of reconstruction. 41 Antonio Cordon, supported
by his colleagues Jose Martin Blazquez and Jose Ceron, directed a kind
of emergency substitute known as the Secretariado Tecnico (Technical
Secretariat) from the war ministry which began the long task of recon-
structing crucial war-related services such as communications, transport,
munitions, suppliesand medical services. 42 The first full new general staff
was announced by Largo on 5 September. There were few names from
the pre-war body, while it included those - most notably Casado, Rojo
and Cordon (the latter at the head of the operational section) - who would
later be key figures in the new Republican army. But in September, with
the continuing removals of officers whose
was considered doubt- loyalty
ful and the ongoing batde to bar the militia’s direct access to the war

ministry, it would be the end of the month before any real semblance of
a new order of command appeared.

pp. 38-40, 377 (onMarenco himself - who was a (not previously prominent) officer from the
Cuerpo de Intervention Militar). On the battalions, see Alpert p. 42 and appendix 3, pp. 320—2;
and on the building of the new Republican army in autumn 1936 ‘en tierras manchegas’,
R. Salas Larrazabal, Historia del ejercito popular de la republica (4 vols., Madrid: Editora Nacional,
1
973 )> v°l- b
>
545 P- -

40 Asensio had been serving as a militia commander at the start of the war on the Guadarrama
sierra front, north of Madrid, where he met and impressed Largo Caballero: Bolloten, SCIV,

p. 280.
41
For biographical details on these officers, see Alpert, El ejercito republicano, appendix 13, pp. 359-

88; M. Suero Roca, Militares republicanos de la guerra de Espana (Barcelona: Peninsula, 1981). For
T.
Cordon see also the very positive description in J. Martin Blazquez, I Helped to Build an Army
(London: Seeker and Warburg, 1939), p. 279. Cordon, a professional army officer of one-time
monarchist convictions, had, like Casado, conspired against the Primo dictatorship and then
retired, along with many other pro-Republican officers, from the active scale under Azana’s
military reform law of 1932: Alpert, El ejercito republicano, p. 12.
42
Cordon, Trayectoria, pp. 233-4, 2 60; Martin Blazquez, I Helped to Build an Army. The latter was
a professional officer who eventually left Spain in spring 1937, exasperated with what he saw as
the irremediable chaos and incompetence of the Caballero government.
i
44 The Spanish Republic at war

The daunting task of military reorganisation was made more difficult

precisely because of the rising’s political implications. The distrust of
military power meant that civilian-political vigilance of its new forms
was inevitable. 43 Indeed, the rebellion had problematised the notion of
the military having any autonomous sphere of power - even in technical
matters. Witness the many highly laboured references to the subordinate
function of the military staff in the overall process of wartime decision
making — for example when the government publicly announced its ob-
jective of a single Republican command ( mando unico) on 16 October. 44
The widespread civilian incumbency in posts which required specialist
military knowledge, and which thus necessitated the additional alloca-
tion of a ‘military adviser’, has often been interpreted as the imple-
mentation of some abstract revolutionary goal. In fact it was a direct
response to the military coup. Post-rebellion, a certain duplication of
functions was as inevitable as it was understandable at various levels in
the Republican zone. But the enduring effects were, nevertheless, nega-
tive. First, because such duplication used up time and human resources
that could scarcely be spared. Second, because it reinforced the military-
civilian divide which the Republic, gearing itself for war, could also ill

afford.
This psychological divide placed tremendous pressure on the hun-
dreds of professional army officers serving the Republican war effort.

For them the rising had produced an enormous culture shock. They had
been betrayed by the conspiratorial actions of their own comrades, which
had ripped away their social and professional terms of reference. 45 On
top of this came the Republic’s own post-coup purge of officers, which
further increased their sense of alienation. At any time, problems caused
by shortage and dislocation could be turned into accusations of treason
and crypto-fascism. The feeling of being continually under suspicion
and required to prove oneself daily (at the same time as risking one’s

43 Thus in November Largo reformed the Consejo Superior de Guerra to include wider ministerial
participation for all the main political groupings in the Republic.
44 The Gaceta de la Republica for 16 October 1936 stated that ‘the Chief of Staff (Estado Mayor Central)
would not function as an executive body itself but rather as a consultative one assisting the
executive power’, cited in Alpert, El ejercito republicano p. 72. There are echoes here of the post-
,

coup fears ofindependent military action. A further, related problem would be the refusal of
Republican governments to pass from a ‘state of alarm’ to a ‘state of war’ because - as a result of
the non-occurrence of pre-war public-order reform, this would still have meant ceding supreme
control to the army. See chapter 1 above and chapter 6 below.
45 Cordon, Trayectoria, p. 253; J. I. Martinez Paricio (ed.) Los papeles del General Rojo, 2nd edn (Madrid:
Espasa-Calpe, 1989), pp. 76-7, 81.
 ,

Building the war effort 145

Even 46
life at the front) caused extremely high levels of stress and illness .

for those whose Republican political identity was clear, the world they
knew had gone The new one was either hostile or offered few known
47 .

points of reference in terms of organisation or values. The state was still

in chaos, its sustaining political forces (republicans and socialists) either
consumed or badly fragmented by the crisis. Yet there was one party —
albeit still on the sidelines - with a very clear line on military and rear-
guard organisation which spoke directly to their sense of order and
discipline. It is in this very particular and desperate context that we
have to understand the appeal of the Spanish Communist Party to reg-
ular army on the centre front - whether or not they actually joined the
officers
party. What the PCE offered them was an alternative family/ a refuge of

structure in culturally dysphoric, structureless and traumatic post-coup

times, and a source of collective protection; and, not least, the party reaf-
firmed their patriotism and sense of civic value, both severely damaged
by the rebellion 48 These motives predominated in the gravitation to the
.

party of officers whose social and political values were usually quite con-
servative. Antonio Cordon, who would become more closely identified
with the PCE than many others, also shared these reasons for joining
the PCE at the start of the war 49
.

While this could be seen as a constructive trend, much less fortunate

was the hostile disdain of the militia displayed by many of the officers who
remained to serve the Republic at war. Indeed, this included convinced
republican officers and even some of those who joined the PCE
(although not Cordon). Whether because of their professional training or
the trauma of post-coup disarticulation, these officers — including some
of high rank - were insufficiendy flexible to adapt to the irrefutable
reality: that the militia represented the only available material out of
which to construct the army needed to resist the rebels and their fascist

46 Alpert,
El ejercito republicano, p. 87; Cordon, Trayectoria p. 233; Rojo’s letter to Indalecio Prieto,
September 1937 in Martinez Paricio, Los papeles del General Rojo documentary appendix (no page
,

number); also cf. p. 73.

47 Cordon, Trayectoria , p.233.
48 For the sense
of ‘family betrayal’ which professional officers serving in the Republican zone felt

towards the rebel military, their sense that the latter had caused their own plight, see Los papeles
del General Rojo , pp. 77, 81.
49 Cordon, Trayectoria , pp. 236—8; Martin Blazquez, I Helped to Build an Army, p. 241; I. Falcon, Asalto
a los cielos (Madrid: Temas de Hoy, 1996), p. 166, according to all of which he joined the party
in July 1936. Also corroborated by the little-known memoir by Colonel Eduardo Cuevas de la
Pena, who would serve as Republican Director General of Security in 1938, cited in Bolloten,
SCW, p. 607.
 46 The Spanish Republic at war

backers. It ifrould seem’ that this was a failing of the otherwise talented

General Asensio, Largo Caballero’s anchorman in the war ministry 50 .

Another case in point was the irritation which some officers displayed
at the ubiquitous mediation of party and union representatives in the
Republic’s decision-making processes. But as Cordon replied to his ex-
asperated friend and colleague Jose Martin Blazquez, this was simply the
way and one had to operate accordingly 51
things were since the coup, .

But this response indicates in Cordon a degree of psychological flexibil-

ity which was comparatively rare among Spanish officers of that time —

and unsurprisingly so if one considers the nature of their professional

education. But the prejudices and abrasive behaviour of some army offi-
cers only reinforced the hostility of the militia volunteers, which, in turn,
‘confirmed’ military antipathy and antagonism. This vicious circle was
of itself wasteful of the energy and potential of both sides. Moreover, it
made the function of the political commissariat an absolutely vital one.
Although the political commissariat was only formally constituted by
Largo in mid October 52 the very nature of the conflict meant that in
,

fact political commissars had existed in everything but name almost from
the very start. They provided the vital link between military and miliciano
(since, as already mentioned, it was common practice for militias to

have a professional military adviser attached). The commissars explained

the rationale of orders, looked after the practical welfare of the men
and reminded them of the raison d’etre of the war. In the initial stages

especially, the function of the commissar was particularly important as

a shock absorber. Even though the commissar (or political delegate, as
they were initially often called) never intervened in matters of military
policy, given the militias’ antagonism towards the professional military
the delegate’s presence was seen by the men as a kind of guarantee of
their own interests 53 .

As the war went on, the commissars’ other functions of practical wel-
fareand political education would become paramount as the Republican
army, forged to meet the needs of modern warfare, was forced to engage
in a long, attritional conflict in conditions of persistent, morale-sapping
material inferiority. In spite of the commissariat’s objective importance,
however, it never ceased to be vehemently denounced both by many

50 For more on this see the discussion of the fall of Malaga (Feb. 1937) later in this chapter.
51
Martin Blazquez, I Helped to Build an Army pp. 207, 240-2.
,

52 Gaceta decrees of 16 and 17 Oct. 1936; also see Alpert, El ejercito republicano, p. 182.
,

53 The commissars provided a kind of temporary ‘quarantine’ service for professional officers. The
expression is from Guillermo Cabanellas, La guerra de los mil dlas, p. 529.
i Francisco Largo Caballero (seated, first on left), veteran socialist leader and
prime minister of the Republican wartime gov ernment from
September 1936 to \lay 1937 (Ministerio de Education, Cultura y
Deporte, Archivo General de la Administration)
 just

capital.
military
is the

of London)

‘front’
the

edge
between
of
The

Museum,
streets

lefthand

encounter

buildings.

War

damaged top

the the
Imperial

abrupt
the of on

front
the
the
through sandbags

in of

conveys

of
shelters

courtesy

stretcher set

which
the

by
sandbagged

image beyond
Photograph

away

An
borne block
two

1937. 71664)

is and
one

(HU
casualty gunfire
1936-January barely

A by Collection
2(b)),

Madrid.

scarred
plate

December Elkan
in (see
shops

space
(Vera

frame
front
and

urban
the
houses

Madrid
photograph.

and
outside

see

2(a)
front
We
 Collection

righthand

the
Elkan

on

(Vera

sandbags

2(a)
London)

plate
The

of
2(a).
Museum,

edge
plate

War
of
lefthand

range Imperial

top
the
the
the

on
beyond of

appear

just
courtesy

lies
which

which

ones Photograph

‘front’

same

71662)

the the

of
are

(HU
shot

photograph

panoramic

this

A of
2(b)

side
 3(a) ‘Emulate the hero of the people’ (‘jlmitad! al heroe del Pueblo’)
Wall poster mobilising military and homefront effort by appealing to the memory of
the legendary anarchist leader Buenaventura Durruti, killed on the Madrid front in
November 1936 and rapidly converted into an icon of Republican resistance
(Ministerio de Educacion, Cultura y Deporte, Archivo General de la Guerra Civil
Espanola (Kati Horna Collection))
3(b) Wall posters, the most prominent reading ‘jUnion! jDisciplina! jSocialismo!
(‘Unity! Discipline! Socialism!) (Abraham Lincoln Brigade Archives Collection,
Tamiment Library, New York University, Albert Harris Collection)
 Elkan

(Vera

Brigades
London)

Museum,

International

War

the

by Imperial

use

for the

of

destined

courtesy

1937

Photograph
January

c.
1512)
car

7
armoured
(HU

Collection

Home-produced

4
in
Republican

London)
battalion

inside

Museum,

Thaelmann

observers

War

German
Imperial
political

the
the
key
of
from

Stalin’s

courtesy

of
commanders

one

was

Photograph

two

Koltsov

by
71579)

flanked
1937.

(HU

Koltsov

1936-January

Collection

Mijail

Elkan

journalist

December

(Vera

Soviet
c.

Spain

5 Madrid
6 This photograph shows Santiago Carrillo (centre), the Socialist Youth leader who at
twenty-one became general secretary of the mass socialist-communist youth
organisation (JSU) and Councillor for Public Order on the Madrid Defence Council
(November 1936). The JSU epitomised the accelerated popular mobilisation
occasioned by the war, as a result of which many young people were catapulted into
positions of major political and military responsibility in Republican Spain. Not least
because of this, the JSU also focused the bitter organisational rivalry between socialists
and communists. In the post-war period Carrillo would become the long-serving
general secretary of the exiled Spanish Communist Party (PCE). On Carrillo’s left is
the PSOE journalist Julian Zugazagoitia, Minister of the Interior from May 1937 until
April 1938 and one ofjuan Negrin’s would be
close political collaborators. In 1940 he
handed back to Franco by Vichy France and shot. To Carrillo’s right (in profile) is
Fernando Claudln, then a Communist Youth leader and member of the JSU executive
and later a leading member of the PCE executive and party theorist until his public
dissent in 1964. (Fred Copeman Collection(HU 34724) Photograph courtesy of the
Imperial War Museum, London)
Collection

Copeman

(Fred

group
London)

youth

Museum,

Front

War

Popular

Imperial

Pioneers,
the

of

Catalan
courtesy

the

of
Photograph

1936

late

(33003)
in

Barcelona

in

gathering

A
7
 8 Parade in support of the construction of the Popular Army, Barcelona, c. February

1937 (Fred Copeman Collection (HU 33009) Photograph courtesy of the Imperial War
Museum, London)
Espanola)

Civil

Guerra

la
de

General
 y
minister),

Cultura

(foreign

Educacion,

Giral

de
Jose

left, (Ministerio

the

From
Miaja

1937.

Administracion)

General

November

by
la
de
in
accompanied

General

front

Madrid

Archivo
Azana,

the

Manuel
Deporte,

visiting

President

ministers

Negrin,

government

Juan

minister

Republican

prime

io
ii Shattered houses (with suspended bicycle) after air raids in Barcelona, January 1938
(FredCopeman Collection (HU 33 151) Photograph courtesy of the Imperial War
Museum, London)
 12 Residents of a Republican home for child refugees in 1938 demonstrating
co-educational principles in action. The boys were required to fulfil their share of the
domestic tasks. (Fred Copeman Collection(HU 33143) Photograph courtesy of the
Imperial War Museum, London)
 the

of (Fred
newspaper

and

right)

wall
London)
(centre)

(top,
the

Ibarruri) Museum,
workers,

Companys

War
and
(Dolores
Luis

nurses
Imperial

leader

Pasionaria

soldiers,
the

Gencralitat
of
of La

courtesy

pictures leader

and

as
centre)
communist
well Photograph

As
(top,

33062)
Spanish
Italy).
Ncgrin

and
(HU
Juan

charismatic

Germany

Collection
premiers,

of the

of
forces

Gopeman
Republican

images

invading

features two

the
after

London)
down

laid

Museum,

bridges

War

pontoon

Imperial

the

of the

one of

of
courtesy

means

by
1938 Photograph

July
17)
in
331

Ebro

(HU

River

Collection

the

crossing

Copeman

Army

(Fred

Republican

crossing

the
boat

of

initial

Soldiers

the

14
15 Wounded Republican soldier on an ambulance train during the battle of the Ebro,
1938 (FredCopeman Collection (HU 34628) Photograph courtesy of the Imperial War
Museum, London)
 1 6 War memorial, Serra de Pandols (Catalunya) built by International Brigaders
(under the supervision of Percy Ludwick, chief engineer of the 15 th Brigade) to
commemorate a number of their comarades killed during the battle of the Ebro in
1938. Unlike other Brigade and Republican war memorials, it survived destruction by
the victors because of its remote position in mountainous territory The monument was
restored in 2000 by a team of Spanish volunteers led by the historian Angel Archilla.
(Photograph from the private collection of David Leach, director of the documentary
film Voicesfrom a Mountain: British Volunteers in the Spanish Civil War (2001) and reproduced
here with his kind permission)
 Building the war effort x
47

professional army officers who disliked what they saw as interference

in their sphere 54 and from various civilian political quarters, mainly on
account of the accusation that the commissariat was a vehicle of Com-
munist Party aggrandisement.
This view was obviously shaped retrospectively by the clientelist po-
and organisational hostilities between the PCE and other
litical rivalries

groups in the Republican camp which would mount across the war pe-
riod for a variety of reasons. But the function of the commissar was
not invented by the PCE, for all that it was strongly advocated and first
implemented in communist units. 55 Nor did the party have any partic-
ularly overriding executive influence in the formation of the commissariat
proper in October. If one looks back at 1936, the relationship between
the PCE and the political commissariat can be seen for what it mainly
was - an organic one emerging from the PCE’s understanding of the
limitations of the militia forces and the imperatives of Republican mili-
tary defence. There would be more communist commissars not because
of any conspiracy but because in 1936 no other political group really
grasped (and certainly Largo did not) the fundamental importance of
the institution to a viablewar effort. Moreover, this lack of understand-
ing stemmed from the resistance among both the republican
logically
and the socialist leaderships in the pre-war period to new forms of poli-
tics involving mass mobilisation. It was, then, only the PCE that initially

prioritised the appointment of high-quality and experienced personnel

as commissars. 56
Much less controversial, indeed highly welcome to army officers, re-
publicans and PSOE leaders, was the PCE’s vocal commitment from
the start to the creation of a unified political-military command ( mando
unico) and militarisation. 57 The PCE stressed the need to put in place new
training structures for both officers and soldiers. As a working example
of what it advocated, the party created a prototype of the new militarised
force in the Fifth Regiment (Quinto Regimiento). 58 Its nucleus was the

54 Cordon, Trayectoria, p. 263.

55 The idea had a pedigree stretching back through the Russian Civil War all the way to the French
or American revolutionary armies of the eighteenth century, as various authors have indicated -
seeCordon, Trayectoria, pp. 262-3 and C. Blanco Escola on Rojo’s ideas, ‘El centenario del
General Rojo’, El Pais, 14 Sept. 1994. For PCE advocacy through Milicia Popular (the newspaper
of the party’s military formation, the Fifth Regiment) see Alpert, El ejercito republicano, pp. 176-8.
56
For a sensible summary of views on the commissariat’s genesis and development, see Alpert, El
republicano, pp. 180, 183.
ejercito

57 D. T. Cattell, Communism and the Spanish Civil War (New York: Russell & Russell, 1965), pp. 86, 230
(n- 5).
58
On the Regiment’s name and origins, see Alpert, El ejercito republicano, pp. 48-9.
 . ,

148 The Spanish Republic at war

*4 ,

PCE’s militia, the MAOC, 59 run since 1933 by Juan Modesto, who had
served in Morocco as an NCO. It was the only party militia that had
undertaken any real military training before i8July 1936. To the MAOC
were added forces from the fifth of the original volunteer battalions, plus
some extra numbers from the united socialist-communist youth organ-
isation, the JSU. The autonomous form of the Fifth Regiment’s own
fighting units (each carrying its entire military infrastructure within it-

self) was adopted Mixed Brigades, the basic unit

as the blueprint for the
of the new army. (These would be built up into the first of the Repub-
lic’s operational armies, the army of the centre, whose model was then

‘exported’ to build the armies of the east, Levante and Extremadura

6o
in the course of 1937. At its peak in October it was estimated to be
)
61
some 20,000 strong. But although the Fifth Regiment contained fight-
ing units, its key function would be as a transitional training formation
which then generated other battalions. 62 Apart from its extremely real-
istic attitude to the needs of war, the Fifth Regiment’s utility lay in its

stress on discipline and in the organisational model it provided. The

PCE had always been extremely clear about the need for the primacy
of military authority in the army and for new forms of hierarchy and
discipline. 63 The difference between the ancien regime army and the new
Republican one lay not in the removal of these principles but in their
qualitative transformationthrough a process of broader military reor-
ganisation which would be catalysed by the desperate defence of Madrid
in November and December.
was the kind of work undertaken so singularly by the Fifth Regiment
It

that underpinned Largo Caballero’s two decrees of 30 September 1936

formally militarising the militia. Under its terms all politically sound

59 Milicias Antifascistas Obreras

y Campesinas.
60
Alpert, El ejercito republicano, p. 82 For an estimate
. of the strengths of the various components of
the Republican army (Centre, Aragon, East (Valencia region/Levante), South (Andalusia and
Extremadura), Basque Country, Santander and Asturias) in the winter of 1936-7, see Thomas,
Spanish Civil War, p. 542 (based on Salas Larrazabal’s figures in Historia del ejercito popular de la

republica, vol. pp. 528-30). Whether or not these figures were initially inflated, Michael Alpert
1,

suggests that by spring 1937 reality had caught up with them and that by July 1937 the Republic
had over half a million men under arms, El ejercito popular, p. 87
61
Report by Marty: Elorza and Bizcarrondo, Queridos camaradas, p. 328.
62
The Regiment probably trained in the region of 25,000 men during the five months when it
was active - a significant achievement given the fact that the PCE was coming from the political
margins: Alpert, El ejercito republicano, p. 54.
63
See, for example, Pasionaria’s explanation of what militarisation involved (for the home front
too) in Mundo Obrero, 25 Sept. 1936, reprinted in D. Ibarruri, En la Lucha pp. 54-6 and Carlos
Contreras’ articles in Milicia Popular, Aug. 1936, cited in Alpert, El ejercito republicano, p. 51.
(Contreras was the pseudonym of the Italian communist Vittorio Vidali.)
 Building the war effort *49

officers and NCOs ( suboficiales) would pass to active service in the army, 64
while men would be distributed to companies and services as required
by the general staff. The militia battalions were to form the basis of the
Mixed Brigades. The decrees also established new minimum and max-
imum ages for military service of 20 and 35 years respectively. From 10
October on the central front, and from 20 October elsewhere, the new
65
militia battalions were subject to military justice. They were also num-
bered in an attempt to erase the political identity of the constituent mili-
tias, whose names were forbidden to be used in official documentation.

In practice, the militarisation process would take months to effect,

given the scale of reorganisation necessary, the severe lack of middle-
ranking officers/NCOs directly to command men and the ingrained
hostility of many (but far from all) milicianos to the centralisation inherent
in the process and, especially, to the loss of militia names. These problems
were worse in Catalonia and Aragon, where, as we shall see, it would
be over half way through 1937 before the militia began to be properly
66
integrated. Even on the central front, militia identities within the new
military formations endured well into spring 1937 67 — encouraged by the
ambivalent messages coming from Largo Caballero himself.
Paradoxically, this situation was also exacerbated by the inflexibility
and overly bureaucratic style of the new general staff. While perfectly
competent in conventional military terms, 68 it failed to make the nec-
essary organisational and procedural allowances for the unconventional
circumstances in which it was operating and for the resulting shortages of
officers and trained men. Its only concession was itself distinctly bureau-
cratic: the general staff provided lengthy and fairly empty ‘ideological’
preambles when circularising (even fairly simple) orders to demonstrate
that these did not offend against soldiers’ rights or egalitarian princi-
ples. What would have been more useful would have been a greater de-
gree of organisational But the legacy of the coup itself worked
flexibility.

against this as the professional military, under pressure as such and aware
of the mutual lack of trust between themselves and the popular forces
they were called upon to mould, took refuge in conventional forms and

64 ‘Pasaran a las escalas activas del Ejercito’. 65 Alpert, El

ejercito republicano, pp. 72-4.
66
The problem of non-integration in the Basque north was somewhat different since it had to do
with the competition for authority between the central government and the Basque government
rather than with a conflict over the type of structures per se. The Basque nationalist militia were
subject to military discipline from the chapters 4 and 5 below.
start, see
67
An April circular reminded units that numbers, not names, must be used to designate themselves
at all times, Alpert, El ejercito republicano , p. 75.
68
For further general staff reorganisations in October and late November 1936, see ibid., pp. 70-1.
150 The Spanish Republic at war

procedures. Their almost fetishistic drive for rigidly centralised bureau-

cratic structures outstripped that of the rebels in practice 69 and, beyond a
certain point, was not necessarily the most effective means of transform-
ing the hybrid Republican forces - a resistance movement sui generis -
into a functioning army that could take on a ‘modern’ opposition. 70
(Indeed, it seems quite likely that some of the tensions between officers
and troops that tend to be interpreted as ideological in origin could in fact
have been mitigated by a more flexible and intelligent form of military
organisation.)
In order to function, this army-in-the-making not only had to be
trained, however; it also had to be equipped and armed. The fact that
the Republic was, from the start, being obliged to undertake this from
a position of clear material inferiority put an absolute premium on
the maximisation of its internal resources. (On the Madrid front the
rebels noted that they were picking up Republican munitions that had
been produced in the Toledo arms factory only days before. 71 ) As Non-
Intervention had deprived the Republic of any hope of credits from the
Western democratic governments, as well as effecting an arms embargo
which was forcing it to procure arms at extortionate prices on the black
market, the first priority was to consolidate existing financial resources
and to develop a strategic plan for their use.
The major task here, as the capital Madrid came under greater
first

and greater threat from rebel forces advancing from the south, was for
the Republic to secure its gold reserves in a place where they could read-
ily be converted to meet war needs. The — inevitably secret — decree to

mobilise the gold reserves was one of the last made by Giral’s republican

69 M. Alpert, ‘Uncivil War - the Military Struggle’, History Today, 39 (March 1989), 14.
70 On the vexed question of the Republic’s ‘failure’ to wage a guerrilla war, see Ronald Fraser’s
judicious summary, Blood of Spain, p. 330 (n. 2). The new Republican Army would include
a guerrilla unit trained by one of the Soviet Union’s premier advisers, Alexander Orlov. It
carried out sabotage and other specific tasks (for example freeing prisoners) behind enemy
lines: see A. Orlov in Forum fur osteuropaische Ideen und geitgeschichte, 4 (2000), 235-7; E. P Gazur,
Secret FBTs KGB General (London: St
Assignment.The Er min’s Press, 2001), pp. 57-78. But the
Republic never waged a full-scale partisan war (even though Orlov would claim a high (possibly
exaggerated) level of partisan activity). That it did not do so indicates something important
about the politics of the Popular Front alliance and (even more crucially) its perception of the
international political environment. But one should also remember the practical context here
and keep hindsight at bay. The Republican government began by throwing all its resources into
war on Franco’s terms in the belief that it could win. It hoped that Non-Intervention
fighting a
would not hold. It had no way of knowing either that it would or that Germany and Italy would
massively escalate their aid to the rebels in spring 1937. After that, all the Republic’s resources
were locked into a different kind of war just to stay alive.
71
Sanchez del Arco, El sur de Espana en la reconquista de Madrid, pp. 134-5. He also refers to Madrid
troops using taxis, which rather belies the otherwise carefully worked-up impression that the ‘red
army’ was awash with hardware, p. 112.
 1

Building the war effort 15

cabinet on 30 August 1936 before it gave way to Largo Caballero’s. After

consultation with high-ranking representatives of the Bank of Spain.
Giral and his ministers, though cautious and conservative by instinct,

came to realise that the escalation of the war made imperative the rapid
mobilisation of the country’s major convertible resources. Otherwise, the
Republic would simply not have the means to continue waging war. 72
With the arrival of the new cabinet, it then fell to Juan Negrin, as treasury
minister, formally to carry out this policy, overseeing the transfer of the
gold from Madrid to the Republican-held naval base of Cartagena on
the south-east coast. The first transportation occurred on 15 September,
part of the reserves being sent to France (either directly or indirectly)
as part of the Republic’s ongoing strategy of gold sales to the Bank of
France. 73
As the on Madrid, the im-
rebels increased their military pressure
pact of the arms embargo began to bite. The Madrid government was,
moreover, experiencing real difficulties with western banks. They were
delaying the transfer of funds urgently required by its agents and diplo-
mats to purchase war material. 74 From this the Republicans drew the
only sensible conclusion possible: that were they to place their finan-
cial resources — the lifeblood of the future war effort — in the W estern
capitalist sector, these would risk being frozen. The partisan logic of
Non-Intervention scarcely gave the Republicans cause for optimism.
From mid September it was also becoming apparent that the Soviet
Union was inclining towards the option of aiding the Republic in order
to avert immediate military defeat. This saw a concerted effort by
its

the Republican government to consolidate diplomatic communications

in order to reinforce this inclination. Marcel Rosenberg, the So\iet am-
bassador, arrived in Madrid in the third week of August to establish the
first embassy in Spain since the act of diplomatic recognition back in

June 1933. 75 (It would be the end of 1936, however, before the embassy

72
Angel Vinas' exhaustive, technical research on the gold's shipment is published in his study El
oro espanol en la guerra civil (Madrid: Instituto de Estudios Fiscales, 1976) and summarised in El oro

de Moscu. Alfa y omega de un mito jranquista Barcelona: Grijalbo, 1979), see pp. 74-86, no—11;
also A. Vinas. ‘Gold, the Soviet Union and the Spanish Civil War’, European Studies Reviv'd. 9 (1) ,

Jan. 1979,; 107—9 ‘The Financing of the Spanish Civil War', in Preston, Revolution and tiar
in Spain, pp. 266-83.
73 Vinas underlines this continuin' of policy, Ed oro de Moscu, pp. 27-41, 66-94. By die end of March
1937 the Bank of France had purchased at least 26.5 per cent of the gold reserves held in Madrid
,

at the start of the wan Vinas, ‘Gold, the Soviet Union and the Spanish Civil War’, p. 108.
/4 Vinas, El oro de Moscu, pp. 91—2, 218—27: E. Moradiellos, Seutralidad benevolo (Oviedo: Pentalfa,
l
99<>'h PP- 207-9.
75 This act of recognition was effectively frozen thoughout the period of conservativ e gov ernment
1 933-5 . Nor would the republican administration elected in February 1936 get around to
!52 The Spanish Republic at war
t

was functioning, and acute staff shortages remained a constant problem

throughout the war. 76 ) On 16 September, as the first consignments of gold
headed for Cartagena, the Republican government approved the setting
up of an embassy in the USSR. Marcelino Pascua, whose name had
already been suggested for the post three years earlier, was appointed to
head the mission. 77 A medical doctor like Negrin and a former colleague
of the finance minister’s in their days in the Residencia de Estudiantes, 78
Pascua arrived in Moscow in early October to face the daunting and
difficult task of setting up a functioning embassy from scratch. 79
Pascua’s brief was to promote more substantial Soviet aid to support
the Republic. But he found Stalin and his lieutenants already persuaded
of this need. By the end of September it was manifest that the military
situation was threatening the Republic’s very survival, and for Stalin
Republican collapse was an alarming prospect. It meant that France
would have three fascist states on its borders and would be less likely than
ever to make a wholehearted commitment to an anti-German alliance. It
also meant that German firepower would be freed up for an attack against
vulnerable Soviet frontiers. This context is crucial to any assessment of
the motives for Stalin’s decision to intervene at precisely this point. A
Soviet military historian, Yuri Rybalkin, has suggested that Stalin’s offer

unfreezing diplomatic relations prior to the war. Some of Rosenberg’s Soviet colleagues doubted
that he had the personal qualities for the job: Gazur, Secret Assignment. The FBI’s KGB General ,

PP-55~6 -

76 Elorza and Bizcarrondo, Queridos camaradas, p. 329; Marchenko, the charge d’affaires, would still
be complaining to Litvinov about this more than a year later. A boost to 35 had been promised -
but it never materialised. For most of the civil war there were only four people working in the
embassy, or five if we include Marchenko’s wife: Schauff, ‘Hitler and Stalin in the Spanish Civil
War’, p. 20. Rosenberg’s second-in-command was Gaikis, who would take over from him in
March 1937. Rosenberg and Gaikis did not get on and it seems clear that Gaikis reported back
to Moscow on Rosenberg. But there was no obvious political reason for their disagreement.
Gaikis may simply have been ambitious. But he too would be removed after several months:
Schauff, ‘Hitler and Stalin in the Spanish Civil War’, p. 19.
77 Pascua was one of the members of the Fabian-esque Escuela Nueva (New School), which had
argued for the PSOE’s alignment with the Third International at the beginning of the 1920s.
Losing this argument, they formed the PCOE, but many, including Pascua, returned to the
PSOE. Falcon, Asalto a los cielos p. 138. On the New School see Meaker, The Revolutionary Left in
,

Spain 19 14-1923.
78 Madrid University’s student residence, which was also an elite powerhouse of Spanish intellectual
and artistic development in the inter-war period.
79 Vinas, El oro de Moscii, pp. 154, 156. Pascua was bitterly critical of republican inaction after
February 1936 because the total absence of Spanish diplomatic infrastructure in Moscow made
his difficult brief even harder - especially in a political culture as distinct as the Soviet Union’s.
Cf. Falcon, Asalto a los cielos p. 138. The purges also made Pascua’s work more difficult. He
,

would complain that he could never reach the same Soviet personnel on any two consecutive
occasions - which made it very hard to complete any piece of business: Schauff, ‘Hider and
Stalin in the Spanish Civil War’, p. 20.
 Building the war effort *53

of more substantial and technologically advanced aid was the result

of an informal assurance at the end of September that its cost would
80
be covered in full by the transfer to Russia of Spanish gold reserves.
Given the Republic’s clear international isolation and war needs, it may
be that such an informal assurance was given. Intervening to save the
Republic would require the dispatch of precious war material from Soviet
factories. So, given the Soviet Union’s imperative of national defence, it

may also be that the gold reserves were seen as constituting a guarantee
(and important psychological reassurance) that these could be replaced
rapidly should a crisis situation occur. But whether the gold was the
single deciding factor is still clear. We
do know, however, that once
not
the Soviet Union had the gold, it would subsequendy manipulate the
exchange rate in its favour, ensuring that the Republican government
81
paid high prices for its aid. But this was not necessarily a preconceived
plan. Soviet intervention in Spain was marked by contingency — so there
is no reason to suppose that it did not apply in this respect too. Moreover,

nor was Soviet intervention particularly consistent in terms of narrow or

pecuniary self-interest. In 1936 the Soviet government would dispatch
at least 50 per cent (and probably more) of its precious total annual
production of military aircraft to Republican Spain. Later in the war too
the Soviet government would provide substantial credits to the Republic
when it knew that it had virtually no chance of recouping them. 82
We do not know whether it was a Soviet or a Spanish Republican
representative who first uttered the words proposing the transfer of gold.
But its significance is less than many commentators have supposed. In
so far as the Republican government was under pressure, this derived
not from its Soviet interlocutors but from the knowledge of its precarious
international situation. 83 The Republic’s objective was to accelerate the
mobilisation of its gold reserves in such a way as to guarantee the for-
eign exchange necessary to fight the war. Given that Non-Intervention
had already produced worrying restrictions on the Republic’s economic
room for manoeuvre, instant convertibility could only be assured if the
gold left was much
Spain. In the existing political climate nor was there
80
Howson, Armsfor Spain, pp. 127-8 cites Lt-Col. Rybalkin. (Stalin had already permitted Repub-
lican purchases in the Soviet Union of low-grade armaments which were shipped to Cartagena,
the first arriving on 4 October: Howson, p. 126.)
81
Ibid., pp. 146—52.
82
The aircraft tally may have been higher because Soviet production targets for arms were rarely
fulfilled in the second half of the 1930s as a result of the disorganising effects of the terror: Schauff,
‘Hitler and Stalin in the Spanish Civil War’, pp. 10, 17. On the credits see chapter 6 below.
83
Vinas, ‘Gold, the Soviet Union and the Spanish Civil War’, pp. 111-12; Howson, Armsfor Spain,
p. 128.
154 The Spanish Republic at war
•

choice of destination other than the Soviet Union or Mexico. In terms of

its financial and infrastructural capacity for assisting the Republic, then
the Soviet Union’s resources were obviously incomparably greater than
those of a small country like Mexico. Thus it was that Largo Caballero
formally proposed the transfer of Republican gold reserves in a letter
dated 15 October which was couriered by the Soviet trade attache,
Stashevsky. 84
The gold transfer obviously took place in the utmost secrecy - both
because it constituted the crux of the Republic’s resistance strategy
and propaganda potential it inevitably pro-
also because of the hostile
vided. Negrin, in overseeing the transfer, was implementing Republican
government policy, not formulating it himself. Much has been made
of his frequent communication with Stashevsky. But both Negrin and
Indalecio Prieto (as navy and air minister) were in close consultation
with Soviet representatives (in Prieto’s case, particularly with ambassador
Rosenberg). Nor is this surprising in the circumstances, given the weighty
cabinet responsibilities borne by the two Spaniards. 85 In actioning the
gold transfer, Negrin was, moreover, acting with the knowledge full

and support of the most senior representatives of Republican constitu-

tional and governmental authority - which obviously included President
Azana, premier Largo Caballero and Prieto himself. 86 The mobilisation
of the gold also involved senior Republican civil servants, including both
of Negrin’s deputies in the finance ministry, the left republican Mendez
Aspe and Negrin’s under-secretary, the socialist Jeronimo Bugeda. As
Angel Vinas also indicates, the shipment was witnessed by represen-
tatives of the Republican executive, legislature and judiciary - among
them the former republican premier Jose Giral, who was then minister
84 Vinas, El oro deMoscu, pp. 166, 288. Thereafter the Republic’s negotiations and purchase of war
material were channelled (and foreign exchange raised) via banks connected with the associates
of the Soviet National Bank (Gosbank) and then charged against the Spanish gold deposits.
85 Cf. Bolloten, SCW, pp. 138-44. But there is nothing conclusive here and Bolloten’s perspective
is (like that of so many of the memoir sources he cites) heavily coloured by a post hoc Cold War
reading of Negrin’s premiership. Note, however, that Santiago Garces, a Carabinero captain
and colleague of Negrin’s who would later head Republican military intelligence (SIM) and who
witnessed meetings between Negrin and Stashevsky, was clear about the fact that it was Negrin

who set the agenda and controlled the proceedings: Bolloten, SCW, p. 909, n. 27.
86
Graham, Socialism and War, pp. 132-3; Vinas, El oro deMoscu, pp. 52-5, 112, 251-3, 314, who also
points up (pp. 52-3) Azana’s resounding silence on this issue in what are otherwise voluminous
memoirs of the civil war period; Vidarte, Todos fuimos culpables, pp. 536—9. The lengthy polemic
over responsibility and knowledge has to be understood in the context of the post-war batde -
both for political leadership of the Republicans in exile and for Allied support to dislodge Franco —
in a burgeoning Cold War environment. Prieto’s denials are absurd and, quite literally, incredible,
as even the most Cold warrior-like of his colleagues (such as Luis Araquistain and Rodolfo Llopis)
were aware (see Graham, Socialism and War, p. 279, n. 28).
 ,

Building the war effort 155

without portfolio in the Largo Caballero cabinet 87 Moreover, Negrin’s .

punctilliousness in ensuring that each stage of the transfer was carefully

documented stands in contrast to Largo’s own irritation at these safe-

guards, underscoring Negrin’s own highly developed sense of the great

responsibility incumbent on those serving the state 88 .

Negrin had reluctandy accepted the treasury post at Prieto’s behest

as part of the PSOE national executive’s team in the September cab-
inet 89 Like the rest of his parliamentary socialist colleagues, he had a
.

very low opinion both of Largo’s leadership capacity and of the gen-
eral political calibre and organisational abilities of the PSOE left which
backed him. 9 ° Negrin thus saw himself as an embattled outpost of orderly
government amid the tide of deadly, if well-intentioned, disorganisation
and inefficiency in Largo’s cabinet. He knew that he held a portfolio
which was absolutely crucial to the survival of a Republic struggling un-
der the crushing weight of economic embargo — an ‘iceberg’ of which
the specific diplomacy of Non-Intervention was only the tip. In such cir-
cumstances foreign exchange was lifeblood. Negrin was shrewd enough
to know that ‘the war ends for the Republic the day the last gold peseta
does ’. 91 Accordingly he saw his function as the husbanding of resources

87
Vinas, El oro de Moscu, pp. 250-1 and ‘Gold, the Soviet Union and the Spanish Civil War’, p. 112.
The kinds of fantasy of ‘Negrin’s treason’ recounted in books like Chantaje a un pueblo (Madrid:
Toro, 1974) by veteran Valencia socialist and Caballero-supporter, Justo Martinez Amutio (see
p. 42) are scarcelyworth refuting since the evidence is so clearly against them. But Marcelino
Pascua’s withering assessment rings true - that it was an empty, scandal-mongering book in
which the author attempted to substantiate his claims by the old trick of presenting himself as
something he never was - part of the inner circle of Republican policy makers: Pascua to Vinas,
13 February 1977 (point 6), caja 8 (13), Pascua’s personal archive (AHN).
88
Zugazagoitia, Guerra y vidsitudes, p. 301; Gazur, Secret Assignment. The FBI’s General, p. 87.KGB
89 Negrin’s reluctance
in Zugazagoitia, Guerra y vidsitudes pp. 154-5, confirmed by Negrin in his
speech to the Cortes, 30 Sept. 1938 (Valencia: Ediciones Espanolas, 1938), p. 4, and in the
Epistolario Prieto y Negrin. Puntos de vista sobre el desarrolloy consecuendas de la guerra dvil espahola (Paris:

Imprimerie Nouvelle 1939), p. 40. Also Gabriel Jackson’s comments on cabinet meetings, The
Spanish Republic and the Civil War, p. 366.
90 The negative attitude was mutual. For Largo’s hostility to Negrin and Negrin’s sense of party
discipline (‘estoycon mi partido hasta en sus errores’), see Graham, Socialism and War, pp. 93-4.
91
be victorious, needs a treasury as strong as its army.’ Both opinions were
Cf. ‘A state at war, to
voiced by one of Negrin’s deputies in the treasury, the socialistJeronimo Bugeda, at the important
PSOE national committee meeting held in July 1937 in Valencia. His report is the closest we
come in documentary form to Negrin’s own political-economic testimony. Stenographic record
of Bugeda’s report (pp. 46-7) in the Archivo Historico de Moscu, Fundacion Pablo Iglesias; see
also Graham, Socialism and War, p. 108. On Bugeda’s role in the treasury, see also Negrin’s letter
of 23 June in Epistolario Prieto y Negrin, pp. 41—2; Vinas, El oro de Moscu, esp. pp. 211-16. Also
correspondence on Soviet credits between M. Pascua and Negrin, 22 June 1937-1 1 March 1938,
AHN/MP, caja 2 ,
carpeta 2, including the (undated) report (c. mid 1937) ‘Ante la perspectiva de una
larga lucha’, which makes the case for the necessity of such credits because of the international
economic isolation of the Republic and the likely prolonged nature of the war.
156 The Spanish Republic at war

for a long struggle. In the pursuit of this goal, first in the treasury
and
later as prime Negrin would implement policies designed to
minister,
concentrate all economic and political decision-making within the cen-
tral Republican government. This is the key to Negrin’s immense im-

portance during the civil war and to the controversies which surround
him to this day.
Economic centralisation brought Negrin immediately into conflict
with the Republic’s procurement committee ( comision de compras) in Paris
as he sought arms purchasing under the di-
to bring all channels of
rect authority of the treasuryand thus under his personal control. The
procurement committee was created by Luis Araquistain, Largo’s polit-
ical lieutenant and Republican ambassador to Lrance (until May 1937).

Araquistain’s objective had been to end the chaos he had encountered

in September on arriving in Paris, where scores of individual Spaniards
were milling around attempting to purchase arms in the name of the
myriad Republican committees which had dispatched them. Indeed,
these agents often ended up driving the already high prices higher by
bidding against each other - such was the general dislocation and lack
of communication as a consequence of state fragmentation after the
coup. Araquistain complained that Negrin simply ignored his requests
arms purchases were lost. 92
for funds, with the result that potential
The lack of trust between Araquistain and Negrin was in part the
product of pre-existing political and personal antagonisms. Their es-
trangement dated back to the May 1936 cabinet crisis and the social-
ist left’s obstruction of Prieto’s bid for the premiership. The left’s role
in helping block PSOE which had
access to the cabinet was the issue
finally broken, de facto the organisational unity of the socialist move-
,

ment, 93 and it was something that Negrin and his parliamentary

colleagues held Araquistain primarily to have been responsible for
engineering. 94 But while it is true that funds were on numerous occasions

92 Araquistain’s requests to Negrin, 9, 25 Feb. 1937 and his sarcastic comments to fellow left socialist
Julio Alvarez del Vayo, foreign affairs minister (22 Feb. 1937) in legajo 70/81, Araquistain’s
correspondence (political documentation) AHN. Rivas Cherif wrote on 26 February 1937 to
Azana (his brother-in-law) that Araquistain was fuming and had proclaimed that Negrin needed
shooting, Archivo de Manuel Azana, caja RE 137 (16) (Archivo de Barcelona).
93 Graham, Socialism and War, pp. 34-41 and ‘The Eclipse of the Socialist Left 1934-1937’, in
Lannon and Preston, Elites and Power in Twentieth-Century Spain, pp. 134-9; see a so chapter
l 1

above.
94 Araquistain’s letter to Negrin, 2March 1937 accusing him of sabotaging his ambassadorial work
,

and claiming that this was because he disapproved of Araquistain’s having ever been appointed,
legajo 35, 8-20, Araquistain correspondence (political documentation), AHN. For the pre-war
conflict in the PSOE, see chapter 1 above.
 Building the war effort i57

not released to the Paris committee, seems unlikely that Araquistain’s

it

accusation of a vendetta against himself has any truth in it, however un-
derstandable his irritation at the delays .
95 Structural difficulties and frac-
tured communication channels played a far greater role than Araquistain
5
supposed. But Negrin’s ‘parsimony was also reinforced by his awareness
of the staggering waste involved in arms procurement via the commit-
tees. While the Paris-based one may have constituted an improvement
on the state of affairs Araquistain had encountered, the committee was
still unwieldy since its membership, like that of Largo Caballero’s cabi-

net, had to reflect the full configuration of the Popular Front. This meant
that there was a great deal of internal politicking and dissension which
significantly reduced its efficacy, while the problem of the committee
duplicating material purchased through other sources remained. In De-
cember, in a joint bid by Negrin and Prieto to overcome these problems,
the Paris committee would be wound up and its brief transferred to
the newly created armament and munitions department (Comisaria de
Armamento y Municiones) within Prieto’s navy and air ministry 96 .

But we should remember that, ultimately, the inefficiency and waste

involved in Republican procurement was a structural consequence
of the embargo itself. Non-Intervention ruled out most of the use-
ful government-to-government aid channels (the upcoming Soviet aid

notwithstanding). In so doing it obliged the Republic to submit to the

vagaries of the arms market and to the dealings of some extremely dubi-
ous private intermediaries. We should also remember that the Republic’s
leading cadres - most of whom were liberal professionals — were ill-
equipped to deal with this jungle, having neither adequate contacts
nor the requisite technical expertise. The Republican purchasers were
made doubly vulnerable to a startling array of middlemen and as-
sorted opportunists, who invariably demanded grossly inflated rates of
commission, precisely because of their lack of alternative channels. The
fact that Non-Intervention forced the Republic to rely on a series of ever
more byzantine routes and procedures to obtain arms and war material
also meant cumulative qualitative, as well as quantitative, disadvantages.
When the purchased material arrived (which was not always the case),
weapons were often found to have incompatible ammunition, or their
instructions to be in some obscure foreign language. On other occa-
sions, what arrived bore no relationship to the ‘specification’. Moreover,
the material invariably lacked any technical or logistical back-up, vastly

95 Vinas, El oro de Moscu, pp. 230—6. 96 Ibid.,

p. 231.
1 58 The Spanish Republic at war

reducing To add insult to injury, the ratchet effect produced

its utility.

by Non-Intervention - which had effectively created a black market

in arms - meant that these substandard goods habitually cost the Re-
public hugely ‘over the odds’. The haemorrhaging of time and money
that Non-Intervention inflicted at the start of the war, solely upon the
Republic, was absolutely devastating. 97 Moreover, nor would Soviet as-
sistance release the Republic from the obligation of paying these crip-
pling prices for the duration of the war. For much of that assistance came
not through the supplying of domestically produced armaments (which
though good were also very expensive), but through Soviet procurement
of arms for the Republic on the international market - where, thanks to
Non-Intervention, black-market prices obtained just the same.
It is in this context that we have to understand Negrin’s ongoing bat-
tle to maximise foreign exchange and to establish a central government

monopoly on the deployment of state economic resources. This was not,

however, a politically neutral agenda. Through treasury centralisation
policies Negrin was seeking to re-establish an orthodox capitalist eco-
nomic order in Republican Spain. He began by overhauling the structure
and technical working of the treasury itself, although there was a continu-
ity of personnel - the ministry staff being chiefly republican technocrats

and senior state functionaries. 98 From the start, Negrin’s aims would
bring him into conflict with union and neighbourhood committees over
control of resources, supply policy and the thorny question of requisition-
ing. 99 And the same objectives would later mean that he clashed with a

variety of regional and sectoral institutions - radical/ proletarian, but also

bourgeois - whose economic and political prerogatives challenged the
realisation of Negrin’s goal and ideal: the reconstruction of the central
liberal state.
This preference for had been evident in Negrin long
state building
before his arrival at the treasury in September 1936. Believing as he
did in the idea of a ‘rational’ or ‘guardian’ state, Negrin had never
been interested in parliament as an oratorical arena which derived from
Mediterranean-clientelist practice. From the beginning, when Negrin
had first been elected to the Republican Cortes in 193 1 ,
he had committed

97 The complex history of Republican arms purchasing under the vicious conditions of Non-
Intervention is unravelled in Gerald Howson’s excellent study, Armsfor Spain. See also his Aircraft
ofthe Spanish Civil War (London: Putnam Aeronautical Books, 1990), which, in spite of its specialist
format, has a considerable amount on the effects of Non-Intervention embedded within it.
98 Names and details in F. Vazquez Ocana, Pasion y muerte de la segunda Republica espahola (n.p.:
Editorial Norte, (1940?)), p. 61; Vinas, El oro deMoscu, pp. 81, 210-18.
99 Vazquez Ocana, Pasiony muerte , p. 59.
 Building the war effort J
59

himself to the technical, behind-the-scenes work in parliamentary com-

mittee.(He was a member of the treasury budgets committee.) In terms
of both Negrin’s personality and political objectives, even then he was
more concerned with renovating, with establishing underlying structures
100
in order to make things work .

It was Negrin’s experience in the very first stages of the post-coup

conflict which galvanised his understanding of the need to concentrate

political and military as well as economic authority. When not engaged
directly, along with his parliamentary socialist colleagues, in the task of
holding together the remnants of government fabric from the presidential
office or navy and air ministry in Madrid, Negrin would often hitch a
ride in a militia lorry in order to observe at first hand the Republican
military defence on the Guadarrama sierra, north of the city, often in far
from secure conditions 101 From these trips he returned unequivocally
.

convinced of the need to galvanise Republican defences by means of a

thoroughgoing programme of militarisation.
An even stronger drive to state building in Negrm, however, came
in response to the explosion of popular violence detonated by the mil-
itary rising. The paseos had posed a serious dilemma for Negrm, even
leading him commitment to
to question for a time the validity of his
102
the Republican cause What confirmed that commitment was, first,
.

Negrin’s awareness of the rebels’ dependence on large-scale foreign fas-

cist assistance both to launch and to fight the war and, second, a growing

realisation that the rebels were committing genocide 103 But the eradica- .

tion of the paseos in Republican territory remained for him the essential
104
precondition of constructing a legitimate state /political order .

100
Even later, as wartime premier, Negrin would never be very ‘visible’. There are very few pho-
tographs of him - at any stage in his life - since he actively discouraged even his children from
taking them: Jackson, The Spanish Republic and the Civil War, p. 392; Vazquez Ocana, Pasiony
muerte, pp. 78-80. He also refused to sit for Jo Davidson, the American artist who sculpted many
leading Republican figures during the war: C. de la Mora, In Place of Splendor, p. 366. Even the
grave in which Negrin was buried in November 1956 in Pere Lachaise (Paris) bore no name.
Interestingly, Negrin’s future military opposite number, Vicente Rojo, displayed similar traits:

Martinez Paricio, Los papeles del general Rojo, p. 16.

101
Zugazagoitia, Guerra y vicisitudes, pp. 122-3; Anso, To fui ministro de Negrin, pp. 140, 15 1; also

M. Pascua, cited in S. Alvarez, Negrin (Documentos), p. 280.

102
Negrin to the French ambassador Morel, J. Marichal, El intelectualy la politico (Madrid: CSIC,
1990), p. 100.
103
For a discussion of rebel violence, see chapter 2 above. Negrin refers (via Bugeda) to the threat
of Spain’s economic colonisation by Germany and Italy in the economic report to the PSOE
National Committee, July 1937 This then becomes a major plank of Negrin’s public campaign
.

to maintain popular resistance: see chapter 6 below.

104
Cf. Negrin’s reported comments in early September 1936 on the importance of constitutionality
to the Republic, ‘No tenemos, no podemos tener otra divisa de guerra’ (‘we simply cannot allow
any “war currency” but this’), Anso, To fui ministro de Negrin, p. 251.
1 60 The Spanish Republic at war
• • , • • •

The killings of detained rightists in Madrid’s Model Prison on 23

August brought the situation to crisis point. More than any other sin-
gle incident, the prison killings have come to symbolise the loss of state
control in Republican territory. President Azana was so devastated by
the events, which for him vitiated the very essence of the Republican
cause, that he considered resigning. 105 But even without his departure
the incident did significant damage to the Republic’s standing interna-
tionally - and precisely in the western democracies whose support was
so desperately being courted.
In an attempt both to limit the political damage and to channel popular
anger, the Republic immediately set up the Popular Courts (Tribunales
106
Populares) to try those accused of supporting the rebellion. Com-
posed of a jury of representatives from all the Popular Front parties
and presided over by a professional judge, assisted by two others, these
tribunals were designed to put an end to the period of paseos and sum-
mary ‘justice’. Gradually over a period of months these would indeed
107
subside. But it was an uphill struggle. The coup had capsized the ju-
diciary as it had every other instrument of state action, and there was
a resulting shortage of career judges and magistrates. Nevertheless, the
popular court system constituted the first step in returning to the state
its and basis
defining function of power - a monopoly on
‘legitimate
violence’.While the juries found for or against the defendant, the judges
were responsible for sentencing. But even though harsh sentences were
passed, including capital sentences, most of those brought before the
Popular Courts escaped with their lives. 108 Punishment most frequently
took the form of social disciplining via the prison system (which in-
cluded work camps where prisoners were deployed in fortification and
other war-related work 109 ). Nor was it government policy that drove
the capital sentences passed in the Popular Courts. This would change

105
Among the dead was his old colleague, the founder of the Reformist Party, Melquiades Alvarez.
Fraser, Blood of Spain, p. 176.
106
A. Reig Tapia, Violenciay terror. Estudios sobre laguerra civil espanola (Madrid: Akal 1990), p. 121;

‘Justicia republicana’, in Justicia enguerra (Jornadas sobre la administracion dejusticia durante laguerra
civil espanola: instituciones y juentes documentales) (Salamanca, 26-8 November 1987) (Madrid: Min-
isteriode Cultura, 1990), pp. 19-245; G. Sanchez Recio, Justicia y guerra en Espana. Los Tribunales
Populares (1936-1939) (Alicante: Instituto de Cultura ‘Juan Gil-Albert’, 1991); Fraser, Blood of
Spain, pp. 177-8. For the equivalent within the system of military justice, see Alpert, El ejercito
republicano, pp. 213-14.
107
Between 50 and 70 per cent of the civilians who died violently in the Republican zone did so
during the summer of 1936: Julia et al., Vktimas de la guerra civil.
108
Julian Casanova gives an overview of the operation of the courts in various areas, ibid., pp. 1 61 -8.
109
Sanchez Recio , y guerra en Espana, pp. 176-9;
Justicia (same author) Justicia republicana’, pp. 30-
6; Arostegui and Martinez, La Junta de Defensa, p. 241.
 Building the war effort 61

later in the war Republican courts 110 as deteriorat-

in other sorts of
ing internal conditions and external circumstances led to an increase
in fifth column activity and other sorts of individual behaviour harm-
ful to the war effort. But in the Popular Courts, in addition to jury-led
motives of popular retribution, it also seems probable that harsh sen-
tences were sometimes the result of the personal agendas of individual
judges - either because they feared for their own lives or, in some cases,

in order deliberately to accelerate the discrediting of the Republic on a

111
public-order ticket. Although we still lack a comparative analysis of
rebel and Republican penal policy, it is clear that there was no equiv-
alent inRepublican Spain of the mass regime-sanctioned killing that
occurred throughout the rebel zone. Nevertheless, the Republic did pass
(and implement) death sentences from the time of the Popular Courts
onwards, which reversed the regime’s pre-war abolition of the death
112
penalty in the civil code.
Although this was a retrograde step in humanitarian terms, not all lib-
erals in the Republican camp were equally disturbed by the restitution
of the death penalty. For Negrfn, ‘legitimate violence’ had always in-
cluded the state’s recourse to capital punishment for exceptional crimes
which threatened its stability. For example, in 1932 Negrin had argued
strongly that the Republic was too weak to do other than execute the mil-
itary ringleaders of the failed August coup attempt." 3 (These included
the man who became the original titular head of the July 1936 rising,
General Jose Sanjurjo, then leader of the Civil Guard.) For Negrin, as
a liberal, legitimacy in this context resided in the fact that violence as a
facet of state power was not arbitrary but limited by constitutional law
subject to revision by collective consent.
By this was con-
criterion, the ‘popular justice’ of the irregular patrols
sidered to be entirely beyond the pale because it was bound by no due
process. As a result, from mid September 1936 the Republican govern-
ment promulgated a series of measures intended ultimately to disarm
and disband all non-state police forces operating on the civilian front. As
a first step, the Milicias de Vigilancia were set up" 4 to replace the myr-
iad and multi-form ‘patrols’ which had emerged during the July days.

110
See chapter 6 below for summary forms ofjustice in the workings of the Tribunales de Guardia
and the Tribunales de Espionaje y Alta Traicion.
111
Fraser, Blood of Spain , p. 178.
112
The Republic had abolished the death penalty in the civil code in 1932.
113
Vidarte, Todosjuimos culpables, voL 1, p. 213.
114
Gaceta de la Republica, 17 Sept. 1936. In theory’ these militias (Milicias de Vigilancia de Retaguardia
(MTVR)) were also constituted to reflect the membership of the Popular Front alliance.
 62 The Spanish Republic at war
« i

These patrols, like some of the militia columns discussed earlier, had a
reputation for violence and looting which terrified the urban and ru-
ral smallholding middle classes of Republican Spain - even those who
had no direct experience of the patrols. Indeed, the insidious threat of
the patrols 115
was probably more corrosive over time than more dra-
matic incidents such as prison massacres and sacas - even though these
had more international impact. The lack of any external control on the
patrols facilitated their infiltration by numerous undesirables including
crypto-rebels bent on mayhem to discredit the Republic, as well as a
variety of lumpen elements, some of whom had been released when
116
the prisons emptied in the wake of the rising While the Milicias de.

Vigilancia represented a certain concentration of coercive force in a

quasi-governmental direction, the forms of repression for which they
were responsible (including executions) were highly contingent, constitu-
tionally dubious and no doubt sometimes driven by particularist political
agendas 117 This situation made it easier for the Republican government
.

to avoid any debate about radical, innovatory forms of justice, allowing

it to present full judicial and police normalisation as a purely humani-

tarian campaign against the arbitrary abuse of power which would be

equally beneficial to all citizens.

The one of the other consequences of the rebellion had

fact that
been an intensification of inter-union violence between sectors of the
CNT and UGT bases also reinforced the government’s case here .
118

Even though the meanings of this intra-union violence were somewhat

different from that practised by patrols on shopkeepers and smallholders,
its very occurrence allowed the state to claim that the order it sought
to reconstruct was ‘class-blind’. In October the government began its
attempt to remove weapons from the hands of private citizenry, which,
119
in theory, included both the committees and workers’ patrols In reality, .

little happened. Many of the municipal authorities that were supposed to

115
Fraser, Blood of Spain, p. 177.
116
Vidarte, Todos juimos culpables, vol. 2, p. 531. See chapter 2 above for similar problems affecting
some militia columns.
117
For the important case of Madrid, see I. Gibson, Paracuellos como jue (Barcelona: Argos Vergara,
1983), pp. 42, 224-5.
118
For a discussion of this intra-union violence in terms both of antagonistic social constituencies of
workers and specific organisational rivalries, see chapters 1 and 2 above. For the way in which
UGT-CNT clashes - above all in Barcelona - were about conflicts between middling- and
working-class sectors (or white- and blue-collar workers), see chapters 4 and 5 below.
119
Decree of 27 October 1936 required the handing over to the municipal authorities of all ‘long
arms’ (i.e. rifles and machine guns): Morrow, Revolution and Counter-Revolution in Spain p. 58. ,
 Building the war effort 163

drive the process scarcely existed. Moreover, the memory of the military
rising (frequendy seconded by sectors of the police forces) meant that
there was significant worker resistance to police normalisation. It was to
be a very slow process, above all in Barcelona, where social resistance was
120
greatest .Nevertheless, the October decree established an important
precedent for subsequent state action.
For many in positions of Republican authority, however, not only were
the patrols and popular justice questionable, but so were the myriad
party' and union committees criss-crossing Republican territory. Largo
Caballero himself came, like many veteran union leaders in the UGT

and the CNT, to adopt the pragmatic view that unless the fragmentation
of military, economic and political power was corrected, the Repub-
lic would never be in a position to hold its own against the rebels and

their backers. It is in these terms that we can explain both his public
support for militarisation (the September and October decrees provid-
ing for the formation of regular regiments), his implementation dur-
ing October of decrees limiting the scope of industrial and agrarian
121
collectivisation and, in the same month, the formal (if often not yet
defacto) re-establishment of municipal authorities. But other ministers —
and most notably Negrfn - were opposed to the popular committees on
principle because they were the antithesis of liberal constitutional order.

Whether the committees had proved themselves good, bad or indiffer-

ent at coping in the period of the Republic’s emergency defence was
ultimately beside the point, since their very existence was an affront to
governmental legitimacy and authority.
Although the months of September and October were a time of con-
tinuing political contradiction and confusion, in retrospect we have to
understand them as a necessary and unavoidable stage in the slow recu-
peration of a liberal Republican order. While we should not discount
humanitarian motives, Republican government measures obeyed a
particular political logic. They were aimed above all at reassuring mid-
dling social constituencies whose confidence was crucial to the restitution
of a functional level of cohesiveness on the Republican home front. With-
out this there could be neither an adequate level of domestic political sup-
port for the Republic nor the practical mobilisation of human resources
imperative for the war effort. But for all their clear intent, government
initiatives were as yet extremely ‘weak’ in practice. For all Negrfn’s iron

120
See chapters 4 and 5 below.
121
For more on industrial and agrarian collectivisation, see chapters 4 and 5 below.
164 The Spanish Republic at war
'+ .

resolve, there was still a fundamental lack of working structures and per-
sonnel to enact them. The government’s authority continued to suffer the
serious damage and dislocation deriving from the rebellion. Moreover,
the military crisis rapidly supervening on the Madrid front was about to
deal it a further substantial blow.
By the end of October 1936 the rebels were on the outskirts of Madrid.
This was somewhat later than it might have been owing to Franco’s ear-
lier detour to Toledo to relieve the rebels holding out in the Alcazar
(atriumph which greatly consolidated his own leadership position 122 ).
This delay gave the Republicans vital time to begin organising the city’s
defences. Soviet aircraft and tanks made their first, crucial appearance
on the central front (24 October), to wild popular enthusiasm. But the
mobilisation forcivil defence was slow and knowledge of its real require-

ments lacking. The Republic’s military situation remained critical. In line

with cabinet guidelines, the press had previously been largely triumphal-
ist, but this was changing in Madrid and, anyway, the unpalatable reality

was increasingly common knowledge. 123 In this context, on 4 November

Largo Caballero announced the entry into his government of four mem-
bers of the CNT. 124 The result was an unwieldy cabinet of eighteen, 125 no
less internally fragmented than before. But in other important ways the

inclusion of the CNT was a positive sign. The military crisis had forced
the realisation that - notwithstanding British and French government
opinion or President Azana’s personal opposition 126 - a viable Republi-
can war effort depended upon worker mobilisation and this, in
full-scale
turn, required bringing the CNT fully on board. Although the CNT had
four titular cabinet portfolios, these really amounted to two politically
significant posts: the moderate syndicalist Juan Peiro and his colleague
Juan Lopez took over industry and trade between them, 127 while the
strong man of the Catalan FAI, Juan Garcia Oliver, was given the justice

122
Preston, Franco pp. 169-80.
,
The conquered Toledo on 28 September.
rebels
123
Arostegui and Martinez, La Junta de Defensa p. 40: on press triumphalism, see Zugazagoitia,
,

Guerray vicisitudes, pp. 173-5 an<3 G. Woolsey, Malaga Burning (Paris, Reston: Pythia Press, 1998),
p. 51 (first edition entitled Death’s Other Kingdom (London: Longmans, Green & Co., 1939)).
124
For the best account of the prior negotiations both inside the CNT and between it and the
government, see Lorenzo, Los anarquistas espaholes y el poder, pp. 177-90. Also, Casanova, De
la calle al /rente, pp. 18 1-6. For a full discussion of the implications for the CNT of entry to

government, see chapter 4 below.

125 126
For a full list see Bolioten, SCW, p. 203. Azana, Obras completas, vol. 4, p. 592.
127
For the pre-war syndicalist-radical anarchist division, see chapter 1 above. For its impact during
the war, see chapters 4 and 5 below. Peiro was one of those handed back to Franco by Vichy
France. He was executed in 1942. See his collected writings, Escrits igij—iggg, ed. P. Gabriel
(Barcelona: Edicions 62 ,1975) and a smaller pre-war selection in Trayectoria de la CNT (Sindicalismo
y anarquismo) (Madrid: Ediciones Jucar, 1979).
 Building the war effort 165

128
portfolio. His fellow FAI member, Federica Montseny, daughter of the
anarchist intellectuals Federico Urales and Soledad Gustavo, became
health minister and thus the first woman in Europe since Alexandra
Kollontai to occupy a ministerial post of cabinet rank. It seems likely

that the appointment of a CNT minister to the justice portfolio was

made with the intention of reining in the patrols, thus bolstering govern-
ment authority while tying the CNT’s credibility to this most sensitive
of tasks. Above all, however, the CNT’s presence in government was
intended to reduce - or at least equally apportion - the political fall-out
as the government abandoned Madrid. 129
Although the capital was not crucial to the military prosecution of the
war, its loss would mean a major and unaffordable blow to the Republic’s
already precarious legitimacy both in the eyes of its own population but
even more importantly with the very political establishments in Britain
and France whose support the regime was so desperately seeking. (Thus
Largo’s own view of Madrid as litde more than a ‘stomach’ was rather
130
too reductive. )
Indeed, the rebels’ eagerness to conquer Madrid
sprang precisely from an understanding of its political significance.
Nevertheless, on 6 November the Republican government took the deci-
sion to leave because virtually no one believed the capital could be held.
The under-secretary of war, General Asensio, had come to be highly
pessimistic, which doubdess explains why the prime minister chose this
moment to offer the combined defence portfolio to his rival Indalecio
Prieto. Unsurprisingly, Prieto refused, seeing it as an attempt to saddle
him with the defeat. 131 In the circumstances, a strategic case could
be made government departure from the capital, 132 but the secret
for
and rushed manner in which this occurred on 6 November created an
overwhelming impression of panicked flight. The political damage done
to Republican government credibility - and the PSOE’s in particular -
would prove considerable. Others too paid the political price. It cost the
CNT’s tminence grise and chief architect of its ministerial participation,
128
The choice of ministerial incumbents in all cases was the CNT’s own - and concretely Horacio
Prieto’s as general secretary.
129
This is conveyed very well in J. Garcia Oliver, El eco de fos pasos (Paris: Ruedo Iberico, 1978),
pp. 303-4.
130
Arostegui and Martinez, La Junta de Defensa, p. 66.
131
Zugazagoitia, Guerray vicisitudes, p. 178.
132
For example, that it would be militarily more efficient to let Madrid go, thereby gaining more
time to build up the new Republican army under construction on the centre front (‘en tie-
rras manchegas’). See Koltsov, Diario de la guerra espafiola, p. 19 1, who makes it clear that this
view originates with Asensio; Salas Larrazabal, Historia del ejercito popular de la Republica, vol. 1,
P- 545-
 66 The Spanish Republic at war

Horacio Prieto, his post as general secretary. Whether or not he directly

advised the CNT ministers to accept the government’s departure re-
mains unclear. But the CNT ministers must themselves have felt under
great pressure to accept government discipline givenwhat was at stake.
Horacio Prieto nevertheless stepped down to appease those in the CNT
who believed that it should have abandoned the government rather than
acquiesce. 133
In the event, the government relocated to Valencia rather than to
Barcelona, where President Azana had taken up residence towards the
134
end of October. The choice of Valencia seems largely to have been
decided personally by Largo Caballero subsequent to the cabinet meet-
ing of 6 November, which approved — somewhat stormily - Madrid
the
departure. No official explanation was ever given for the choice, and
contemporary commentators offer us little clear information. 135 The key
could very well lie in the notorious treatment to which several govern-
ment ministers (though not Largo himself) were subjected just outside
Madrid by militiamen belonging to the del Rosal column. Its checkpoint
at Tarancon refused to let them pass, threatening them with execution as
cowards for abandoning Madrid. 136 The ministers in question (including
Julio Alvarez del Vayo, thePC E’s Jesus Hernandez and the CNT’s Peiro
and Lopez) were forced to retreat, finding their way out to Valencia by
the back roads instead. Not only was this deeply humiliating for the indi-
viduals concerned, but it also served as a brutal reminder of the fragility

of Republican government authority beyond the limits of the capital.

Nor would it bethe last occasion on which PCE leaders were obstructed
by anarchist checkpoints as they went about government business. 137

133
F.Montseny, Mis primeros cuarenta anos (Barcelona: Plaza y Janes, 1987), pp. 105-6; Peirats, La CNT
1, pp. 219-21; Lorenzo, Los anarquistas espanolesy el poder pp. 177-206.
en la revolucion espanola, vol. ,

Lorenzo claims (p. 206) that the CNT ministers accepted the government’s departure without
consulting Prieto. Garcia Oliver says that Prieto was consulted. This discrepancy is discussed by
Arostegui and Martinez (La Junta de Defensa, pp. 59-60), but they draw no conclusion. In spite
of Prieto’s resignation, he continued to be a key figure in the political evolution of the anarcho-
syndicalist organisation during the war: see chapters 4 and 5 below. (The new secretary was
Mariano R. Vazquez: Casanova, De la calle alfrente , pp. 186-7.)
134 For Azana’s departure from Madrid, see his Apuntes de Memoria, pp. 79-83.
135 A summary in Arostegui and Martinez, La Junta de Defensa, pp. 57-61, which also indicates how
the manipulation of the issue in subsequent political disputes makes it even harder to clarify the

original decision.
136
Zugazagoitia, Guerray vicisitudes, pp. 180-1; Vidarte, Todosfuimos culpables vol. 2, p. 531; Bolloten,
,

SCIV, p. 206; Carrillo, Memorias, p. 190. A

full resume of memoir sources is in Arostegui and

Martinez, La Junta de Defensa, pp. 60-1.

137
See Arostegiu and Martinez, La Junta de Defensa, pp. 236—8 for the cases of Mije and Pablo
Yagiie; Gibson, Paracuellos, pp. 207-8. Later on also Pasionaria and others were prevented by an
anarchist frontier committee from crossing the Catalan border into France (in spite of carrying
passports), where they were due to embark on a propaganda and aid-seeking tour.
 Building the war effort 167

In these terms, then, Barcelona represented a veritable lion’s cage

-
the heardand of radical anarcho-syndicalism and hostile territory par
excellencewhere the batde against governmental authority was being
waged more fiercely than anywhere else. The Madrid government
would, moreover, be doubly a target: not only was it a government,
but it was the central government and as such symbolised Castile’s histor-
ical controlling pretensions. The very fact that the Catalan government

(Generalitat) was at this stage prepared to accept its presence offers some
indication of the parlous state of its own political authority in Catalonia in
the early months of the war. Valencia, on the other hand, also offered the
government a base deep in the home front, 138 but one where the CNT
moderates ( treintistas were strong and thus the anarcho-syndicalists - at
)

least at leadership level - would be significandy less unfriendly.

Meanwhile, in Madrid, rumours of the government’s departure spread
like wild fire, eliciting responses of popular anger and consternation, but

also of some cynicism and fatalism. These latter feelings were intensified
by the evident paralysis of government functions in the city: once the
cabinet had gone, ministerial buildings emptied rapidly as terrified civil
servants packed their bags. Even those not set upon immediate departure
seemed locked into a panic-induced torpor. 139 The mood amongst those
charged with Madrid’s defence was one of quiet desperation. Largo
Caballero had left the task in the hands of General Jose Miaja, appointed
for the purpose as commander of the Madrid military area. He had
instructions to liaise with General Sebastian Pozas, the new commander-
in-chief of the centre army, and to establish a defence council. In the
precipitousness of the government’s departure the sealed written orders
toMiaja and Pozas became mixed up and each initially received the
140
other’s envelope. This chaos reinforced the general feeling that Miaja
was being sacrificed to a hopeless task - as did the otherwise inexplicable
instruction (ignoredby Miaja) not to open his envelope until 6 o’clock
on the following morning of 7 November. Indeed, these orders from
the government specified guidelines for troop withdrawals on the fall of
the city but made no mention of defence provisions. 141 The government

138
Valencia was an oasis, with plentiful food in the early months of war - a marvel to the hungry
eyes of refugees from Madrid: Abella, La vida cotidiana, pp. 134, 160-2.
139
Zugazagoitia, Guerra y vicisitudes, pp. 182-3.
140
The accusation that this was a deliberate act of sabotage by Asensio has no real basis, although
it is repeated in the memoirs of those hostile to Asensio such as Santiago Carrillo and Politburo
member and ex-minister Jesus Hernandez. Arostegui and Martinez, La Junta de Defensa, p. 64.
141
A. Lopez Fernandez, Defensa de Madrid (Mexico: Editorial A. P. Marquez, 1945), pp. 147-50;
Arostegui and Martinez, La Junta de Defensa, p. 72. A
copy of Miaja’s orders is in Rojo, Asi file la
defensa de Madrid, pp. 255-6.
 68 The Spanish Republic at war

even gave orders that the anti-aircraft guns be removed to Valencia 142 .

Miaja himself certainly believed that Largo Caballero - and thus really
General Asensio - had chosen him to implement the surrender of Madrid
precisely because they saw him as expendable 143 Since Miaja’s failure .

to recapture Cordoba in August - a failure which reinforced the intense

popular distrust of the professional military in general 144 - his reputation
had been under something of a September he returned to
cloud. In
command came up against
troops in the Valencia region, where he
the reality of state collapse. Military structures were in disarray, and
soldiers and police were being directed by a plethora of party and union
committees - nominally under the authority of the Popular Executive
Committee. Miaja had to contend not only with his personal aversion to
this situation but also with the hostility which greeted him. The general

had already been outspoken about his lack of faith in the Republic’s
ability to resist when he declined to continue as defence minister in Giral’s
August cabinet, at the same time opposing the arming of the popular
militia 145 Moreover, it seems very probable that, before the war, he
.

was a member of the conservative officers’ association, the (Union UME

Militar Espanola), and his wife and children were still in the rebel zone 146 .

Incidents of sabotage by officers on the Madrid front during October 147

further increased popular suspicions, making them an ever-present and
sapping occupational hazard for all those professional officers serving
the Republic.
It was thus in this context of personal and professional alienation,

abandoned, so it seemed, by almost the entirety of the Republican po-

litical class, and with no concrete defence plan in place, that Miaja

and his exiguous military personnel faced an organisational challenge

of staggering dimensions: to build both the military and civil defence of
Madrid while also administering the city, feeding its refugee-swelled pop-
ulation and tackling the running sore of intimidatory violence and killing
by self-appointed patrols and other uncontrollable elements, including

142
Arostegui and Martinez, La Junta de Defensa, p. 60; Santiago Carrillo interviewed in Gibson,
Paracuellos, p. 202.
143 For a resume of the available memoir sources and an extremely sharp assessment of Miaja’s
situation, see Arostegui and Martinez, La Junta de Defensa, pp. 70-2; General Pozas also recalled
Miaja’s rage over this in an interview with B. Bolloten (Mexico, 1939), SCIY, p. 285.
144 Zugazagoitia, Guerra y vicisitudes, p. no.
145 Lopez Fernandez, Defensa de Madrid, pp. 63, 65. (Pozas, on the other hand, accepted its necessity.)
146
The evidence for UME membership is summarised in Arostegui and Martinez, La Junta de
Defensa, pp. 68-9 and Bolloten, SCW, p. 291 After five months in a rebel gaol Miaja’s family
.

was freed in a prisoner exchange.

147
Cox, Defence ofMadrid, p. 29.
 Building the war effort 1
69

crypto-rebels. Miaja was, moreover, facing these demands with frag-

mented and debilitated state agencies and a shortage of civilian person-

148
nel. In such circumstances, it is, then, unsurprising that he accepted
with alacrity the organisational support and man and woman power
which the Spanish Communist Party offered.
The PCE had resisted the government’s evacuation longest, giving in
only when it became clear that further opposition risked breaking up the
cabinet. 149 Individual PCE leaders may have had private doubts about
the feasibility of defending Madrid, but Comintern discipline determined
that the party as a collective entitywould concentrate its organisational
efforts and best leadership cadres in the capital. The key importance of
PCE support for Miaja was prefigured in the pressing demands facing
him on the night of 6 November: the military defence, maintaining calm
amongst the civilian population and containing the fifth column. Miaja
immediately consulted Major Vicente Rojo, who was to act as his chief of
staff. Rojo was a practising Catholic, but he had a pre-war reputation in

the army as a liberal and educated moderniser. Throughout November

Rojo would be, from behind the scenes, the real architect of Madrid’s
military defence: an achievement which would take him on to be prob-
ably the single most important military figure inside the Republic and
from May 1937 chief of its general staff. 150 On the Madrid front military
authority lay with Rojo and his officers who were direcdy responsible
to Miaja. The general also contacted and briefed all the political forces
that had militia active on the Madrid front. Among these briefings the
most crucial was with Mije, Santiago Carrillo and Jose Cazorla, the lat-
ter two leaders of the youth organisation, JSU. 151 From the very start

148
While Arostegui and Martinez’s very thorough analysis suggests that there were still basic
governmental structures in place, these still needed reactivating and staffing: La Junta de Defensa ,
pp. 62-3.
official PCE history of the war simply refers to cabinet unanimity in the decision to leave
149
The
Madrid, Guerray revolution en Espana, vol. 2 pp. 140-1 There is little specific on the PCE ministers’
,
.

interventions in memoir accounts by their cabinet colleagues, but Largo quite correctly saw the
PCE suggestion that the government’s departure be publicised beforehand as an implicit blocking
tactic: ‘Notas historicas de la guerra en Espana 1917-1940’ (Ms, Fundacion Pablo Iglesias,

Madrid), p. 482. The PCE also leaked news of the government’s departure in an attempt to
galvanise the formation of immediate defence measures.
150
Rojo’s function in Madrid was very much an anonymous one until an article by the Pravda
correspondent Koltsov in the PSOE newspaper El Socialists (21 Dec. 1936) drew attention to
his key role: Arostegui and Martinez, La Junta de Defensa p.
, 77; Martinez Paricio, Los papeles del
general Rojo , pp.
86-7; see also Koltsov’s diary entry for 10 Nov. 1936, Diario de la guerra espahola,
p. 236; also C. Blanco Escola, Franco y Rojo. Dos generates para dos Espahas (Barcelona: Editorial
Labor, 1993). For more on Rojo see chapter 6 below.
151
This preceded a further meeting on 7 November with both Mije and the socialist deputy and
commissar general, Crescenciano Bilbao. See Arostegui and Martinez, La Junta de Defensa ,
170 The Spanish Republic at war
** *

the communists had argued

and consistently for militarisation
strongly
and had constructed the Fifth Regiment as an exemplar. The party had
an important war commissar in Mije and experienced military com-
manders such as Juan Modesto, and had long demonstrated its general
support for professional army officers in the difficult circumstances of
the times. It is clear that the party endorsed the principle of separating
military and civil functions in the defence of Madrid. The PCE’s uncom-
promising line on public order was also crucial to Miaja. As elsewhere,
this appealed to those socially conservative middling sectors whom oth-
erwise the Republic ran the risk of permanently disaffecting. But even
more importantly, given Madrid’s position on the front line with the
enemy at the gates, a tough public-order policy was a crucial weapon
in the battle to neutralise an acutely effective and dangerous fifth col-
umn that had been finding it easy to pass military and political intelli-
gence to the enemy on the perimeter. 152 Thus the core responsibilities
given to the communists on the still-to-be-configured Defence Council
reflected Miaja’s belief that they would deliver what he needed.
The Defence Council was formally constituted on the evening of
7 November. At this meeting the component organisations of the Popular
Front nominated their own representatives for the portfolios allocated to
them by Miaja as a result of his consultations during the previous twenty-
four hours. Under the presidency of Miaja, the PCE had war, public
order (Carrillo) and - through the UGT incumbent - supply. Mije’s
function in war was essentially that of coordinating Council resources in
order to meet Miaja’s orders and requirements for the military defence.
The supply portfolio went to the communist Pablo Yagiie of the Madrid
UGT’s bakers’ and confectioners’ union (Artes Blancas). Here the party’s
known preference for centralisation and the reconstruction of govern-
ment control in the economic sphere must certainly have recommended
itself to Miaja after his own experience of fragmented committee power in

Valencia. The Madrid CNT (including Libertarian Youth) was allocated

war industries and information, while the republican groups Republi-
can Union and Republican Left covered finance and communications
respectively Ex-CNT
moderate Angel Pestana’s tiny Syndicalist Party
was responsible and the PSOE for the Council’s admin-
for evacuation,
istrative post (Secretariado). 153 Thus constituted, the Council would run

pp. 66-8 and pp. 75-9 for the complex set of meetings and consultations over the period 6-7
November before the Council was formally constituted.
152
See Council minutes, ibid., p. 447.
153
For a full breakdown of Council membership, see ibid., pp. 76-7.
 Building the war effort l
7

the capital and organise its civil defence throughout November - thereby
overseeing not only the first but also the most intense process of civil mo-

bilisation seen in the Republican zone during the war. References abound
to the extraordinary atmosphere of exhaustion and elation in Madrid
during November 1936 - with the seventh as the day of ‘iron in the soul’.
It was an experience which went deep, many participants remember-

ing those days as the most intense experience of their lives. 154 The days
of urgency in Madrid generated a high political profile for the Council
which would bring it rapidly into mounting conflict with the Valencia
government, whose own reconstructing authority had been set back by
its departure. Matters were made worse by the fact that the government,

convinced that the Council would not survive, had neglected to specify its
own understanding of the Council’s political role. The ensuing conflict
was multifaceted, as we shall see. It stemmed from the power fragmenta-
tion that still plagued the Republic. But unlike Barcelona or even Bilbao,
‘Madrid’ did not in any real sense represent a separate or antagonistic
political project to that of the Republican government in Valencia.
General Miaja’s function at the head of the Council was essentially
a symbolic one. He was not a gifted military strategist or technician.
(Indeed, his expendability was part of the reason why he had been
chosen.) Nor was Miaja possessed of particular administrative talent. But
he turned out to be a reasonably good manager of people and he had the
good fortune to have allocated, and to choose for himself, some gifted and
efficient collaborators. On the military side, he could rely on the vision

and abilities of Vicente Rojo. Miaja’s political collaborators were, for rea-
sons already explained, often communists. But this was not invariably
so: his aide-de-camp was a cenetista. Nor was Miaja blind to the party polit-

ical/clientelist implications of his strong reliance on PCE cadres. 155 The

fact that Miaja took the party card proffered by the PCE tells us little more
than that political membership was vital in a traumatised Republican
polity otherwise lacking adequate or credible organic structures. 156
As Miaja himself commented, faux-naively, in November to the Italian
socialist leader, Pietro Nenni, ‘I’m further to the left than you. You’re in
154
Thisis remarkable in most memoirs, but see the best example in Margarita Nelken’s preface to
Lopez Fernandez, Defensa de Madrid.
155
R. Gullon, ‘Justice et guerre civil: souvenir d’un procureur’, in C. Serrano (ed.), Madrid 1936-
I939- Un peuple en ambigue (Paris: Editions Autrement, 1991), p. 235.
resistance ou I’epopee
156
Miaja collected party and unions cards in the spirit of the true stamp collector, according to
R. Malinovski, Bajo la banderade la Republica (Moscow, 1975), cited in C. Zaragoza, Ejercito popular

y 1936-1939 (Barcelona: Planeta, 1983), p. 278. (The (rare) Malinovski

militares de la Republica

volume is a collection of testimonies by the most important of the Soviet advisers who served
in Republican Spain.)
 -

172 The Spanish Republic at war

** ,

the Second International, but, me, I’m a member of the Third - even
though I’m a political illiterate!’ 157 In fact, Miaja was an old-style mil-
itary conservative who came to keep a faith of sorts with the wartime
Republic - even though he had once sworn to Azana (when the latter was
war minister in 1933) that there was nothing for it but to shoot the social-
158
ists. But Miaja never acquired much grasp of the liberal democratic
principles underlying the Republican war effort: witness his half-joking
but nevertheless illuminating — retrospective comments that a Republi-
can victory could potentially have made a political career for him akin to
Franco’s. Santiago Carrillo would also later claim that in 1937 Miaja even
suggested it would be no bad thing for front-line morale if the PCE ‘took
things in hand’ and put an end to the internal political ‘wrangling’ in the
Republican government - an opinion which rings true because it reflects
the underlying cultural comfort which many army officers derived from
democratic centralist discipline. 159
Although many have passed judgement on Miaja’s (very real) mil-
itary and political limitations - his ‘provincialness’ — he was shrewd
enough to understand that as head of the Council his key function was
to navigate and adjudicate as best he could its internal organisational
political conflicts in order to deliver the wherewithal of military and
civil defence. Miaja’s vanity was certainly titillated by the popular accla-

mation accorded him as the ‘hero of Madrid’ by thousands of ordinary

madrilehos- especially women. (Miaja’s commitment to the Republic was,
- who may ties with the UME - in
i6 °
like Rojo’s also have had pre-war
161
a sense ‘made’ by his experience of popular resistance in Madrid. )

157
P. Nenni, Laguerra de Espaha (Italy, 1958; 4th Spanish edn, Mexico: Ediciones Era, 1975), p. 124.
158
Azana, Obras completas, vol. 4, p. 589.
159 The (unsourced) post-war comments are cited in Cortada, Dictionary of the Spanish Civil War and
Carrillo’s anecdote in his Memorias, p. 246; Azana also transmits this current of opinion in Obras
and abilities have always excited strong opinions
completas , vol. 4, p. 603. Miaja’s personality
for and and all have to be understood in the context of their political provenance and
against,
chronology. For a resume of sources with a sharp analysis of the interpretative difficulties, see
Arostegui and Martinez, La Junta deDefensa pp. 68-75. But, on a c l° se reading, these differences
,

of opinion rarely amount to outright contradictions about Miaja - something I have tried to
indicate in my own analysis here.
160
Rojo always denied it. See resume of evidence in Arostegui and Martinez, La Junta de Defensa,

p. 69. Barring the outright forgery of documents by the rebels (given that they were produced
in the context of the Causa General ), this would seem to suggest that there was a connection
-
unless Rojo’s ‘ficha’ (record card) resulted from his being ‘signed up’ by military colleagues -
not beyond the bounds of possibility, particularly in the case of the ‘list’ of members apparently
seen by Largo Caballero.
161
For a sense of this, see the text of the dedication (to anonymous Spanish women of the home
the
front) in Rojo’s book, Ast jue la defensa de had a
Madrid, written in exile (although the experience
stronger impact on Rojo, who was altogether more serious and deeper thinking).
 Building the war effort i73

But Miaja understood the underlying importance of his function as the

‘necessary hero’ in whom the population could trust, projecting a confi-
dent but empathetic authority and an absolute faith in victory. This is not
own myth, for all that
the same, however, as saying that he believed in his
he perceived - and was prepared to use - the opportunity it afforded for
personal advancement. Indeed, he displayed a robust and ironic sense
of humour in regard to his ‘myth’, awarding his ‘hero’s’ medals to a field
162
hospital . Miaja’s bluff bonhomie rang hollow toinformed military
and political observers, including Prieto and, later, Negrin 163 But it was
.

an important part of Miaja’s appeal to Madrid’s civilian population and

to its soldiery - as Azana recognised. Troop morale was a material condi-
tion of Republican survival, and for many of them Miaja remained a ‘un
gran jefe’ (a leader they believed in and felt loyalty to). The importance
of this myth lay in its ability to galvanise and channel the collective en-
deavour of defending the capital. By the same token, the myth or image
was not - as is often suggested - uniquely created by the PCE 164 But .

the communists’ particular understanding of the requirements of mass

mobilisation led them to disseminate it very energetically.
The especial importance of the PCE
to Miaja came from its unique
uncompromising position on militarisation, po-
profile: the fact that its

litical and economic centralisation and public-order policy was married

to an especial organisational dynamism. Although one must be careful

not simply to take the party’s own propaganda at face value, there is no
doubt that there was an aura of vitality and positive thinking (a sense of
‘can do’) about the PCE which made a great impression in a situation
where such qualities were at a premium. Miaja himself summarised it
excellently when he commented - again to Nenni - ‘the communists
are more capable and determined. The socialists have to talk about it,
weigh up the pros and cons, before they act. They have an enormous
capacity for sacrifice but little sense of initiative. With the communists
it’s actions rather than words — or at least if they discuss it’s after they’ve

taken action. In military terms that is a distinct advantage 165 The PCE’s .’

profile derived to a great extent from party discipline. But communist

dynamism was also about numbers. For even before the Madrid defence

162
Gullon, ‘Justice et guerre civil’, p. 236.
163
For the views of both Prieto and Negrin (the latter thought that Miaja was a dolt), see Azana,
Obras completas vol. 4, pp. 639, 678, 767.
,

164
Arostegui and Martinez, La Junta de Defensa, p. 75.
165
Nenni, Laguerra de Espana, p. 124; other witnesses also reinforce this view, for example the News
Chronicle journalist Geoffrey Cox, letter to the author, 3 Dec. 1999.
 1

174 The Spanish Republic at war

accelerated tfie PCE was expanding. A wide variety

process, the wartime
of new members - young - were joining party organisa-
especially the
tions, especially the joint communist and socialist youth organisation,
the JSU, whose leadership was fast moving into the PCE orbit.
The PCE’s connection to the Soviet Union gave it a tremendous boost
in October. For Soviet aid offered the besieged population of Madrid
crucial psychological support at a devastating moment
of complete in-
ternational isolation. It is this sudden craze among the
that explains the
city’s population for Soviet objects and images - hats, badges, mobil-
isation posters - and the warm reception given to Soviet
and the like
166
films. Films about the Bolshevik revolution and the civil war would
be used to boost troop and civilian morale throughout the battle for the
167
capital. But most effective of all here were the Soviet tanks and planes,
in action on the Madrid front by late October. Also important were the
military advisers sent under Soviet auspices. There would never be that
many of them — between six and eight hundred at any one time across
the whole Republican zone and a maximum of 3,000 personnel during
168
the war. But they played a vital role. Contrary to the impression often
given, however, this was predominantly a subordinate, technical role: the
3,000 included many engineers, interpreters and technicians of various
kinds and not least the pilots and tank drivers. 109 For the Republican
general staff, led by Rojo, was need of those with a practical
in desperate
knowledge of how to wage war
modern, mechanised conditions.
in
This need, and the gratitude and appreciation it initially stimulated,
did not of course preclude the emergence of tensions between Spanish
commanders and Soviet military advisers. In the course of time there
would be real conflicts over organisational prerogatives and protocol. But
166
Cox, Defence of Madrid, pp. 51, 72, 153; Lopez Fernandez, Defensa de Madrid, p. 162; Carrillo,
Memorias, pp. 185, 200; Zugazagoida, Guerra y vicisitudes, p. 128, who memorably conveys what
Soviet support meant in his image of a pure oxygen shot. Cf. M. Azana, Causas de la guerra de
Espana (Barcelona: Critica, 1986), p. 50 (text written in 1939).
167
R. Cruz, ‘jLuzbel vuelve al mundo! Las imagenes de la Rusia Sovietica y la accion colectiva en
Espana’, in R. Cruz and M. Perez Ledesma (eds.), Culturay movilizacion en la Espana contemporanea
(Madrid: Alianza Universidad, 1997), p. 302. Alpert, El ejercito republicano, p. 53 refers to the
freneticpace of such showings on various fronts and how - on one day alone (27 December
1936) - there were six showings in the vicinity of Madrid.
168
These figures derivefrom an estimate of Soviet human war losses 1935—41 published by the
Russian Ministry of Defence in 1998 am grateful to Frank Schaufffor providing me with this
.

material and for making the important point that the Soviet leadership’s main concern was that
its technicians gain valuable military experience from their engagement with the Axis in Spain.

See also Schauff, ‘Hitler and Stalin in the Spanish Civil War’, p. 9. Cf. Azana’s figure (provided
by Largo) of 781 Russians in Republican Spain in spring 1937 Causas de la guerra de Espana, p. 50.
,

169
Cox, Defence ofMadrid, p. 168; Zugazagoida, Guerra y vicisitudes, p. 128; cf. P. and A. Abramson,
Mosako roto (Madrid: Compania Literaria, 1994), especially pp. 67-87.
 Building the war effort i75

it is misleading to explain these in terms of a fundamental disagreement

over military strategy,still less in terms of Stalin’s desire to control the

Republican war effort. This would have run counter to his whole strategy
in Spain and, in any case, would have been physically impossible. First,
because Soviet personnel in Spain were too sparse (particularly political
‘advisers’ proper - as opposed to specialist technical personnel). Second,
because there was also a tremendous discontinuity of personnel. Staff were
constantly being recalled and replaced as a consequence of the vast
purges then gripping the Soviet Union. 170 The disagreements between
Soviet military advisers and Spanish officers were often a complex mix
of the cultural, personal-psychological and military-procedural. Many of
the Spaniards were ambivalent about the presence of foreigner advisers
per se - especially ones whose ideas and strategies might reveal their
own shortcomings. Some of the advisers lacked tact, failing to grasp this
human dimension, while others were high-handed, but others too were
reasonably discreet in handling prickly Spanish male, military amour-
propre. 171
The Soviet Union did not, of course, send fighting troops to Spain
in the way that Germany and above all Italy would, increasingly, for
the rebels. But the Comintern’s organisation of a volunteer force, the
International Brigades, which began to arrive in Spain in October 1936,
provided the Republic with a core of experienced fighters who could
be thrown into the breach to gain time. 172 The International Brigades
symbolised the Popular Front in arms. Their bulk was constituted by so-
cialists, communists, trade unionists - politically conscious workers from
across the world, but in the main from the continental European core of

170
Schauff, ‘Hitler and Stalin in the Spanish Civil War’, pp. io-ii, 19-20. For a useful attempt to
explain the sociology of the purges, see F. Schauff, ‘Company Choir of Terror: The Military

Council of the 1930s - The Red Army between the XVII and XVIII Party Congresses’, Journal
of Slavic Military Studies 12(2) (June 1999), 123-63.
,

171
on military advisers, doc. 77
Cf. Col. Sverchevsky (‘Walter’) to Voroshilov, retrospective report
(undated but c. August 1938) in R. Radosh, M. R. Habek and G. Sevostianov (eds.), Spain Betrayed.
The Soviet Union in the Spanish Civil War (New Haven/London: Yale University Press, 2001),
pp. 491-4; Bolloten, SCW, p. 278 cites Rojo’s endorsement of Republican military supremacy
against other testimonies which he seems to think argue the opposite. As cited, however, they
fit entirely with the analysis I propose here. It is also interesting to compare the 1945 and 1975
versions of Defensa de Madrid, the memoir by A. Lopez Fernandez, Miaja’s personal secretary. The
earlier version talks about various tensions between Spanish commanders and Soviet advisers
(for example p. 298), while the 1975 edition converts everything into a single patriotic reaction
against Stalin’s bid for hegemony in Spain. Observations on the cultural differences between
Spanish officers and Soviet advisers are in Gullon, ‘Justice et guerre civil’, in Serrano, Madrid
I 93^~I939->PP- 238-9. See also the illuminating anecdote in L. Crome, ‘Walter (1897-1947): A
Soldier in Spain’, History Workshop Journal 9 (spring 1980), 121.
172
Elorza and Bizcarrondo, Queridos camaradas, p. 324.
176 The Spanish Republic at war

France, Italy *and Germany. Initially, the Republic’s military command

hoped on rebel lines while the mili-
to use the Brigades in flank attacks
tia defended the capital itself. But such was militia demoralisation after

the tide of previous defeats that this plan had to be abandoned and the
Brigades used in the defence of Madrid itself. 173 But crucial though they
were, neither this emergency use of the Brigades nor Soviet military
technology/ advice could in itself have saved Madrid. To hold the six-

mile front along which the rebels were attacking required an enormous
effort of domestic militarisation as well as mass mobilisation on the civil-

defence front. And as Miaja’s comments intimate, it was here that the
174

PCE’s superior discipline and organisational dynamism would really tell.

One of the most important sources for the sustained mobilisation
which the PGE began via the Fifth Regiment was to be found in the
united youth organisation ( JSU). Expanding rapidly after 8 July, it would 1

provide vital forces not only for the Fifth Regiment but also then for
other battalions. Eventually some 70 per cent of the JSU’s total force 175
would be sustaining the Republican war effort at the front. JSU mem-
bers contributed significantly to the political commissariat too. These
developments need to be set in the context of a wider process of political
and social modernisation already in train and further accelerated by the
rebellion of 18 July. As Comandancia de Milicias records show, recruit-
ment to the militia, far from deriving from the ‘people in arms’ as a whole
or from the organised proletariat, came overwhelmingly from the young
(in this case male), unskilled and previously unmobilised sectors.
176
A similar
phenomenon of rapid youth mobilisation (of both sexes) underpinned
the vertiginous growth of the JSU across late 1936 and into 1937, upon
which an important part of the PCE’s strategy of permanent political
mobilisation was built.
The PCE used Soviet images to stimulate both military and civilian
mobilisation in Madrid. Propagandistic parallels were drawn between
the achievements of the Bolsheviks’ revolutionary army and the potential
of the new Spanish Republican army under construction. But there was
no lack of awareness among the party’s leaders of the great differences

173
Cox, Defence ofMadrid, pp. 69-70.
174
There were about 57 militiamen for every one brigader on the Madrid front initially. With
brigader reinforcements this became 32 to one. But the ratio would have increased again with
the arrival of militia from Catalonia and other areas: ibid., p. 106.
175 Carrillo, La juventud, factor de la victoria (speech to PCE Central Committee 6-8 March 1937)
(Valencia, 1937).
176
Alpert, El ejercito republicano, pp. 41 (n. 25), 62. The political as well as military mobilisation of
youth was a cross-organisational phenomenon: Fraser, Blood of Spain, p. 371.
 .

Building the war effort i77

between the two historical situations that scarcely favoured the Spanish
Republicans. Most was the disparity of military training. In the
critical

case of the Russian Civil War the mass base of the revolutionary army
had already received basic training for (First World) war service, which
also meant that arms were available. Spain had been neutral in the 1914-
18 war. And in 1936, apart from the lack of arms, there were very few
soldiers integrated in the militia. Most milicianos had never handled a
weapon before or received any training - hence their vulnerability, as we
have seen, as they faced rebel tanks, aviation and the Army of Africa’s cav-
alry and troops in open terrain. The defence of Madrid too meant regular
warfare - although in streets, this time, not in open spaces - which at
least reduced the militia’s disadvantage. But military training remained
the vital ingredient - hence the importance of the Fifth Regiment, and
also of the International Brigades, from whose example the Spanish
troops learned tactics as well as discipline. Trained and, above all, highly
mobile, the Brigades took the lead in November in holding up the Army
of Africa’s troops in the open scrubland of the Casa de Campo outside
Madrid. Then, by mid month, they were confronting them in bloody
hand-to-hand fighting in the buildings of the University City on the
western perimeter of the capital. Among the rapidly learning Spanish
troops, the political commissars played a vital role at this stage in prevent-
ing retreat under fire. Meanwhile, as well as training, the local fighting
forces on the Madrid front were also receiving reinforcements from other
areas of the Republican zone. They came via Valencia. As one contem-
porary commentator pointed out, Franco committed a major strategic
error in not making the closure of the Madrid—Valencia road his target
from the outset. Disdainful of his opponents, he spoke of letting the ‘red
rats flee’. But ‘the tide which flowed on it was not a rabble moving to
Valencia, but an army to Madrid’. 177
Among the reinforcements came several thousand Catalans, includ-
ing anarchists of the Durruti column. 178 Madrid was the The battle for
forcing house of anarcho-syndicalist pragmatism. Durruti himself came
rapidly to understand that the Republic’s survival depended upon its
ability to put an army in the field and that all such armies required disci-
pline and a command structure. 179 Among the CNT militia active on the

177
The quotations are from the News Chronicle and Daily Express correspondent, Geoffrey Cox,
Defence ofMadrid, p. 1 1 1

178
R. Sanz, Los quejuimos a Madrid (Toulouse: n.p., 1969).
179
See also the revealing Defence Council minutes for 14 November 1936 (in Arostegui and
Martinez, La Junta de Defensa, pp. 301-2). The Comintern representative, Andre Marty, opined
178 The Spanish Republic at war

Madrid front there was relatively little opposition to this, as they learned
the reality of these needs through their own hard fighting experience. The
Madrid CNT as an organisation also proved adept at producing mobil-
ising propaganda as excellent as the PCE’s. The issue which confronted
anarchists and communists in Madrid was never in fact militarisation
per se: it was political power, including that which flowed from control
of the militarisation process, as well as from the parallel one of recon-
180
structing civilian police forces behind the lines. It is simply wrong to
see the CNT - anywhere in Spain - as an anti-militarist monolith.
181

Nor indeed was libertarian anti-militarism itself a single, homogeneous

phenomenon. There were, of course, many instances of militiamen leav-
ing their columns rather than submit to a militarisation with which they
profoundly disagreed. 182 But in a rural society like Spain, resistance to
it was by practical needs (militarisation would pre-
also often driven
vent peasant-soldiers dividing their time between their land and the
front).

The death of Buenaventura Durruti on 20 November saw the birth

of an anarchist icon but also the death of a living symbol which the
Republic at war could ill afford to lose. He died in the environs of the
University City while returning from a tour of duty, probably as the re-
sult of the accidental discharge of a weapon - either his own or that of
183
one of his guards. The tragically arbitrary circumstances of Durruti’s
demise were suppressed, however, because all sections of the Republi-
can leadership were fearful - not only of feeding internecine political
feuds, but also, quite simply, of demoralising the troops battling to hold
Madrid. For the CNT his death would prove a watershed. 184 But the
liberal Republic also needed heroes like Durruti. The protection of the
legendary anarchist’s reputation was also a means of protecting what

favourably in October 1936 on the pragmatism of both Durruti and Garcia Oliver: Elorza and
Bizcarrondo, Queridos camaradas, p. 327 as did ,
CNT
newspaper editor Jacinto Toryho, No eramos
tan malos (Madrid: Toro, 1975), p. 136. This acceptance of discipline also holds true for other
on the Madrid front, such as Cipriano Mera: Bolloten, SCW, pp. 326-9.
anarchist leaders
180
The same was true for the POUM, which consistendy argued for a centralised army with a single
command - as copious articles in its own and the CNT press indicate. The real debate was over
who should control the army, how it should be organised, and, above all, what kind of social,
economic and political order it should be defending: Morrow, Revolution and Counter-Revolution in

Spain p. 71.
,

181
Ibid., p. 7 ;
Casanova, Anarquismoy revolucion, pp. 1 1 1 -13 (who also indicates the multiple meanings
of ‘anti-militarism’).
182
B. Bolloten gives several examples, including that of the Iron Column: SCW, p. 420.
183
For a forensic resume, see chapter 5 below.
184
For a discussion of Durruti’s death in the context of the internal crisis of the CNT, see chapter 5
below.
 ,

Building the war effort l

79

the Republican authorities chose to present as Durruti’s endorsement

of state, as well as military, reconstruction - encoded in the famously
resonant (but apocryphal) slogan ‘we will renounce everything except
185
victory’.
Victory — or indeed survival — at Madrid also demanded a major effort
of organisation on the civilian front. The waves of refugees arriving in
the capital had begun the process whereby the war entered popular con-
sciousness. This was reinforced by civilian involvement in fortification
work that accelerated in November. But it was the devastating impact
of the civilian-targeted waves of air raids between 14 and 23 November,
led by the German Condor Legion, which brought home the reality of
modern war to the population of Madrid. 186 But far from undermining
civilian morale, as Franco had hoped, the bombing reinforced it. Many
saw it as proof of what the refugees told of rebel atrocities to the south and
thus acquired a furious determination never to submit to the barbarism
of their compatriots. There would be no gas attacks, however, on either
was too indiscriminate a weapon. The front
military or civilian targets. It

lines were too close, winding their way in irregular, snake-like fashion,
while the political topography of the city meant that - even without
a strong wind - Franco would necessarily have risked gassing his own
supporters. 187 Madrid’s trade unions acted as agents of government, im-
plementing war measures according to their specialist competences - in
communications, transport, supply, fortifications, war-material produc-
tion and other civil-defence tasks. and communists took
Both socialists

the lead in establishing house or block committees on a neighbourhood

basis to organise civil defence. These committees also had responsibil-
ities for investigating government housing decrees to deal with refugee

housing needs and other war-related contingencies, and they also played

185
Although Durruti himself never spoke these words, the slogan was invented by the CNT: see
C. Ealham, editor’s introduction to J. Peirats, The CNT in the Spanish Revolution vol. i (Hastings:
The Meltzer Press, 2001), pp. xii-xiii. Cf. Geoffrey Cox’s view of Durruti as a ‘binding force’
for Republican/Popular Front unity; Defence ofMadrid, pp. 139-41. A classic defence of Durruti
as a radical anarchist and intransigent opponent of militarisation is to be found in A. Paz,
Durruti. The People Armed (New York: Free Life Editions, 1977) and Durruti en la revolucion espahola
(Madrid: Fundacion Anselmo Lorenzo, 1996). But this interpretation is contested: see Casanova,
Anarquismoy revolucion , p. 144 and De la calle alfrente, pp. 186-7, 251.
186
Cox, Defence ofMadrid pp. 114-31; G.
,
Hills, The Battlefor Madrid (London: Vantage Books, 1976),
pp. 105-8; Fraser, Blood of Spain, p. 175 (n. 1).
187
For more on gas, see Cox, Defence ofMadrid, p. 199. Franco did, nevertheless, request poison gas
bombs from Italy in August 1936: P. Preston, ‘Mussolini’s Spanish Adventure: From Limited
Risk to War’, in Preston and Mackenzie, The Republic Besieged, p. 49. For more on Franco and his
contemplated use of gas, see A. Vinas, Franco, Hitlery el estallido de la guerra civil (Madrid: Alianza,
2001), pp. 29—112.
180 The Spanish Republic at war

188
a role in checking the activities of fifth columnists. In an important
way too they represented the beginning of the reconstruction of the rela-
tionship between government and populace shattered by the rebellion.
But again it was the PCE in Madrid which led the way, quantitatively
and qualitatively, with organisational initiatives to link the home front to
the military front. The party would attempt to develop such initiatives
in other parts of the Republican zone, but these were arguably never
as successful as the ‘prototype’. 189 Catalysed by the Madrid campaign,
at the heart of the PCE’s strategy was the Fifth Regiment. Integral to
its functioning were welfare and support services. The Regiment had

medical and auxiliary training facilities (a school for nurses) and forged
links with women’s organisations, most notably the Popular Frontist As-
sociation of Anti-fascist Women (AMA) to provide support for soldiers’
families (which included provision of a nursery). The Regiment also de-
veloped links with specific war production factories through an ‘adoption’
scheme. It initiated literacy and general educational courses, organised
talks, filmshowings and exhibitions, and through its press and poster
campaigns promoted both its ideas on the new army and enhanced
communication between military and civilian fronts. The Regiment also
originated the idea of using loudspeakers to transmit Republican propa-
ganda to the rebel trenches. In keeping with the philosophy underpin-
ning its work, by the end of 1936 part of the Regiment’s barracks was
also converted to create the Casa del Combatiente — recreational and
educational facilities for soldiers’ use which anticipated the Republican
Army’s own Hogar del Soldado. 190 But by then the Regiment would be
winding down, as some 70 per cent of its base had been incorporated in
the Mixed Brigades and its separate support functions reverted, where
appropriate, to the Republic’s central general staff. (Its purpose served,
the Fifth Regiment would be formally dissolved on 27 January 1937.)
Through the Fifth Regiment, the PCE demonstrated not only its con-
siderable practical organisational skills but also that it understood, as
no other Republican group did, that the looming challenge of total war

188
For a vivid depiction of Madrid’s population waking up to the war, see Cox, Defence ofMadrid,
~
PP- 3 l 7-
189
See Mary Low’s comments on the'gaps’ in cultural mobilisation in Barcelona, where there was
a rapid resurgence of commercial popular entertainment: M. Low and J. Brea, Red Spanish
Notebook London, 1937; San Francisco: City Lights Books, 1979), pp. 224-7.
{

ejercito republicano p. 54. On the PCE’s cultural and educational work, see C. Cobb,
190
Alpert, El ,

Los milicianos de la cultura (Bilbao: Universidad del Pais Vasco, 1994) and ‘The Educational and
Cultural Policy of the Popular Front Government in Spain 1936-9’, in Alexander and Graham,
The French and Spanish Popular Fronts, pp. 246-53.
 Building the war effort i8i

required new forms of organisation. These had, by addressing welfare

and morale, permanent physical and psychological mobil-
to ensure the
isation of both military and civilian fronts so that each understood its
relation of mutual interdependence and the overarching reasons for the
fight. It was this understanding and the PCE’s ability to act efficiently

upon it which gave it the edge over both the CNT and the socialists.
The appeal of the communists, in Madrid first and foremost, was to
young people across a range of social classes. As the then JSU leader and
later (dissident) Communist Party theorist Fernando Claudin acutely
remarked: ‘they were attracted by the party’s military virtues and by a
simplified ideology in which the idea of revolution was identified with
191
anti-fascism mingled with patriotism’. To this should also be added the
attraction of the modern and novel. For the ‘Soviet Union’, whatever else
it may have represented in 1930s Spain, conjured up a powerful image

of political and cultural modernity which was especially appealing to

young (but also to some not so young) Spaniards seeking a model, or
at least examples, of change which might offer pointers for their own
situation and aspirations. Or, as a cultural theorist might put it, the PCE
was appealing to (and influencing) the Republican popular imaginary.
This also needs to be understood in the broader context of the cultural
and social reception of images of the Soviet Union across the 1920s and
1 93 os in Spain - an important topic that still awaits its researcher.
192

Nor was the phenomenon of youth protagonism confined to the com-

munist base. The JSU’s leaders were also by and large in their late teens
and early twenties. This reflected a post- 19 18 generational shift. But the
Republic at war saw this generational shift become a veritable revolution
youth leaders were given major national political respon-
as these ‘young’
Contemporary observers remarked on the youth of many of the
sibilities.

Madrid Defence Council members: for example, JSU general secretary

Santiago Carrillo was only twenty-one years old when he was appointed

191
F. Claudin, The Communist Movement. From Cominternto Cominform (Harmondsworth: Penguin,

I was generally important - not just to young people - as PCF leader

975), pp. 230-1. This
Maurice Thorez remarked at the start (September 1936). The PCE offered ‘very simple for-
mulas, popular formulas’ which, above all, ‘appealed to a sense of national pride’: Elorza and
Bicarrondo, Queridos camaradas, p. 318.
192
‘Russia together with other expressions of modernity such as “aviation, the radio, the telephone
gave life great interest” ’, E. Montero, ‘Reform Idealised: The Intellectual and Ideological
Origins of the Second Republic’, in Graham and Labanyi, Spanish Cultural Studies , pp. 13 1-2.
A suggestive preliminary study is Cruz, ‘jLuzbel vuelve al mundo! Las imagenes de la Rusia
Sovietica accion colectiva en Espana’; also E. Ucelay da Cal, ‘Socialistas y comunistas en
y la
Cataluna durante la guerra civil: un ensayo de interpretacion’, in S. Julia (ed.), Socialismoy guerra
civil, Anales de Historia, vol. 2 (Madrid: Fundacion Pablo Iglesias, 1987) p. 306.
182 The Spanish Republic at war

its Councillof*for PublicOrder 193 The war also saw an important gen-
.

der shift as large numbers of young women joined the JSU and took
up active war-related roles on the civilian front. (This provoked another
‘civil war within the civil war’ waged in many Spanish families in the

Republican zone as young women defied social convention to stay out

late at night on account of their JSU responsibilities.)
The most striking feature of communist membership in the war pe-
riod, however, was its hybridity. This is true whether one looks at younger
or older constituencies - the party proper or the youth movement, or
the various other Popular Front organisations, for example and most im-
portantly that for women (the AMA), which the party originated. This
hybridity is true for the Republican zone more or less as a whole -
although the configuration of motives leading people to join the party,
or at least to participate in Popular Front organisations, varied from area
to area. In Madrid and on the centre front the impact of the war was
naturally paramount. We have already discussed the particular case of
professional army But given the general, ‘neo-clientelist’ con-
officers.

text 194
which encouraged political membership of some kind or other as
a form of political and social protection in uncertain and chaotic times,
many men at the front opted for the Communist Party because it was the
best-organised, most disciplined and most efficient formation. This was
also true for many already subscribed as republicans or socialists as well
as those (a majority) who had not previously belonged to any political
195
organisation .

In Catalonia, where the military conflict remained a more distant

prospect for longer, the appeal of the Communist Party was still very
considerable (although here it was a separate party, the PSUC, which
retained a strongly Catalan nationalist steer 196 ). Communist appeal was
notable too in the eastern coastal region of Valencia, which always re-
mained ‘behind the lines’. In both places, and especially Catalonia, the
appeal was predominantly because the communist parties offered the
most robust defence of private property and ‘law and order’ in areas
which had been more affected than had Madrid by challenges to the
social and economic status quo: thus peasant smallholders and the own-
ers of industrial workshops and commercial premises joined the PCE or

193 Cox, Defence of Madrid, p. no; Lopez Fernandez, Defensa de Madrid, p. 207; Arostegui and
Martinez, La Junta de Defensa, p. 80.
194
Cf. Azana’s shrewd if cynical assessment, Causas de laguerra de Espana, p. 51.
195 Martin Blazquez, I Helped to Build an Army, p. 205. For the organisational chaos in the PSOE
caused by enthusiastic rank-and-file ‘fusionism’, see Graham, Socialism and War, pp. 75-6.
196
See chapter 4 below.
 Building the war effort 183

the PSUC By the same token, the total lack of any such
in droves .
197

threat in the Republican Basque Country, as well as the overwhelming

nationalist political loyalties of the lower-middle classes (encadred in the
PNV ), 198 meant that the PCE remained a marginal party The same was
true in the other area of the Republican north, Asturias. There, the only
significant middling class of property holders to whom the PCE might
have appealed was located in the regional capital of Oviedo, which was
held by the rebels.
But for all its new appeal to middling social constituencies in various
areas of Republican Spain, the PCE retained its pre-war working-class
membership. Indeed, it expanded this. The inter-class nature of the
wartime Spanish Communist Party - evident too in the JSU and the
women’s organisation, AMA, is the crux of the party’s importance during
the war. For what the PCE was able to achieve, at least for a time,
was the recreation of the inter-class Popular Front alliance exploded
by the military rebellion - within itself as a mass organisation. The
hypothesis proposed here about the importance of the PCE as a producer
of integrating, patriotic images and symbols for a fractured Republican
society/polity (in which, otherwise, there were far too many conflicting
symbols) needs more theoretical exploration than can be afforded it here.
Indeed, it also needs more empirical research. " But what is clear is that
1

the PCE’s ability to address and incorporate a range of different social

and political constituencies, and to vary political discourses accordingly,
meant that the PCE was the first party on the left to go a significant way to
realising the fundamental challenge of Spanish politics since 1931: that of
achieving mass political mobilisation across class boundaries. This pursuit
of a politically and socially modernising (but not a socialist) national
project had been prefigured in the pre-war speeches of the PSOE leader
200
Prieto, who was challenging his own party to take the lead here But .

197
The fact that, unprecedentedly, the Comintern allowed two (organisationally separate) commu-
nist parties - the PCE and the PSUC - to hold official representation for one country, Spain, is
indicative of the crucial importance of Catalonia to the Popular Front alliance /industrial war
effort as the basis of a viable Republican defence strategy.
198
See chapter 4 below.
199
There is some - as yet unpublished - theoretical work on the PCE’s cultural-political functions
during the war by M. T. Gomez, viaje/The Longjourney: The Cultural Politics of the
‘El largo
Communist Party of Spain 1920-1939’, Ph.D. thesis, McGill University, Montreal, 1999, and
much that is open to this interpretation in C. Cobb’s Los milicianos de la cultura. But we need to
ground this in an empirical social history of the wartime Spanish Communist Party - something
we as yet lack.
200
Graham, Socialism and War, p. 16. Prieto’s speeches in Discursos Jiindamentales; see, for example,
p.185 (Cine Pardihas (Madrid), Feb. 1934), p. 279 (Ejea de los Caballeros (Aragon), May 1936)
and pp. 255-73 f° r his visionary speech, ‘La conquista interior de Espana’ (Cuenca, 1 May 1936).
184 The Spanish Republic at war

it was to be tfie wartime’PCE that finally acted upon Prieto’s strategy

of ‘thinking as republicans’. As a result, there was increasingly little
that we would commonly identify as ‘communist’ in the content of either
201
the wartime PCE’s political discourse or its cultural policy, which was
liberal democratic through and through. 202 What was radical (i.e. new
or even ‘communist’) about the PCE in the war was not the content of
its policies but its organisational techniques. By marrying the two, the
wartime PCE would thus become the best republican party Spain had
ever known.
The inevitable question that arises is why the Spanish Socialist Party
was not fulfilling this key role. Until the military coup of 18 July it had had
a much higher profile than the PCE. Indeed, the socialists (party and
union) constituted Spain’s only mass parliamentary political movement
of the left in the pre-war period. Madrid especially had been a socialist
203
fief in the political sphere. The image of the government’s abandon-
ment of Madrid in November 1936 damaged the PSOE most of all the
political forces in Republican Spain, since it formed the nucleus of the
‘fleeing’ government. Moreover, the PSOE national executive commit-

tee also decamped to Valencia, thus giving the impression of systematic

leadership withdrawal. The socialists further erred in sending relatively
unknown rank-and-file members (Maximo de Dios and Fernando Frade)
to represent the party on the Defence Council - because it was perceived
204
as an ephemeral body. The PCE, in contrast, appointed national and
provincial leaders: Mije, and PCE’s Madrid ex-
Isidoro Dieguez of the
ecutive, as well as members of the JSU executive. (The fact that the
CNT’s national leadership had also ‘departed’ Madrid was crucially
compensated for by its ministers in the Valencia cabinet, Montseny and
Garcia Oliver, constandy moving between the two cities.) Many indi-
vidual socialists - known and unknown - did stay behind in Madrid to
fulfil important public responsibilities related to the city’s defence and

201
See Pasionaria’s speech of 23 May 1938 to the PCE’s central committee in which she reiterates
the policy with unprecedented force in the light of the acute military crisis of those days. See
also the comments of Vincent Sheean, an American journalist who heard the speech, that she
‘was asking these people to stop being Communists altogether, at least until the war was won’.
Sheean is cited in Preston, Comrades p. 302.
,

2°2 This is the conclusion drawn in Gomez’s thesis. In my opinion, it is also the conclusion to
be drawn from Christopher Cobb’s work. The PCE’s sectarianism was organisational, not
ideological - a point which some of Cobb’s reviewers have failed to understand (see, for example,

J. M. Fernandez Urbina in Historia Contemporanea 13-14 (1996)).

,

203
Although, as we have seen in chapter 1, the in Madrid was being challenged by the CNT
UGT
in certain sectors in the 1930s: Julia, Madrid, 193 1-1934.
204 Maximo
Arostegui and Martinez, La Junta de Defensa, p. 77. Frade leaves Madrid so de Dios
becomes the acting councillor: S. Carrillo interviewed in Gibson, Paracuellos, p. 203.
 Building the war effort 185

administration: for example Carlos Rubiera, who served as civil gov-

ernor and then in a variety of war-related committee functions such as
evacuation and supply 205 and Julian Zugazagoitia, the editor of the most
influential socialist newspaper. 206 But this presence did litde to counter
the prevailing impression of PSOE paralysis.
An obvious major contributory factor to the erosion of PSOE prestige
was the lack of support for the Republic from the organisations of the
Second (Socialist) International. By contrast, the optimism generated
by Soviet military aid procurement and Comintern support - in the
shape both of the International Brigades and of a high-profile European
solidarity movement gathering humanitarian aid and lobbying for the
lifting of Non-Intervention - rubbed off heavily onto the PCE. But the

passivity of the Second International is far from the whole explanation

for PSOE eclipse. For the socialists were already in massive internal crisis
before the war.
The adult party was and the resulting paralysis had
seriously divided,
encouraged the departure of the Socialist Youth (IJS), which, as we have
seen, decided, without the approval of the PSOE, to merge with its Young
Communist counterpart in spring 1936. 207 This was a step that marked
the beginning of the end for the party’s control of its youth movement.
But although the PSOE would blame much of its wartime plight on
this loss, the departure of the IJS was itself a symptom rather than the

underlying cause of socialist crisis. The reasons for this were bound up
with the very identity of the PSOE/UGT as a political entity. In sum, as
we have seen in the introduction and chapter 1 above, the very values
that had shaped the Spanish socialist movement historically meant that
neither the PSOE nor the UGT was able to respond effectively to the
pressing political challenge of mass political mobilisation. Historically,
the socialistshad sought to guarantee what they saw as the organisational
health of the movement by demanding a high level of political awareness
of their militants. The mass political mobilisation on the agenda in the
1 93 os meant that the UGT was obliged to accept new recruits with

a lower level of both general and political education. But there was

205
Rubiera was appointed civil governor on 8 October 1936. He was a stalwart of the skeletal
Madrid socialist organisation during the war who fulfilled various war-related functions and
was later part of the PCE-PSOE liaison committee in the capital. On Rubiera, see Graham,
Socialism and War pp. 123, 201, 239, 242.
,

206
On Madrid, as on the war in general, Julian Zugazagoitia’s memoirs are among the three or
four most important we possess. (Both Rubiera and Zugazagoitia were executed by the Franco
regime in 1939 and 1940 respectively.) On the PSOE in Madrid, Santiago Carrillo is right about
the impression but wrong on the details - interview in Gibson, Paracuellos, p. 217.
207
See chapter 1 above.
i86 The Spanish Republic at war

no understanding of how to integrate these constituencies into the

real
socialist movement and, moreover, there was an unwillingness to do so

(although this was largely unconscious) for fear of the organisational

consequences. At root, Spain’s socialist leaders - whether from party or
-
union saw politics as still the province of an elite inside parliament rather
than of a rapidly mobilising population pressuring for social change
from the streets. This underlying tension in the socialist movement was

already sapping its energies by the spring of 1936, and that, in turn, had

encouraged the departure of the Socialist Youth.

What was pressing before, however, became imperative after 18 July.
Mass mobilisation was now the sine qua non of Republican survival. Yet
the socialists became even more fearful of party expansion at such an
‘abnormal’ time. Hence the endlessly repeated advice from veteran lead-
208
ers not to ‘pescar en rio revuelto’ (‘fish in troubled waters’). As a result,
the wartime PSOE lost militants to the PCE - including some very high-
profile ones like the socialist deputy for Badajoz, Margarita Nelken, the
soul of Madrid’s defence. 209 Indeed, that her allegiance to the PCE was
made through the transformative experience of the siege makes her case
emblematic. In the end, the war would see the final crystallisation of the
PSOE’s internal political crisis in the emergence of two opposing ratio-
nales among socialists. There were those, like Largo Caballero and many
of his old union retainers (both reformists and radicals), for whom even
the war effort had to come second to the protection of socialist organi-
sational ‘heritage’ (although no doubt they never consciously mooted it

in such terms). 210 What this meant was keeping out threats new and old:
fifth columnists/ crypto-rebels, communists and the ‘ill-fitted’ of all kinds.
Alternatively, there were PSOE/UGT members - including leaders such
as Prieto, Ramon Lamoneda (the wartime PSOE general secretary) and
Juan Negrin himself - for whom the party was an instrument for achieving
national political reform. For this group the war was being fought as a
last-ditch attempt to salvage the very possibility of such change, and thus,
if need be, it would be worth sacrificing the Socialist Party to the battle.
Unfortunately for the PSOE, many of the socialists who believed this also
came to the conclusion that they could best ensure the maximisation of
their own wartime efforts from the ranks of the Spanish Communist
208
Graham, Socialism and War pp. 118-19.
,

209 She was famous for her rousing radio speeches: Zugazagoitia, Guerra y vicisitudes, pp. 186-8;
Arostegui and Martinez, La Junta de Defensa, p. 63 and P. Preston, ‘Margarita Nelken’, in Doves
of War. Four Women of Spain (London: Harper Collins, 2002), pp. 297-407.
210
For more on this see the discussion of the May 1937 cabinet crisis in chapter 5 below. New
members did join the wartime PSOE, but its growth was much slower than the PCE’s.
 Building the war effort i 87

Party. This was a decision based primarily, if not exclusively, on organ-

isational criteria rather than ideological differences. Indeed, one could

see the wartime PCE here as picking up the socialists’ ‘historic’ objec-
211
tive of renovating the state. And nowhere was the PCE’s contribution
to the renovation of government authority earlier evident than in the
besieged city of Madrid in winter 1936.
One of the Defence Council’s major objectives during this period was
the evacuation of the civilian population from the capital. Evacuation
in some form had been occurring since October, given the need to dis-
perse the accumulated waves of refugees from further south who had
fled there ahead of the advancing rebel forces. But from 7 November the
situation became acute. Madrid’s population of 1 million now stood at
some 1,300,000, swelled by troops and refugees. The additional pressure
on food and increasingly exiguous housing increased further with the
siege-induced displacement of Madrid’s own residents from the south
and parts of the west of the capital. The shock of the initial rebel assault
did persuade some to depart the city. But the bulk of madrilehos were ex-
tremely reluctant to leave or indeed to let their children go to unknown
conditions and anonymous carers — even though
this meant the greater

safety of the Valencia and Catalan regions behind the lines. (Although
212
as time went on and the food supply and living conditions deteriori-
ated this reluctance would decrease, especially when the children were
known to be heading for the children’s residences ( colonias infancies) set

up by various Spanish and foreign refugee agencies.) The only really

willing adult evacuees were those who followed family members taking
up government-related employment in Valencia. 213 People had a strong
attachment to their patria chica 2I4 This intensified what were also very
.

real fears about the fate of their houses and remaining belongings if they
left. The Madrid authorities were technically responsible for ensuring

that these came to no harm. But, in an emergency war situation with so

much else for the hard-pressed authorities to oversee, ordinary people
were, quite understandably, dubious. Moreover, everyone remembered
211
It is for these reasons that Burnett Bolloten’s presentation of wartime Spanish socialists turned
communists as ‘submarines’ isfundamentally wrongheaded. He ignores the context and tex-
ture of the 1930s - that is, historical chronology - interpreting the actions of the political
actors through a post hoc (and therefore anachronistic) Cold War interpretative schema. SCW,
passim.
212
As Quaker sources indicate, it rapidly became difficult to procure the range of foods needed to
meet the nutritional needs of growing children; also Cox, Defence ofMadrid, p. 155.
213
Lopez Fernandez, Defensa de Madrid, pp. 227-8.
214
There was some ironic humour too about the Junta trying to get the inhabitants to ‘do a bunk’
like the government: ibid., pp. 225ff.
i88 The Spanish Republic at war

the mayhem bf the early* weeks when ‘patrols’ and, indeed, individuals
had been free to appropriate virtually whatever they liked. This was no
longer the case, but the memory was still a vivid one. 215 Then there
were mothers, sisters and wives who were reluctant to abandon male
family members on the Madrid front or whose vital source of employ-
ment was in Madrid. (Adult males required express permission to be
216
evacuated. )
The Junta and its dependent bodies waged a continuous
propaganda campaign in favour of civilian evacuation, covered the cost
of the journey and guaranteed accommodation at the other end. But
popular reluctance was hard to diminish. Moreover, there were also se-
rious material restrictions on the rate of evacuation: vehicles and petrol
were in extremely short supply and there existed intractable organisa-
tional impediments deriving from the fragmentation of state power. 217 In
the course of November (when the rate of evacuation was at its highest)
some 250,000 people left. By 9 December the Junta reported that
218
the figure stood at over 300, 000. The evacuation was ongoing into
J
937 but
,
at a decreasing rate. Transport still presented problems, but
the decrease was mainly about people’s reluctance to leave - in spite
of numerous deaths in the air raids. In view of the increasing pres-
sure on resources (including human ones), the difficulty of guaranteeing
civilian safety and the growing fear of contagious disease, Council mea-
sures were passed at the end of 1936 to require the evacuation of non-
essential groups. But these were only ever seriously enforced for transit
refugees. 219 Towards the end of March 1937 the Council would claim
a total of 700,000 evacuations. But this was undoubtedly an exagger-
ation, being based on the statistics for valid ration cards issued in the
220
city.

From the Council’s perspective, a greater level of civilian evacuation

would have been useful. Apart from important humanitarian objectives,

215
A. Jacob toj. Reich, 10 April 1937, FSC/R/Sp/Box 1, file 2.
216
Originally interpreted as males between sixteen and fifty-five, this was amended to between
twenty and forty-five at the beginning of 1937: Arostegui and Martinez, La Junta de Defensa ,

p. 183.
217
Ibid., pp. 176-7. The railway system was used, but road was the predominant means.
218
My calculation is based on Arostegui and Martinez’s Council-based figures, ibid., pp. 178, 180.
219
Ibid., pp. 1 80-1, 183-4.
220
Ibid., p. 188. The number 700,000 is an overestimation even taking into account those who
organised their own departure without accessing Council facilities (and thus did not figure in the
statistics). It may be that transit refugees have inflated the figure. The analysis is based primarily
on Arostegui and Martinez’s work and on material from the Friends Relief Council (Quaker)
archive. (The 1930 census registered a population of 952, 832 for Madrid capital and 1,383,951
for the whole province.)
 Building the war effort 189

itwould also have made it easier to identify and contain the serious
danger posed in a besieged city by the clandestine activities of spies and
saboteurs. It was the presence of such a fifth column, with the enemy
at the gates of the city, which made public-order policy so serious and
fraught an issue. In particular, there were fears that, should the rebels
break through to the capital, their military capabilities would receive
a significant boost from the incorporation of the expertise of officers
currently imprisoned in Madrid’s gaols and, above all, those in the Model
(Modelo) Prison, which was right on top of the rebel lines west of the city.

As a result, Largo Caballero had instructed his interior minister, Galarza,

to implement the transfer of such high-risk prisoners (and especially army
officers) out of Madrid to gaols further inside the Republican home

front. But, with the typical lentitude afflicting all process in the Largo
government, no transfer had yet been effected when the government
itself left Madrid on 6 November.

It was thus the Madrid Defence Council which inherited responsibil-

ity for the transfer. But, by definition, this meant that it did so at precisely

the moment when the Madrid military situation went critical. The city
was under siege, with an enemy at the gate that had already shown it
possessed remarkably accurate military and political intelligence on the
Republic.
221
A number of foreign enemies and legations were sheltering
armed pro-rebel refugees. 222 Even without the fifth column’s contribu-
tion the Republican defence was highly precarious. Madrilehos had al-
ready experienced air raids and were fearful of more. If the rebels broke
through, would there be atrocities in Madrid like those in the south re-
ported by the refugees? The mood in the city was tense. Nor were such
apprehensions limited to the ordinary population. The nerves of those
political cadres who had remained to staff the Council were also taut:
the government had gone, they were alone with so much to organise
and with so few means. Maybe tomorrow they would wake up to the
rebels in the city and their own executions? The sense of living on the
edge, of there only being two choices in the fight — survival or oblitera-
tion - enveloped the November days, hugely increasing the fear of- and
animosity towards - the enemy within.

221
Cox, Defence ofMadrid, p. 175; Arostegui and Martinez, La Junta de Defensa, p. 447.
222
Defence Council minutes 14 and 24 November, reproduced in Arostegui and Martinez, La
Junta de Defensa, pp. 303, 326-7; see also memoir material: J. de Galindez, Los vascos en el Madrid
sitiado (Buenos Aires: Ekin, 1945); A. Nunez Morgado, Los sucesos de Espana vistos por un dipbmatico

(Buenos Aires, 1941); E Schlayer, Diplomat in roten Madrid (Berlin, 1938); Claude Bowers, My
Mission to Spain (New York: Victor Gollancz, 1954).
190 The Spanish Republic at war

The initialtransfer or prisoners, as arranged by the government before

its departure, took place over 7-8 November, during which period there
were evacuations from Madrid’s four Modelo. 223
gaols, including the
A number of convoys set out for gaols beyond Madrid - in Alcala de
Henares, Chinchilla and Valencia’s San Miguel de Los Reyes. But most
never reached their destination, and none at all made it from the largest
single evacuation - that of 970 prisoners from the Modelo. Some 1,200
prisoners were shot in the villages of Paracuellos del Jarama and Torrejon
de Ardoz, situated on the Aragon road on the outskirts of Madrid. 224
Throughout the rest of the month and into early December - effectively
until the end of the siege — there was a series of further smaller prison
evacuations that each time ended in the killing in this same area of
a percentage of the prisoners under transfer. 225 There had been sacas
before in Madrid and in other Republican towns, and these continued
into December 1936. They were invariably the result of popular violence
when angry crowds assaulted prisons following some provocation, such as
news of atrocities in the rebel zone or - most frequently - after air raids. 226
But Paracuellos was different. Those implicated here in extrajudicial
murder were the security forces charged with the prisoner transfers —
that is, the very representatives of Republican political authority whose
re-emergence was supposed to halt such abuses.
Precisely because of this, Paracuellos would become a major plank
in the Crusade martyrology confected to legitimise the Franco regime
after the war. The events were presented as proof of an orchestrated
killing machine sanctioned at the highest level by those exercising
power Given the anti-communist tenor of the Crusade,
in the Republic.
the accusation was levelled particularly at Santiago Carrillo, leader
of the PCE from i960, who, on 6-7 November 1936, as JSU secre-
tary, had become the Madrid Defence Council’s Councillor for Public

223 The others were San Anton, Porlier and Ventas; for gaol locations see map in Gibson, Paracuellos,
p. 128.
224
Because the rebels held the first part of the Valencia road on the exit from Madrid, Republicans
had to make a detour along the Aragon road and then descend to join the Valencia road further
along - see map, ibid., p. 128.
225
The material used in this section comes mainly from Gibson, Paracuellos and, to a lesser ex-
tent, from Carlos Fernandez, Paracuellos del Jarama: < Carrillo culpable? (Barcelona: Argos Vergara,
I983) -

226
The gaol in Guadalajara was assaulted in December with fatalities, and a similar attack was
narrowly averted in Alcala de Henares: Gibson, Paracuellos, p. 178. For a discussion of this
phenomenon and other examples, see chapter 2 above. Rebel air raids even succeeded in
breaking the social peace prevailing in the Republican Basque Country: there was a saca in
Bilbao at the end of September 1936 and again on 4 January 1937, when some 200 prisoners
were killed. For the war in the Republican Basque Country see chapters 4 and 5 below.
 Building the war effort I
9I

Order. 227 Also targeted was Carrillo’s then deputy, the twenty-four-year-
old Segundo Serrano Poncela, like Carrillo a former Socialist Youth
leader and JSU executive member. This targeting occurred in spite of
the fact that until 22 November the orders for prison transfers were
signed by interior minister Galarza’s deputy director of security (DGS),
228
the policeman Vicente Girauta Linares. In spite of the sound and
fury of Francoist denunciatory propaganda, however, there is no hard
evidence that Carrillo, Serrano Poncela or anyone else with ministerial
229
or political responsibility issued orders for the prisoners to be killed.
Certainly there is no solid basis for the conjecture (originating in Crusade
literature) that in signing evacuation orders Republican officials were in
230
fact giving coded orders to kill. Moreover, if the prisoner deaths at
Paracuellos were the result of a ‘well-oiled’ killing machine,how was it
that, on each occasion (excepting 7-8 November) a proportion of the
evacuated prisoners reached their end destination of prison? 231
On the existing evidence it seems no less possible or likely that what
occurred was the result of the anger and outrage felt by the mix of
Milicias de Vigilancia and regular police forces (the, then militarised,
Assault Guards) who undertook the various transportations. 232 These

227
Such was the perceived potency of Paracuellos that the episode would be reactivated in 1977 by
those hostile to the process of democratic transition then under way - see list of Paracuellos dead
republished, with a highly inflammatory commentary, by the extreme rightwing newspaper El
January 1977 The ensuing polemic led to the production of a number of examinations,
Alcazar, 3 .

including J. Bardavio, Sabado Santo Rojo (Madrid: Ediciones Uve, 1980) and those by Gibson and
Fernandez.
228
Girauta Linares, who left for Valencia on 22 November, was a member of the Cuerpo de
Vigilancia: Gibson, Paracuellos, p. 261.
229
See the accusations against Girauta Linares’ superior, the army officer and freemason Munoz
Martinez, who was Director General of police in Galarza’s ministry: R. Casas de la Vega,
El terror: Madrid 1936 (Madrid: Fenix, 1994), pp. 193-5 (also PP- 78-9, 105-12). For Girauta and
Koltsov, see pp. 198-206.
230
Gibson forgets his own caution over the limitations of testimonies from the Causa General (the
Franco regime’s (inchoate) state lawsuit against the defeated) as source material (see Paracuellos,
pp. 20-2) when he then adduces such a testimony as the sole proof of Serrano Poncela ’s involve-
ment, Paracuellos, p. 97. On the lack of evidence against Carrillo in the Causa General see also
Bardavio, Sabado Santo Rojo.
231
The expression is Gibson’s in Paracuellos, p. 233, but the idea comes from the Causa General.
232
Gibson never really addresses this crucial issue. But it is clear from his account and others that
as well as militia forces, regular police were also involved in some capacity - certainly in the
later sacas and quite probably in the ones of 7-8 November too. Schlayer, the German diplomat
who investigated the prisoner disappearances, refers to ‘state police’ recruiting militia for the
task -
as Gibson notes on p. 114. Gibson’s suggestion that it was Assault Guards is corroborated
by other testimonies such as Gullon, ‘Justice et guerre civil’, in Serrano, Madrid 1936-1939,
p. 235. The fact that the Assault Guards were militarised is also probably reflected in Carrillo’s
reference to the evacuee prisoners being in military custody: interview with Gibson, Paracuellos,
T
P- 97 -
192 The Spanish Republic at war

were the days of the major battlesin the Casa de Campo. On 15 Novem-
ber rebel forces crossed the River Manzanares into the University City,
where there was fierce hand-to-hand fighting. From the fourteenth, heavy
air raids were again being targeted directly against civilians. There were
also gruesome episodes such as the one in which the body of a Re-
publican fighter pilot, dismembered by the rebels, was dropped back
over Madrid. 233 Fears of the enemy within (and we should remember
that General Mola had named them as such) increased as the threaten-
ing presence of the enemy ‘without’ loomed large. To understand why
Paracuellos happened we have to return the events to the historical cir-
cumstances of November 1936 and the unique and acute vulnerability
of Madrid under siege. At one point the rebel front was only some 200
metres from the Modelo Prison.
In so far as any influence was exercised by Comintern and Soviet
political representatives over what happened to the prisoners, it was also
an indirect one. The well-documented concern of the journalist and
high-ranking political observer Mijail Koltsov about Madrid’s hostile
prison population derived from the Russian experience of revolution and
war under siege conditions. 234 But
concern registered only because
his
it resonated with the anxiety already felt by the Republic’s defenders.

Francoist sources repeatedly claimed there was intervention by the Soviet

secret police (NKVD). But this remains highly speculative. 235

233 Koltsov, Diario de laguerra espanola, Gibson, Paracuellos records Carrillo and other witnesses
p. 250;
on this, also citing the contemporary press reports, p. 219.
234 Koltsov, Diario de la guerra espanola,
pp. 191, 206-7.
235 Such an intervention by the NKVD in 1936 risked damaging the careful strategy of Popular
Front and attempted rapprochement to Britain and France of which Stalin still had real hopes.

But given the acute Madrid in November 1936 we cannot rule it out. As Burnett
military crisis in

Bolloten relates, there was a Comintern presence in the Modelo Prison. Members of the First
International Brigade had been sent in at the height of the danger of rebel army breakthrough
to Madrid on 7 November in order to replace the ordinary prison warders in the guarding of
military inmates: see Schlayer, Diplomat in roten Madrid, pp. 118-19. The Spanish police doubtless
listened to the opinions of Soviet intelligence advisers. But, as with Koltsov, these would have
been additional influences, not ‘directives’ which determined the course of events. Alexander
Orlov, who was NKVD chief in wartime Republican Spain, shed no light on his role to his
US debriefers of the 1950s. Nor does the NKVD material available to date refer to the Madrid
and O. Tsarev, Deadly Illusions (London: Century, 1993 ), passim. The authors
events: see J. Costello
assume Republican Spain was a blank screen to be ‘written’ on by the Soviet operatives. But
that
the archival evidence they present often undermines this assumption. Indeed, they cite Orlov
himself complaining precisely about the lack of efficient internal police security in Republican
Spain (see pp. 255-6, 264-6). It is also worthy of note that although ‘Soviet influence’ is the
background mantra in Francoist material, the real focus is always on the Spanish protagonists.
A more recent study of Orlov offers (with some lapses) a slightly more nuanced view of Soviet
intervention and Orlov’s role: Gazur, Secret Assignment. The FBI’s KGB General, pp. 29, 49-51,
332 4
- -
 Building the war effort l
93

By dint of their office, Carrillo and Serrano Poncela obviously bore a

general political responsibility for what happened. But this responsibility
was one of default. Instead of intervening to prevent - or at least to re-
visit the supervision of - further transfers as soon as it became apparent

that something was amiss (that is, within a day or so of the first trans-
236
fers of prisoners failing to arrive at their destination), Carrillo and
Serrano Poncela turned a blind eye. 237 Carrillo’s
subsequent insistence
that he only even heard the name ‘Paracuellos’ after the war is clearly
238
not credible. Nevertheless, in one important respect, Carrillo consis-
tendy relates a central emotional truth about how it felt inside Madrid in
November 1936: given the magnitude of what people were undergoing
in the defence of the city - the gargantuan effort of organisation, the
miltary struggle, the deaths in batde and in air raids behind the lines -
when the rumours came of prisoner killings on the outskirts they seemed
somehow remote and, in the immediate scheme of things, of secondary
importance. 239
The fact that Defence Council measures on public order brought a
rapid general improvement in the security of the population at large was
no doubt a contributory factor here. For it had been the widespread
and unpredictable rough justice of the night-time ‘patrols’ inside the city
which most fed public anxieties. Although it would take until the early
months of 1937 properly to eradicate such phenomena, the Council
at least reduced their occurrence inside Madrid to isolated, sporadic

236
R. Gullon, ‘Justice et guerre civile: souvenir d’un procureur’, p. 234.
237 I have chosen not to discuss Carrillo’s subsequent denunciation of Serrano Poncela here. First,
because it does not bear on the main thrust of my argument. Second, because it remains
difficult to substantiate and highly controversial in view of Serrano Poncela’s increasing political

alienation from theJSU during the civil war. (By 1938 he was helping the Socialist Party executive
inits attempt to re-establish a separate Socialist Youth: see Graham, Socialism and War, pp. 30-1,

226-31 - Serrano Poncela was the author of an important report on theJSU passed to the
PSOE executive in 1938 (Graham, p. 297, n. 31).)
238
Interview with Ian Gibson in 1982, Paracuellos, p. 230 and in Carrillo, Memorias, pp. 208-11.
On 1 1 November Carrillo reported on the prison transfer question at a special meeting of the
Defence Council chaired by Mije in Miaja’s absence. But the minutes are extremely brief and
oblique. Even from these, however, there is a sense that Carrillo and the others present have
some sense of the looming problems of prisoner security: see text of minutes in Arostegui and
Martinez, La Junta de Defensa, pp. 295-6 (see also pp. 230-3).
239 Interview with Gibson, Paracuellos, pp. 219-21. Note also the format of the Defence Council
minutes for 16-30 November (reproduced in Arostegui and Martinez, La Junta de Defensa).
Before and after these dates, Council meetings are identified by day, month and year. But the
entries for these dates - the most intense of the siege /fighting - are almost always indicated just
by the day. This was moment zero: there was no sense of an ‘after time’. None of this meant,
however, that the Republican government considered the Paracuellos killings to be unimportant.
Even so, there was no attempt either at cover-up or justification by recourse to raison de guerre :

Cox, Defence ofMadrid, p. 183.

194 The Spanish Republic at war

240
instances. This cumulative, collective improvement in public order
also saw the ending of the prison sacas. The CNT’s new prison delegate,
Melchor Rodriguez, who took up his duties on 4 December, increased
security measures around prison transfers and forbade nocturnal ones. 241
But crucially too, the ending of the siege itself reduced the tension in
the city enormously. In all, fifth column activity was largely brought
under control by the actions of the Council and would only reappear
as a notable problem when the material conditions on the home front
began seriously to deteriorate in late 1937, thus providing a new series
of opportunities.
But it is also vital to understand that the Council’s public-order mea-
sures, including those specifically directed at the control of fifth colum-
nists, represented something more than themselves. They were also part
of the dense battle for political power being waged inside the Republican
polity. In Madrid, the Council’s drive against the uncontrollables and the
fragmentation of political power saw it close the private prisons and in-
terrogation centres previously operated by various political organisations
and, in particular, the communists and anarchists. But this consolidation
of public-order machinery, in accordance with central government di-
rectives, did not end the organisational struggle; it simply reconfigured
it. It is in these terms, therefore, that one needs to understand much
of the conflict between the PCE and the CNT on the Council right
up to its dissolution by the Valencia government towards the end of
April 1937.
The JSU’s Jose Cazorla, who had replaced Carrillo as Public-Order
Councillor in December 1936, consolidated control of the police and
implemented rigorous and punitive measures against those
investigative
suspected of pro-rebel subversion and fifth column activities — either
sabotage or, at the worst end of the scale, conspiracies in the army or
violence aimed at discrediting the Republic internally and externally. In
the latter cases, he instituted a system of preventive detention of suspect
individuals pending the outcome of further police investigations lead-
in £ either to punishment (often in work brigades) or evacuation from
Madrid. 242
240
Arostegui and Martinez, La Junta de Defensa, pp. 228, 230, 444; Cox, Defence ofMadrid, p. 179;
Lopez Fernandez, Defensa de Madrid, pp. 245-9; Gibson, Paracuellos, pp. 194-6; Fraser, Blood of
Spain, p. 177.

p. 261 (n. 102). He remains a controversial figure,

241
Arostegui and Martinez, La Junta de Defensa,
however. See the startling instance recalled by Cazorla in the Defence Council minutes of
15 April 1937, cited in Arostegui and Martinez, p. 446.
242
Ibid., pp. 234, 241,447.
 Building the war effort J
95

These measures were rapidly denounced by the CNT on the Council

as a covert means of persecuting its members. It accused Cazorla of il-
legal detentions and, later (in April), of trumping up espionage charges
against cenetistas. This would explode into a damaging slanging match in
the press - in spite of censorship regulations, which were heeded by none
of those involved. The CNT complaints were not always without sub-
stance. 243 But Cazorla was strongly driven by the need to eradicate a real
and present danger to Republican security: the case he made powerfully
to the Council in session (on 15 April) was, indeed, unanswerable in its
essentials - hence the significant personal embarrassment of the CNT’s
representative (Marin) in having to insist on the contrary. 244 In particular
there was the delicate matter of pro-rebel embassies in Madrid serving
as a base for fifth column conspiracy. The Popular Courts often simply
acquitted people on whom they had no case notes without conducting
any independent investigations (since they did not have the resources). A
proportion of these indi\iduals, it was well known, subsequently holed
up in the embassies in question. Cazorla, prepared to take swingeing
action and deal with the constituticnal issue later, had many of them
rearrested at the doors of the Courts. 245
But although Cazorla could justify his public-order policy, the po-
licepersonnel (new or old) who implemented it may sometimes have
had rather more mixed motives when cenetistas were involved. Some no
doubt had directly party -political motives. But many of the Madrid po-
licemen involved would have been ‘new’, not ‘old communists — i.e. they
5

would have acquired a PCE membership card after 18 July 1936. They
were, thus, exactly the same policemen who had dealt with the CNT
before the war, with exactly the same attitudes. Cazorla was open about
the occurrence of some grave irregularities - including killings, kidnap-
pings and other forms of coercion. But he wason record as acting against
every case brought to his knowledge. Cazorla put his finger on the root
cause of the problem when he remarked that ‘today it is easier to get
away with being a thief than it ever was before — because the state is in

243 For a detailed and suggestive analysis of internal conflict on the Defence Council see ibid.,

pp. 136-42, pp. 226-42.

244 Ibid.,
pp. 440-54, especially pp. 445—9 for Cazorla ’s major intervention and p. 449 for Marin’s
discomfiture: ‘I’m in an awkward position, I have to represent the collective position of the
Organisation [Madrid CNT]; I can’t go and say to them that Cazorla’s explanation has con-
vinced me.’
245 See ibid, for Cazorla to the Council, minutes of 19 Feb., p. 41 1 and 27 Feb., pp. 421-2, where
he gives convincing reasons for some of these detentions. The embassy problem is present
throughout the minuted discussions: see also p. 446 (15 April 1937).
196 The Spanish Republic at war

disarray 246 Unfortunately that disarray also facilitated dubious be-

5
.

haviours inside the reconstructing agencies of the state. But nor did the

CNT occupy the moral high ground here. There were cases in Madrid
and elsewhere of illegal detention and the extortion of funds by individ-
uals who claimed libertarian affiliation. 247
The PCE representatives on the Council also retaliated, predictably,
but also with some truth, by recalling how, whenever there was insub-
ordination to government instructions (of whatever kind), there would
always be cenetistas involved somewhere. Indeed, the Madrid CNT rather
disarmingly confessed that it could not change overnight the deeply in-
grained anti-authoritarian attitudes of many of its supporters — war or no
war. 248 As a result, in this instance the CNT was unable to avail itself of
the support of republican and socialist members who, otherwise, sought
to maintain the PCE and CNT in equilibrium on the Council.
Although there was an ideological backdrop to the confrontation be-
tween the CNT and PCE in Madrid, there was no equivalent of the
confrontation looming by early 1937 between anarchists and commu-
nists in Catalonia over government directives to disarm the home front

and centralise food supply policy. The Madrid CNT held the key Coun-
cil portfolio of war industries. This required it to oversee the industrial
mobilisation upon which the supplying of the crucial central front de-
pended. This direct and immediate responsibility for maintaining the
front - in particular guaranteeing the rapid and reliable movement of
supplies and personnel - made the centralisation of such functions a
much less polemical issue in Madrid than it was in Catalonia (at least
as far as local CNT leaders were concerned). 249 In Madrid the root of
PCE-CNT conflict was an organisational struggle for power in the cap-
ital, as the socialist Councillor Maximo de Dios would openly identify in

246
‘No tenemos un Estado organizado’, ibid., p. 446. Although the Council gave Cazorla a public
vote of confidence in April (with the CNT abstaining), the very occurrence of such incidents
in his sphere of responsibility led to some private criticism from the other political groups
(Arostegui 255, n. 112). The ensuing enquiry into the Cazorla affair exposed
and Martinez, p.
some dubious activity by individuals in his employ: Broue and Temime, The Revolution and the
Civil War in Spain pp. 273-4. Nothing was proven against Cazorla personally. Cazorla, who
,

stayed behind as part of the PCE’s first cadres in clandestinity, was imprisoned and executed by
the Franco regime: see the recent memoir by his widow, Aurora Arnaiz, Retrato hablado de Luisa
Julian. Memorias de unaguerra (Madrid: Companla Literaria, 1996). Cazorla emerges as scrupulous
(see, for example, p. 34 )
though prepared to be tough and to take unpopular decisions if his

political responsibilities required it.

247 Arostegui and Martinez, La Junta de Defensa, p. 448.

248
Ibid., p. 238 and p. 416 for similar intervention by Marin (CNT), minutes of 19 Feb. 1937.
249 For the situation in Catalonia, see chapter 4 below.
 Building the war effort 197

February 1937. 250 Indeed, the very' fact that political parties and trade
unions were required to fulfil crucial public and social sendee functions
because of the deficiencies of the state was another factor exacerbating
the organisational competition between them. 251
The comments of another witness on the ‘rough’ or unpolished qual-
ity' of the political discussion at Madrid Council meetings also offer a
252
key' here: it is important to remember that, in addition to the pressing
circumstances of the war, democratic politics in Spain w ere very recent.
The impact of the war massively accelerated popular participation in
government process and fora. A good proportion of those involved in
the Council had had little or no pre\dous jobbing political experience. It
is important to factor in this ‘immaturity’, as well as the accompanying
hangover from clientelist political modes, in order to understand the or-
ganisational confrontations in council chamber, police stationand street
in wartime Madrid. 253 The production of mobilisation propaganda w as
also part of this organisational contest. Along with the PCE, the CNT
also participated energetically in the task. Here the PCE used Soviet
images and symbols in order to compete with the CNT for popular sup-
port. But this was a domestic war of symbols in which the Soviet Union
had no role other than as a depository" of images.
Although the PCE was a major player in the internal conflicts in
the Madrid Council, this w as not as some mythic instrument of Soviet
pressure. The disputes had little to do with Stalin or the Comintern.
Indeed, pushed to an extreme, this fierce organisational contest preju-
diced Stalin’s ov erarching objective of keeping the Republican coalition
afloat and the CNT on board the war effort. Moreover, once the Soviet
Union had begun to offer support to the Republic, the main medium
of Soviet-Spanish relations shifted to diplomatic channels broadly con-
strued. 254 By 1937 v arious kinds of liaison related to military support/
advisers, Soviet intelligence
Z
activitv; and even some matters concern-
ing the International Brigades were occurring via channels other than

250
‘Lo que aqui cxiste es una lucha de organizaciones Arostegui and Martinez, La Junta de Dtfensa
' .
,

p. 238.
251
For example, the unions oversaw various kinds of benefit - unemployment, sickness and acci-
dent - while both parties and unions provided canteens and nurseries, organised civil defence,
and ran searches for refugees /displaced persons on behalf of families.
2y2 ?
According to Miaja s secretary, the General had a bell which he would ring furiously to bring
meetings to order: Lopez Fernandez, Defensa de Madrid p. 259. ,

253
The stenographic record of the Councils deliberations - reproduced ibid. - offers invaluable
insights into this and its deleterious effects in wartime): see, for example, minutes of 19 February
especially Cazoria. p. 410 and 15 April 1937, pp. 440-54.
254 C£ Bolloten, 5Clf , p. 589; Schauff. 'Hitler and Stalin in the Spanish Civil War, p. 18.
198 The Spanish Republic at war

the Comintern’s. This was in part a consequence of the heavy purge

going on inside the Comintern itself. As a result, its role in Republican
Spain was in a sense shrinking as it shifted back to more specifically
PCE-related matters, although again with the emphasis strongly on how
best to achieve practical wartime mobilisation. 255 This shift was also her-
alded in the one instance where the Soviet Union did make a political
preference strongly known in regard to the Madrid Defence Council.
Although the PCE was clearly in agreement, 256 it was primarily through
diplomatic channels (including ambassador Rosenberg) that represen-
tations were made in early November 1936 to debar the POUM, as a
dissident communist (and for the Comintern a ‘trotskyist’) party. As a re-
sult, the POUM’s exclusion was accepted without even being discussed

by the Council in session. 257

But one must keep the Soviet veto in perspective. While it indicated
Spanish Republicans’ preparedness to take on board Soviet political
sensitivities because of their own military and defence requirements (that
is, to ensure Soviet support), it is important to remember that excluding
the POUM was not in itself a source of particular conflict or difficulty
for any of the groups. The POUM was a very marginal political force
inMadrid. 258 Certainly no other group on the Madrid Council had
any political empathy with it. Socialists and republicans were hostile
and the CNT largely indifferent. The Madrid POUM’s exiguousness
ruled out any real political hostility such as existed in Catalonia, where
CNT-POUM organisational rivalry was an important feature of the
Republican landscape. But even so, the Madrid CNT clearly felt no
need to object on political principle to the POUM’s exclusion. Indeed,
the only criticism of this, or indeed any, kind was made (implicitly) by
the newspaper of Angel Pestana’s Syndicalist Party when it carried a

255 Elorza and Bizcarrondo, Queridos camaradas, pp. 14, 329. This interpretation is also sustained by
recent research on Soviet-Republican Spanish relations during the war: Frank Schauff, doctoral
thesis, ‘Sowjetunion, Kommunistische Internationale und Spanischer Biirgerkrieg 1936-1939’,
University of Cologne, 2000. Cf. also the clear division of labour between Soviet representatives
charged with military, diplomatic and Comintern functions: Gazur, Secret Assignment. The FBI’s
KGB General 334.
, p.
256
Arostegui and Martinez, La Junta de Defensa, p. 130 (and n. 76).
257 Ibid.,
pp. 79, 80-1.
258 Statistical information about POUM
membership outside Catalonia (i.e. in Madrid and
Valencia) is no doubt that the party was tiny in Madrid. Some statistics,
sparse, but there is

including a figure of 70 for Madrid (area) membership in 1936, are in Durgan, B.O.C. 1930-
1936, pp. 556, 559. The Madrid POUM was to the left of the Catalan party - being mainly
constituted from Andreu Nin’s tiny Left Communist component (Izquierda Comunista): Mor-
row, Revolution and Counter-Revolution in Spain p. 85. For the
,
POUM in Catalonia see chapters 4
and 5 below.
 Building the war effort l
99

note from the POUM executive criticising the Council for weakening
the fight for a revolutionary victory over fascism by excluding it. 259 (The
Syndicalist Party was itself represented on the Council, in spite of being
even smaller than the POUM and having no militia presence.) Across
November and December, as the POUM also came under attack in
Catalonia, its public criticisms of the Soviet leadership, the Republican
government and the Madrid Council became increasingly vehement.
In response, it was made harder and harder for the to function POUM
politically in Madrid, and this culminated in the Council’s closure (in

January 1937) of POUM

press and radio in the city.
260
But given the
full-time demands of the war effort and the significant energy required
even to keep relations between the Coucil’s constituent groups on an
even keel, no one had the time or inclination to worry overmuch about
the fate of a group as marginal as the POUM.
The inherent source of conflict in the Madrid Council was not Soviet
interference but the fraught nature of relations with the central Republi-
can government in Valencia — a conflict which would eventually lead to
the Council’s dissolution in April 1937. This difficulty was partly struc-
tural in origin and Both aspects were encapsulated in
partly personal.
the deep mutually antagonistic relationship between Largo Caballero
and General Miaja. The Council’s inevitably high political profile dur-
ing the November days seemed to threaten reconstructing government
authority, already dealt a further blow by the cabinet’s departure. In Mi-
aja Largo saw both an insubordinate general and a popular hero whose
fame detracted from his own prestige. Both men had a wide streak of
vanity which increased the antagonism between them. In their excessive
concern to protect their own reputations and bask in popular acclaim
they were very alike. The subjects of many of their running disputes
seem scandalously petty given what was at stake in November. (And in
261
this respect the prime minister’s capacities outdid the general’s. )
But at
root the fundamental question was one of hierarchies of power. Once the
immediate danger to Madrid had passed by the end of November 1936,
259 Arostegui and Martinez, La Junta de Defensa, pp. 130-1.
260
Broue and Temime, The Revolution and. the Civil War in Spain, p. 254; Arostegui and Martinez,
La Junta de Defensa, p. 376 (minutes of meeting of 29 Jan. 1937); Bolloten, SCW, p. 298 and
n. 83, p. 835. Bolloten looks for a hidden, conspiratorial explanation of the measures against
the POUM and entirely ignores the force of the reasons given - the destabilising impact of its
vehement anti-government criticisms in wartime.
261
Lopez Fernandez, Defensa de Madrid (1945), pp. 169-70, igoff., 281, 285ff. The first of these
references describes Largo’s outraged sense of protocol when Miaja authorised the purchase
of a consignment of zips for pilot uniforms - which, in theory, required the war minister’s
approval.
200 The Spanish Republic at war

Largo insisted that the Council be restructured. The ensuing change of

nomenclature was highly significant: henceforward the Council was to
be called the Under-Council (Junta Delegada) and its members were no
262
longer ‘councillors’ (
consejeros) but ‘delegates’ delegados).
(<

But the conflict of competences with Valencia would be ongoing and

evident across the range of the Council’s responsibilities. In the vital func-
tion of supply, for example, the government was determined to reimpose
its control in the Madrid area over and above the Council’s. Supply cen-

tralisation was opposed by many of the local neighbourhood committees

in Madrid. But, unlike the contemporaneous conflict over the same issue
in Catalonia, Madrid’s again lacked an overtly ideological dimension.
Although there were some tensions deriving from people’s commitment
to patria chica there was no head-on clash over supply policy between the
,

CNT and the PCE on the Madrid Council.

Nevertheless, all the internal conflicts and those between Council and
government involved an expenditure of energy and time that could be
ill afforded in the circumstances. Against such a background of gov-

ernment dissatisfaction, it would be the conflict over public order that

eventually brought matters to a head. Cazorla’s confrontation with the
CNT exploded into the press, thus giving the rebels a useful opportu-
nity to make damaging propaganda about Republican disorder with an
international audience in mind. This probably accounts for the precise
timing of Valencia’s move against the Council. Nevertheless, the deci-
sion to replace it with a reconstituted municipal authority was entirely
in line with zone-wide Republican government directives first issued in
December 1936 as part of the ongoing process of state reconstruction. 263
Neither the PCE nor the CNT would express surprise. Indeed, the PCE
was already publicly calling for the Council to be dismantled in the cause
of the mando unico (single military and political command) it so strongly
advocated. 204
For all that premier Largo Caballero accused the Council of having
pretensions to autonomy, in fact it signified a crucial intermediate stage
in the process of Republican state reconstruction (from the bottom up as
well asfrom the top down) which would then be ‘exported’ with greater
other parts of Republican territory in the course of 1937.
difficulty to
For although in Madrid they knew that there was a full-scale war on

262
The basic distribution of functions remained the same, however: for a list of Council members
as of 15 January 1937 (and until its dissolution), see Bolloten, SCW, p. 296.
263 Largo gave no
The decree dissolving the Madrid Defence Council came on 21 April 1937.
particular explanation.
264
Arostegui and Martinez, La Junta de Defensa, p. 142.
 Building the war effort 201

by the end of 1936, few elsewhere fragmented Republican zone

in the
perceived this imperative. In Barcelona, the experience of emergency
defence - street fighting and the storming of rebel garrisons - had not,
of itself, produced any sense of the need to build a war machine. And
given the strength of nationalist feeling in Catalonia, still less was it

accepted that had necessarily to see the centralisation of political

this

and military power in order to organise total mobilisation of domestic

(human and material) resources. In the Republican Basque Country a
rigid nationalist political agenda would also conflict with the needs of
the war effort, as we shall see in the next chapter.
The battles to defend Madrid continued through the first quarter of
1937. There was bitter fighting and enormous casualties resulted, espe-
cially among the International Brigades in their role as shock troops. The
battle of Jarama (5-24 February), fought to keep the Madrid-Valencia
road open, would see the decimation of its British contingent and savage
losses among the American brigaders. The price was great, but so was the
prize: Madrid was to be a major defeat for the rebels. As their forces dug
in around the capital’s perimeter, the war turned into one of attrition. But
for all the psychological boost which the successful defence of Madrid
provided, in the harsh light of day the Republic was badly isolated by the
economic blockade which underlay the diplomacy of Non-Intervention.
It was facing the onslaught of Violent modernity’ in the form of full-scale

warfare, courtesy of the rebels’ fascist backers. Moreover, Soviet aid was
insufficient to do more than keep the Republic afloat. In Valencia, some
minds were concentrating ever more on the growing imperatives of state
centralisation in order to accelerate the maximum mobilisation and co-
ordination of the Republic’s internal resources. Only by so doing could
it engage in and survive the looming war of attrition.

In order to provide an infrastructural basis for such a programme of

political and economic centralisation, the Popular Front was formally
reconstituted in Valencia in January 1937 on the basis of an alliance
between the parliamentary PSOE and the PCE. It was agreed that it
should work towards establishing a network of inter-party liaison com-
mittees ( comites de enlace). The idea was that these would function in all
areas of home front war organisation at local, provincial and national
level, providing personnel and coordination to compensate for the weak-

nesses and gaps in Republican governmental machinery. 265 But immedi-

ately Largo Caballero and his supporters accused the PSOE executive of

265
For the origins and political dimension of the comites de enlace , see Graham, Socialism and War,
pp. 74-8. The circular was published in El Socialista, 7 Jan. 1937.
202 The Spanish Republic at war

factional manoeuvring against themselves. Although Prieto, the execu-

tive’s eminence grise, may have had some idea of this, far more important in

his calculations was the role of the inter-party liaison committees in sup-
porting the war effort. Indeed, the Caballeristas’ reaction here prefigured
what would rapidly become their obsessive trait: they would interpret
every policy or strategy of the PSOE executive solely in terms of its tac-
tical impact on the internal war which they were determined
socialist
to play out, come what may. They quite forgot that, beyond the relative
tranquillity of their bolt hole in behind-the-lines Valencia, there was a
real and very brutal war being fought for stakes which made those of the
266
internal organisational ‘war’ appear trivial indeed.
The understanding between the national leaderships of the socialist
and communist parties was, then, based on a set of shared political
preferences and the common objective of centralisation to serve the war
effort. The PSOE was as keen as the PCE to accelerate militarisation at

the front and liberal normalisation on the home front. The need for these
things was brought home forcibly to Prieto by his experience in the navy
and air ministry. He was constantly having to manage crises without any
real organisational back-up or even decently articulated communication
channels. In terms of the home front, the target in the spring of 1937
remained the regularisation of the state police forces. These were crucial
in the battle to centralise economic power being waged by Negrfn in the
treasury.
Negrfn spent the last months of 1936 and the beginning of 1937 build-
ing up the Carabineros. Recruiting heavily from PSOE affiliates, he made
the force directly responsible to himself as minister. 267 It was deployed to
curb the activities of the (mainly CNT-related) internal checkpoints and
frontier controls on the French-Catalan border in order to recoup state
control over foreign exchange. These, usually self-appointed, controls
also interfered with the free movement of ministers and personnel on
268
government business. In the medium term, the Carabineros would

266 This story is told in Graham, Socialism and War. For the ‘peacetime’ atmosphere in the Valencia
region see various memoirs including Lopez Fernandez, Defensa de Madrid, p. 145 (‘ciudad de las
fibres’ — ‘city of flowers’); Cox, Defence ofMadrid, p. 212.
267
Zugazagoitia, Guerra y Morrow, Revolution and Counter-Revolution in Spain, p. 68.
vicisitudes, p. 179;
The Carabineros were first brought under treasury control as a result of General Sanjurjo’s
attempted coup against the Republic in August 1932: see Payne, Spain’s First Democracy, p. 100.
On Negrin’s use of them, see Vazquez Ocana, Pasiony muerte, p. 59; Alpert, El ejercito republicano,
p. 306. The PSOE ‘ring fence’ around the Carabineros would only be breached after the division
of Republican territory in April 1938: see chapter 6 below.
268
This particularly enraged Negrin, for whom state prerogatives were an overriding value. See
F. Vazquez Ocana, Pasiony muerte, p. 59, also his article in Alvarez, Negrin (Documentos), pp. 252-3;
 Building the war effort 203

also be deployed to reinforce Assault Guards and National Republican

(ex-Civil) Guards in Negrin’s bid to impose state control over socialised
industries. The economic strategy was thus consonant
thrust of Xegrin’s
with the liberal statist political values which he. probably most of all the
PSOE leaders, endorsed. 269 But he also understood that accelerating the
centralisation of domestic industrial resources was crucial to compensate
for the increasing scarcity of external supply which Non-Interventionist
embargo would invariably mean.
It was in the logic of the PCE’s commitment to an inter-class policy

of Popular Front that the party would grow closer to the PSOE’s par-
liamentary wing and away from its former (pre-war) socialist left allies
grouped around Largo Caballero. \Vhile Prieto was delighted by the
transformation in the PCE’s policies, the socialist left, acutely aware of
the PCE's rapid wartime growth, and smarting from its loss of control
over the socialist youth, was becoming increasingly hostile. The power-
ful discourse of unity suffusing the Popular Front period in Spain had

intensified after 18 July, and by 1937 the PCE w as calling for the PSOE
to merge with it in a single party of the left (Partido Unico). Caballero’s
supporters reiterated their historic complaint that the communists were
intent upon absorbing the socialist movement. But the case of the JSU
notwithstanding, the Caballeristas’ anti-communist broadside, like their
criticism of the PSOE. ignored the wartime imperatives fuelling the
unit\' drive which also made it an initially appealing prospect to mam-
rank-and-file part)' and union members.
The Caballeristas. immensely shrewd and self-protective in matters of
organisational prerogative, were nevertheless right to discern a streak of
competitiveness in the PCE. This was epitomised in Mttorio Codovilla.
the Argentinian communist who had been Comintern representative in
Spain since 1932. By the end of 1936 Codo\-illa had come to be Largo
Caballero’s bete noire so insistent was he that the premier should give a
.

strong steer to socialist-communist unification. 270 But Codo\-illa’s obses-

sive ambition for the PCE derived far less from his Comintern function
than it did from a trait of personality. Codovilla had effectively ‘gone

Largo Caballero. \Notas historicas de la guerra en Espana 1917-1940*. p. 986. Xegrin's strong
opinions on the obligations of individ uals to the state for example in the payment of taxes
come through cleariy in the economic report to the PSOE National Committee. July 1937.
269
Xegrin's treasury work represented in microcosmic form his larger goal of putting the state ‘hack
on track* — cf. his hatred of the ‘arbitrary spirit' of the July days, Vazquez Ocana. Pasiony mucrte.
pp. 59-60: Xegin to Prieto, letter of 23 June 1939, Epistolano Prieto y \egrin. p. 42.
270
Elorza and Bizcarrondo. Queridos camaradas. pp. 329-30: Largo Caballero. Xotas historicas'.
p. 265 and Mis recuerdos. pp. 210-12.
204 The Spanish Republic at war
*4
# ,

and looked upon the PCE as his personal creature. 271

5
native in Spain
This became a particular problem after the Comintern’s adoption of
the Popular Front strategy. For while democratic centralism remained
unbreachable, there was, nevertheless, an attempt to foster the organisa-
tional capacitiesand talents of indigenous Communist Party leaderships.
The effects of the war itself would to some extent counteract
Codovilla’s influence. The huge scale of practical demands meant that
the PCE’s own leadership cadres had to take the strain. And, in a fast-
moving war, that meant they were also taking political initiatives and
gaining experience, becoming, in some ways, a real leadership. 272 Com-
munication between the PCE and the Comintern was, moreover, slow
and frequently out of synch. The Spanish Politburo often found itself
‘on its own in situations which demanded rapid political responses.
5

Codovilla’s influence was great but, in the maelstrom of war, with all
5
its myriad demands, not even he could be the ‘one-man executive often

claimed. Nevertheless, Codovilla’s overbearing behaviour and ambitions

were a problem, and from the beginning of the war he was increasingly
the target of Comintern criticism for a variety of reasons - although he
would not be removed from Spain until September 1937. 273
Largo Caballero interpreted Codovilla’s insistent demand for party
unification as an expression of Soviet leadership intent. But in this he
was mistaken. Throughout the winter of 1936 and spring of 1937 Stalin’s
primary concern was not to upset Largo Caballero and his wing of
the socialist movement since they held crucial cabinet posts and this
would have risked destabilising the Republican alliance and thus the war
effort. Stalin knew that Largo was the best guarantee of keeping the CNT

on board the war effort. Internationally too, Largo was saying the kind
of things about the parliamentary democratic nature of the Republic
and its war aims which Stalin believed should bring Britain and France
to reconsider their position on Non-Intervention. 274 Codovilla’s insistent

271
Elorza and Bizcarrondo, Queridos camaradas, p. 337; Falcon, Asalto a los cielos, pp. 144-6; Carrillo,
Memorias, pp. 262-3.
272
T. Rees, ‘The Highpoint of Comintern Influence? The Communist Party and the Civil War in
Spain’, in T. Rees and A. Thorpe (eds.), International Communism and the Communist International
igig-43 (Manchester: Manchester University Press, 1998), pp. 143-67.
273 See chapter 6 below. A. Elorza and M. Bizcarrondo suggest that the delay was because the
Comintern was very short of suitable representatives, Queridos camaradas, pp. 319—20. But
Frank Schauff has suggested that it was simply evidence of the arbitrariness and irrationality that
afflicted Soviet procedures: ‘Failure inEmergency: The Spanish Civil War and the Dissolution
of the Comintern’ (unpublished research paper, presented at the University of Bristol, 1 Mar.
2001), p. 25.
274 Graham, and War, Elorza and Bizcarrondo, Queridos camaradas,
Socialism p. 89; p. 329.
 ,

Building the war effort 205

pursuit of party unification from November 1936 was, then, going directly
against Stalin’s policy. But the reciprocally slow and cumbersome pro-
cess of communication between the Comintern and Republican Spain
obscured this fact for a time. 275 Eventually, however, on 8 January, the
Comintern telegraphed Stalin’s urgent instructions that the PCE should
backpeddle on party unity and instead promote unity of action, via
276
the mooted liaison committees and other avenues. Particularly to be
discouraged were high-profile cases such as that of Margarita Nelken,
the PSOE leader and parliamentary deputy who had joined the PCE
in November 1936 during the siege of Madrid. 277 Stalin’s instructions
would also seem to shed some light on why Nelken never rose to promi-
nence in the wartime PCE in spite of her undoubted talent, intelligence
and passionate commitment to the Republican cause. 278
But if Stalin saw the fortunes of the PCE in contingent terms, this was
precisely because he was enormously exercised by the need to ensure a
viable Republican war effort. And the primacy of this objective meant
that he instructed his diplomatic representatives, and in particular the
Soviet ambassador in Valencia, Marcel Rosenberg, to visit Largo daily
to discuss practical matters of war policy. But this was a thankless task,
as the PCE leader Mije had already discovered. Rosenberg and Largo
were soon as much at loggerheads as Largo and Mije had been earlier. 279
The prime minister would accept no advice on war ministry matters
even though such advice was seriously necessary. When it came from a
foreigner it was even easier for Largo to deflect it by a display of outraged
,

patriotism. No doubt the emotion was genuine. But that did not resolve
the underlying problems in the war ministry.
Then came the devastating news of the collapse of the southern city
of Malaga, taken by Franco’s Italian allies on 7 February. The repression
that ensued in the city - in spite of the lack of military resistance — saw
some 4,000 people shot in the week after conquest (and executions on

275 Cf. Carr, The Twilight of the Comintern pp. 3—32.

276
Elorza and Bizcarrondo, Queridos camaradas, pp. 338—9 (n. 80 cites the Comintern archive refer-
ence for the telegram); Schauff, ‘Hider and Stalin in the Spanish Civil War’, pp. 10-11.
277 Zugazagoitia, Guerra yvicisitudes, p. 187; also the less famous case of Francisco Montiel (PSOE

deputy for Murcia) and author of the February 1937 agitprop pamphlet ‘Por que he ingresado
en el Partido Comunista’ (Barcelona, 1937).
278
Preston, Palomas de guerra (Barcelona: Plaza & Janes, 2001), pp. 315-18 (and in English in Doves
of War, pp. 360-6).
279 Elorza
and Bizcarrondo, Queridos camaradas, p. 338 have critical details from Codovilla’s De-
cember 1936 report (Largo didn’t understand the concept of a Popular Army or mando unico\
he didn’t have the required sense of urgency about creating war industries; he communed only
with his ministry bureaucrats).
206 The Spanish Republic at war

a massive scale would continue for months ). 280 Refugees fleeing the
repression eastwards along the coastal road to Almeria were bombed
and machine gunned from the air, shelled from the sea and pursued
by a motorised column 281 All the reasons why Malaga had fallen en-
.

dorsed the criticisms of Largo’s leadership. Above all, it brought home

the vulnerability of the Republic’s military defence for as long as
the civilian front was not brought fully under centralised government
control.
Malaga’s geographical isolation had exacerbated the problem of arms
shortage, which, in turn, intensified various forms of internecine politi-
282
cal conflict among the Republicans holding out there . In particular,
anarchists and communists clashed over the forms of defence strategy to
be adopted. Both the arms shortages and the internal political conflicts
damaged morale, as, no doubt, did the awareness that the hard-pressed
government was concentrating its organisational energies mainly on the
central Madrid front. During the second half of January, as rebel troops
advanced towards the city, Malaga suffered severe bombing raids. There
were no anti-aircraft guns (rare anyway in Republican cities), yet it was
the end of the month before the government sent planes and (some) ar-
tillery. Although shortage rather than political malice was the underlying

reason for the delay, it is also true that the government felt that materiel
sent to Malaga would be poorly deployed.
Indeed, much more could have been achieved by way of military or-
ganisation had it not been for the fragmentation of political authority
in Malaga. The political committee and civil governor were neither ac-
tive in practical, war-related matters nor in contact with each other. This
was exacerbated by the total disconnection of political bodies in the town
from the military authorities based just outside it 283 These various sorts .

of internal disarticulation led to fatal inefficiencies: crucial infrastruc-

tural repair work was not undertaken (for example of a bridge), which
28° ranco told the Italian ambassador, Roberto Cantalupo, that Malaga was a particularly ‘red
jr

Jackson, The Spanish Republic and the Civil War, p. 345. On the repression and Malaga’s
city’:

place in the rebels’ pathological firmament, see Richards, A Time ofSilence, pp. 41, 62; Julia et al.,
Victimas de la guerra civil, p. 201.
281 - - madres que matan a sus hijos’, Azana, Apuntes de Memoria, p. 73;
‘100,000 fugitivos suicidios
foran oral history account see R. Fraser, In Hiding. TheLfe ofManuel Cortes (London: Allen Lane,
1972), pp. 149-52; other eye-witness accounts in T. C. Worsley, Behind the Battle (London: Robert
Hale, 1939), pp. 179-208; A. Bahamonde, Un aho con Queipo, pp. 126-36.
282
As elsewhere, of course, these labour conflicts had pre-war histories. At the end of 1936 the UGT
in Malaga had written to the national executive begging for arms - which meant prestige
-
and informing the committee that they had agreed to fuse with the local CNT since they had
arms. Actas de la UGT, 9 December 1936. (The UGT executive forbade the merger, of course.)
Vidarte, Todosfuimos culpables, vol. 2, pp. 649-50.
283
Borkenau, The Spanish Cockpit, p. 219.
 Building the war effort 207

hindered the delivery of arms even when these were eventually made
available.Nor were there enough fortifications - a situation worsened by
the desertion to the rebels of the two commanders in charge. The lack of
anti-tank trenches of adequate quality explains the very rapid progress
of the Italian forces. The isolation of the civilian and military fronts from
each other, which the visiting Austrian writer Franz Borkenau remarked,
was in part a result of the fragmentation of political authority. But it was
also a result of prevalent military attititudes.
Malaga, like other peripheral fronts, had seen less progress towards
militarisation. The defence forces were no longer the ‘militia’ of summer
1936, but in terms of training and mentality they were still a body of
men in transition. 284 Yet military professionals, like Colonel Villalba,
who was in charge, were far too inflexible in their understanding of
what constituted and ‘strategy’ to be able to respond effectively.
‘troops’
The tragedy of Malaga was that the resources the military commanders
wanted were simply not available, while those they had on hand they
proved unable or unwilling to use. Villalba did not understand his ‘troops’
and they did not trust him.
After Malaga’s collapse, questions were immediately asked about the
responsibility borne by the Republic’s military strategists and the south-
ern commanders on the ground. In particular, there was a redoubling
of the criticism of General Asensio, Largo’s under-secretary of war and
the main impeller of military strategy in the ministry The PCE had
long been critical of him, but now so too were socialists and republicans,
including President Azana. 285 These assessments all carried a heavy
political charge since they implicitly cast doubt on Largo Caballero’s
286
judgement in choosing his ministerial collaborators. On 14 February
a public demonstration in support of Largo arranged by the Valencian
UGT was transformed by the organisational efforts of the PCE into a
show of support for mando unico , a clean-out of the military command
structure, boosting the programme and, perhaps above
war industries
all, for the genuine implementation of military conscription. 287
The vol-
untary recruitment favoured by Largo was falling off and Malaga served
to clarify the urgency here. The PCE’s use of street mobilisation was

284
220 and cf. pp. 215-16.
Ibid., p.
285
Bolloten accepts that ‘[Asensio] was defeated by a broad opposition spanning the political
spectrum from the Anarchosyndicalist to the left Republican ministers’, SCW, p. 355.
286
peg criticism of the pernicious influence of Largo’s personal entourage was constant through
to the May crisis of 1937: see Manuel Azana’s diary entry for 20 May 1937, Obras completas,
vol. 4, p. 592.
2 7
R. Llopis, Spartacus, 1 October 1937, pp. 6-7; El Socialista 14 Feb. 1937; Bolloten, SCW, pp. 360,
345 -
208 The Spanish Republic at war

deprecated as cheap po’pulism by republican and, especially, socialist

leaders - even those hostile to Largo. But the fact remained that the
PCE had not invented popular lack of confidence in the professional
military - with distrust inevitably surging on occasions of defeat - and it

was less the need for military resignations that the PSOE disputed than
the PCE campaign itself. 288
In response to mounting pressure inside and outside the cabinet, to
which the CNT also contributed, 289 Largo finally agreed to remove Asen-
sio on 21 February along with Generals Martinez Monje (head of the

southern army) and Martinez Cabrera, the chief of the Republican gen-
eral staff. 290 Asensio was a far from incapable officer, if conservative and
conventional in matters of military organisation. He was also very ambi-
291
tious. This would explain his brutal contempt for fleeing militiamen.
Asensio took no account of their lack of training and on one occasion, in
the crisis conditions of autumn 1936 on the Madrid front, had had thirty
of them shot for leaving their positions. 292 This brought him into conflict
with the PCE. The party wanted to impose as strict a military discipline
as Asensio, but understood that first it was necessary to train and pre-
pare men who had started out as militia volunteers. Asensio ’s abrasiveness
had certainly been an important contributory factor in Largo’s decision
to move him from his field position to the war under-secretaryship in
October 1936 when the Madrid front was in grave disarray and militia
morale low because of past defeats. 293
In the case of Malaga, Asensio probably believed that it could not be
held with only the militia to defend it. He even seems to have considered,

288
As the PSOE press makes clear - see El Socialista between 12 and 20 February 1937.
289
Bolloten, SCW, pp. 355-9.
290 Martinez Cabrera’s critics saw him as the epitome of a fossilised Republican General Staff:

Alvarez del Vayo to Araquistain Feb. 1937, Araquistain correspondence, Leg. 23 no. 112b.
11
291
Cordon, Trayectoria , pp. 261-2. Cordon was a professional army officer under the pre-war
Republic, and he had also served in Africa. In spite of his PCE affiliation, his assessment
of Asensio is astute. There is a party-driven zeal to the criticism and also a puritanism that
may or may not be related. But the basic assessment fits with other (often at least partially
positive) assessments of Asensio: Zugazagoitia, Guerra y vicisitudes, pp. 166-9 ( al so PP- 117-18,
160). Biographical sketches of Asensio are in M. Teresa Suero Roca, Militares republicanos de la
guerra de Espana, pp. 27-52; Zaragoza, Ejercito populary militares de la Republica, pp. 221-3.
292
H. E. Knoblaugh, Correspondent in Spain (London/New York: Sheed & Ward, 1937), pp. 45, 210.
It was to this reputation that Indalecio Prieto referred in February 1937 when he remarked that

Asensio’s ‘antecedentes’ (record) made him disinclined to support him against his critics.
293 The more emollient General Pozas replaced Asensio on 22 October 1936. Rojo, Asi jiie la
defensa de Madrid, pp. 27-8, 40-1; Cox, Defence ofMadrid, pp 69-70. Bolloten claims that Asensio
was moved exclusively because of PCE pressure on Largo, SCW, pp. 280-1. But this does not
tally with Largo’s general recalcitrance in the autumn of 1936 in all matters of war policy and
especially where the PCE was concerned.
 Building the war effort 209

somewhat consequent shortening of the Repub-

controversially, that the
lican front would be an advantage. 294 Certainly his appointment of the
expendable Villalba, with his chequered professional history and ordi-
nary military record, lends credence to this interpretation (and indeed
one is reminded of Asensio’s earlier thinking in respect of Miaja on the
Madrid front). 295 Although Asensio thus laid himself open to charges
of negligence in respect of the Malaga front, there is nothing which
necessarily indicates that he was a saboteur. On the other hand, Colonel
Villalba’s precipitate departure from Malaga more than twelve hours be-
296
fore the rebels arrived left a seriously negative impression. His role in
the defeat - along with that of Asensio, Martinez Cabrera and Martinez
Monje - would be the subject ofjudicial investigation. The case against
them was eventually quashed and all would be rehabilitated to a greater
or lesser extent. But given the especial fragility of civil—military relations
in Republican Spain, none would serve again in pivotal, active military
commands. 297
But the removal in February 1937 of those held to be militarily respon-
sible for the loss of Malaga did litde to resolve the underlying tensions
between the Republican alliance and Largo Caballero over his run-
ning of the war ministry and the need to accelerate military, political
and economic centralisation. The attack was spearheaded by the PCE.
But, contrary to what is frequently suggested, its criticism of Largo’s
performance as war minister was not simply a ‘cover’ for communist
organisational ambitions. In any case, there was a rather more complex
picture of organisational rivalries being played out between socialists, an-
archists and communists. When Asensio was sacked after Malaga there
were also several other sackings of PCE members and sympathisers in

294 Cordon, Trayectoria, pp. 291-2.

2 95
Villalba had been pro-rebel in July 1936. But, at the head of the Barbastro barracks (Huesca),
he decided at the last moment not to join the rising, Cordon, ibid., pp. 292-5; Thomas, Spanish
Civil War, p. 239; Bolloten, SCW, p. 846, n. 6.
296
Villalba was apparendy allowed back into the Francoist army reserve after the war: Alpert El
pp. 387-8. Given his behaviour in July 1936 (see preceding note) it can only
ejercito republicano,

be assumed that he made an excellent retrospective case for himself as a fifth columnist.
297 Asensio wrote a memoir, El General Asensio. Su lealtad a la Republica (Barcelona, 1938) while in prison
awaiting trial over the Malaga affair. He had powerful supporters, including Rojo, by then Chief
of the Republican General Staff, whose letters are cited in Asensio’s book, pp. no-11. Negrin
also supported him: see Zugazagoitia, Guerra y vicisitudes, pp. 453-4. After his rehabilitation
Asensio was appointed military attache to Washington in January 1939 and would later be
identified with Negrin and the SERE in the immediate post-war period: Graham, Socialism and
War, p. 272, n. 21; Alpert, El ejercito republicano, p. 361. Martinez Cabrera was later appointed
military governor of Madrid. He would support the Casado rebellion at the end of the war and
was shot by Franco: Alpert, El ejercito republicano, pp. 377-8.
 , ,

210 The Spanish Republic at war

the war most ndtably of Cordon, who was sent to the Cordoba
ministry,
298
front. Moreover, it is also clear from the Comintern archival sources
now available that war policy, rather than socialist-communist unifica-
tion, was the nub of the dispute between PCE and premier and, above
all, between Largo and the Soviet Union’s diplomatic representatives throughout
the first four months of 1937. 299 In Catalonia, the fall of Malaga also
galvanised the PSUC. It organised a big demonstration in Barcelona on
28 February, with the participation of soldiers, to mobilise political and
popular support for the immediate militarisation of the militia on the
Aragon front and the call up of the 1934-5 military roll. To the same
end it also created the Committee for the Popular Army. 300
It was only later that the issues of war policy and party unification came
to be seen as two sides of the same coin in Largo’s confrontation with the
PCE. This interpretation has persisted largely as a result of historians
(including me) too readily accepting the conflation of these two questions
in both Largo’s own memoirs and those of his supporters. 301 Instructions
from Stalin had been received by the PCE Politburo on 8 January that
Largo was to be conciliated. It is then highly unlikely that Codovilla
would have been belligerent over party unification after this date. He
could have continued to be belligerent over war policy, but Largo refers

298
Bolloten presents intra-political conflict as if it were uniquely the result of PCE (and/ or Com-
intern) ambitions (i.e. a kind of ‘one-way traffic’), but his own evidence contradicts this: SCW,
pp. 355-8; also Cordon, Trayectoria pp. 296-8.
299 Elorza and Bizcarrondo, Queridos camaradas p. 337 The same point is made in the memoirs of
.

Pasionaria, Carrillo and Irene Falcon. See also Koltsov, Diario de la guerra espahola, pp. 352-3:
‘Everyone speaks ill of Largo Caballero: his enemies shout from the rooftops while his supporters
whisper criticism .The old man is inflexible too, he shouts, won’t let anyone gainsay him, he
. .

“decides” military matters as minister of war and all the rest as premier. If only that were true!
The problem is his decisions don’t resolve anything. Crucial documents on military operations
just pile up without being looked at, still less dealt with.’ I would like to make the immensely
novel suggestion that we take Koltsov literally here. The subtext of PSOE-PCE rivalry should
never be taken as the only message.
300
Bolloten, SCW, p. 419; see 228-9 and Low and Brea, Red Spanish
Abella, La vida cotidiana, pp.
Notebook, p. 217 for the previously voluntaryform of much defence work in Barcelona.
301
Bolloten, SCW, p. 352; Graham, Socialism and War, p. 90. Innumerable memoirs and secondary
sources do the same. In the end, however, we have all relied on Largo’s interpretation. Largo
Caballero, Mis recuerdos, pp. 210-12 and Notas historicas, pp. 264-5, 698. The former is Largo’s
published memoir. It was written up after his release from a German concentration camp in
1945 when he had no access to his papers. Largo Caballero died in 1946, but the memoir
was only published in 1954 (first edition). Its brief, elliptical and highly impressionistic form
makes it a problematic source for historians. Largo had, however, written a much fuller set of
memoirs between 1937 and 1940 while he had access to his papers. These (‘Notas historicas’)
c. 1,500 pages of manuscript can now be consulted in the Fundacion Pablo Iglesias in Madrid

(to date only the pre-war section has been published). At times the case made by Largo and his

supporters is so exaggerated that it appears as if the war itself is just an excuse for the PCE’s
campaign against themselves.
 -
1

Building the war effort 2 1

only to one wartime meeting with Codovilla - and that in relation to

the party question. This must have taken place before 8 January and
302
probably at the end of 1936. Largo gives no date in his memoirs, and
precisely because he does not the incident blends into his increasingly
heated interchanges with Soviet ambassador Rosenberg, which led to
theirfamous confrontation at the end of January 1937. 303 The subject
here was, once again, war policy (and, most particularly, Asensio). Largo
much
in fact says as later in his memoirs. 304
What makes Largo’s conflation of the issues seem something more
than forgetfulness - or even telescoping - is that it occurs not only in his
published recollections, which were constructed from memory after 1945
without the benefit of access to his papers, but also in the documentary-
based manuscript version written between 1937 and 1940. In this Largo
also presents his February 1937 meeting with the Republican ambas-
°
sador to Moscow, Marcelino Pascua, 3 5 as yet a further attempt to
force unification between the PSOE and the PCE. 3 ° 6 But this seems
highly improbable. Between the Rosenberg incident and Pascua’s visit to
Valencia, Malaga had fallen. The viability of Republican defence policy
broadly understood - was, from the point of view of the Soviet leader-
ship, now the issue eclipsing all others and surely the main reason behind
Pascua’s exceptional trip. Ostensibly he had come to hand-deliver a let-
°
ter from Stalin. But this was merely an introductory note 3 7 to a verbal
presentation for which we have no record other than the reference to pro-
posed party unification in Largo’s memoirs. 308 Pascua may have raised

302
Largo Caballero, ‘Notas historicas’, p. 265 and Mis recuerdos, pp. 210-12.
303 Dated by Bolloten, SCW, p. 350; Thomas, Spanish Civil War, p. 533 also gives the meeting as the
end ofJanuary (but offers no source).
304 The infamous meeting between Largo and Rosenberg (Alvarez del Vayo in attendance) which
ended with the ejection of the ambassador from the premier’s office is described by Largo
himself as a marathon session about Asensio, Mis recuerdos (1976), pp. 180— 1.
305 El Socialists, 12 Feb. 1937.
306
Largo Caballero, ‘Notas historicas’, vol. 2, pp. 264, 698.
307 Dated 4 February, it is reproduced in L. Araquistain, Sobre la guerra civily en la emigracion (Madrid:
Espasa-Calpe, 1983), p. 240.
308
Pascua’s own archive contains some private notes written following the May 1937 crisis. He
comments that Largo was a notoriously incapable prime minister in a war where the element of
international great power intervention was decisive and that he wasted crucial and irreplaceable
time in both the military and internal political spheres. Pascua also says of Largo that ‘no
correspondia la suma de sus calidades a lo que reclamaba/requeria el mito limitado por la
. . .

formacion intelectual que habia recibido’ (‘the real Largo’s talents did not match up to what the
myth demanded ... he was limited by his background and education’) - all of this in Pascua’s
private notes on ‘Largo Caballero’s socialism’ and ‘the fall of May 1937 Pascua correspondence,
’,

caja 5 (6) (AHN). There is an interesting echo too in the comments on Largo made by a News

Chronicle journalist in March 1937: ‘I had shaken hands with that shabby myth, Largo Caballero,
212 The Spanish Republic at war

the matter of*Asensio, or more generally the loyalty and competence of

Republican military personnel. But it also seems likely that he would
have been bearing a message that reiterated Stalin’s concern that the
Popular Front alliance should adhere to parliamentary politics and mar-
ket economics both to reassure Britain and France and also to keep
middling social sectors in the Republican zone on board the war effort.
On 2 1 December Stalin had sent a letter to Largo outlining the impor-
tance of these matters. But in his reply of 12 January Largo expressed
himself rather lukewarmly about the ‘parliamentary institution’ in Spain
(no doubt thinking back to the conservatives’ philibuster of social reform
pre-war period). 309 After the military defeat at Malaga
legislation in the
it became crucial to clarify Largo’s mind for him. Surely the real point of

Pascua’s was that he, as a Spaniard, might be able to convince where

visit

Rosenberg had failed. But given Largo’s previous berating of his own for-
eign office minister Alvarez del Vayo for supporting Rosenberg 310 this ,

tactic was scarcely likely to work. Indeed, it is clear from Largo’s mem-
oirs that he was immensely irritated by any reminder - from whatever
quarter - that the wartime Popular Front alliance had to be politically
and economically liberal or else it would not survive. Largo also refers
indirectly to his tiredness at being told ‘the lessons of the Russian Civil
War ’. 311
No
doubt there was an element of prickly patriotic pride and
‘culture clash’ here.But it is worth bearing in mind that what this ‘lesson’
also amounted to was being told that his great rival Prieto’s pre-war strat-
egy of republican-socialist alliance was indeed the only valid one 312 .

War policy would become the absolutely critical issue from the middle
of March - ironically, as a result of the Republic’s success at Guadalajara.
In retrospect, the victory was one of imposing a stalemate rather than

whose vanity did so much to dissipate the early strength of Republican Spain’, Philip Jordan,
There Is No Return (London: Cresset Press, 1938), p. 18.
309 Text of both letters reproduced in Guerra y revolucion en Espaha, vol. 2, pp. 10 1.-3.
310
Largo Caballero, Mis recuerdos, pp. 180-1; Bolloten, SCW, pp. 348-9 cites the published post-war
recollections of the Caballerista deputy, Gines Ganga.
311
Largo Caballero, ‘Notas historicas’, vol. 2, pp. 691, 694.
312
See chapter 1 above. In his public pronouncements during the war Largo had already accepted
this. Cf. his speech to the Cortes on 1 Feb. 1937 calling for social and economic normalisation

on the basis of the market (‘ya se ha ensayado bastante’) and the UGT executive’s manifesto fol-
lowing the fall of Malaga: ‘Ha llegado el momento de dejar en suspenso el logro de nuestros mas
preciados ideales Igualmente hay que suspender, por ahora, toda inovacion en los metodos
. . .

economicos y sociales con vistas al futuro’ (‘The moment has come for us to postpone our most
cherished political goaL Likewise, all innovation in social and economic affairs - that will
. . .

have to wait’), reproduced in El Socialista, 12 Feb. 1937. But the private political ‘reckoning’
in someone like Largo, who did not relinquish grudges easily, was quite another matter. The
evidence of his rancour fills the pages of the ‘Notas historicas’ for the war period and may well
be one reason why their publication seems to be indefinitely delayed.
 Building the war effort 213

making a positive advance. But it meant that the Madrid front was
now stable, with no possibility of a rebel breakthrough to the capital
in the foreseeable future. On realising this, Franco’s fascist backers drew
the fateful conclusion that only would
by massively escalating their aid
the rebels be assured of victory. This escalation meant that from late
March 1937, Germany and (especially) Italy were effectively at war with
the Republic. 313 The stakes in the Republic having a centralised and
rationalised strategic plan for military, economic and civilian resistance
had just risen a hundredfold. And no single minister understood this
better than did Negrin from the treasury. Rosenberg was replaced as
ambassador by his more emollient charge d’affaires, Leon Gaikis. 314
But time was ever more at a premium, and it was becoming clear that
Largo Caballero’s control of the defence ministry was an anachronism
the Republic could increasingly ill afford.
With the batde for Madrid indicating that the Republic could suc-
cessfully resist, even the notoriously pessimistic President Azana began
to think in terms of an outcome other than defeat. His deduction would
lead him in the direction of exploring the possibilities of a mediated
peace. In fact, Franco’s rebels would never be interested in mediation.
But Azana’s reactions in the wake of Guadalajara constitute the remote
origins of the polemic over a putative mediated peace that, in time, would
become the explosive centre of political strife inside the Republic.
The PCE’s rise to become a political force of the first magnitude in
the first year of the wartime Republic has to be understood within the
context of the internationalisation of the Spanish conflict and the growing
importance of Soviet aid to the Republic. But, as we have seen, there were
also important structural political reasons for PCE ascendancy. The PCE
assumed the mantle of progressive republicanism. Above all, it served to
bring the previously unmobilised - both middle and working class - to
the state. ‘Hybridity’ - or class heterogeneity - is the most oustanding
characteristic of the wartime Spanish Communist Party as it constructed

313 While Hitler remained relatively cautious about committing German resources to rebel Spain -
sending important equipment and the crack, technologically advanced, but small Condor
Legion - Mussolini ploughed in money and material: ‘financial and physical resources [were]
deployed on a scale which severely diminished Italian military effectiveness in the Second World
War’, P. Preston, ‘Mussolini’s Spanish Adventure: From Limited Risk to War’, in Preston and
Mackenzie, The Republic Besieged pp. 22, 48-51; I. Saz Campos, Mussolini contra la 11 Republica:
,

hostilidad, conspiraciones, intervencion (igji -igg 6) (Valencia: Edicions Alfons el Magnanim, 1986).
314
Bolloten, SCW, p. 383. On Stalin’s general instructions that Soviet political/ diplomatic person-
nel in Spain should be conciliatory, see Schauff, ‘Hitler and Stalin in the Spanish Civil War’,
pp. 10-11. On Rosenberg’s lack of the appropriate social skills, see Gazur, Secret Assignment. The
FBI’s KGB General , p. 56.
214 The Spanish Republic at war

itself during 1936-7 This can be seen as part of a broader modernisation

.

process speeded up by the war itself Of particular importance here was

the communist mobilisation of youth in the united youth organisation
(JSU) and its associate bodies. As a result, the PCE was creating within its
own organisational structure a model for the inter-class alliance sought
by republicans and socialists since 1931 in order to provide the social
base of a reforming regime. It is in this frame that we have ultimately

to understand the growing organisational antagonism between the PCE

and Largo Caballero’s sector of the socialists. But the issue of war policy
remained the primary source of conflict between these two groups during
J 93^
— a fact ^at has been somewhat obscured by the overwhelming
7 :

emphasis placed by the Caballeristas on the Comintern’s putative agenda

of absorption. An organisational rivalry certainly existed, and it would
become more acute as time went on. But this would have less to do
with the Comintern than it did with the struggle for power between
competing groups inside the Republican polity.
In spite of this emergent conflict, however, the early months of 1937
saw the Popular Front, as the ‘least weak option’, being reconstructed
as the basis of the Republican war effort. With the eclipse of repub-
licanism, its new axis was provided by the PSOE— PCE alliance. The
attitude of the socialist who led the cabinet was deeply antagonistic to
this alliance. In the medium term, this was a problem that would require

resolution. In the immediate term, however, the viability of the Popular

Front project depended on the full incorporation to the war effort both
of the Republican Basque Country (Vizcaya) and of Catalonia. In the
latter, however, the battle was still being waged to establish government
authority. It is thus to the situation of revolution and war in Catalonia
that we now turn.
 CHAPTER 4

Challenges to the centralising Republic: revolutionary

and liberal particularisms in Catalonia, Aragon

and the Basque Country

To articulate the past historically [. .

. ]
means to
seize hold of a memory as it flashes up in a moment of danger.
1

In Catalonia as elsewhere the collapse of the state, combined with the

protagonism of organised labour in resisting the military rebels, saw the
emergence of worker committees to rearticulate crucial supply, transport,
defence and public-order functions. But unlike anywhere else in Republi-
can Spain, the Catalan anarcho-syndicalists channelled the power which
their pivotal defence role - and the armed strength underpinning it —
2
gave them to spearhead in Barcelona and in other urban centres a wide-
ranging programme of industrial and commercial collectivisation in a
bid to reinvent not only the economy but also social and cultural life on
anti-capitalist lines. 3
The unparalleled range of what was attempted in Barcelona in the
months after the July days cannot be explained purely in terms of
Barcelona’s greater distance from the active front of the war 4 com-
pared to Spain’s other capital, Madrid. Certainly an emergency war
footingwould have required a different prioritising of radical energies
in Barcelona. But no amount of distance from the war could have
made 1936 Madrid a revolutionary city. What occurred in Barcelona
was rooted in the unique scale and richness of its popular and proletar-
ian cultures. Formed in the crosscurrents of resistance to a multi-layered

1
Walter Benjamin, ‘Theses on the Philosophy of History’, VI.
2
In some of these other industrial towns — for example in Lleida, Tarragona and Girona - the
dominant political force was the dissident communist POUM
rather than the CNT. Relations
between the two organisations are discussed later in this chapter.
3 ‘
“The revolution of July 19” was incomplete, but that it was a revolution is attested to by its
having created a regime of dual power’, Morrow, Revolution and Counter-Revolution in Spain p. 23. ,

4 Barcelona militias went to attempt the taking of Zaragoza (in rebel hands) and then some
(including Buenaventura Durruti) to fight on the Madrid front.

215
2l6 The Spanish Republic at war
• »

liberal commercial and artisan-industrial 5 development evolving over

decades and unparalleled anywhere else in Spain, these worker cultures
were shaped by the radical heterogeneity of a cosmopolitan port city,
but also by the federal political traditions and civic dynamism of the
bourgeois metropolitan culture against which they defined themselves 6 .

But for all that Barcelona was uniquely advanced within Spain, it
was still a place where ‘modern’ and ‘pre-modern’ worlds of popular
and labour cultures merged or, indeed, were actively linked through
the direct-action practices of the CNT, which constituted the dominant
form of labour organisation in the city. The Catalan CNT’s praxis had
been forged in the brutal labour wars waged by industrialists and the
Restoration state police on an unskilled and overwhelmingly immigrant
workforce in the years after the First World War. But the flexibility of the
CNT’s organisational structures and direct-action tactics meant that its

mobilising capacity extended beyond strictly libertarian-identified work-

ers through street sellers, the itinerant and the unemployed via a grey
area of semi-illegality to the underworld of petty crime: from the po-
conscious and organised to the ‘lumpen’, the CNT’s influence
litically

reached out across the myriad overlapping worlds of the urban poor of
‘outcast Barcelona ’. 7
Heterogeneous and amorphous though some of these CNT con-
and economic dispossession all shared
stituencies were, in their political
a common resistence — whether visceral /intuitive or consciously ideolo-
gised — to liberal economic order and the machinery of the liberal state.
What the Spanish state meant to the poor and marginalised was basi-
cally still the police and the army. It was thus associated with punitive
functions: with conscription and indirect taxation, when not with direct
persecution, especially of the unionised. The general brutality of daily
life - example a highly exploitative private housing market or the
for
ever-present problem of food procurement - also generated neighbour-
hood support networks (often centred around women’s activities) into
which the CNT could plug and which it, reciprocally, politicised: the
5 Although there were some very large concerns, industry in Barcelona - textile in the main -
was still mostly organised on a small scale in the 1930s as family-run businesses and workshops:
Shubert, A Social History ofModern Spain pp. 15-16, 19-21.
,
1

6
For details of the city’s industrial development, demographic growth, urbanisation, and more
complex sociological stratification compared to elsewhere in Spain, see Ealham, ‘Policing the
Recession’, pp. 27-30; also E. Ucelay da Cal, ‘Catalan Nationalism: Cultural Plurality and Political
Ambiguity’, in Graham and Labanyi, Spanish Cultural Studies, pp. 145-7. See also chapter 1 above.
7 The term is used by Christopher Ealham in his thesis ‘Policing the Recession’, and borrows from
Gareth Stedman Jones’ study of the world of nineteenth-century London’s poor and dispossessed,
whose image so terrified the affluent classes.
 Challenges to the centralising Republic 2 17

libertarians and the inhabitants of ‘outcast Barcelona’ were, for a time

at least, in their social war against the liberal state.
one
In its essentials, that relationship had remained intact, as we have seen,
during the Republican years between 1931 and 1936. The conditions of
daily existence for many reinforced the arguments of radical anarchists
8
in the FAI was but a new facade for the old order.
that the Republic
By 1936 many in the CNT doubted whether liberal governance (that
is, constitutional politics and a market economy) could effect structural

social and economic change. 9 But even after the coup-induced implo-
sion of the liberal state, it was only Barcelona that possessed sufficient
‘collateral’ to confront liberal social and economic order with anything

approaching an alternative project.

But although Barcelona was a unique proletarian centre in 1930s
Spain, and the CNT’s articulating political power singular therein, the
imperatives of the post- 18 July 1936 situation would deepen ideological
divisions and reveal organisational flaws in the anarcho-syndicalist move-
ment which together seriously undermined it. These, in turn, opened
up opportunities to the political opposition. For even in the days of
emergency defence against the military rebels, Catalonia - and even
Barcelona capital - never ceased to be contested political space.
When Barcelona’s workers faced down the military rebellion in the
streets, they did so in a context where the rural and urban middle classes
were far more uniformly hostile than any other sector of Spain’s bour-
geoisie to the ultra centralism of the military insurgents. The Catalan
bourgeoisie was more subde, but equally
also resistant to the relatively
real, centralist power in
values of the Spanish liberal republicanism in
Madrid since 1931. But neither of these hostilities meant that there was
any less of a contradiction between the anarchist and liberal federalist
visions of Catalonia’s politico-social order after the rebel military had
been defeated. While the worker militia had support from both munici-
pal and state police in Barcelona and while Catalanist values sometimes
influenced friendly officers to hand over weapons, we must remember
that Luis Companys, president of the Generalitat (the Catalan regional

The Anarchist Federation of Iberia (Federation Anarquista Iberica): see introduction and
chapter i above. There was an overlap between male youth gangs engaged in petty crime and
the FAI’s ‘grupos de afinidad’ (cells of young libertarians dedicated to direct action - such as rob-
bery and, sometimes, assassination - as acts ‘against the state’), Ealham, ‘Policing the Recession’,
pp. 17, 327-8, 346-70 and ‘Pimps, Politics and Protest’ (unpublished article), pp. 4-5.
9 Although most still voted for the Popular Front coalition in the February 1936 elections, this was
overwhelmingly motivated by die desire to secure the release from gaol of thousands of political
prisoners to whom the Front’s electoral programme promised immediate amnesty.
2l8 The Spanish Republic at war
• %

government), and leader of the (until then) hegemonic liberal-left Cata-

lanist party, the Esquerra (ERC), never formally sanctioned the arming
of Barcelona’s militia forces.
Once the anarchist movement-in-arms had, nevertheless, subdued the
military rebellion in Barcelona, the expropriationary new order emerg-
ing implied immediate jeopardy for the social and economic interests
of Catalonia’s urban and rural middle classes, large sections of which
constitued the Esquerra’s own base. 10 The coup-induced collapse of the
region’s liberal political structures and state institutions, in particular
related to law and order, meant that there was no physical means of
repression available to protect private property/ capital. (Once the city
was secure against the rebels, both Civil and Assault Guards would be
dispatched to the front - such was the distrust of the corps among worker
constituencies.) This is the context that explains the purposeful humil-
ity of Companys’ comments to the CNT-FAI leaders who met with the
Generalitat on 20 July:

Today you are the masters of the city and of Catalonia You have conquered . . .

and everything is in your power. If you do not need me or want me as President of

Catalonia ... I shall become just another soldier in the struggle against fascism.
If, on the other hand, you believe in this post ... I and the men of my party can . . .

11
be useful in this struggle.

Companys was performing a strategic retreat in order to achieve by

lateral means two key objectives on behalf of the political class and social
groups he represented: first to keep the concept of government legal-
ity formally in play and second to persuade the anarcho-syndicalists
that their revolution needed a central governing body. In both, Com-
panys was successful. The CNT agreed to the formation of a Central
Anti-fascist Militia Committee (2 July 1936) 12 whose legitimacy was thus
1

implicitlydetermined by Generalitat sanction. In the circumstances, this

was a staggering concession on the CNT’s part. Through it the Catalan
government was able repeatedly to assert its legal existence. Nor should
we consider this a question of mere form or rhetoric: it constituted the

10
The Esquerra contained both urban professionals and white-collar sectors as well as the Unio
de Rabassaires or rural tenants’ and sharecroppers’ union. The Rabassaires were the especial
base of Companys himself. See chapter i above.
11
The report of Companys’ address is to be found in a number of texts and memoirs. These are
listed in Bolloten, SCW, p. 389 (n. 17), from whom I also take these translated extracts.
12
The Committee had 15 members: CNT/FAI (5), UGT (3), POUM (1), Rabassaires (1), PSUC
(1), liberal republicans (4): full details are in Solidaridad Obrera, 21 July 1936. See also Morrow,
Revolution and Counter-Revolution in Spain, p. 21; Broue and Temime, The Revolution and the Civil War
in Spain, pp. 13 1-2.
 Challenges to the centralising Republic 2 19

first material stage in the battle to re-establish the Generalitat as the in-
strument through which liberal order could be reimposed. The CNT’s
13

ascendancy was reflected in its control of the central committee’s key

departments of defence, transport and public order. But the last of these
would rapidly become the supreme focus of tension in the battle for po-
litical control waged over the next twelve months between the CNT and
its liberal Catalan opponents. For the latter understood, quite correctly,
that if they could reconstruct their coercive force, then all other forms of
liberal reconstruction would be possible — as indeed they proved to be.
All of which leads one to ask the fundamental question of why the CNT-
FAI agreed in the first place to share power on the Central Anti-fascist
MilitiaCommittee.
The answer to that question lies in the ideological and organisa-
tional specificities of the CNT-FAL Its historical trajectory had scarcely
equipped it with either a political blueprint for the seizure and exercise
of power or the organisational structures through which to realise this.

At the time and since cenetistas have represented their behaviour as a con-
scious/willed rejection of ‘bolshevik methods’. 14 But, in reality, it was the
CNT’s ‘invertebrate’ organisational forms that clinched
limitations in the
matters here. The CNT still had very few vertically structured industrial
unions in 1936, and the CNT’s own structure was a highly decentralised
one. As neither the national committee nor the regional confederal com-
mittee in Catalonia had executive power over its constituent union sec-
tions, nor did they have particularly good communication channels. So

the dissemination from top to bottom of information - even leaving aside

the vexed question of instructions — remained a difficult if not impossible
15
task.

13
For a suggestive comment on the cultural dimension of this consolidation - via the amalgamation
of worker demonstrations and the civic symbols of Catalan nationalism (music, flags, police
wearing nationalist insignia) - see H. D. Freund, in ‘The Spanish Civil War. The View From the
Left’, Revolutionary History , 4(1-2) (winter 1991 -2) 321.
14
As late as May 1936 radical libertarian sectors of the CNT were still stressing ‘revolutionary
spontaneity’ as the means and autonomous rural communes as the goal. Cf. the editorial line of
the influential anarchist theoretical review La Revista Blanca, and the lack of any detailed strategy
(linking ideas and CNT’s Zaragoza congress (May 1936). On this see
practice) evident at the
‘
W. Bernecker, “Accion directa” y violencia en el anarquismo espanol’, in J. Arostegui (ed.),
‘Violencia y politica en Espana’, Ayer (Madrid), 13 (1994), 181 ff.
15
Cf. ‘Only a party with an iron discipline could have taken power - a party organised as if it were a
military unit, with its revolutionary general staff, its centralised and hierarchised structures ... In
the CNT ... to get something accepted, a militant had to argue at length and do the rounds
convincing people [in the various regional organisations] . How on earth - in such conditions -
could the CNT have taken power, even if its ‘leaders’ had wanted to? The anarchists had no effective

organisational machinery with which to fight the risings of 1932 and 1933 had already demonstrated
that’ (my italics). Lorenzo, Los anarquistas espaholesy el poder, pp. 192-3. See also chapter 1 above.
220 The Spanish Republic at war
•

The lack of industrial unions and of centralised CNT executive power

also made lateral communication between unions difficult. None of
these things had ever been absolutely crucial before. 16 The libertar-
ians’ strength - particularly in Barcelona - lay in their ‘bottom-up’
mobilising ability. But after 18 July 1936, the libertarian order needed

‘top-down’ political articulation and a political ‘head’ in order to de-

fend itself. Ironically, the lack of such facilities was itself an indication of
how successful purist anarchists had been in blocking the initiatives of
pragmatist-syndicalist currents in the CNT right up to 1936.
In July 1936, the radical anarchist leaders in Catalonia, for want of
any alternative political blueprint and also because, by confusing armed
strength with the totality of power, they believed that they were more
powerful than they were, opted to utilise the political experience, per-
sonnel and central apparatus of the Catalan government. In all of this, the
CNT were ceaselessly encouraged by Companys, an immensely shrewd
politician who rapidly assimilated the rhetoric of ‘the revolution’, using it
with consummate skill throughout the long summer, autumn and winter
17
of 1 93 6, even as he denounced libertarian ‘excess’ and campaigned
for regional government control of industry and the municipalisation of
services.
Companys could take heart from the rich base of social constituen-
cies supportive of liberal order to be found in Catalonia: ranging from
tenant farmers and sharecroppers - the majority of the rural Catalan
population — through urban white-collar workers, state functionaries,
liberal professionals and the owners of small businesses and industrial
workshops to the security forces (police and Republican army officers).
Although the existing political institutions of Catalonia had been bat-
tered by the coup, they still had this strong social base beneath them.
These constituencies had not often been on the front line against the
rebels, to be sure, but nor had they evacuated the zone or passed to tacit
support of the rebels. The situation in Barcelona thus resembled that of
‘dual power’ in Petrograd in 1917. But Catalan liberalism (represented
by the Generalitat Committee) was stronger
inside the Central Militia
than its Russian counterpart since it rested on a sounder social base.

16 -
Although they had hindered effective strike action in the 1930s see chapter 1 above.
17
‘Intoxicated with their control of the factories and the militias, the anarchists assumed that
had already disappeared
capitalism in Catalonia. They talked of the “new social economy” and
Companys was only too willing to talk as they did, for it blinded them and not him’, Morrow,
Revolutionand Counter-Revolution pp. 42-3. Companys’ credibility derived from his special rela-
,

tionship with the CNT - forged in the political battles against the central state in the pre-war
decades when he had acted as a labour lawyer for arraigned cenetistas.
 Challenges to the centralising Republic 221

But Companys also had to recognise that his own party, the Esquerra,
was an insufficient instrument to deliver the goal of liberal political and
economic reconstruction. The coup attempt had dealt an immense shock
to both its dynamic and the confidence of the leader-
organisational
ship. Paradoxically, the impact had been so great precisely because the
Esquerra’s hegemonic status in Catalonia between 1931 and the 1936
coup meant that no clear distinction had ever been made between the
18
functions/ apparatus of the party and those of the Catalan government.
In addition, Catalan liberals, although relatively more ‘modern’ than the
rest of progressive republicanism in Spain, still lacked adequate experi-
ence and understanding of what the arduous task of grass-roots political
organisation required. What was needed, then, on the wild new frontier
of political life in post-revolutionary Catalonia was a new kind of party,
able to mobilise its base ‘bottom-up’ (taking a leaf out of the CNT’s
book) and also unafraid to enter the fray because it had no pre-existing
organisational stakes orpower base to protect - only everything to gain.
The political force which emerged to fill this political space was the
PSUC (United Socialist Party of Catalonia), formed on 23 July 1936
from the merger of four smaller parties: (in ascending order of size), the
minuscule Catalan Proletarian Party, the Communist Party of Catalonia,
the Catalan section of the PSOE and the Socialist Union of Catalonia
(USC), a Catalanist social democratic party. 19 The most important force
in the new party, both quantitatively and qualitatively, was the USC, led
by the ambitious Joan Comorera. At the time of the merger the party
was still on the margins of political life, unable to compete with the
Esquerra’s umbrella appeal to rural as well as urban constituencies. In
urban centres outside Barcelona the USC had also been contending with
serious competition from the Workers and Peasants Bloc (BOC), which

18
Esquerra dominated from the 1931 elections that brought the Second Republic to power. The
Catalan autonomy statute was passed in September 1932. For Catalan politics 1931-6, see
chapter 1 above.
19
The Catalan Communist Party was the Catalan section of the official, Comintern-affiliated
Spanish Communist Party (PCE). Estimates of the precise membership levels of the four groups
vary, with PCE sources tending to give a slighly higher figure for the communists over the
PSOE’s Catalan section. But all figures indicate the minuscule size of the Proletarian Party
(PCP) and the USC’s numerical dominance. The likely approximate figures for July 1936 are:
PCP (c. 80); Catalan communists (max. 400); Catalan PSOE (c. 600); USC (c. 2000). Bolloten,
SCW, p. 397 has a selection of figures and sources. The unified PSUC still only had somewhere
between three and five thousand members at its creation (and almost certainly nearer three),
as compared to a POUM membership of between six and eight thousand. For PSUC, see Joan
Comorera (USC /PSUC leader), ‘Catalonia, an Example for Unity’, Communist International, April
x
93 ^> P- 376- On the PCP, see Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la
guerra civil: un ensayo de interpretation’, p. 309.
222 The Spanish Republic at war

in September 1935 had amalgamated to form the POUM - a process

that made it the largest socialist party in Catalonia. 20 Comorera saw
the merger as a major opportunity for the USC to overtake the POUM
(or, more exactly, its major component, the BOC) as well as offering a
chance to compete with the Esquerra.
But it was the exceptional agenda precipitated by the July days and the
common fear of CNT armed dominance which served both to clarify
Comorera’s own mind about the PSUC merger and to overcome Catalan
PSOE resistance. The prospect of leadership of the new party doubtless
also influenced Comorera. While the Comintern’s delegate to Catalonia,
Erno Gero had exercised his good offices in the discussions
(Pedro),
en route it is not credible to suggest that Gero’s intervention was the
,

determining factor in Comorera’s course of action. Far more influential

was the overwhelming instrumental sense among the entire political class
of the liberal left in Catalonia that a common defence of social ‘normality’
was imperative in the aftermath of the July days. 21 An alliance reinforced
by the organisational expertise and resources of a Comintern espousing
Popular Frontist moderation seemed a very small price to pay to facilitate
22
this process. But only by taking on board the thrust of Comorera’s
policies since 1931can we appreciate the crucial explanatory context for
the formation of the PSUC. The emergence of the PSUC responded to
an underlying logic which went back deep into the interstices of Catalan
Republican politics of the pre-war period - although this seldom seems
tobe recognised in the dominant explanations of the party’s origins. 23
What this means, moreover, is that the tendency of most of the Anglo-
American historiography to present the PSUC as the Communist Party
of Catalonia tout court is highly problematic. Nevertheless, the fact that
the PSUC
had come into being largely around the USC’s Catalanist
agenda, but that the Comintern had admitted the PSUC simply as ‘the
United Socialist Party of Catalonia’ hints at the serious tensions which
would subsequently arise. 24

20
See chapter i above.
21
Although - as we saw in chapter i - the USC congress had ratified entry to the PSUC in
May 1936, the other component parts’ ratification conferences were still ongoing when the war
erupted. The final PSUC ‘unity congress’ scheduled for the end of August would never take
place.
22
It is odd that Burnett Bolloten’s own empirical analysis broadly supports the argument I elaborate
here, yet his conclusions - a prime example of the mythologisation of Gero’s influence - then
largely contradict his own analysis: SCW, pp. 386-404.
23 See chapter 1 above for a discussion of the politics of the Catalan left 1931-6.
24 Cf. Togliatti, Escritos sobre laguerra de Espana, This was the time the Comintern had
p. 247. first

ever recognised more than one party in a single country.

 8

Challenges to the centralising Republic 223

For Comorera, with his Catalan-focused agenda, the PSUC’s dy-

namism made it an ideal vehicle to attract sectors of the Esquerra base
unmobilised sectors of Catalonia’s middling classes.
as well as previously
Comorera’s aim was to overtake the POUM
and permanently eclipse
the Esquerra by demonstrating that the PSUC was better able to defend
middling economic interests. This task was facilitated by the fall-out from
the coup. As the PSUC issued robust press attacks on CNT ‘disorder’,

recruits flocked to the party (and concomitantly to the Catalan UGT 25 ).

Nor was there an urban predominance here, for the PSUC also recruited
well among small and medium owner-farmers, tenant farmers and share-
26
croppers, many of whom had previously held Esquerra membership.
All these sectors - which together constituted the majority rural pop-
ulation of Catalonia - had in common a sense of unease provoked by
the apparent libertarian ascendancy in the region. This also intensified
the importance of being able to ‘prove’ one’s political loyalty, which, as
elsewhere in Republican Spain, saw a generalised rush after 1 July to
acquire a party or union membership card as a means of reinforcing
one’s personal security.
Requisitioning by the CNT’s supply committees was the main focus
of peasant hostility in Catalonia, as we shall discuss later. But collectivi-
sation also provoked anxiety. This was the case even though many had
no direct experience of it. For in Catalonia, where large estates were
the exception, the CNT-FAI had tacitly accepted the strength of the ru-
ral majority of smallholders, tenants and sharecroppers and had largely
respected their property and individualist forms of farming. 27 In neigh-
bouring (eastern) Aragon, meanwhile, some three-quarters of the land
was collectivised. 28 Catalan collectivisation, like Valencia’s, 29 tended thus
to focus on industry - starting with those of Francoist supporters 30 - as
well as artisan, commercial and service-sector activity in and around

25
See more detailed discussion later in this chapter.
26
For a breakdown of PSUC social constituencies and membership figures, see Bolloten, SCW,
P-399-
27
‘Oases in the middle of small estates, [the collectives] were the exception rather than the rule’,
Broue and Temime, The Revolution and the Civil War in Spain, p. 158; Bolloten, SCW, p. 395.
28
Broue and Temime, The Revolution and the Civil War in Spain, pp. 158-9; Fraser, Blood, of Spain,
pp. 348-50; Casanova, Anarquismoy revolucion, pp. 119-29.
29
Bosch Sanchez, Ugetistasy libertarios, pp. 23-31, 383-6 - this also stresses the heterogeneity of
such ‘intervened’ forms.
30
Some sectors of the Catalan big industrial bourgeoisie did support the rebels. But compared
to what happened in the Basque Country (discussed later in this chapter) Catalonia’s very
particular and gradual process of industrial and commercial development had produced a much
stronger liberal autonomist bourgeoisie (in both cultural and political terms) which, consequently,
supported the Republic.
224 The Spanish Republic at war

urban nuclei (hence the PSUC’s appeal to groups like the shop and of-
fice workers’ union, CADCI). As the conflict over requisition indicates,
where agriculture was concerned, the CNT intervened in distribution
systems rather than in those of production. 31 This is not to say that
confrontations over rural collectivisation did not sometimes occur - the
CNT could not entirely control the actions of pro-collectivist enthusiasts
in its cadres. 32 Nor did it mean that the very existence of alternative
economic forms worry and demoralise the
in Catalonia did not itself
rural middle classes.But what close scrutiny of rural Catalonia reveals is
the extent to which the CNT’s own power base was far from secure.
The situation of hung power in Catalonia was precisely what made
the adjacent territory of Aragon so important to the CNT. Not only was
it Catalonia’s war front; it was also the agrarian hinterland of its urban

revolution. But ‘revolutionary Aragon’ was not quite the bulwark it first
appeared. The CNT’s pre-war strongholds had been in the urban cen-
tres of western Aragon -
Huesca and the anarcho-syndicalists’
Teruel,
-
second ‘capital’, Zaragoza which had all fallen to the rebels in July. 33
The UGT had also had its real strength in the western zone. Paradoxi-
cally, it was eastern Aragon, with no large centres and, consequently, a
much lower level of political mobilisation in the pre-war period, which
remained to the Republic. Wartime collectivisation in eastern Aragon
occurred largely through the initiative of Catalan (and some Valencian)
anarchist militiamen who carried the new order to its villages. 34 Although
the CNT had little by way of previous roots in the area, nor did they
encounter any organised political opposition, for the obvious reason that
Republican political institutions/ the state had collapsed. 35 The republi-
cans,who had been a significant presence in pre-war municipal politics,
had no means of opposing anarcho-syndicalist initiatives in the villages.

31
This pattern also holds true for the Valencia region: Bolloten, SCW, pp. 57—9.
32 There were clashes and Castile. One bad example occurred in
in the Valencia region, Catalonia
January 1937 in La Fatarella (Tarragona, close by the Ebro River/ Aragon border), where thirty
peasants resisting collectivisation were killed: Broue and Temime, The Revolution and the Civil War
in Spain, p. 228. For Cullera (Valencia), see Bosch Sanchez, Ugetistasy libertarios , p. 122.
33 See map, Casanova, Anarquismoy revolucion , p. 99.
34 Ibid.,
pp. 127-88, 218 and 32-9; M. Cruells, Mayo sangriento. Barcelona igjy (Barcelona: Edito-
rialJuventud, 1970), p. 12. Some CNT
leaders who managed to escape from western Aragon
also participated. See the nuanced and richly informative oral testimony-based account of the
Aragonese collectives in Fraser, Blood of Spain, pp. 348-71.
35 In spite of the version in the PCE’s official history of the war, it was not so much the CNT as the
of the military coup that disarticulated Popular Front committees and closed party premises
effects

in Aragon: Guerra y revolucion en Espaha, vol. 1 pp. 29-30. Hence also the inoperativeness of the
Republican agriculture minister’s measures of 10 August 1936 putting abandoned land under
municipal control: Casanova, Anarquismoy revolucion, p. 129.
 Challenges to the centralising Republic 225

For in Aragon, as elsewhere, the republicans lacked a mass base. None of

this necessarily meant that collectivisation was imposed on a uniformly

unwilling rural population. There were rural labourers who took little

persuasion. Moreover, collectivisation emerged, as it did elsewhere, as

a response to the desperate need to re-establish shattered communities
and economic functions.
essential
But in Aragon (east and west) there was also a very high percentage of
small farmers who already possessed land - albeit that many had quan-
36
tities too small to be economically self-sustaining. They would have
preferred ‘the revolution’ to have improved their economic situation by
consolidating their private landholdings. But, equally, they felt unable to
remain outside the collectives - even though this was in theory possible
for Aragonese small farmers - because of a deep sense of insecurity in
the political environment in which they were living. 37 After all, during
August and September 1936 the CNT was the only organising political
presence in eastern Aragon. Moreover, ithad also prohibited salaried
38
labour there. This meant not only that some more prosperous individ-
uals could not recruit labour (formally or informally), but also that other
small farmers were deprived of working as such — a means by which they
had previously supplemented (whether in cash or kind) their meagre
incomes from their own land. Given the levels of hardship endured by
many of the peasant farmers in Aragon, they were not all necessarily,
a priori, opposed to collectivisation - although they were certainly igno-
rant about it. But once the militias began to engage in arbitrary requi-
sition - as they did almost immediately in Aragon — then disaffection

among the peasantry occurred rapidly. 39 And with disaffection came the
seeds of opposition which would then give other political groups their
opportunity to recruit. 40
Eastern Aragon was, moreover, a weakness the Catalan cenetistas could
ill afford, in view of their isolation in their own region. Not only were
they up against liberal Catalonia, bolstered by Comorera’s PSUC, but
they were also increasingly estranged from the POUM. Overlaying the
36
Casanova, Anarquismoy revolution, pp. 40-50.
37
Blood of Spain, pp. 348-9, 352-3. In villages for which we have testimonies
Ibid., p. 126; Fraser,

(and probably in many more for which we do not) the very real, practical difficulties of making
the collectives viable meant that moral pressure was exerted to ensure everyone joined.
38
Broue and E. Temime, The Revolution and the Civil War in Spain, p. 159; Casanova, Anarquismoy
revolution, p. 125; Fraser, Blood of Spain,
p. 355.
39 Lorenzo, Los anarquistas espaholesy el poder, p. 120; Casanova, Anarquismoy revolution, p. in.
40
As Casanova’s research (in Anarquismoy revolution) makes quite clear, pro- and anti-collectivist
positions were part of a broader struggle for political power in Republican Aragon. For more on
this see chapter 5 below.
226 The Spanish Republic at war
• .

ideological rift, and organi-

there were strategic political disagreements
sational rivalries between the CNT and the (and its forerunner POUM
BOG) that stretched back years. The BOC/POUM were highly criti-
cal of the libertarians’ dogged anti-politicism, which, they argued, left
Barcelona’s industrial proletariat defenceless. Nor had the CNT been
able to ignore the BOC since, as we have seen, its mix of Catalanism
and radical politics had given it a solid base among skilled workers and
other urban and rural lower-middle class sectors. But the BOC also
set its sights beyond them. With the Catalan CNT’s membership cri-
41
sis after 1931 and the internal dissension over treintismo and the FAPs
violent direct action, the BOC rather over-optimistically thought that it

had spied an opportunity to break the CNT’s hegemonic relationship

with the immigrant, unskilled industrial workforce of Barcelona capital.
In 1932 the BOC had called upon the CNT dissenters to unite with
them in a new political union leadership in Catalonia, whereupon the re-
gional CNT responded by expelling the BOC-led trade unions in Lleida,
Tarragona and Girona. 42 These inter-organisational antagonisms were
doubtless also fuelled by the fact that many of the key socialist leaders
in Catalonia (and not a few beyond) had served their political appren-
ticeship in anarcho-syndicalist ranks before moving on or, as the cenetistas
understandably interpreted it, becoming renegades. 43
During the July days of 1936 the CNT and POUM militia fought in
unison. (While the CNT’s control was uncontested in Barcelona capital,
the POUM had strongholds in other urban centres of the region, most
notably in Lleida. 44 ) But there was clearly a connection between the
pre-war tensions and the fact that the CNT, by agreeing to such a high
level of representation on the Central Anti-fascist Militia Committee for
the Catalan liberal republican parties, the UGT and the PSUC (some
eight posts in all), allowed the POUM
to be relegated to more of a
minority voice than its strength and role in the July days elsewhere in
the region warranted. In acting thus, the CNT may have believed that
it was ridding itself of a troublesome competitor. But the CNT was also
41
Tavera and Vega, ‘L’afiliacio sindical a la CRT de Catalunya: entre l’euforia revolucionaria i

l’ensulsiada confederal 1919-1936’. See also chapter above on the CNT crisis.
1

42 Durgan, BOC iggo-igg 6 pp. 163-6. Barcelona capital remained the CNT’s fief: Durgan,
,

‘Trotsky, thePOUM and the Spanish Revolution’, p. 59. The union expulsions led to the forma-
POUM
tion of a separate union federation (FOUS) which in May 1936 merged - disastrously -
with the Catalan UGT (see chapter 1 above).
43 For more on the phenomenon of the transjuga (political renegade), see the discussion on the May
days in chapter 5 below.
44 This was basically the BOC’s old base. Lleida was dubbed ‘Mauringrad’ after the BOC leader,
Joaquin Maurin.
 Challenges to the centralising Republic 227

blocking out the POUM’s valuable critique of the shortcomings in its own
revolutionary structures and strategy. It is important to keep in mind this
troubled relationship between the CNT and POUM
because it was to
have material effects on the development of the Catalan political scene
between July and December 1936.
Within ten days of the creation of the Central Anti-fascist Militia Com-
mittee, the CNT-FAI had also consented to Companys’ formal reconsti-
tution of an all-liberal republican Generalitat (on 31 July). 45 Companys’
attempt to strengthen his hand by including three PSUC members in

the economy, supply and agriculture portfolios had to be abandoned

because of CNT-FAI opposition. But the more important point here is
that the anarcho-syndicalists did not object to the formation of the cab-
inet per se or to the fact that, thereafter, the
,
Central Militia Committee
6
was formally responsible to it as a subcommittee of the Generalitat d It was
as if the old Catalan anarchist mentality of the ‘division of labour’
between bourgeois ‘politics’ and the syndicalist ambit of the workers
was still operating to blind the CNT to the dangers of permitting the
power - es-
further consolidation of its antagonists in a situation of dual
pecially when the sudden, unexpected bonus of freedom from Madrid’s
control made the reconquest of political power an even more seductive
prospect for the Catalan bourgeoisie. At best the CNT’s armed strength
in the July days had given it a temporary advantage in Catalonia at a mo-
ment when power - in the fullest sense of the term - hung in the balance.
But dual power is a situation that always begs resolution. By allowing its
opponents back into the political game, which the CNT did when it had
initially agreed to Esquerra, Rabassaire, PSUC and UGT representa-

tion on the Anti-fascist Militia Committee, the anarcho-syndicalists were

undermining their own position. 47
At the end of September 1936 this conflict entered a new phase when
the CNT agreed to the dissolution of the Central Anti-fascist Militia
Committee and joined the Generalitat. Two months of war — even if
45 Headed by Companys’ party colleague Joan Casanovas, the cabinet contained six Esquerra
members (including one from the Rabassaires), one representative of Accio Catalana and one
‘independent’ - an army officer, Jose Sandino - in the defence portfolio. (Sandino would continue
in this function in the Generalitat cabinet formed on 28 September with CNT membership.)
For full details of the 31 July cabinet, see Guerray Revolution en Espana, vol. 2, pp. 18-19.
46
Freund, ‘Dual Power in the Spanish Revolution’, p. 324.
47 Nor can these concessions be attributed to some disembodied/ abstract sense of political magna-
nimity on the part of the libertarian movement (cf. Garcia Oliver’s analysis in Abad de Santillan’s
Por queperdimos
la guerra (Buenos Aires: Editorial Iman, 1940), p. 255). This is not to suggest that

such sentiments were absent from his mind or from those of the other CNT-FAI leaders in
Barcelona on 21 July 1936, but rather that the underlying reasons for the outcome lie elsewhere,
as will be discussed below in chapter 5.
228 The Spanish Republic at war

that war was not on Catalonia’s threshold - had taken their toll on the
CNT’s leaders 48 The militia defeats in the south and on the approach
.

to Madrid as well as the bleak international horizon weighed heavily

on libertarian hopes everywhere 49 As we have seen too, prime minister.

Largo Caballero was determined to bring the CNT into government. As

a result, in late September the CNT in plenary session sanctioned the
principle of an anarcho-syndicalist governmental presence by reference
to the overwhelming necessity of the But the simultaneous situation 50 .

attempt to justify the decision on the grounds that the emergency situ-
ation had transformed the very nature of government and the state is
indicative of how the intense crisis of
Republican defence had exposed
the CNT’s underlying lack of political acumen:

[t] he government has ceased to be a force of oppression against the working

class, just as the state is no longer the entity that divides society into classes. Both
will stop oppressing the people all the more with the inclusion of the CNT .
51

The were the words of Diego Abad de Santillan, one

fact that these
of the leaders of the Catalan FAI, also nullifies the argument that the

controversy over entry to government neatly split purists from prag-

matists (or treintistas) along the lines of the pre-war conflict 52 In fact, .

48 The organisation publicly admitted that they had been moved by ‘the difficult situation on certain
[military] fronts’: quoted byjellinek, The Civil War in Spain, p. 497.
49 Cf. ‘Those who talk of implanting a perfect economic and social system are friends who forget that
the capitalist system has international ramifications and that our triumph in the war depends
. . .

gready on the warmth, sympathy and support that reaches us from outside’, Juan Peiro, in talk
on Radio CNT-FAI, 23 Oct. 1936, quoted in Broue and Temime, The Revolution and the Civil War
in Spain, p. 207 (Peiro, a moderate syndicalist and close colleague of Angel Pestana, had become

industry minister in the central government formed by Largo Caballero on 4 November 1936).
Peiro criticised radical anarchists’ practice in the war for having alienated the small peasantry:
Abella, La vida cotidiana durante la guerra civil, pp. 89-90. For the pre-history of this intra-CNT
dispute, see chapter 1 above.
50 ‘The CNT considers as essential its participation in a national body (organismo nacional)
equipped to take on the task of war leadership and political and economic consolidation.’ Extract
from plenary resolution in Peirats, La CNT en la revolucion espahola, vol. 1, p. 200. Cf. ‘The respon-
sibility both before History and their own consciences of those who, being in a position to

facilitate the creation of an instrument of national defence (organo nacional de Defensa) fail

to do so, is enormous.’ Extract from post-plenary statement, Peirats, p. 202. The fact that the
word ‘government’ is studiously avoided indicates the CNT’s awareness of its acute political
dilemma. See also V Richards, Lessons of the Spanish Revolution, 2nd edn (London: Freedom Press,
1972), p. 68. A slighdy different, but not incompatible assessment of the September plenum is in
Lorenzo, Los anarquistas espaholesy el poder, pp. 184-5.
51
The commentary appeared on 4 November 1936 in the CNT press (
Solidaridad Obrera) and is

cited in Peirats, La CNT en la revolucion espahola, vol. 1, p. 220. Translated extracts appear in Broue
and Temime, The Revolution and the Civil War in Spain, pp. 207-8.
52 This remains true, in spite of the fact that Abad de Santillan was the first to criticise this ‘short-
sightedness’ after the war in Por que perdimos la guerra, p. 116 (a translated extract is in Broue and
Temime, The Revolution and the Civil War in Spain, p. 208).
 Challenges to the centralising Republic 229

all sections of the movement were understandably overwhelmed by the

Republic’s increasingly desperate military (and international) situation. 53
Although a Catalan regional plenum at the beginning of September had
agreedCNT entry to the Generalitat, the Catalans would continue to
oppose CNT entry to the Largo Caballero government until mid
October 1936. 54 The logic here derived from the belief that whereas the
Catalan anarchists were collaborating with the Generalitat from a posi-
tion of strength, the CNT’s inclusion in the central government would
lead only to endless compromises and the defeat of its revolutionary po-
litical goals. But it was precisely to this that the CNT’s presence in the
Generalitat too would lead.
In the composition of the 28 September cabinet we can see embedded
the strategy of the CNT’s opponents. The anarcho-syndicalists’ eco-
nomic control was now formally ‘contained’ by the Esquerra. (Josep
Taradellas, appointed premier by Companys, also held the finance port-
folio, while agriculture was in the hands of the Rabassaires. 55 ) The

dissolution of the Central Anti-fascist Militia Committee meant that

the CNT committees had effectively been decapitated. Moroever, in the
press and public fora, liberal commentators /politicians from the
Esquerra and the PSUC (as well as other, smaller parties) had started to
attribute all economic dislocation and inefficiency to the forms of CNT
control per se, even though many of the problems they identified were
intractable ones, deriving from the macro-economic and territorial dis-
locations of the war — industrial regions cut off from their suppliers of raw
materials, productive regions from their markets and so on. It was the
overarching crisis of the war which allowed the liberal political agenda
underpinning this campaign to be submerged beneath high-minded de-
nunciations of committee and collectivist inefficiency and abuses (which
certainly existed) and impassioned exhortations to a - very necessary -
unification of economic production. But what had opened the door to
this liberal counter-attack was precisely the CNT’s inability to articulate

the committees politically. The fact that the only centralised forms of or-
ganisation available belonged to their political enemies was now forcing
53 Cf. the comments of the Catalan FAI’s premier leader, Juan Garcia Oliver, also quoted in Broue
and Temime, The Revolution and the Civil War in Spain p. 208. Ironically (if unsurprisingly), the
,

most voluble critics of the CNT’s change of stance came from those national sections in the
AIT (Anarchist International) with minuscule organisations: Casanova, Anarquismoy revolution,

p. 147 . Ideological differences aside, therean interesting parallel, I think, in the criticisms of
is

the POUM’s governmental collaboration made by the exiguous Bolshevik Leninists of Trotsky’s
Fourth International.
54 Lorenzo, Los anarquisias espaholesy el poder, pp. 188—9.
55 Full cabinet list in Bolloten, SCW, pp. 402-3. The PSUC had two portfolios - public services
and labour/ public works.
 ,

230 The Spanish Republic at war

• •
%
• • • ,

anarcho-syndicalists into one of two positions, neither of which was fea-

5
sible: a political ‘compromise with liberal order or an all-out defence of
the collectives. But the latter implied a defence of economic decentral-
isation that was difficult to sustain in the conditions in which the civil
war was being fought. Moreover, it laid pro-decentralisation currents of
anarcho-syndicalism wide open to the attacks of their political enemies.
Agrarian collectivisation was slowly but surely being circumscribed.
The famous Republican decree of 7 October 1936 only legalised collec-
tives when they occupied land belonging to those who had supported the
military rising - this, according to Republican law, was the only legally
expropriable land. 56 All Republican agricultural decrees, including that
of October, functioned to appease and protect (mainly small and mid-
dling) landholding sectors that had not actively demonstrated themselves
to be pro-rebel. Tenant farmers of six years’ standing acquired the right
to buy their land outright or by means of redemption payments. 57 By
February 1937 the Generalitat would also make all Catalan tenant farm-
ers owner-occupiers. In that these transactions reinforced the concept
of private property, they fell full square within the logic of the inter-
class Popular Front alliance and, as such, found their most energetic
upholders in the PSUC
and the PCE. The latter was also responsible
for the Republican agriculture ministry and its agrarian reform institute,
the IRA. While the IRA would provide finance and technical support
for legally constituted collectives in the whole of the Republican zone
throughout the war, the price was state supervision, and many CNT
collectives held out against registration. 58 Again, the fact that politi-
cal control belonged not to the anarcho-syndicalists but to their oppo-
nents would prove fatal. Without access to state funds (banks, gold etc.),

credit or external trade, the viability of the collectives reduced over time.
Lack of financial control meant that there was no means of capitalis-
ing them - so new machinery, fertiliser and specialist agronomist advice
were hard to come by. Lacking access to resources and in the face of the
inevitable increasing dislocation of wartime, even the ideologically com-
mitted minority would become weary and disillusioned by what all too
often ended in such circumstances as the collectivisation of shortage and

56 Republican government thinking on all other collectivisation (in industry, the public services and
transport) was codified in its 24 October 1936 decree.
57 Central (Giral) government decree of August 1936 in Broue and Temime, The Revolution and the
Civil War in Spain pp. 164-5; Fraser, Blood of Spain, p. 372. For the Generalitat ’s January decrees

indemnifying individuals (other than pro-rebel sectors) for property lost to collectivisation since
18 July 1936, see Payne, The Spanish Revolution, pp. 290-1.
58 Fraser, Blood
of Spain, p. 373 (n. 1).
 Challenges to the centralising Republic 23

poverty. 59 The dissolution of the Central Anti-fascist Militia Committee

also loosened the ties between Barcelona and Aragon. The Anti-fascist

Militia Committee had had titular responsibility for overseeing the re-
60
gion’s political and military affairs. But a plenum called on 6 October
by Aragonese anarcho-syndicalists to discuss the CNT’s recent entry to
the Generalitat decided - in the face of opposition from the Catalan rep-
resentatives - to establish their own (all-anarchist) regional council - the
61
Council of Aragon. This was a response to Aragon’s very real isolation
from the Republican government in the aftermath of the coup. There
was too a clear element of regionalist sentiment reacting against the per-
ceived ‘colonisation’ by ‘Catalonia’ and, most concretely, by Catalan and
62
Valencian anarchist columns. The Council’s primary purpose was to
coordinate Aragon’s highly fragmented collectivised economy. And in
spite of its provenance, the Council would play a crucial interventionist
(that is, quasi-governmental) role - fixing prices and
and organ- salaries
ising the exchange of commodities within the region. But by far the most
urgent task facing the Council at the time of its birth was to combat the
plague of irregular militia requisition — amounting often to sheer armed
despoliation - which was fast alienating the Aragonese peasantry even
as it undercut the fragile viability of the new collectives. (In spite of the
Council’s efforts, however, such ‘requisition’ would continue even after
the militarisation of the Aragon front in spring-summer 1937. 6s )
Inevitably, there were also practical and political problems within the
64
collectives. In a region where average educational standards were low,
there was, from the start, a lack of personnel with the requisite adminis-
trative or technical expertise. This created an opening for PCE criticisms
that the CNT was accepting those of dubious political loyalties provided
59 Cf. ‘As an economic system it was producing a conglomeration of self-contained barter markets
grinding slowly towards stagnation as the general economy ran down’, R. Carr, Spain 1808-igy5
(Oxford: Oxford University Press, 1982), p. 660. It may be also that the departure of more engage
sectors to the activated front from August 1937 also contributed to this stagnation in Aragon as
those (older men) left behind had neither the will nor the enthusiasm to confront the increasing
material problems: Fraser, Blood of Spain, p. 371 (n. 1).
60
Although in practice there was virtually total military decentralisation on the Aragon front:
Casanova, Anarquismoy revolution, p. 109.
61
Pleno de Bujaraloz: Casanova, Anarquismoy revolution, pp. 133-40. Nominal military responsibility
for the Aragon front remained with the Generalitat (p. 134).
62
Fraser, Blood of Spain, p. 350.
63
Casanova, Anarquismoy revolucion, p. in. (In practice, each military unit continued to have its own
autonomous supply section.)
64
About which CNT internal documentation - as opposed to its press and political propaganda -
was quite frank: Casanova, Anarquismoy revolucion, pp. 3-4; see also Fraser, Blood ofSpain, pp. 354-
71 and Broue and Temime, The Revolution and the Civil War in Spain, pp. 164-5 f° r an astute and
perceptive summary of the complex issues surrounding collectivisation.
232 The Spanish Republic at war
»

65
they had the requisite At the same time, for socially conservative
skills.

sectors of theAragonese peasantry, revolutionary goals of social equality

were harder to deal with than the practical rigours of wartime requisition.
But equality - even for those who sought it - proved difficult to achieve in
practice, and under the dislocations of war the adequate distribution and
exchange of products was a fraught process. All of these things opened
up space within the Council for socialists, communists and republicans
to compete for the political support of the rural constituencies of east-
ern Aragon. Although there were ideological differences between these
organisations, much of the mutual recrimination that everyone else was
harbouring political undesirables merely indicates the problematic het-
erogeneity of the social base that the Republic was required to integrate.
Before the Largo Caballero government would recognise or commu-
nicate with the Aragon Council, it insisted that it be reconstituted to
reflect the political configuration of the Popular Front alliance. (Neither
the central Republican government nor the Generalitat had been re-
motely happy with the Council as an entirely anarchist body. But while
the Generalitat saw its very existence as an infringement of its own
powers, the anarchist ministers in the central government managed to
persuade Largo Caballero to take the path of conciliation rather than
confrontation with the Council.) Agreement was reached by the end of
1936 and henceforward the Council was composed of an equal number
of CNTand Popular Front representatives, though presided over by the
cenetista Joaquin Ascaso, cousin of the famous anarchist leader Francisco
Ascaso, who had been killed in action during the July days. 66 But the
erosion of the CNT’s ‘hegemony’ on the Aragon Council occurred so ra-
it had never had more than an
pidily thereafter as to suggest that, in fact,
appearance of control. The collapse of the state had not entirely dislodged
the principle of its legitimacy - as we have seen in the Catalan case. And
this was something upon which the CNT’s political opponents in Aragon
could build from the beginning of 1937 Moreover, in addition to the on-
.

going internal political struggle for influence and members, there was also
the overriding issue of war needs that pointed to centralised political and
economic control. The Council’s raison d’etre was as an instrument of po-
litical and (above all) economic articulation. But this still only responded

to regional needs. The Council controlled resources that it converted

into foreign exchange, spending it in ways that, while not improper, did
not necessarily respond to larger war priorities. For example, its purchase

65 Fraser,
Blood of Spain, pp. 347, 362, 363-6.
66
For a full list of members and posts, see Casanova, Anarquismoy revolucion, p. 14 1 (and on Joaquin
Ascaso, pp. 134-5).
 Challenges to the centralising Republic 233

of tractors and other agricultural machinery may have increased pro-

duction in those collectives that received them, and certainly there was
67
an improved quality of life for the But it was a decision
collectivists.

taken without reference to overall Republican need/ deployment of re-

sources. In the end, then, the Council was a localism ‘writ large’. Above
all, there was the as yet unresolved problem of properly articulating and

supplying Aragon’s military front. Once it became necessary to activate

the front, then the Council’s days would be numbered - as would be the
68
Generalitat’s titular military responsibility for it.

Meanwhile, the CNT’s political control went on unravelling inside

Barcelona capital. On 9 October 1936 the Generalitat decreed the re-
constitution of all local committees as municipal bodies with the same
composition as itself in a bid to liquidate anarcho-syndicalist power on
the ground. 69 The Esquerra was amazed to find the CNT’s ministerial
representatives acquiescing. But by this stage it was too late for them to
oppose what were, in effect, the consequences of power sharing in July
From late September to December, Comorera and the PSUC spear-
headed an increasingly bitter propaganda campaign against the em-
battled CNT supply committees. Keen to re-establish the free market
in staple goods sought by their supporters (smallholders, traders and
shopkeepers), the PSUC publicly blamed the supply committees (and,
implicitly, all forms of collectivisation) for the increasingly acute food

shortages facing the population.

The Catalan liberals’ primary objective, however, remained full con-
trol of public order in Catalonia, since a monopoly of coercive force was
the sine qua non of all other change. From the Generalitat’s Home Office,
Artemi Aiguader, the Esquerra’s risk-taking political operator, set his
sights on the CNT’s de facto control of the defence/ security portfolio. 70
At the same time, a number of decrees were issued with the aim of dis-
arming the worker committees. On 27 October the central Republican
government decreed that all ‘long arms’ (i.e. rifles and machine guns)
held by ‘private citizens’ were to be handed over to the ‘municipal
authorities’. 71 But this would inevitably be a slow process — especially
in Catalonia. It would take months for the municipalities to start

67
Fraser, Blood of Spain p. 356. 68
,
See chapter 5 below.
69
An indication of the difficulty of implementation is in M. Vilanova, ‘L’Escala y Beuda'. Historia

_
FmM Oral, 3 (1990), 53.
70
Bolloten, SCW, p. 858, n. 42. On Aiguader, see Cruells, Mayo sangriento, p. 49.
71
Morrow, Revolution and Counter-Revolution in Spain, p. 58. For the continuing ‘unofficial’ use of
petrol as part of this conflict (illegal detentions occurred by car), see E. Ucelav da Cal, ‘Cataluna
durante la guerra’ in (various authors) La guerra de Espana (Madrid: Taurus, 1996: 1 st edn 1986),

P- 327 -
234 The Spanish Republic at war

functioning, and neither these, nor the end of the Anti-fascist Militia
Committee, nor CNT representation in the new Generalitat had, of
themselves, liquidated dual power. A network of defence committees as
well as factory and neighbourhood committees still functioned through-
out Barcelona, even if they had been decapitated. These committees

generally chose to interpret the arms decree as still permitting the re-
tention of such weapons for their own collective use. In the long run,
however, the October decree would set the precedent for a withdrawal
of worker patrol and committee arms from the home front. Liberal gov-
ernments in Madrid and Catalonia were seeking to return a monopoly
of firepower to the police as the agents of the state - although this process
was explained and justified purely in terms of garnering all resources for
the military front.
To this end, the police were being consolidated. After having been
formally dissolved in the aftermath of the coup, both the Assault Guards
and National Republican (i.e. ex-Civil) Guards were by the end of 1936
subject to central government reforms - the same reforms which saw the
building up of the Carabineros. 72 By the end of 1936 they were involved
in skirmishes — which inside three months would escalate into bloody
confrontations — with the CNT’s control committees on the Franco-
Spanish/ Catalan border (at places such as Figueras and Puigcerda) as
treasury minister Negrin sought to exert economic control over crucial
foreign exchange for the central Republican government. 73 Although
this encroachment by ‘Madrid’ was far from welcome to the Catalan

nationalists in Esquerra and PSUC, such was their desire to see social
and economic normalisation that it ensured their interim support for
Negrin’s initiative. However, it was to be the looming political crisis over
the POUM’s ministerial presence in the Generalitat in November 1936
that provided the first golden opportunity significantly to consolidate

governmental authority in Catalonia in terms of both public order

liberal
and economic policy.
This opportunity was provided by the eruption inside Republican
Spain of another dispute - the increasingly bitter one raging in the
72 C. Semprun-Maura, Revoluciony contrarrevolucion en Cataluna (1936-1937) (Barcelona: Tusquets,
1:978), p. 237. On the Republic’s - essentially failed - pre-war reform of the police /public
order (they were not demilitarised), see Ballbe, Orden publico y militarismo en la Espana constitucional,

pp. 317-96 and chapter 1 above. For the Carabineros, see chapter 3 above.
(partic. pp. 391, 393)
73 Factory delegates in socialised industries would go abroad to arrange imports and exports direct:
Morrow, Revolution and Counter-Revolution in Spain, p. 20. Another major focus of the batde between
government and unions over control of foreign exchange was waged in the Valencian region over
the collection and export of oranges/citrus fruit by a UGT-CNT consortium, CLUEA: Bosch
Sanchez, Ugetistasy libertarios, pp. 380-3.
 )

Challenges to the centralising Republic 235

communist movement. In late August, the

international publicly POUM
denounced the execution by Stalin of Zinoviev, Kamenev and other old-
guard Bolsheviks. 74 It did so in spite of a strong current of opinion in
the party that urged caution so as to avoid any further deterioration of
their own relations with the Comintern-aligned PCE and PSUC.
75 By

November, the POUM’s press criticism had escalated, in significant part

goaded by the party’s exclusion from the Madrid Defence Council. 76
The Catalan POUM
now openly accused the Comintern of pursuing
the containment of the Spanish Revolution because it was out of step
with the Soviet government’s defence needs - especially as the revolution
offered no sectarian advantage since the Comintern did not control it
politically. 77 These public criticisms are enough to explain the hostility to

the POUM exhibited by Moscow and the Comintern. Moreover, while

the party was not Trotskyist, the fact that Trotsky’s former (albeit now
Andreu Nin, and his small Communist
politically estranged) secretary,
Left Party alsoformed a minor component of the POUM clinched the
Comintern’s determination to remove it from the political scene in Spain.
This might have been rather more difficult to achieve, however, had
the POUM’s political position not already been weakened by its am-
biguities. 78 The party publicly espoused a radical anti-capitalist ideol-
ogy for furthering both the July revolution and the war effort. Yet since
the February 1936 elections, the POUM had supported the liberal-left
Popular Front alliance. Moreover, its own party base in Catalonia also
depended significantly on sectors of the urban and rural lower-middle
classes who, while they were Catalanist and politically to the left of the
Esquerra, were far from revolutionary or socialist in their outlook. 79 It

74 V Alba, El marxismo en Espana igig-iggg. Historia del B.O.C. y del P.O.U.M. (2 vols., Mexico:
Costa-Amic, 1973), vol. 1, 316 quoting La Batalla (27 August 1936).
p.
75 V Alba, Histona de la segunda republica espafiola (Mexico: Libro Mex, i960), p. 255 and El marxismo ,

vol. i,p. 317.

76
See chapter 3 above.
77 La Batalla , 14 and 18 November 1936 (respectively for two criticisms) - cited in Bolloten, SCW,
pp. 408 (
n. 25), 410.
78
The Madrid POUM was to the left of the Catalan party - being mainly constituted from
Andreu Nin’s tiny Communist Left component plus a few Bolshevik-Leninists: Durgan, B.O.C.
I93 °~I936 PP- 556, 559; Morrow, Revolution and Counter-Revolution p. 85. The POUM’s Valencia
, ,

branch constituted the ‘right’ of the party in that it entirely supported the Republican government
alliance: Morrow, Revolution and Counter-Revolution p. 66; Bolloten, SCW, p. 860 (n. 12).
,

79 The POUM justified its support of the Popular Front for tactical reasons, but it caused a definitive
break between Trotsky and Nin. For the tensions inside the POUM and between the POUM
and (Spanish and non-Spanish) Trotskyists, see Durgan, POUM and the Spanish
‘Trotsky, the
Revolution’. For the importance of the POUM’s Catalanist discourse, see the way in which the
party press (La Batalla frames its criticism of measures to bring both the Generalitat and Catalan
militia under the authority of the central government’s war ministry in late March 1937, article
236 The Spanish Republic at war

was the concerns of these middling sectors that lay behind the POUM’s
post-July days professions that it would ‘uphold [the middle classes’] eco-
nomic claims within the framework of the revolution’. At the same
. . .

time the POUM sought to distinguish itself from the PSUC, which it
denounced as guilty of capitulating to its middle-class constituency be-
coming its mere instrument. 80 But it is difficult to see how the POUM
could have squared this particular circle. The party’s dilemma after 18
July 1936 arose from the conflict between the radical ideological project it
enunciated in its wartime publications and the nature of its social base. 81
This conflict was probably much more acute because of the absence
of the POUM’s undisputed leader, Joaquin Maurin, who was in prison
82
in the rebel zone. The former leader of the BOC, Maurin was an
experienced politician who understood his Catalan base. Those who re-
placed him were activists and theorists, but none had Maurin’s strategic
talent. The POUM’s new helmsman, Andreu Nin, was not able to invest
the party’s radicalised discourse with a matching practice. Indeed, some
have gone as far as to argue that Nin’s dogmatism wasted six years of
the BOC’s rich populist political practice and thus facilitated the PSUC’s
stealing of POUM’s political discourse, strategies and base. 83 Nor did the
POUM have real political allies. Indeed, the POUM’s very ambiguities
meant that groups which were otherwise politically antagonistic to each
other (i.e. the CNT, the PSUC and the Esquerra) could come together
temporarily over their common hostility to the POUM.
For all his perceived radicalism, it was Nin who took the POUM into
84 (Like CNT,
government in spite of the reservations of the party left. the

SCW, p. 420. (The editorial was written by the

cited in Bolloten, POUM (and ex-BOC) leader,
Enric Adroher (Gironella).)
80
‘It is one thing to attract the middle classes to the revolution and another to form a coalition giving

them a decisive role as a governing force but we uphold their economic claims
. . . within the . . .

framework of the revolution’, La Batalla, 23 Feb. 1937.

81
There is little to suggest the POUM’s crucial political ambiguities in George Orwell’s diary
account, Homage to Catalonia (first pub. 1938) - an omission which problematises his explanation
of later political developments upon which so many accounts still ultimately rely.
82
J.Maurin, Como se salvo Joaquin Maurin. Recuerdosy testimonios (Madrid: Jucar, 1980); Bolloten, SCW,
P- 4 °5 -

83 Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 312; see also
J. Miravitlles, Episodis de la guerra civil espanyola (Barcelona: Editorial Portic, 1972), p. 178. Interest-
ingly,another Catalan historian, Borja de Riquer, has recently commented upon the resemblance
between the POUM
and the PSUC during the war in the matter of nationalism. Both had party
bases that were substantially Catalanist, yet had leaders (or at least some leaders) who did not
want to recognise it. Borja de Riquer during the Escorial summer school, Madrid, July 2001
(‘Entre la etica y el extremismo: los personajes de la guerra civil’).
84 Fraser, Blood SCW, p. 861 For a bolshevik-leninist critique, see
of Spain, p. 341; Bolloten, (n. 15).

Morrow, Revolution and Counter-Revolution, pp. 54-7.

 Challenges to the centralising Republic 237

he too was up against the enormous political responsibility of the war.)

Once there, Nin took a conciliatory line over the question of unions
versus soviets as the appropriate instrument of revolution. But the CNT
still remained hostile. Nin’s policy in the Generalitat justice portfolio was
also a source of friction with CNT radicals. 85 Moreover, the POUM’s
numerous criticisms of CNT excess in the implementation of wartime
86
collectivisation As
introduced yet further tensions to the relationship.
we have already seen, the CNT viewed the POUM as a potential rival. At
the same time, both the Esquerra and (more realistically) the PSUC were
interested in attracting sectors of the POUM’s urban base. The old USC
Catalanist agenda, as much as the new Comintern one, was now driving
the PSUC to compete with the POUM for the political loyalty of lower-
middle-class sectors in the region. The hybridity of the Popular Frontist
model no doubt also sharpened Comorera’s expectations of a parallel
success among industrial workers in Catalonia. The PSUC could now
seek to recruit simultaneously from both Esquerra and POUM bases just
as, elsewhere, the PCE was acquiring a cross-class membership through

the adoption of the same model.

After the POUM’s public criticism of the Soviet Union, the PSUC
demanded its exclusion from the Generalitat on 24 November. great A
deal of emphasis has been laid here on the intervention of the Soviet
Consul-General in Barcelona, Vladimir Antonov-Ovseenko, in persuad-
ing Companys and the Esquerra to accept this. 87 That such intervention
occurred is But given that this intervention and influence were sug-
clear.

gesting a political direction which all Catalan liberals were in any case
keen to take, it is something of a nonsense to argue that the outcome was
purely the product of Soviet ‘duress’. What Antonov-Ovseenko offered
Companys was a largely superfluous reminder of the liberal Republic’s
precarious international position. Nor did Companys need to be re-
minded that the POUM’s presence in the Generalitat was an additional
obstacle to liberal normalisation.
The PSUC’s demands for POUM exclusion opened the way for a
vehement Esquerra denunciation of ‘committee chaos’. Companys made

85
V Alba, ‘De los Tribunales Populares al Tribunal Especial’, in Justicia en guerra, pp. 226—7,
229-30. For the underlying ideological hostility of the to the CNT POUM - which was rooted
in the memory of Trotsky’s role in the repression of anarchists during the Bolshevik Revolution -
comments on the POUM’s impracticable plan to bring the exiled Trotsky
see Bolloten’s to
Barcelona, SCW, p. 859 (n. 3).
86
The POUM press between July and December 1936 attests to this: see examples /summary in
Bolloten, SCW, p. 863 (n. 48).
87
For example, ibid., p. 411; Cruells, Mayo sangriento, p. 29.
238 The Spanish Republic at war

no specific reference to the POUM, but demanded ‘a strong government

with plenary powers, capable of imposing its authority on everyone’. 88
The CNT instantly rejected this, well understanding that it rather than
the POUM was Companys’ real target. Indeed, the PSUC made this
absolutely explicit. After echoing Companys’ declarations on the need to
concentrate all power in the government, Comorera went on to demand
that dual power in Catalan defence and public order now be ended
defacto as well as dejure by the dissolution of both the CNT’s defence and
security committees.
Faced by this Esquerra-PSUC line-up, the libertarians cannot have
failed to perceive the political dangers for themselves. But this did little to
reduce their ambivalence towards the POUM. Moreover, in terms of the
hard currency of political power in Catalonia in 1936, both the POUM’s
increasing marginalisation (as the PSUC expanded) and its internal
political tensions meant that the CNT did not interpret the POUM’s
presence or absence as making any strategic difference to its own posi-

tion. So, when the PSUC offered a deal, the CNT accepted: it would drop
its opposition to POUM exclusion from the cabinet and the PSUC would
drop its call for the dissolution of the defence and security committees.
On 1 6 December, after four days of cabinet crisis, the new line-up was
announced. Out of a total of eleven portfolios, the Esquerra had four
(including the Rabassaire-held post of agriculture), the PSUC three and
the CNT four. There was still a stand-off in public order: the Esquerra’s
Aiguader retained the Home Office portfolio, but the CNT committees
continued to function. A crucial shift had nevertheless occurred in the
economic posts. Why the CNT allowed this constitutes far more of a
conundrum than does its position on the POUM. 89 Not only was the
CNT’s economy brief overseen, as before, by Taradellas’ control of
finance, but now Comorera had control of the crucial supply port-
also
folio.
90 The battle to Tree the market’ was about to begin.
This was a battle in which the Esquerra wanted the same outcome
as the PSUC - the restitution of liberal economic order. But it was less
prepared than the PSUC to take the strain in the bruising confrontation
that loomed. 91 In part this was because the Esquerra was still recovering
88
Solidaridad Obrera, 9 Dec. 1936.
89 The ostensible reason was the CNT’s prioritising of the war portfolio in the Generalitat. So in
the December crisis they relinquished supply in exchange for this: see Fraser, Blood ofSpain, p. 375
(and n. 1) for the recollections ofjuan Jose Domenech, the CNT’s outgoing councillor for supply.
90 Comorera and Domenech had in fact swapped posts from the previous (September) cabinet,
with Domenech now in public services.
91
During 1936 the Esquerra press was far less explicit in its anti-CNT stance than was the PSUC’s,
Bolloten, SCW, pp. 410, 862 (n. 39).
 Challenges to the centralising Republic 239

from the erosive impact of the July coup. But it seems highly likely that
it was also a conscious tactic of Companys to allow the PSUC to do
5

its ‘dirty work in the expectation that this would, in turn, erode the

fast-growing PSUC 92 and allow the Esquerra to reclaim its hegemonic

position in the Catalan arena. Increasing tensions in the ‘hybrid’ PSUC
between centralist and Catalanist currents would give Companys and
the Esquerra both the will and the opportunity seriously to recontest the
political arena.
The fall of Malaga in early February 1937 and the hard fighting contin-
uing around Madrid intensified Republican government demands that
all political, economic and military decision-making power be invested in
itself. The Spanish Communist Party was also spearheading demands for

the proper implementation of conscription - again to be controlled by the

central government. Companys had already experienced this pressure
first hand in December 1936 when Antonov-Ovseenko reinforced the

Republican government’s agenda by linking Soviet military aid to

Catalonia not only to the POUM’s political exclusion, 93 but also to
the Generalitat’s acceptance of a single Republican political and mil-
itary command - something far more politically unpalatable. 94 The
Esquerra’s highly developed sense of its nationalist prerogatives was se-

riously offended. In turn this would lead the party into a complicated
political balancing act: supporting the CNT at strategic moments, albeit
at the risk of alienating some of their own middle-class supporters - to the
benefit of the PSUC. The Esquerra was of course keen to see the PSUC
drive for economic normalisation, but it also needed a means of resisting
that party’s centralising tendencies. The PSUC too trod a fine line here,
however. Its dynamic defence of liberal order had brought it the political

92
Although I know of no figures for the first six months of the war alone, the PSUC grew from
little over 3,000 to 50,000 members by March 1937. This compares with c. 30,000 claimed by
c.

the POUM in December 1936. See Bolloten, SCW, pp. 399, 405 for PSUC and figures POUM
respectively.
93 There may well also have been a psychological link between the worsening climate of denuncia-
tion inside the Comintern, the simultaneous fall of Malaga and the intensification of Comintern
concerns about ‘the enemy within’ in Republican Spain (interpreted as both the and POUM
potentially traitorous army officers): Elorza and Bizcarrondo, Queridos camaradas, p. 340.
94 The evidence for this is hearsay (i.e. there are no corroborative testimonies from participants -
Companys, Taradellas, Antonov-Ovseenko - see Cruells, Mayo sangriento, p. 29) and comes,
moreover, from hostile sources, for example Rudolf Rocker (the German anarchist and AIT
representative in Republican Spain), in Extranjeros en Espana (Buenos Aires, 1938), p. 91. Nor
are the textual extracts from Bolloten’s interview with Companys’ lieutenant, Serra Pamies,
in Mexico in 1944 necessarily as unambiguous as Bolloten’s deductions imply: Bolloten, SCW,

pp. 41 1, 863 (n. 49). However, that a Soviet diplomatic representative should have made strong
representation in favour of a single command is consonant with Stalin’s preoccupation with the
viability of the Republican military defence.
240 The Spanish Republic at war

support of the Catalan* middle classes. But whereas in the rest of the
Republican zone the PCE’s appeal was to a great extent built upon its
identification with mando unico for the PSUC the pursuit of such policies -
,

necessary though they may have been to the war effort - risked alienating
their recently acquired party base. Partly these differences in perspec-
tive were about the nearness or distance of the war/ military front. But
it was the old liberal rivalry of unevenly developed Spain where
at root
economic and industrial power in Catalonia confronts political authority
in Madrid (or, in this case, Valencia).
The pressure applied by less Catalanist-minded sectors within the
PSUC (i.e. those deriving from the former Catalan sectors of the PSOE
and the PCE and maybe some new membership constituencies too)
was itself part of the longstanding battle between centralist liberals and
Catalan nationalists, even if its most efficient conduits during the civil
war were the peninsular communist parties and the Comintern. 95 But
the Republic’s worsening military situation was an objective pressure
affecting all Republican political sectors alike. Comorera’s increased
criticism of the POUM’s public anti-Sovietism in February 1937 also
has to be understood as a response to this. Nor were these conflicts

and pressures inside communist ranks materially different from those

that increasingly divided the CNT’s national leadership from sectors of
the anarcho-syndicalist movement in Catalonia. For wartime impera-
tives were also turning the CNT’s ministers and its national committee

into a conduit for central government policies which sought to keep the
Catalan anarchists compromising inside the Generalitat. 96 Thus far the
Catalan liberals could agree. But what they failed to recognise was that
while it was feasible to ‘contain’ the CNT’s regional leadership, anarcho-
syndicalist social constituencies were quite another matter. It was in the
daily lived experiences of these groups and the meanings they ascribed
to such experiences that we find the crucial motive force for the build-up
of social and political tension in urban Catalonia across the winter and
spring of 1937.
Meanwhile, the prosecution of the war under the highly unfavourable
conditions of Non-Interventionist embargo was also increasing the pres-
sure on the central government to bring the Basque Country more closely

95 The three PSUC leaders in the December 1936 Generalitat were appointed simultaneously to
the PCE’s central committee: Pedro Checa, A un gran partido unagran organization (PCE agitprop
pamphlet, Valencia, 1937), p. 23, which lists Central Committee members.
96 For this ‘assimilation’ of CNT leaders and its political and organisational consequences, see
chapter 5 below.
 Challenges to the centralising Republic 241

under its control - in accordance with mando unico. Indeed, until the
government properly controlled Basque military forces and, crucially,
its iron- and steel-producing industrial plant, then ‘the Republican war

effort’ would remain, in a crucial sense, unrealised. But against this aim

was ranged a powerful regional particularism - no less recalcitrant for

being entirely liberal, in contrast to Catalonia’s liberal and revolutionary
varieties.

The military rebellion of 18July 1936 had opened up the internal cleav-
ages in the Basque Country. Those areas least affected by modernisa-
tion - those which were agricultural, least urbanised and the strongholds
of political traditionalism - split off from the industrialised, urbanised
and more socially heterogeneous parts. Instantly removed from the
Republic’s orbit was Navarre, as the centre of Carlism, a bastion of the
Most of Alava was also in rebel hands, as
ultra anti-Republican right. 97
well as the majority of Guipuzcoa by September. The Republic retained
the Basque industrial heartland Vizcaya - and some adjacent territory,
-
also of a mainly industrial character. 98
But this internal separation did not much facilitate an easier fit be-
tween the Republican Basque Country and the rest of the Republic. The
military defence of Vizcaya was undertaken by a combination of social
groups: Basque and immigrant workers encadred in the trade unions
STV and UGT respectively, and the strongly nationalist lower-middle
classes represented by the Basque Nationalist Party (PNV). But it was
the PNV that dominated the politics of Vizcaya. It did so, moreover,
in the name of a provincial social base significantly more conservative
than the profile projected by the ‘capital’ city, Bilbao, whose secular
and workerist components chimed more easily with the rest of Popular
Frontist Spain. 99
In Bilbao had been no military rising. Both army and
itself there
state security forces remained loyal to the Republic. Nevertheless, in
the Basque Country as elsewhere, political power fragmented under the
impact of the rebellion. Defence Juntas emerged in a number of places,
including Bilbao and San Sebastian (the capital of Guipuzcoa), each
of which initially acted independently. The PNV had relatively more

97 There was, nevertheless, a significant repression in Navarre which suggests that the appeal of the
rising was far from universal: see Altaffaylla Kultur Taldea, Navarra 1936. De la esperanza al terror
(Tafalla: Altaffaylla Kultur Taldea, 1986) and E. Majuelo, Lucha de closes en Navarra (1931-1936)
(n.p.: Gobierno de Navarra, 1989).
98
For example Eibar, which was Guipuzcoa’s main industrial centre.
99 The PNV was electorally dominant in Vizcaya by dint of its provincial control. It did not have a
majority in the industrial capital of Bilbao.
242 The Spanish Republic at war

influence in*Bilbao’s. And with Republican attrition in Guipuzcoa and

the fall of San Sebastian in the
first weeks of September, the Defence

Junta of Vizcaya emerged as the main body, overseeing both military

defence and organisation of the civilian front.
Unlike elsewhere, however, political fragmentation in Vizcaya never
produced a viable challenge to liberal economic order or any revolution-
ary experimentation with social life. Private property was guaranteed by
the continued political dominance of the Basque Nationalists, backed
up by the moderate social democratic Basque component of the PSOE.
Unlike what happened elsewhere in Spain to the political representatives
of the middling classes, the PNV’s influence survived the formal break
in institutional political continuity provoked by the military rebellion.
But the PNV in Vizcaya was doubly cut off from Madrid and the Giral
government — by dint of both its territorial separation and the state cri-
sis triggered by the coup. In the absence of government directives and

structures, the immediate needs of Basque defence saw the formation of

militias on the basis of political parties and unions. But here too the PNV
was soon in control. At the beginning of August, the party, along with its
sister formation the ANV and the Basque labour union, STY formed na-

tionalist volunteer battalions (obeying military-style discipline from the

outset), in order to wage a war on two fronts: first against the rebels and
second against the radical left minority inside the Basque Country, which
the PNV’s Manuel de Irujo described as ‘extremist elements brought into
100
the house There were some (mainly anarchist) workerist manifesta-
’.

tions in San Sebastian in the immediate post-coup days, but they were
101
quickly contained By August, liberal order reigned to such an extent
.

that it appeared almost as if Basque society had experienced no break

with the pre-coup world. The most striking indicator of this was the
degree of religious normality in Vizcaya. Two churches were burned
down in San Sebastian in July. But this was an exceptional occurrence.
The PNV ensured that the churches remained open and that religious
102
observance was still a public matter Indeed, it was common for each
.

Basque nationalist battalion to have its own chaplain.

100
F.de Meer, El Partido Nacionalista Vasco ante la guerra de Espana (1936-1937) (Baranain-Pamplona:
Eunsa, 1992), pp. 128-9; f° r Irujo? see A. Lizarra, Eos vascosy la Republica espahola (Buenos Aires:
Ekin, 1944), p. 95.
101
Broue and Temime, The Revolution and Civil War in Spain, pp. 138-9. San Sebastian fell to the
rebels on 13 September.
102
‘The Basque question 193 1-7’, in Preston, Revolution and War in Spain, pp. 183-4.
J. P. Fusi,
(Leading Basque prelates were caught between their desire to defend Basque nationalist rights
and their hostility to Republican anti-clericalism. The Castilian hierarchy protected them from
rebel ire but also pushed them to line up with the rest of the Spanish Church.)
 Challenges to the centralising Republic 243

This social normality in the Basque Country was very welcome to the
Madrid government. But other tensions between it and the PNV would
soon surface in relation to the imperatives of the war effort, making
for an uneasy relationship right through to the fall of Bilbao in June

1937. In the first instance, Madrid was not even sure that the PNV
would bring Vizcaya into the Republican alliance. The PSOE - and
in particular Prieto with his extensive Basque contacts - was aware of
the fragility of the PNV’s allegiance, especially given the breakdown
of liberal order and eruption of anti-clerical violence in much of the
rest of the Republican zone. After all, the commercial middling classes
constituting the heart of Basque nationalism were largely defined by their
Catholicism. It cemented a set of social values which, while not those of
the rebels, were still clearly conservative. As such, the collectivisation of
industry in Catalonia had also delivered a considerable shock in Vizcaya.
Largo Caballero had originally designated the PNV’s leader, Jose
Antonio Aguirre, as his minister for public works on 5 September 1936.
Aguirre rejected the offer, whereupon there followed labyrinthine ne-
103
gotiations over the price of Basque participation. The military pres-
sure on the Basque defence was temporarily eased in mid September,
when Mola’s troops came up against the mountain chain separating
Vizcaya from Guipuzcoa and Alava. But Madrid still had grounds to
fear that the PNV might strike out alone and convert Junta power into
an independent Basque Republic. 104 Indeed, the PNV made a pact with
Largo Caballero only because his government was prepared to offer
the PNV’s key goal: the autonomy statute — something the rebel al-
liance would never have ceded, as Aguirre well knew. This awareness
was reinforced, moreover, by the fact that the repression in Alava and
Navarre included nationalist priests among its victims. Lor the Basque
nationalists the statute was interpreted as a device that would permit
the Basque Country after the war to ‘secede’ culturally and to some
extent politically from the Republican centre. 105 But in addition to pass-
ing the stalled Basque autonomy statute, 106 the PNV also demanded

103
De Meer, El Partido Nacionalista Vasco, pp. 137-56.
104
See Irujo’s comments, Lizarra, Los vascos y la Republica espanola, p. 81.
105
There were, however, a minority in the PNV
(the most doctrinaire followers of party founder
Sabino de Arana) who refused to take sides in the civil war because they considered it a ‘war
between Spaniards’ and thus none of their concern. If there had to be a Basque representative
in the Madrid government they insisted that it should not be a PNV representative: de Meer,
El Partido Nacionalista Vasco, p. 138.
106
Indeed, this had been Prieto’s strategy since April 1936 in order to tie the PNV to the Republic:
ibid., pp. 67-72. Prieto as minister of air / navy negotiated the last stages of the statute. For the

process which had led to the stalling of the statute, see chapter 1 above.
244 The Spanish Republic at war

that Madrid permit the formation of a Basque government. This was a

harder condition for the centre to meet in view of the desperate fragmen-
tation of political power against which it was already struggling. Nor did
PNV personnel have any obvious leaders with experience of the crucial
areas of government: defence, economic policy and international rela-
tions. But in spite of the reluctance of Madrid (and on this issue Largo
Caballero and Prieto entirely agreed) in the circumstances, the central
government had to show willing in order to maintain Basque morale.
Moreover, there were also pressing practical considerations. The fall of
Irun on 5 September — only six weeks into the conflict — meant that
the Republican Basque Country was isolated not only from the rest of
Republican territory but also from France. Whatever Madrid’s ambiva-
lence concerning regional nationalist pretensions, sense to have it made
an organisational structure that could fragmented power
articulate the
of the Juntas. Irujo joined the central cabinet as minister without port-
107
folio in late September. The ensuing hardening of rebel attitudes and
the bombing of Bilbao also contributed to the sense of the PNV having
crossed the Rubicon into the Republican alliance. The Basque Statute
was approved by the Republican Cortes on October and a Basque 1

government formed on 7 October under Aguirre. It was configured to

reflect the Republican alliance, though excluding the CNT, which re-
mained a marginal force in the Basque Country. A majority of the posts
went to the PNV, the real political power behind the new government. 108
At the same time, the Republic dispatched part of the Republican navy
to Bilbao with arms and munitions to seal its pact with the Basque
Nationalists. 109
Given the fact that the Republican government was also instrumen-
talising the statute to PNV fully to the war effort, the political
commit the
transaction was in fact a reciprocal one. But even though the PNV would
strongly associate itself with the principles of constitutionalism underpin-
ning the Republican alliance, its other members still regarded Basque

107
Gaceta de Madrid, 26 Sept. 1936.
108
For constitution of the Basque government, see de Meer, El Partido Nacionalista Vasco, p. 181.

On the evidence here (pp. 14 1—2) the author seems to exaggerate the significance of Basque
socialist and communist support for a regional government. But the presence of local loyalties
was, nevertheless, a generic political factor across the north and thus, as elsewhere, an obstacle
to the centralisation of the Republican war effort.
109 XXI,
M. Gonzalez Portilla and J. M. Garmendia, La guerra civil en el Pais Vasco (Madrid: Siglo
1988), p. 69; de Meer, El Partido Nacionalista Vasco, p. 176. Such aid would prove scanty, however:
Salas Larrazabal, Historia del ejercito popular de la republica, vol.
1 pp. 3 69-7 2,
In general, Republican
.

ships were handicapped by the lack of naval bases in Republican hands on the Cantabrian coast.
M. Chiapuso, Elgobierno vasco y los anarquistas (San Sebastian: Ed. Txertoa, 1978), p. 79.
 Challenges to the centralising Republic 245
110
loyalty as fraught and conditional. For one thing, it was clear that the
Basques envisaged a defensive war of position to protect the territory of
Euskadi. The question of contributing manpower or economic resources
111
to other Republican fronts would be much more fraught. Moreover,
it was known that part of the PNV leadership favoured negotiating sep-
arately with the rebels.
112
Their reasons were in part pragmatic - they
win against Franco’s professional
did not believe that the Republic could
army backed by and Germany.
Italy But ideological reservations about
the Republic were also present, and indeed increased when the CNT
joined the Madrid government in November 1936. The knowledge that
this pactist group inside the PNV had taken soundings with the rebels

through intermediaries in Alava, and that it was attempting to involve

113
the Vatican in a mediatory role, reinforced Republican suspicions. Nor
did it help that the brunt of the fighting in the north up to the fall of San
Sebastian was borne by socialist, anarchist and communist militias rather
than the Basque Nationalist forces." 4 San Sebastian was evacuated on
11 September by retreating anarchists who operated a scorched earth
policy before crossing into France. The fact that the Basque Nationalist
militia were not prepared to do the same increased suspicions of their
disloyalty among the workers’ militia. In addition, tensions arising over
and economic jurisdiction would lead to
questions of military, political
increasing antagonism between Aguirre at the head of the Basque ad-
ministration and the reconstructing power of the central Republican
government under Largo Caballero.

110
Cf. Aguirre’s claim that the PNV had civilised the war, inj. A. Aguirre, Veinte anos de gestion

del Gobierno Vasco (1936-1956) (Paris, 1956). Jackson, The Spanish Republic and the Civil War,
p. 387 summarises the substantive political differences between the Basque Nationalists and
rebels/ traditionalist right (Carlists). However, constitutionalism as a war banner was stronger
in Aguirre and (above all) Irujo than in other PNV leaders. See de Meer, El Partido Nacionalista
Vasco, pp. 153-4 an d P- 254 for the view that their promotion of humanitarian policies, such as
prisoner exchanges, had greatly increased the Republic’s international standing as a whole.
111
This has been a controversial point, but de Meer offers a convincing circumstantial testimony,
El Partido Nacionalista Vasco, pp. 151-2. Aguirre would offer to send Basque troops to the Catalan
front in July 1937, but that offer was part of a different and complex political context when the
defence of Vizcaya itself was no longer a possibility: Chiapuso, El gobierno vasco y los anarquistas,
pp. 225-8.
112
De Meer, El Partido Nacionalista Vasco, pp. 236-7.
113
Ibid., pp. 137-56 is illuminating on the differences of political perspective within the PNV
leadership; see also Spanish Civil War, p. 431. Nor did these attempts at procur-
Thomas,
ing mediation ever properly stop (de Meer, p. 240). The attempts are usefully summarised
inM. Tunon de Lara, Algunos problemas historiograficos de la guerra civil en Euskadi’, in
M. Tunon de Lara (ed.), Gernika: 50 anos despues (1937-1987) (San Sebastian: Universidad del
Pais Vasco, 1987), pp. 136, 144-5.
114
De Meer, El Partido Nacionalista Vasco, pp. 113, 129-30.
246 The Spanish Republic at war

Largo dispatched the efficient though abrasive Captain Francisco

Ciutat as the central government’s Commander-in-Chief to coordinate
the ‘Army of the North’. But Ciutat had to accept the reality of separate
militia forces in Vizcaya,Santander and Asturias, which he consequently
advised separately. Relations between Ciutat and Aguirre were poor. The
many professional officers, had joined the
fact that the captain, like so
PCE guaranteed Aguirre’s enmity. 115 Ciutat’s relatively lowly rank also
made it easier for Aguirre to resist. But although the PNV did not like
the idea of an integrated Northern Army, the underlying point at this
stage was that it saw no reason to cede military control in Euskadi’s
defensive war of position to a central government whose own authority
was so weak. The PNV’s uncertainty as to what the Largo government
represented politically also reinforced the Basque Nationalists’ idea of
a defensive war, as did the low priority given by Madrid to the Basque
front. Aguirre was determined that the PNV should control all areas
of military authority even when this meant removing those who had
acquired some real practical experience in the early months. 116 By late
October, all the Republican militia within the Basque Country were mil-
dependent on the Basque defence ministry - also
itarised into battalions
held by Aguirre. By the same token, the Basque concept of defensive
war also militated against its economic and military cooperation within
the rest of the Republican north. This was particularly the case as the
PNV disliked what it saw as the workerist tenor of the other Republican
117
authorities in the north.
But while the PNV’s quite agenda did not help mat-
specific political
ters, intense localist allegiances existed across the whole of the Republican

north.
118
There would be some instances of troop exchanges - for ex-
ample of Basque forces to Asturias in February 1937 and of brigades
from Asturias and Santander to the Basque front the following April. 119

115
J. Ambou, Los comunistas en la resistencia nacional republicana. La guerra en Asturias, el Pais Vasco y
Santander (Madrid: Editorial Hispamerca, 1978), p. 145. Some in the CNT militia also distrusted
Ciutat, but whether this was about more than the abiding military-civilian tensions of the
post-coup period it is difficult to gauge: Chiapuso, El gobierno vasco y los anarquistas, p. 59.
116
Chiapuso, El gobierno vasco y los anarquistas, p. 63 (and p. 192 for Aguirre’s sense of himself as
a ‘providential leader’). The Basque socialist and later government minister, Paulino Gomez,
was removed from the war post that he had held under the Junta de Vizcaya.
117
Ibid., p. 195.
118
‘Localism (el espiritu de la patria chica) prevailed to the point of aberration’, Ibid., p. 60.
In this respect, the Basque CNT agreed with the constant refrain of its northern communist
counterparts: Ambou, Los comunistas en la resistencia nacional republicana, pp. 58, 59, 75—6, 135,
I4 3, 248.
119
Tunon de Lara, ‘Algunos problemas historiograficos de la guerra civil en Euskadi’, p. 138.
 Challenges to the centralising Republic 247

But the reciprocal effect of these enduring localisms, combined with

an insufficiently strong central government, cumulatively impeded vital
economic cooperation between Asturias, Santander and the Basque
Country in the key period of Largo’s premiership (which was precisely
when the Republicans had preparatory time), just as they impeded the
120
coordination of military planning and strategy. The Communist Party
in Asturias argued for just such a coordination. But it was far less influen-

tial within the Asturian Popular Front alliance than were its counterparts
121
in Catalonia or Madrid. Moreover, its calls for mando unico were neu-
tralised by the increasing hostility towards it of sectors of the Asturian
socialist movement with whom it was competing for members. Nor, cru-

cially, was there any force in the Basque government at this time to argue

against the PNV’s concept of a purely defensive war. The Basque social-
ists - the second most influential political force in Vizcaya - accepted the

PNV’s war policy, at least in part to consolidate the PNV’s Republican

allegiance. But so too did the Basque Communist Party, until, under the
pressure of the spring 1937 rebel offensive, it was obliged to reform its
policy line in the direction of greater support for mando unico and the
122
incorporation of Basque resources to the greater war effort. But by
then it would be too late.
In early November 1936 Ciutat was replaced as commander in the
north by General Llano de la Encomienda, as preparations were put
in train for military action designed to take the pressure off Madrid. Its
specific objectives included the retaking of Vitoria (the capital of Alava)
and of Irun - in order to reconnect the Republican Basque Country
with France. When the action began in late November 1936 Llano de
la Encomienda was in command. But control of the Basque forces was

120
Chiapuso, Elgobierno vascoy los anarquistas, pp. 63,125,195, which suggests Basque political reserve
towards ‘socialist’ Asturias and indicates the free market nature of the economic exchange within
the north. For Basque factories supplying the Asturian front, see Ambou, Los comunistas en la

resistencia nacional republicana, p. 129.

121
From December 1936 Asturias was administered by an inter-provincial council, the Consejo
Interprovincial de Asturias y Leon: Ambou, Los comunistas en la resistencia nacional republicana p. 75.
,

122
Ibid., pp. 60-1; Thomas, Spanish Civil War, p. 541. Pressure came from the rest of the PCE in the
north. The Basque communist general secretary, Juan Astigarrabia, who had also represented
the party in Aguirre’s government, was obliged to perform an ‘autocriticism’ inJuly 1937 and was
expelled from the party: Elorza and Bizcarrondo, Queridos camaradas, p. 334. Text ofAstigarrabia’s
‘self-criticism’ in PCE civil war archive, microfilm XVII. He was, in a sense, taking the blame
forhaving pursued an ultra Popular Frontist line but one which ended by conflicting with overall
Republican war needs. Thereafter the PCE in the Northern Council of Asturias and Leon would
clash increasingly with the rest of the Asturian Popular Front (republicans and anarchists, but
predominantly the PSOE/UGT) over their tolerance of localism and insufficiently energetic
pursuit of centralising war measures: Fraser, Blood of Spain, p. 245.
248 The Spanish Republic at war

mediated by Aguirre. He was fixed upon the idea of a military victory

to bolster the PNV’s authority Yet once again the defensive mentality
infusing the Basque nationalist forces militated against this. Nor had
militarisation modified the homogeneous political identity of the various
Basque battalions. The notion of Basque political separateness and the
war on two fronts meant that Aguirre insisted on maintaining the political
purity of his nationalist soldiers (the gudaris). 123 For the same reasons, until
early May 1937 Aguirre managed largely to resist the appointment of
political commissars — who elsewhere were an important generator of
discipline and fighting cohesion. 124 Nor was the morale of the equally
homogeneous Basque CNT battalions improved by the PNV’s insistent
exclusion of the organisation from government and wartime decision-
making processes. 125 Political tensions between the PNV and the CNT
would have an even more deleterious effect in spring 1937 as we shall see. ,

Aguirre was clearly determined to maintain PNV control of the Basque

government. This being the case, for reasons of morale and military
efficiency, he should have accepted the need to break up the political
homogeneity of all the Basque units - even at the (inevitable) cost of
breaking up that of the gudaris. That he would not meant that Aguirre’s
political agenda undermined the overall fighting efficiency of the forces
in Euskadi. The military objectives of the Basque campaign of winter

1936 remained unachieved in what was the only real action prior to the
Francoist offensive against Vizcaya launched in late March 1937 By that .

time, however, it would be too late to transform the Basque battalions

into an offensive force, even though a central Republican government
would soon emerge in which Aguirre had greater political confidence
and which, in turn, better understood the strategic importance of the
Basque/ northern front.
But there was also an underlying clash of economic priorities be-
tween successive central Republican governments and the PNV This
clash ultimately derived from the cleavage the military rebellion had
exposed within the Basque bourgeoisie. As we have seen, it was the
less economically powerful sectors - the commercial middle classes,

123
Tunon de Lara, ‘Algunos problemas historiograficos de la guerra civil en Euskadi’, pp. 133, 137;
de Meer, El Partido Nacionalista Vasco p. 233. Left party members were also excluded from the
,

Basque motorised police.

124
Chiapuso, El gobierno vasco y los anarquistas, p. 197; Ambou, Los comunistas en la resistencia nacional
republicana, pp. 60—1; Tunon de Lara, ‘Algunos problemas historiograficos de la guerra civil en
Euskadi’, p. 139.
125
Chiapuso, El gobierno vasco y los anarquistas pp. 64, 117, 122.
,
 Challenges to the centralising Republic 249

professional /service sectors and small-scale producers - who were

prepared to bargain support for the Republic in return for the auton-
omy statute. Aguirre himself is a good example of the kind of people
the PNV represented. His father owned a chocolate factory in Getxo.
In contrast, the Basque industrial grande bourgeoisie, which had always
been politically and culturally centralist {espaholista) because of its long
economic integration with Madrid through the banking system, was en-
126 127
tirely pro-rebel. It had thus retired (as had its Catalan counterpart )

to southern France or rebel-held territory in the north - predominantly

Biarritz and, after 13 September, San Sebastian. There they planned the
economic reconstruction of industrial Vizcaya while awaiting its ‘libera-
tion’ by Franco’s forces. But, in economic terms, the PNV demonstrated

by its actions (or, rather, the lack of them) that it remained more commit-
ted to preserving the overall ‘normality’ of the Basque economy - that
is, including the property rights and profits of pro-rebel big industrial
capital - than it was to contributing Basque economic resource to the
broader Spanish Republican war effort.

For a start, the Aguirre government sought to maintain a peacetime

economy in the Republican Basque Country. Its predecessor, the Junta
de Vizcaya, on which the PNV had not been quite as influential, had
128
been relatively more assertive with regard to wartime imperatives. It

had ‘intervened’ arms factories in Eibar, Durango, Guernica and Bilbao.

But Aguirre was not prepared to militarise heavy industry. By spring 1937
the PNV’s public discourse would finally be moving in the direction of
such intervention. But it was only in June, when Bilbao was already in
grave peril, that the PNV finally agreed measures to intervene Basque
129
industries.
From October 1936 the PNV tried desperately to find alternative ex-
port markets for Basque industrial goods to replace those lost abroad or
through the region’s isolation from the rest of Spain. But the party’s ef-
forts were often coverdy sabotaged by the very industrial concerns it was
126
See introduction and chapter i above.
127
The Lliga’s Francesc Cambo and other leading figures were also resident in (or visited) San
Sebastian, planning for a Catalan industrial ‘liberation’. On this group, see Richards, A Time of
pp. 74-5, 123, 124-5. See also Borja de Riquer, El ultimo
Silence, Cambo igj6-ig4y (Barcelona:
Grijalbo, 1997).
12
Constitution of the Junta in Chiapuso, El gobierno vasco y los anarquistas, p. 43. Chiapuso (p. 69)

accuses Basque socialists and communists of paying too litde attention to economic organisation
at the start - thus the PNV was unopposed in government.
129
Gonzalez Portilla and Garmendia, La guerra civil en el Pals Vasco, pp. 89-91; Chiapuso, El gobierno
vasco y los anarquistas, p. 199.
250 The Spanish Republic at war

attempting to assist. The PNV’s inexperienced laissez-faire combined

with this industrial sabotage (effectively amounting to a fifth colum-
nist activity) has been assessed as being more responsible for Vizcaya’s
plummeting industrial production during winter 1936 and spring 1937
than was the evident shortage of raw materials stemming from its iso-
lated and blockaded position. 130 If the PNV had been entertaining hopes
of winning over some of the big industrial interests by demonstrating
their probity as caretakers of capital and property, then they were to be
disappointed. 131
The Catalan and central governments had to buy Basque steel ‘cash up
front’ during the war — as if they were ordinary commercial customers. 132
Nor were pesetas acceptable. The Basque government demanded gold
or foreign exchange of Valencia and Catalonia in exchange for industrial
goods and materials. In part this was an attempt by the PNV to secure
the wherewithal for its own purchases abroad - whether of food supplies
or war-related materials - which also had to be made in hard currency.
The central Republican government paid credits to the PNV to sup-
port its contribution to the war effort, but it refused to allow Aguirre
directly to deploy a portion of central gold or foreign reserves. 133 In
part this was because they were at a premium and needed to supply the
Madrid front. But it was also because allowing the Basque government
to deploy such resources directly was perceived as further fuelling PNV
pretensions to sovereignty for Euskadi. It is clear that such pretensions ex-
Moreover, the wartime situation had allowed an extension of the
isted.

Basque government’s statutory political and military responsibility. 134

But the complexities of assuring transactions and their delivery were
130
Gonzalez Portilla and Garmendia, Laguerra civil en el Pais Vasco, pp. 84-9 Rebel control of the air
1 .

and occupation of Oviedo with its rail heads also put Asturian coal largely out of reach: Jackson,
The Spanish Republic and the Civil War, p. 378 - although Chiapuso suggests that Republican vessels
could have been deployed here but were not for lack of planning and foresight, El gobierno vasco
y los anarquistas, pp. 79-80. (For Basque industry after rebel take-over, see Richards, A Time of
Silence, pp. 1 10-13.)
131
See revealing anecdote about Aguirre’s propping up the publication of El Noticiero Bilbaino,

Chiapuso, El gobierno vasco y los anarquistas, p. 78.

132
This is of course implicit in the peacetime economy status outlined by Gonzalez Portilla and

Garmendia in La guerra civil en el Pais Vasco. Chiapuso, El gobierno vasco y los anarquistas, pp. 102,
hi, 122 and Broue and Temime, The Revolution and the Civil War in Spain, p. 167 refer as well to
Asturian coal also having to be bought - which reminds us of the general drag of particularisms
on the war effort.
133 Gonzalez Portilla and Garmendia, Laguerra civil en el Pais Vasco, pp. 73-4; Chiapuso, El gobierno
vasco y los anarquistas, pp. 1 14-17.
134 De Meer, El Partido JVacionalista Vasco, pp. 253-4; Chiapuso refers to the printing of Republican
Basque paper money which was used to pay their soldiers (as also occurred in Catalonia): El
gobierno vasco y los anarquistas, p. 208.
 Challenges to the centralising Republic 25 1

135
indeed greater in the case of the isolated and blockaded north. Thus a
practical case could have been made for the Basque government’s direct
deployment of state capital resources. A solution might have been for the
central government to bargain this against the PNV’s militarisation of
its war industries. Both the Basque and greater Republican war efforts
needed the Basque economy to be fully militarised in order to control
and stimulate production. But this question went straight to the heart of
the contradiction between the economic and political imperatives of the
Republican war effort. For reasons of both internal and international
politics, it demanded an integrated, inter-class alliance. But this, in turn,

precluded breaking the economic deadlock in the Basque Country. The

fall of Malaga in February 1937 impressed upon the Basque authorities

the need to consolidate their defences. But the resistance to a Republi-

can single command remained. And while the PNV’s political discourse
began recognising the need for interventionist measures in the economy,
in practice, nothing would happen until too late, when Vizcaya was on
136
the threshold of military defeat.
In other economic areas too the PNV’s ability to adapt to the needs
of the war was curtailed by its political outlook. It was insufficiently pre-
pared to intervene in the growing war-related problems of the home
front - and, in particular, the hoarding of paper money and coins by
affluent sectors, which was never adequately tackled. 137 The lack of
money in circulation did for a time help to contain inflation on basic
goods and limited to some extent the scope of the black market — as
did Vizcaya’s physical isolation and the consequent (and increasing) dif-
ficulties of importation. The PNV’s supply policy was to some extent
interventionist — perhaps initially in response to the early refugee influx
of at least 100,000 people, mainly from Guipuzcoa. 138 But it balked at
price and rent would have particularly assisted the most
controls that
vulnerable sectors of the urban population — especially as time went on
and the problems of food speculation and price inflation inevitably made
themselves felt. But the sectors worst affected here did not belong to the

135
Tunon de Lara, ‘Algunos problemas historiograficos de la guerra civil en Euskadi’, pp. 135,
J
39> Hi-
136
For example the Basque government decreed the seizure of safety deposit box valuables in
Basque banks in May 1937 - by which time the contents of many of these had probably already
been removed by their owners: Gonzalez Portilla and Garmendia, La guerra civil en el Pais Vasco,
P- 74-
137
Ibid., pp. 70-4.
138
Ibid., pp.94-5, 75 (n. 13). Ration cards were distributed as a preventive measure on 18
September: Chiapuso, Elgobierno vasco y los anarquistas, p. 58.
252 The Spanish Republic at war

PNV’s constituency. They belonged to those of the UGT and CNT. The
CNT was by far the most vocal critic of the PNV’s conservative economic
policy, which would lead to the imposition of increasingly severe censor-
ship on its press and the banning of its meetings. 139 But the CNT’s overall
marginality in Vizcaya meant that it could do little about this state of
affairs. The fact that both the Basque socialists and communists more

or less supported the PNV’s economic policy had the effect of neutralis-
ing the CNT’s criticism. More pressure on the PNV might at least have
forced it to employ innovative measures to solve wartime problems: for
example the deployment of refugees as industrial labour or, at the very
least, as agricultural labour to allow marginal land to be brought under
cultivation in order to ease food shortages -
ever present and increasing
under conditions of Non-Intervention. But such a use of refugee labour
would have involved large-scale trade union mobilisation in civilian front
war organisation, as was happening elsewhere in the Republican zone.
This would, in turn, have required a redistribution of wartime (as well
as, potentially, of post-war) social and political authority in the Basque

Country which the PNV was not prepared to contemplate.

The PNV’s conservative notions of political and cultural order also
meant that it looked askance at the influx into Vizcaya of Guipuzcoan
refugees. 140 The rising xenophobia in bourgeois Bilbao directed against
the perceived migrant and worker tenor of this influx echoed some of the
fears then also being expressed in more conservative nationalist circles in
Catalonia with regard to the CNT’s urban constituencies of ‘immigrant’
labour - although, in the xenophobia as such was less overt.
latter case,
The were fortunate, however, in that the struc-
authorities in Vizcaya
ture and pattern of Basque industrial development meant that they did
not face any problem of mass urban provisioning comparable to the one
challenging the Catalan authorities in Barcelona. (The ribbon develop-
ment of industry and the greater prevalence of the ‘mixed worker’ in
Vizcaya meant that a proportionally greater part of its urban popula-
tion retained ties to the countryside, and thus easier access to alternative

139 Chiapuso, El gobierno vasco y pp. 68-70,117-19, 123. As one moderate anarcho-
los anarquistas,

syndicalist leader put it (p. ‘These gudaris are more reactionary than the Carlist militia.’
117):
(‘Estos gudaris son mas carcas que los requetes.’) See also chapter 5 below for the PNV’s attempts
to obstruct production of the CNT’s main press, CNT del Norte. And while Basque nationalists
did suffer repression at rebel hands, one should note that there were those who went on to
fight for Franco and even to staff the prison brigades in which Basque leftists served (Chiapuso,
p. 242).
140
Ibid., p. 79.Also pp. 136-7 for the insularity of the bulk of the rural nationalist base which
reinforced these xenophobic sentiments.
 Challenges to the centralising Republic 253

sources of food - something which immigrant industrial worker popu-

lations in urban Barcelona did not have.) Nevertheless, the onset of the
rebel offensive against Vizcaya at the end of March 1937 saw a signifi-
cant build-up of refugees in Bilbao, whose population had by this point
more than quadrupled as a result - from 120,000 to 500,000. Under this
pressure the Basque government initiated refugee evacuation in early
May 1937, thus contributing to the growing subsistence crisis in urban
141
Catalonia.

141
Ibid., p. 206 refers to the initial evacuation of women, children up to 15 and persons over 65.
It is from the profile we have of the Basque child refugees sent to England that agency-
also clear
and government-assisted refugees represented predominantly less affluent social sectors: see
J. Fyrth,
The Signal was Spain. The Aid Spain Movement in Britain 1936—39 (London: Lawrence and
Wishart, 1986), pp. 220—42. Other sectors had long since made private arrangements which
usually took them to France.
 CHAPTER 5

The Barcelona May days and their consequences

(February—August 1937)

I am not fighting this war so a stupid, provincial separatism can

sprout up again in Barcelona I’m fighting for the sake of Spain.
. . .

For her past greatness and future possibilities. Those who imagine
1
otherwise are mistaken. There is only one nation: Spain .

Whether or not Negrin realised what he was saying when he spoke

of having delivered better public order than any government of the
2
previous fifty years ... is irrelevant: neither do pigs know they stink .

By early 1937 living conditions in Barcelona and the other urban centres
of Catalonia were coming under strain from the economic dislocations
occasioned by the war. 3 The region’s high pre-war population density
(already double the rest of Spain’s) was exacerbated by the major and
relatively continuous refugee influx from Madrid, from Malaga in the
south (from February 1937) and by late spring from the north. 4 By the
end of 1936 there were already somewhere between 300,000 and 350,000
refugees in Catalonia. 5 In addition, there were many evacuees who had
1
Juan Negrin in conversation with Julian Zugazagoitia in July 1938, cited in Zugazagoitia, Guerra

y vicisitudes,
454.p.
2
G. Munis, Jalones de derrota, promesa de victoria. Criticay teoria de la revolucion espahola (1930-1939)
(Mexico, 1948; Madrid: Zero, 1977), p. 502.
3
J. M.
Bricall, Politica economica de la Generalitat (1936-1939), 2nd edn (Barcelona: Edicions 62, 1978),

pp. 33-40 - although the situation was not necessarily worse in Barcelona capital than in the
other urban centres of Catalonia. Moreover, in absolute terms, the food situation at the start of
1937 was worse in Madrid ( Friends Service Council (hereafter FSC)/R/Sp/4).
4 The first refugees (women and children) left Asturias for Catalonia in February 1937, but the big
influx began in the spring with Basque refugees and continued via the Gijon convoys of October
when Asturias was on the verge of defeat.
5 Bricall, Politica economica, p. 93 gives the lower figure (based on Generalitat estimates). The higher
estimate is drawn from Quaker sources which cite National Refugee Council (i.e. central Repub-
lican government) sources. The Republican government defined a refugee as: ‘anyone, with the
exception of combatants or men in good health aged between 20 and 45 years, who has been
obliged to change residence on account of the present war and is not hostile to the regime but
lacks the means to support themselves and is not sheltered by family or friends’. Figures and
definition in FSC/R/Sp/2, file 4. The refugee influx from Malaga also supposed cultural (and
class) tensions as well as a quantitative problem, FSC/R/Sp/ 1
(files 1-2, reports/corresp. from

254
 The Barcelona May days 255

moved to live with relatives, but who were not dependent on any refugee
agencies - domestic or foreign - and thus did not figure in the formal
But even excluding this group, the refugee population repre-
statistics.

sented some io per cent of the region’s total population by the beginning
of 1937. This precipitated an urban resource crisis which was initially
manifest in the form of infrastructural overload. (Catalonia had the most
developed municipal services in Spain, but this could not prepare it for

the scale of social welfare war would produce.) By Decem-

demand the
ber 1936 there were shortages of basic foodstuffs and other staples which
6
fuelled inflation, in spite of official price controls.
The onset of war
had disrupted industrial production in urban cen-
tres. This led to sectoral unemployment and to the disruption of normal
7

8
rural-urban commercial exchange. Both left the poorer sectors of urban
Catalan society exposed, and the immigrant working class most of all. 9
Without family contacts in rural Catalonia and with the least monetary
resources, they lacked the wherewithal to engage in the barter economy
10
that was already appearing. This situation was mitigated in the early
months after the coup by the emergency provision of communal kitchens
and, more crucially, by collective means of food-procurement - neigh-
bourhood and workplace food cooperatives - organised in the period
of CNT ascendancy.
11
Such grass-roots initiatives often connected up

Barcelona 1936—7). Also FSC/R/Sp/4, reports 1936-8, report of 19 May 1937 describes Malaga
refugees (this time in Murcia) as ‘wild’ and ‘half Moors’ and very frightened of ‘lists’ for fear of
what their exposure to state or public authorities might mean.
6
Both the Generalitat and the central Republican authorities had decreed price controls after
18 July, but these had little effect: Bricall, Politico, economica p. 106 (n. 14), p. in; Abella, La vida
,

cotidiana durante la guerra civil, pp. 192-3.

7 Bricall, Politico economica, pp. 85^
8
Ibid., pp.137-40; E. Ucelay da Cal, La Catalunya populista. Imatge, cultura i politico en I’etapa republicana

(1931 -1939) (Barcelona: La Magrana, 1982), pp. 309-10.

9 The analysis in this section should not be taken as implying that conditions for the poor and
most economically vulnerable were, in absolute terms, worse in urban Catalonia than elsewhere in
the Republican zone. This is a difficult matter to assess, and the situation changes across the war
period. But by December 1936 things were extremely hard for the urban poor - a constituency
more numerous in the Catalan capital than anywhere else and one constandy growing through
the refugee influx. The point of the analysis here is rather to indicate how the acute deterioration
of material conditions (crisis de subsistencias) was subjectively experienced, interpreted and reacted
to by those constituencies at the receiving end.
10
Widespread barter - properly speaking - emerged during the second half of 1937 Bricall, Politico :

economica, pp. 141-2. But by the end of 1936 those with enough resources were certainly using
them to obtain agricultural produce direct - either with money or in exchange for other goods -
often luxury/ specialist or imported (such as coffee or tobacco). As Quaker reports indicate, city
shops still had plenty of goods to in late 1936 and early 1937, only these were often tinned
sell

reserve stocks or luxury products which scarcely addressed the gathering staple food crisis: see
Ucelay da Cal, La Catalunya populista, pp. 315-16.
11
Bricall, Politico economica, pp. 148-9; Ucelay da Cal, La Catalunya populista,
p. 313.
256 The Spanish Republic at war

with the CNT supply committees, which played a major role in feed-
ing working-class neighbourhoods. In the absence of rationing, 12 these
mechanisms were the key to the survival of the urban poor as shortages
had increased over the autumn and winter of 1936.
By December 1936, however, the bread shortage was acute. 13 The
material hardships indicated by food queues and accelerating inflation
fuelled popular support for CNT- and POUM-led campaigns for the
implementation of rationing. But the initial response of PSUC leader
Comorera, now in control of the supply portfolio after the December
cabinet reshuffle, was publicly to accuse his CNT predecessor, Juan Jose
Domenech, of incompetence and to abolish the supply committees. 14
These had become the focus of smallholders’ hostility because they
regarded the prices at which they were required to sell as unfair and
perceived the transaction to be based on implicit (when not explicit)
coercion.
For both the PSUC and the Esquerra, getting rid of the supply com-
mittees had more to do with eroding the political power of the CNT than
it did with economic deregulation/?^^. Nevertheless, Comorera and his
colleagues do seem to have believed that the urban food shortage in
Catalonia was mainly the result of peasant hoarding rather than dearth.
Allowing prices to increase was thus seen as a way of resolving the prob-
lem by giving Catalan smallholders the necessary incentive to sell. 15 The
PSUC referred to its economic deregulation as ‘Catalan NEPV 6 But
the party’s optimism was badly misplaced. Catalonia’s macro-economic
situation was quite unlike that of post-civil war Russia.
Catalonia was a net importer of staple foodstuffs. At least half of the
region’s wheat consumption was normally dependent on imports from
other parts of Spain or abroad. But the wartime division of Spain had

12
Although rationing was formally introduced for Barcelona by Generalitat decree on 13 October
1936, the system had not yet been implemented: Bricall, Politico, econdmica, p. 150.
13
Quaker sources remark on the population’s reluctance to change its eating habits - even though
substitutes for bread were available at this stage - because of the cultural importance of bread at
a meal. FSC/R/Sp/i, correspondence/reports from Barcelona 1936-7, letter to London office,
December 1936. Where poorer constituencies were concerned at least, bread had a value beyond
the material - what has been called ‘the sacralisation of bread’ inherent in societies existing on
the edge of hunger: Simeon Riera, Entre la rebelioy la tradicio, p. 244.
14
By decree of 7 Jan. 1937.
15
By autumn 1936 Generalitat publicity posters were issuing pleas against hoarding and specula-
tion; see, forexample, Fontsere’s November 1936 poster reproduced in Bricall, Politico econdmica,
p. 96 (no. 9). Although the committees were the primary target of smallholders’ hostility, by
spring 1937 the (restored) municipal authorities were also being criticised for paying low prices
for crops in order to generate extra income for local government services (Bricall, p. 148).
16
Payne, The Spanish Revolution, p. 289.
 The Barcelona May days 257

separated food-producing areas from their natural markets. 17 Nor was

the problem susceptible to solution under wartime conditions. Although
grain was imported from abroad on occasions, both the Generalitat
and the government were increasingly limited in
central Republican
their ability to purchase foreign wheat as the value of the Republican
peseta fell and the Non-Intervention arms embargo obliged them to
concentrate virtually all foreign exchange on the covert purchase of war
18
material (‘cash up front’) on the international arms market. The lack of
wheat and other basic foodstuffs in Catalonia was, then, the result of an
absolute shortfall, massively exacerbated by the continually increasing
refugee population. Moreover, given the circumstances just described,
the shortages could only get worse as time went on.
For this very reason the Republican authorities needed strict central
control over domestic economic resources in order to be able to prioritise
their use. In Russia, the Bolsheviks had relaxed their centralised control
to implement the New Economic Policy (NEP) only after the defeat of
the White armies. For the Spanish Republic, its version of that battle
still lay ahead. Moreover, it was up against an enemy far more efficiently

aided and supplied by its European backers than Russia’s White armies
had ever been. The mass, modern war - driven by German technolog-
ical aid - made the urban population of Catalonia and, especially the

industrial workforce at the heart of armaments production, vital to the

Republican war effort. Guaranteeing an equitable rationed minimum
of essential foodstuffs for this sector would probably have been beyond
the organisational capacities of the Republican state in 1936—7. But even
the political will to achieve this was absent because it ran counter to the
liberal economic orthodoxies which still underpinned the Republican
political alliance.
After the abolition of the committees, bread shortages increased and
prices rose further. In theory, overall staple-food prices were now un-
der Generalitat control. But the technical complexities of setting up
price-control mechanisms plus an ambivalence to such controls in the
PSUC 19 made for quite a different daily reality at the stores and stalls of
urban Catalonia. The CNT and the PSUC engaged in mutual recrim-
inations over the cause. The anarcho-syndicalists denounced it as retail

17
Bricall, Politico, economica, pp. 137-40; Ucelay da Cal, La Catalunya populista, pp. 309-10.
18
Howson, Armsfor Spain passim.
,

19
The PSUC had even objected to the Esquerra’s mild streamlining measure back in August 1936
which required the peasantry to sell its produce through a single body in order to control prices,
Guerra y revolucion en Espana, vol. 2, p. 31.
 .

258 The Spanish Republic at war

speculation by the middle-men of the UGT- affiliated Federation of Small

Traders and Manufacturers (Gremis i Entitats de Petits Comerciants i In-
20
dustrials (GEPCI)) founded by the PSUC at the start of the war, while
the PSUC blamed the legacy of CNT inefficiency and poor harvests,
made worse by the collectives. In fact, both the CNT and the PSUC faced
a complex array of economic problems in the supply portfolio, with only
21
limited organisational resources and controls at their disposal. The
CNT had been responsible for supply functions during the period of
greatest fragmentation. The harvest had not long been gathered in, but
although that meant there were stocks to cushion the situation, the peas-
antry was attempting to thwart the purchasing committees. As a result,
by early December the CNT had already begun to campaign against
22
speculation. When the PSUC took over, government authority was
being reconstructed, but it still had to contend both with real shortages
and with administrative disorganisation in Generalitat structures. Spec-
ulation also went on growing, as did food distribution difficulties arising
from a serious lack of transport (and petrol) in the Republican zone.
Food shortages, inflation, speculation and the emerging black market
were all symptoms of a war-induced economic crisis. Tackling that crisis,
however, involved a political choice. The shift from CNT supply commit-
tees (for all their very real shortcomings) to government price controls
signified that the industrial workforce and the urban poor were to be
required to bear the brunt of Catalonia’s growing wartime subsistence
crisison behalf of the rest.
With both prices and the scale of refugee need continuing to increase,
the Generalitat was obliged to implement rationing in Barcelona in
February. 23 But the system was extremely inadequate. In practice, staple
goods were frequently unavailable through the rationing system. 24 More-
over, rationed goods were supplied to neighbourhoods in an extremely
haphazard way that took little account of different population densities.
Inevitably it was the urban poor in the cramped housing conditions of

20 igoo-igg6, pp. 157—8;

A. Balcells, Trabajo industrialy organization obrera en la Cataluna contemporanea

Bolloten, SCW, 397p.

21
CNT control of supply (July-December 1936) saw a 47 per cent price increase, while the next
six-month period under PSUC supervision saw a further 49 per cent increase, according to
figures produced by the Servei Central d’ Estadistica de la Generalitat de Catalunya cited in
Bricall, Politica economica, pp. 137-8.
22
Abella, La vida cotidiana durante la guerra civil, p. 193.
23 It was extended thereafter to other municipalities of the province and region: Bricall, Politica
economica, p. 150.
24 Originally bread, potatoes, sugar and eggs were rationed (dairy produce was generally in short
supply), and soon olive oil was added. The list increased as the war went on: ibid., p. 150.
 The Barcelona May days 259

and the densely populated working-class neighbour-

inner-city Barcelona
hoods beyond the Eixample 25 who suffered most through the system’s
shortcomings. The already scarce resources in such areas were subject
to added pressure too, with the influx of war-displaced persons to stay
26
with family members or friends. The food depots frequently ran out of
food before everyone in the long queues had been served. 27
Nor was it the urban working class alone who suffered from this situ-

ation. There were also large numbers of middle-class families of modest

means — the natural supporters of Esquerra and PSUC — who could not
afford high market prices either and who were equally outraged by the
evidence of speculation and the emerging black market. The manifest
lack of a guaranteed minimum in the life of so many urban dwellers
provoked street demonstrations against the food shortages from early
1937 onwards and, most notably, on 14 April, the sixth anniversary of
28
the Republic’s birth, following a sudden and sharp price hike. Many
of the street protests against bread shortages and high prices, running
through from February to May, were led by women - apparendy replicat-
ing the role they had taken in subsistence crises in many Spanish towns
across many decades. 29 But these wartime protests were different in that
many were direcdy politically mediated. The CNT, FAI and all POUM
instrumentalised such demonstrations as a plebiscite against Generalitat
economic policy, and they also boycotted the Republican commemora-
tion on 14 April. But a greater number of protests were mobilised by
women’s associations affiliated to the Popular Front. In Catalonia this
allowed the food protests to be channelled by the PSUC in order to
bolster the case for ‘strong government’. 30 At the same time as Popular
Frontist mobilisation was accelerating in Barcelona, however, the CNT
remained plugged in, in a rather less ‘modern’ or ‘organised’ way, to

25 This district, built during the nineteenth-century expansion of the city, was where the most
affluent sectors of Barcelona’s bourgeoisie traditionally resided.
26
No figures available - but thousands of such people existed, as Quaker relief reports indicate
(FSC). Thus, in spite of relief agencies’ utilisation of the empty houses of the bourgeoisie, there
was still a lot of extra population pressure in poor neighbourhoods.
27
Long bread queues and the subsequent problems of malnutrition are well documented in the
reports and correspondence of Quaker Barcelona (Sants and San Andres were
relief workers in
estimated to be the neediest neighbourhoods); also Morrow, Revolution and Counter-Revolution in
Spain pp. 64-5.
,

28
Prices rose another 13 per cent on top of a nearly two-thirds rise since July 1936: Fraser, Blood
of Spain, pp. 375-6; Morrow, Revolution and Counter-Revolution, pp. 64-5; Bricall, Politico economica,

P- *37 •

29 M. Seidman, Workers Against Work. Labor in Paris and Barcelona during the Popular Fronts (Berkeley:
University of California Press, 1991), p. 138.
30 Ucelay
da Cal, La Catalunya populista, pp. 315, 316; Cruells, Mayo sangriento, p. 31.
26 o The Spanish Republic at war

other social strata where the daily impact of laissez-faire

- more silent,

but no less profound - was priming a time bomb of social protest of a

different kind.
Extreme material hardship was scarcely a stranger to the urban work-
5
ingclass, or to other poor and marginal sectors of ‘outcast Barcelona.

There had also been shortages when the CNT committees were in place.
Moreover, in so far as these were the result of speculation, it would have
been very hard for the CNT — especially in a Popular Front mode which
precluded or restricted coercion — to have controlled that process even if

it had retained charge of the Generalitat supply portfolio after Decem-

ber 1936. Nevertheless, the continuing deterioration in the city’s supply
situationand increasing inflationary pressure after January 1937 were
interpreted by at least some of the social constituencies who looked to
the CNT as other than merely war-induced incremental hardship. In
the daily battle for survival, shortages, inflation, middle men and the
black market 31 were read by sectors of the poor as the outcome of the
economic normalisation occurring in parallel with the process of recon-
structing state political authority inside the Generalitat. To understand
how and why was made, we need to remember the context
this link

of repressive continuities experienced by poor and marginal social con-

stituencies in their ongoing war with the liberal state.
In the flux following the defeat of the military coup in July 1936 the
possibility of social and economic change had been glimpsed. But with
the reconstitution of the Generalitat the full weight of institutional dis-

approval fell on collective grass-roots food procurement initiatives. The

abolition of the supply committees was the culmination of a larger pro-
cess of eroding such initiatives daily. 32
This also witnessed scenes - famil-
iarfrom the pre-war period 33 - of the police clearing street sellers 34 and
breaking up food protests, as well as now protecting commercial quarters
5
from ‘popular requisition Clearing itinerant vendors could of course be
.

publicly justified as a move against the abuses of the black market, but it

was a government measure which also erupted in unforeseen ways into

the fragile economies of the urban poor.
Popular protest grew across the early months of 1937. The poor of
Barcelona had no rights to assistance from the refugee agencies in their
31
Morrow, Revolution and Counter-Revolution in Spain, p. 78; Semprun-Maura, Revolution y contrarre-
volucion en Cataluna, pp. 24off.; Abelia, La vida cotidiana durante la guerra civil, p. 196.
32 Ucelay da Cal, La Catalunya populista, p. 313.
33 Ealham, ‘Policing the Recession’, pp. 192-3. See also chapter 1 above.
34 For example after the municipal ordinance of 10 November 1936, cited in Ucelay da Cal, La
Catalunya populista, p. 320. See also Abelia, La vida cotidiana durante la guerra civil, p. 188.
 The Barcelona May days 26 i

daily struggle for survival. Even in those industries converted to war

production, workers’ wages failed to keep up with inflation. But many
families had lost their male breadwinner, while some had never had one.
This produced intolerable strains on working women since wartime food
procurement was itself a full-time job, especially for the poorest 35 .

While the poor fought local battles for material survival much as they
had done before the war, at regional level, the Generalitat was preparing
its own final push on public order. The early months of 1937 saw the

imposition of tighter police discipline as Assault Guards and National

Republican Guards were merged into a single Catalan police corps and
prohibited from membership of any political party or trade union 36 .

While such a measure was still difficult to enforce, it did achieve the
major goal of putting the workers’ patrols beyond the law while also
debarring their members from the unified police (and, therefore, from
any legitimate policing functions ). 37 In reality, however, the patrols went
on existing - now in open conflict with the state. The tension mounted fur-
ther on 12 March when the central Republican government ordered
all worker organisations, committees, patrols and individual workers to

hand over their arms - long and small - within forty-eight hours 38 This .

confrontation over public order led to the dissolution of the Catalan

cabinet on 27 March when
the CNT’s representatives withdrew. But the
ensuing three-week concluded in a very similar cabinet configu-
crisis

ration on 16 April. Still entrenched in the Generalitat’s Home Office,

the Esquerra’s Artemi Aiguader stepped up his war against the patrols.
Along the French/ Catalan border, there were fatalities in the escalating
clashes now occurring between Carabineros and CNT patrol commit-
tees over control of customs posts which the committees had held since

July 1936.39
One of these incidents in particular suggests how these confrontations
were many-stranded. They connected up with ancient border disputes
pitting local smugglers against those upholding the fiscal prerogatives of

35 The Quaker wartime reports contain much devastating evidence of this. As a result, children
also had to spend long hours in food queues.
36 Bolloten,
SCW, pp. 417, 865 (n. 25) cites Diari Official of 4 March 1937. The ban on political
membership was originally a central government decree 28 Feb. 1937) applying only to the
(of
Carabineros (Customs Police), but then extended to the Catalan police force at the beginning of
March 1937.
37 As part of March
this normalisation of public order, 3 1937 saw the Generalitat dissolve the
CNT-controlled defence committee, creating another one, under cabinet control and with the
power to dissolve all local police and militia committees.
38 Morrow, Revolution and Counter-Revolution p. 73. ,

39 Broue and Temime, The Revolution and the Civil War in Spain, pp. 281-2.
262 The Spanish Republic at war

the political ceptre- whether Spanish or French-40 as well as with more

recent conflicts over state control, or stemming from acute social conflicts
between the Catalan labour unions. On a day in late April at Puigcerda,
Catalan police engaged in fire, shooting dead a number of anarchists,
including the influential Antonio Martin, old-guard radical anarchist and
ex-smuggler turned CNT customs chief. 41 When this incident occurred,
the police detachment had been returning from Molins de Llobregat
having arrested its CNT leadership for supposed involvement in the
earlier assassination of UGT leader and PSUC member Roldan Cortada
on the 25th. As he was an ex-treintista (and the secretary of Rafael Vidiella,
another leading PSUC politician who had once been in the anarchist
movement42 ), suspicion fell on the CNT even though it had publicly
condemned the murder and called for a full enquiry. From the time of
these incidents political tension was mounting in the whole of the Baix
Llobregat area of Barcelona province. 43
There were longstanding tensions between the CNT and the UGT
stretching back into the pre-war period, as we have already seen. But
these escalated after 18 July 1936 and led to the assassinations of both an-
archist and socialist leaders and militants in various areas of Republican
Spain, but notably in Barcelona. 44
After the military coup of 18 July 1936 the Catalan UGT began to
expand hugely, for much the same reasons as the PSUC, as lower-
middle-class commercial and white-collar sectors - often previously

40 Cf. P. Sahlins, Boundaries: The Making of France and Spain in the Pyrenees (Berkeley: University of
California Press, 1989), pp. 103-33.
41
An exhaustive account is inj. Pons i Porta and J. M. Sole i Sabate, Anarquiay Republica a la Cerdanya
(1936-1939)- El ‘Cojo de Malaga’ i els Jets de Montserrat, 1991) and
de Bellver (Barcelona: l’Abadia
for Martin especially, pp. 21-46, 133-73. Martin had apparendy belonged to the Los Solidarios
direct-action group, along with Ascaso, Durruti and Garcia Oliver. See also M. Benavides, Guerra

y revolucion en Cataluha (Mexico: Ediciones Roca, 1978), pp. 344, 351-62; Jellinek, The Civil War
in Spain pp. 544-5;
,
Broue and Temime, The Revolution and the Civil War in Spain pp. 282, 293-4
,

(n. 36); Thomas, Spanish Civil War, pp. 31 1 (n. 2), 653 (n. 1); Ucelay da Cal, La Catalunya populista,
p.304 (n. 23).
42 Thomas, Spanish Close to Largo Caballero before the war, Vidiella had re-
Civil War, p. 300 (n. 2).

signed from the PSOE national committee in May 193 6 and gravitated - with some reservations -
towards a working relationship with the Catalan communists in the PSUC: Graham, Socialism
and War, p. in; Bolloten, 398. He replaced Nin as Generalitat Justice councillor in
SCW, p.
December 1936, moving to the labour/public works portfolio in April 1937. He was also a
freemason: Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 306,
n. 12.
43 Peirats, La CNT en la revolucion espanola, vol. 2, pp. 137-8; Ucelay da Cal, La Catalunya populista,
p. 303; Broue and Temime, The Revolution and the Civil War in Spain, p. 282.
44 Graham, Socialism and War, p. 64. The UGT union leader, Desiderio Trillas (shot dead on 29
July 1936), was an ex-cenetista: Balcells, ‘El socialismo en Cataluna hasta la guerra civil’, p. 33;
Jellinek, The Civil War in Spain, p. 343. For the pre-war history of this conflict, see chapter 1 above.
 The Barcelona May days 263

unorganised (or only organised as isolated trade/ technical associations) -

joined party and union 45 in search of protection in a hostile, appar-
ently anarchist-dominated political environment. Indeed, it was the joint
Esquerra/PSUC rearguard action against the CNT that underlay the
compulsory sindicalisation decree issued in Catalonia in December 1936.
After this, the Catalan UGT expanded rapidly, membership extending

to the self-employed and small businesses, in the shape of the Small

Traders and Manufacturers Federation (GEPCI). The identification in
popular perception of GEPCI with increased economic speculation was
a further source of social friction between the Catalan UGT and the
CNT
But everywhere, not just in the GEPCI, the newly politically mobilised
urban lower-middle classes perceived party and union membership as a
form of professional advancement or career opportunity. For the PSUC
in Catalonia (as the PCE elsewhere) was the major articulator of the Pop-
ular Front war effort and provider of government/ service cadres. 46 As
white-collar staff joined the Catalan UGT, so it also became the organ-
isational bulwark of internal resistance to socialised factory systems. 47
(White-collar employees, and even skilled workers sometimes - because
they had more bargaining power, or for other more subjective reasons -
usually felt little sense of solidarity with more precariously placed sec-
tors of labour.) It was consistently the office and professional staff who
saw CNT industrial controls as the ‘problem’ 48 and supported increased
governmental intervention at the factory assemblies that the Generalitat
required in each workplace to ratify its socialised status. 49
Although the CNT’s radical middle-level cadres were dug in on the
factory committees, this was more a case of siege conditions once the
anarcho-syndicalists had lost political control in the Generalitat. (In

particular, the erosion of CNT economic control went back to their

acceptance of Companys’ suggestion of a Council of Economy with an
explicidy centralising brief, although, as it had been government that

45 For example, the shop and office workers’ union (CADCI) joined the Catalan UGT in August
1936: Balcells, Trabajo industrialy organization obrera en la Cataluna contemporanea igoo-igy6 pp. 156,
,

157-8; Bolloten, SCW, p. 416.

46 For
a similar modernising role played by another European Communist Party, see Mark Mazower
on the transition between clientelist structures/networks and mass mobilisation in occupied
Greece, Inside Hitler’s Greece. The Experience of Occupation ig^i-44 (London/New Flaxen: Yale
University Press, 1993).
47 Cf. Balcells, Trabajo industrialy organization obrera en la Cataluna contemporanea igoo-igy6, p.158.
48 Abella, La
vida cotidiana durante la guerra civil, p. 229.
49 C. Vega, A. Monjo and M. Vilanova, ‘Socializacion y hechos de mayo’, in Historiay Fuente Oral,
3 (1990)* 95 -
264 The Spanish Republic at war

had held the gold and controlled the banks from the start, credit star-
vation was always going to be a major problem for all the collectives —
industrial and agricultural. 50 Moreover, the ‘syndical capitalist’ attitudes
which inevitably resulted made the collectives easy targets for their po-
litical enemies, who could, quite truthfully, point to the damage being
done to the war effort by individual collectives seeking to protect their
resources from ‘outsiders’. 51 )The CNT was unable to prevent the Gen-
eralitat from gradually circumscribing socialised industry in Catalonia -
which was the intention behind successive decrees from December 1936
onwards. 52 It was the scale and entrenchment of socialised industry
in Catalonia (and especially urban Barcelona) which created an un-
precedented political problem for the liberal Republican alliance - even
though worker-intervened industrial forms existed in many places: in
Madrid, Valencia, Alicante and Almeria and in Malaga (prior to Febru-
ary 1937). 53 By April 1937 the Generalitat was refusing to certify factory
councils’ ownership of exported goods tied up in foreign ports pending
the resolution of legal suits lodged by former owners.
In its impatience, the Generalitat shifted in April from indirect political
manoeuvring within individual factories. (Government officers would
exploit internal differences in the assemblies between blue-collar and
administrative staff, or they would manoeuvre with UGT officials or,
very occasionally, with syndicalist sectors of the CNT.) Instead they sent
in the police, which, on at least one occasion, saw the surrounding of a
factory at the time of the vote. 54 For politically active sectors of the labour
rank and file, such a blatant use of the police clearly reinforced what they
already knew: at the heart of the battle to control Barcelona’s factories

50 Industrial collectives in financial straits also began to look to state intervention to solve their
problems: Fraser, Blood of Spain, pp. 211, 230, 231-2, 578.
51
Broue and Temime, The Revolution and the Civil War in Spain, pp. 163, 169.
52 Morrow, Revolution and Counter-Revolution, pp. 51, 60-2. Generalitat measures included: classifying
firms into industrial groups (26 Dec. 1936); a statute increasing the powers of the government
representative within the factory (30 Jan. 1937); the introduction of a ‘certificado de trabajo’
(21 Feb.1937) and reintroduction of the dismissal procedure as per contract laws of November
1931 (24 Feb. 1937); the stipulation that the statutes of firms had to be presented to the Juntas
de Control Economico Sindicales for approval (30 March 1937): see Vega et al., ‘Socializacion y
hechos de mayo’, p. 100, n. 4. Creeping governmentalisation was facilitated by the fact that the
CNT had never really developed its ideas on industrial (as opposed to agrarian) collectivisation:

Fraser, Blood of Spain, p. 212. None of the measures fostered by the CNT industry minister, Peiro,
facilitated ‘worker control’ in any real sense.
53 Beyond Catalonia the central government had been more successful in installing appointee
directors and confining factory committees to routine matters.
54 A summary of tactics is in C. Vega et al., ‘Socializacion y hechos de mayo’, pp. 97-8 (dissuasion
at Hispano-Suiza; CNT/UGT manoeuvrings at Casa Girona and coercion at the Trefileria
wirework factory, pp. 94-6 of same article).
 The Barcelona May days 2 65

there was a political agenda. Consubstantial with this, moreover, was the
restitution of power to the very police forces that, prior to the military
coup, had been the instrument of capital and state - whether monarchist
or Republican.
But the perception of police ubiquity also influenced broader social
sectors of ‘outcast Barcelona’. For them too the revanche of the ‘state’ -
or, at any rate, a hostile form of order - was embodied in the repressive

power of its security forces. The police were remembered for their evic-
tion of rent strikers in the pre-war period, and for their front-line role in
the implementation of Republican policies of social control which rigor-
ously invigilated public spaces and criminalised the unemployed. That
many of the Catalan policemen enforcing the ‘new’ liberal order had
gravitated to the PSUC 1936 meant no more than that they
after 18 July
had swapped the moribund conservatism of right-wing republicanism
for a more dynamic brand - just as in 1931 they had turned to republi-
can conservatism upon the eclipse of the dynastic variety. In so far as the
role of political parties had a higher profile, this reflected the relative un-
derdevelopment, historically, of the liberal state in Spain. The strength
of lower-middle-class and white-collar sectors in Catalonia, plus the par-
ticular circumstances in which the war was being fought, conditioned
the means by which private property and liberal order was defended.
Nevertheless, what was primarily significant about the PSUC here was
not that it was ‘communist’, but that it came to be so closely identified
with liberal state building and the economic establishment — not least
by the police operatives who joined it. They were now in the civilian
‘front line’, enforcing liberal order in the factories and on the streets of
Barcelona.
By
April 1937 worker patrols had already been excluded from all
police functions in the other major Republican cities of Madrid and
Valencia. State enforcement in Catalonia, and Barcelona especially, was
bound be more complex given the strength of popular resistance.
to
(The after-effects of political scandal probably also impacted here. The
chief of police, Andreu Reverter (of the radical nationalist Estat Catala),
detained in November 1936 for bribery and corruption, was also ac-
cused of plotting to assassinate CNT leaders and of involvement in a
plan to call upon the French to intercede to procure a separate Catalan
peace with Franco. 55 Although the latter never progressed beyond the

55 Morrow, Revolution and Counter-Revolution reported from Solidaridad Obrera (Nov. 1936);
,
p. 68,
M. Low and J. Brea, Red Spanish Notebook (1937; San Francisco, 1979), pp. 214-15; Ucelay da
Cal, La Catalunya populista, p. 295; C. Rojas, La guerra en Catalunya (Barcelona: Plaza & Janes,
 ,

266 The Spanish Republic at war

political fantasies of assorted Catalan nationalists, even the rumours of

such a thing would have been enough to slow up the campaign against
forms of worker popular power and especially the worker patrols.) But
after the April 1937 cabinet crisis, Aiguader gave orders that the arms
decrees were to be rigorously enforced. During the second half of the
month, workers in Barcelona were disarmed on sight by the police -
except, that is, where the police could be outnumbered and themselves
disarmed.
The political temperature rose further as Roldan Cortada’s funeral
turned into a demonstration of state power in the form of a long march
past of armed police and troops. While this also reflected middle-class
fears that the recent violence might herald a return of the feared paseos,
the blatant rehearsing of the state’s repressive capacity and moral panic-
inducing editorials in Barcelona’s liberal republican press (including
Treball, the PSUC newspaper) 56 were fatal components in the accumula-
tion of social and political tensions ( crispacion). According to one source,
three hundred workers were disarmed in the last seven days of April. 57
It was this escalating confrontation over arms which catalysed the dense
web of conflicts, bringing the city to the brink of street fighting. 58
In an attempt at containment, the traditional First of May (May Day)
labour demonstrations were suspended throughout Catalonia. But then,
on the afternoon of Monday 3 May, a detachment of police attempted to
seize control of Barcelona’s central telephone exchange (Telefonica) in
order to remove the anarchist militia forces inside. (Their presence dated
back to the July days of 1936 when the militia had taken the Telefonica,
along with other key buildings in central Barcelona, from the occupying
But what this amounted to was the ejection of the CNT from
military.)
the union control committee - also comprising UGT representatives
and a government delegate - thereby depriving the anarcho-syndicalists

1979), PP- 120-1; Thomas, Spanish Civil War pp. 524-5. Joan Casanovas (while he was Cata-
lan premier in the autumn of 1936) proposed that Catalonia should try to negotiate a separate
peace with Franco, with international guarantees. But this was a maverick line. It did not reflect
Companys’ thinking or that of the Esquerra in general: J. Benet, La mort del President Companys
(Barcelona: Edicions 62, 1998), p. 119.
56 Cf. M. Cruells in the Diari
de Barcelona, 1 May 1937, ‘En la diada d’avui comencem la neteja del

baix fons de la Revolucio. Cada organitzacio tindra la seva feina.’ (‘Today we must begin the
cleansing operation to remove the dregs of the Revolution. Each organisation will have a share in
this task.’) Cited in M.
Cruells, Els Fets de Maig, p. 40. It is also interesting to compare the subtext
of such editorials with the degenerationist and social darwinist discourses - languages of anxiety
all - which emerged in Catalanist discourse after the First World War, C. Ealham ‘Policing the
Recession’, pp. 17-19, 34-40 (see also note below on the Barri xines).
57 Morrow, and Counter-Revolution, 58 Cruells, Mayo
Revolution p. 73. sangriento, pp. 27-43.
 ,

The Barcelona May days 267

of access to information which was now vital to shore up their fragile

political position. Moreover, the building itself was charged with extra
painful memories since the CNT’s bitter strike defeat of 1931.
News of the attempted seizure spread rapidly through the popular
neighbourhoods of the old town centre and port. By the evening the
city was on a war footing, although no organisation - inside or outside
government - had issued any such command. By 4 May barricades were
up in the city centre, there was a generalised work stoppage and sporadic
street fighting erupted. In its origins, this conflict set the Catalan govern-
ment, and their political defenders against all those who,
state agencies
for whatever reasons, opposed the expansion of state jurisdiction. The
events would be complicated by a further conflict - between competing
centralising and regionalist liberal agendas inside the Republican polity.
But it is important to make the distinction here between the underlying
causes of the May days, and the ‘high political’ opportunities and con-
sequences which then emerged from the evolution of the conflict in both
street and cabinet office.
The key to anti-state popular mobilisation lay in the close relationship
between mid-level cadres (shop stewards and branch activists) of the
CNT’s comites de base (both neighbourhood59 and factory committees) and
militant, mobilised sectors of the industrial working class, many of whom
had been the shock troops of earlier labour wars in the city. 60 Although
the CNT’s committee structure had been decapitated by the process
of state political reconstruction spearheaded from the Generalitat, the
grass-roots committees still existed and could thus provide, together with
the worker patrols, the organisational sinews of collective resistance in
May. 6 '

The assault on the Telefonica focused the resisters’ energies on the city
centre where all the political and economic machinery of government
was concentrated — in close proximity to the most volatile of popu-
lar neighbourhoods, the Barri xines (literally, ‘Chinese Quarter’), which
had long constituted the front line between ‘respectable’ and ‘outcast’

59 These were also known as comites de defensa confederal (CNT-FAI), Peirats, La CNT en la revolucion

espanola vol. 2, p. 192.

60
Obviously this sector had been reduced by its protagonism in the militias. But the infrastructural
needs of the home front, especially industrial war production, meant that many CNT shop
stewards and workers had been ‘mobilised’ in their jobs.
61
Freund, ‘Dual Power in the Spanish Revolution’, pp. 326-7; Low and Brea, Red Spanish Notebook,
pp. 221-2; Peirats, La CNT en la revolucion espanola, vol. 2, p. 192; Broue and Temime, The Revolution
and the Civil War in Spain,
287; Fraser, Blood ofSpain, pp. 381, 382; cf. also C. Ealham’s
p. comments
on the low public profile of pivotal branch activists, ‘Policing the Recession’, p. 148.
268 The Spanish Republic at war

Barcelona. 62 Ijideed, the force of the initial May explosion is explica-

ble only if one bears in mind the longstanding connection between the
‘outcast’ city and the CNT. While the appearance of the barricades
constituted an act of conscious ‘political’ contestation, the CNT’s direct
action was also mediating more amorphous, ‘pre-political’ 63 forms of
popular resistance. The CNT was, once again, functioning as a lightning
conductor in inner-city Barcelona, transforming both a shared history
of persecution and the perception among the city’s marginalised of the
connection between state action (public order, food supply and so on)
and the brutality of daily life into generalised support for street action
as active protest ‘against the state’. 64 This was what confronted liberal
Catalonia and its police force in central Barcelona on 4 May. 65
As evening of 3 May, with an equilibrium of forces inside
late as the
the Telefonica and an armed stand-off around it, the Generalitat could
have averted the explosion in the streets. But it would have required
a climbdown over the police seizure and the removal of police chief
Rodriguez Salas, who had led the assault (as he had also led the earlier
one into Molins de Llobregat when Antonio Martin had been killed). It
would probably also have required the resignation of the ERC’s abra-
sive Artemi Aiguader, who, as Generalitat Home Office minister, had
66
formally authorised the Telefonica raid. But far from backing down,

62
The Barri xines, as the hub of ‘lawlessness’ and moral iniquity inspired anxiety because of its close
proximity to the financial and political centres of bourgeois Barcelona (even ‘Imperial Barcelona’
in the language of early Catalan nationalism — according to which planned urban development
provided a vital key to social control): Ealham, ‘Policing the Recession’, p. 18.
63 Ucelay da Cal, La Catalunya populista pp. 321-3. However, anarchist beliefs were no less a response
,

to ‘everyday experience’ than other forms of political behaviour.

64 These connections also make somewhat questionable the implicit assessments in some work
on the history of industrial socialisation in Barcelona. (There is an example in Vega et al.,
‘Socialization y hechos de mayo’, which is largely based on oral testimony fifty years on.) In
referring to the ‘passive role’ of the bulk of the affected Barcelona work force, this seems to be
taken as meaning such people had no views on state or government actions. I would argue this
ignores the broader context of daily life, above all around the issue of food supply, which must
inevitably have shaped opinions.
65
J. M. Bricall’s Politica economica perhaps underplays the fairly steep learning curve
of Catalan
politicians regarding the need to be interventionist in matters of wartime staple food pricing and
supply (or, at least, this is a consequence of the thematic rather than chronological structure of
his book - see, e.g., p. 105). The social protest of May would be a major political object lesson for
the PSUC and ERC that, in wartime, some level of economic interventionism was required to
protect liberal social order. The PSUC’s supply policy too, presented in full at the third congress
of the Catalan UGT in November 1937, surely owes something to the experience of May 1937:
Bricall, p. 155. Bricall’s own data implicidy endorses this interpretation: see pp. 107-8.
66
Cruells, Mayo sangriento, pp. 48-9; Bolloten, SCW, 431, pp. 869-70. Bolloten makes much
p.
of Rodriguez Salas’ PSUC membership and Aiguader’s contacts with the PSUC. Doubdess
the PSUC’s general political abrasiveness appealed more to Aiguader - for reasons of his own
style and personality - than it did to his other Catalanist cabinet colleagues. But Bolloten’s
 ,

The Barcelona May days 2 69

Companys issued a press communique to the effect that it would be nec-

5
67
essary to ‘clear the streets ,
a sentiment which echoed the Esquerra’s
stock-in-trade denunciations of the dangerous classes threatening the
good order and livelihoods of bourgeois Barcelona. Clearly both the
ERC and PSUC, as the driving forces in the cabinet which had been
in session throughout the evening, wanted to force the issue over the
telephone control committee as part of their onward consolidation of
government power. However, it also seems likely that they did not yet
realise what this would involve. Lulled into a false sense of security by
the CNT leadership’s quiescence, Companys expected at most to have to
engage in political arbitration with CNT representatives after the event
—
in spite of the fact that the police action breached the agreement en-
shrined in the February collectivisation decree. No one in the Catalan
cabinet had reckoned with the force of the ensuing popular explosion
on 4 May. This saw the Generalitat and other government buildings
armoured and barricaded against attack, while whole neighbourhoods
of the city rapidly became no-go areas for the police.
The capital’s industrial satellites were solidly behind the CNT, 68 as
were the worker districts of the industrial periphery and neighbour-
hoods such as Sants, San Marti and Gracia, situated just around the
perimeter of the Esquerra’s fief, the predominantly middle-class Eix-
ample. Although violent incidents occurred in some of the peripheral
areas, there was no serious challenge here to the worker patrols. 69 Police

own research findings indicate how all the Catalan cabinet, minus the CNT, was supportive of
Aiguader’s objectives. (The criticisms made by Companys’ cautious political lieutenant, Josep
Tarradellas, were purely about tactical matters: Azana, Obras completas, vol. 4, pp. 577-8 (Cuaderno

de la Pobleta, entry for 20 May (1937).) In the end, however, the speculation over who in the
cabinet knew in advance about the Telefonica initiative is not particularly meaningful. Aiguader
may have leaked information to the PSUC some days before the police assault: Fraser, Blood of
Spain p. 377.
67
‘There are armed groups on the streets ... we have no option but to clear them’: Cruells, Mayo
sangriento, p. 58.
68
Such as Granollers, Mataro, Terrassa and Sabadell. Also Badalona - a small city to the north of
Barcelona - and Hospitalet de Llobregat, a major immigrant quarter to south of capital. POUM
strength was in the urban centres of Barcelona province - but not in or around the capital itself.
69 Fraser, Blood ofSpain, p. 378. In spite of government decrees, the worker patrols went on function-
ing throughout the May events. It was only after they were over, with responsibility for public
order in Catalonia transferred to the central Republican government, that the worker patrols
were wound down in June: Cruells, Mayo sangriento, p. 77; Sole i Sabate and Villaroya Font, La i

repressio a la reraguarda de Catalunya, p. 205. There were also cases of local loyalties overriding the

dominant political divisions - for example patrol members belonging to state-identified repub-
lican parties, in particular the ERC, withdrawing at the start of May to avoid any potential
confrontation with cenetistas who were also neighbours and sometimes even friends. I am grateful
to Chris Ealham for this information. Fraser, Blood of Spain, p. 294 (n. 1) gives an example of how
some working relationships continued post May days even across the POUM-PSUC divide.
270 The Spanish Republic at war

action was focused on the city centre as the hub of political power. But
the peripheral neighbourhoods were also perhaps left alone for fear of
the resulting violence spilling over into the Eixample, which was sur-
70
rounded by hostile territory. It is certainly the case that throughout
the May street protests the areas of uncontested CNT strength virtually
71
encircled the city.

On 4 May there were violent confrontations across the

city. There

were lengthy cabinet deliberations with the CNT’s national and regional
leaderships. Also present were Garcia Oliver, CNT justice minister in the
central government, and Pascual Tomas and Carlos Hernandez Zancajo,
representing the UGT’s national executive committee. Both were close
to Largo Caballero and keen to find a resolution to the Catalan crisis
that did not unduly undercut the CNT leaders’ political position. For
Largo and his supporters were now looking to the CNT for backing in
their own deepening conflict with the Spanish Communist Party over
organisational control of the UGT throughout the Republican zone. 72
Nevertheless, the UGT leaders also represented the wishes of the
Valencia government, whose overrriding objective remained, necessar-
ily, a ceasefire. By the evening of 4 May, Companys was amenable to
compromise he had refused twenty-four hours
discussing the political
earlier. Companys was effectively recognising
In changing his stance,
the unforeseen gravity of the situation on the streets. For this was now
posing a direct threat to Catalan autonomy.
In his dealings with the central Republican government, Companys
had since July 1936 assiduously cultivated the picture of Esquerra pop-
ulism as a vital device for controlling the CNT. This was intended to
keep at bay any attempt by ‘Madrid’ to recoup the defacto expansion of
Catalan statute powers - especially in regard to the army and finance -
which had occurred wake of the military rising. Inside the Gen-
in the
eralitat too, although the Esquerra had mainly sided with the PSUC,
Companys had nevertheless wanted to keep the CNT in the game as a

70 Although a proportion of the Eixample ’s mansions lay empty after 18 July 1936 - their occupants
never having returned from their summer residences (in the Pyrenean zone or southern France)
where the rebellion had surprised them - the neighbourhood was still home to a substantial
number of lower-middle-class households.
71
See plan of Barcelona, map 3; Josep Costa of the Badalona CNT commented that the CNT had
Barcelona surrounded: Fraser, Blood ofSpain, p. 380; Bolloten, SCW, p. 432 (though he mistakenly
includes Sarria).
72
Carlos Hernandez Zancajo led the UGT’s urban transport federation, whose Catalan contingent
had virtually ceded from the national federation because it supported the PCE. For the wartime
intra-union dispute, see Graham, Socialism and War, pp. 167-218.
 3

The Barcelona May days 271

counter- weight, 7 in spite of the difficulties this posed, 74 in order to en-

sure Esquerra hegemony. In particular, Companys was concerned not to
let the centralising sectors inside the PSUC gain the upper hand for fear
that they would permit Catalan economic resources to pass under the
control of the Valencian government - as the logic of mando unico dictated.
But now, faced with the May days, the central government repeat-
edly refused Companys’ requests for police reinforcements, unless he
also surrendered the Generalitat’s control over public order and military
affairs in the region. The PSOE, republicans and communists in the

Valencia government were all unsympathetic to Catalan particularism

and wanted a public-order crack-down. But Largo himself was increas-
ingly aware of the strategic importance of the CNT given his growing
isolation in the cabinet. As a veteran populist union leader with a rad-
ical reputation, moreover, Largo also wanted to avoid an outcome that

involved his dispatching police to fire on workers.

But by the end of 4 May it was too late for Companys to avert the
acceleration of violent confrontation. The CNT leadership - national
and regional - desperately sought to broker a ceasefire on the basis of the
cabinet compromise. But they were, for a time at least, swept aside by the
sheer force of what was happening down on the streets. When Garcia
Oliver, once the strongman of Barcelona’s July days, broadcast his appeal
for a ceasefire from the Generalitat on the night of the 4th, many cenetistas
hearing him from the other side of the barricades were convinced that he
had been taken hostage. Soon incredulity gave way to a dominant mood
of embitter ment at what was interpreted as the leadership’s betrayal of
core anarchist values. In particular, Garcia Oliver’s attempt to claim as
his brothers all those who had died, on whichever side of the barricades,
was instandy scorned with a savage and sardonic amazement that has
reverberated in the memoirs of the revolutionary left ever since:

Garda Oliver ... in a speech which was teeth-grating to the poumistas and cenetisas
who heard it told, in a voice charged with pathos, how he had arrived at his
. . .

beloved city to find a dead whereupon he knelt and

cenetista lying in the street,
laid a kiss upon which he did the same for a dead poumista,
his forehead, after
and then, ‘crossing over the road he espied another body in police uniform, so
kneeling again, overcome with emotion, he kissed this one too And, so he . . .
’

73 For example, as late as April Tarradellas defended the CNT’s record in administering the war
industries: Morrow, Revolution and Counter-Revolution , p. 77 (this was not, of course, an endorsement
of socialisation itself).
74 It was a difficult balancing act for Companys since the Esquerra ’s membership was, as a whole,
more conservative than he was.
272 The Spanish Republic at war

claimed, he went on, planting kisses willy-nilly on whatever corpses he came

across ... In factlie did no
dead were in the mortuary, and, sufch thing, as all the
anyway, Garcia Oliver took very good care not to stop on his car journey to the
Generalitat - and he was also very careful to avoid the city centre 75 .

In the black humour of the barricades, Garcia Oliver’s speech was chris-
tened ‘The Legend of the Kiss’ after a famous light opera.
mounting anger and incomprehension among their
But, in spite of
Catalan cadres, the CNT-FAI national and regional leaderships repeat-
edly refused to sanction armed action of any kind in Barcelona. The fact
that this included the FAI as much as the CNT, with its treintista heritage,
indicates the enormous impact of the war on the political consciousness
of anarcho-syndicalist leaders. Nor can this be satisfactorily explained
as the corrupting effect of political power, as the consolatory but also
highly reductionist post-war anarchist history tells it. More than any-
thing, people like Garcia Oliver held back in May because they saw the
bigger picture, not only — or even necessarily primarily — in terms of the
overriding imperatives of the war against Franco, but also in terms of
Beyond Catalonia
the overall balance of firepower within Republican Spain.
and Aragon, the CNT had always been a politically subordinate force 76 .

This was even more evident in the war period since the military rebels, in
rapidly conquering western Andalusia (including Seville) and the urban
centres of western Aragon, had deprived the anarcho-syndicalists of key
strongholds.
The CNT could certainly have ‘taken out’ the state in urban Catalonia.
But holding Catalonia as a whole would have required calling upon their
troops from the Aragon front. Moreover, either courses of action would
have brought them up against the central Republican government. Its
more powerful propaganda machine and greater media access could

75 ‘Garcia Oliver . .
.
[en] una alocucion que hizo rechinar los dientes a los del POUM y a los
cenetistas . . . en tono patetico, dijo que al llegar a su querida ciudad y encontrar en la calle un
muerto de la CNT, se arrodillo y le beso en la frente, despues hizo lo mismo con otro muerto
poumista, y cuando “cruzado en la acera vio un muerto con el uniforme de guardia, se arrodillo,
” Y asi, segun el, fue besando a diestro
emocionado, y lo beso . . .
y siniestro, a tantos muertos
como iba encontrando . . . Claro es que no beso a muerto alguno, porque todos estaban en el

deposito de cadaveres, y ademas, en su marcha hacia la Generalitat, “su coche” no para ni un

minuto y se guardo muy de pasar por el centra de la ciudad.’ A. Bueso, Recuerdos de un cenetista
(2 vols., Barcelona: Ariel, 1976, 1978), vol. 2, p. 243. (Adolfo Bueso was a member of the BOC
in the early 1930s and in the CNT printers’ union. Because of the importance he attributed to
political action he sided with the treintistas when the split occurred.) Fraser, Blood of Spain p. 379.
,

76 Even within Catalonia the CNT was now ‘in conflict with all organisations comprising the other
social layers’, as Helmut Ruediger, vice-secretary of the AIT (Anarchist International) and present
in Barcelona during the events, indicated: Souchy et al., The May Days. Barcelona igjy (London:
Freedom Press, 1987), pp. 71-2.
 The Barcelona May days 273

easily have portrayed the CNT as the betrayers of the Spanish prole-
tariat who had
turned their backs on the war. There was inevitably a
lack of knowledge and understanding elsewhere of what was occurring
in Barcelona - even the POUM’s small sections in Madrid and Valencia
were uneasy77 - and the fact that Barcelona was the Republican city
farthest from the battle front was a gift to hostile propaganda. More-
over, the Republican government was already poised to intervene: had it
been faced with an all-out CNT challenge, it would surely have drafted
in far greater numbers of troops and police to take on ‘revolutionary
5
Barcelona Otherwise it could not have guaranteed the Aragon front or
.

retained liberal state control over Catalonia’s war industries - even more
essential now as Basque industry came under massive rebel attack in the
north. The Republic itself might well not have survived such a massive
CNT would
escalation of armed internecine conflict, but, either way, the
certainly have gone down in the blood bath.
Thus the CNT-controlled anti-aircraft guns on Montjuic hill, trained
on the government buildings below, remained silent. The armoured-car
attacks on government buildings occurred only as sporadic and uncoor-
dinated attempts by individual groups of CNT resisters and were easily
repelled. The CNT’s most seasoned and best-equipped fighters, the five
hundred or so men left from the Durruti column (now commanded by
Ricardo Sanz), which had fought on the Madrid front, were instructed by
Garcia Oliver to obey the orders dispatching them to Aragon. 78 Those
CNT militia members on the Aragon front who had shown a willingness
to come to the defence of their comrades in Barcelona were ordered to
remain at the front. The POUM
leadership in Barcelona also sent the
same instructions to its divisions.
What transpired on the Aragon front itself is, however, less clear. The
evidence that exists is fragmentary and at times contradictory. 79 Bitterly

77 Fraser, Blood of Spain p. 384 , (

n. 1). The Valencian POUM was unambiguously Popular Frontist
and therefore tended to be criticised by Nin’s sector in the party.
78 Ibid.,
p. 380. Sanz was one of Fraser’s interviewees. Sanz’s own memoirs, however, make no
specific reference either to his own role in May 1937 or to the situation on the Aragon front: Los
queJuimos a Madrid pp. 137-46.
,

79 Coll and Pane, Josep Rovira. Una vida al servei de Catalunya i del socialisme (Barcelona: Ariel,
J. J.
i 978), pp. 163—75, esp. 17 1
-3; for Rovira ’s own trial testimony and proceedings (he was the head
of the POUM’s 29 th Division on the Aragon front) see Alba and Ardevol, El proceso del RO.UM ,

pp. 493-529; Broue and Temime, The Revolution and the Civil War in Spain, p. 284: Bueso, Recuerdos
de un cenetista, vol. 2, p. 246; Bolloten, SCW, pp. 452, 875 provides a summary of the available
sources. To my reading, official republican, Communist Party and POUM-friendly reports do not
absolutely contradict each other. But sources fav ourable to the government side - and therefore
keen to justify the ensuing repression - suggest that far greater numbers of men abandoned the
front, with the express intention of going all the w ay to Barcelona.
274 77z £ Spanish Republic at war

disappointed and confused though many cenetistas and poumistas were,

remained intact - in spite of the inactive state
overall, military discipline
of the front probably adding to the temptation. A group of CNT/POUM
troops did get part of the way to Lleida. But they returned to the front
once they were assured that government forces would desist from be-
sieging the CNT and POUM
headquarters there — a pattern of attack
being replicated by pro-government forces in a number of other urban
centres in Catalonia.
In Barcelona, however, all CNT action remained defensive. This situa-
tion was the despair of the POUM’s Catalan leadership - and specifically
of Andreu Nin, who during 3-4 May argued to no avail with the CNT’s
regional leaders that they must support the cause of their workers on
the barricades. When it was clear that they would not, however, Nin
too pulled POUM militants back from the brink, refusing to sanction
joint armed action with CNT cadres, even though in some places joint
80
defence and neighbourhood committees already existed. Neither the
POUM nor, still less, the Spanish Bolshevik Leninist groupuscule
81
or-
biting the party nor the handful of radical anarchist activists who made
up the Friends of Durruti (Amigos de Durruti) 82 had the organisational
purchase to intervene in a politically decisive way in the May events.
Indeed, it could be argued that the POUM’s published proclamations of
and its public articulation
ideological support for those at the barricades
of what the May days meant functioned in some ways as a compensation
for the party’s relative political marginality.
The POUM’s intention in calling upon the CNT to back their work-
ers at the barricades was to strengthen the resisters’ hand in subsequent
negotiations with the Generalitat. 83 Nin and his executive colleagues
were as aware as Garcia Oliver that the overall balance of forces within
Republican Spain was against those at the barricades. For this reason,
the POUM leadership, like the CNT’s, rejected outright the Friends
of Durruti’s published manifesto of 5 May, which attempted to rally

80
The POUM youth wing, JCI, was the most prominent sector in favour of joint action with the
CNT cadres: see JCI leader Wilebaldo Solano’s testimony in Fraser, Blood of Spain, pp. 380-1.
81
Estimated at between 10 and 30 individuals: Bolloten, SCW, p. 860 (n. 11).
82
This was a splinter group formed by those from the Durruti militia column who refused to be
incorporated in the Republican Army. For a summary, see Fraser, Blood of Spain p. 381 (n. 1). ,

Also Bolloten, SCW, pp. 420, 866 (n. 49); Casanova, Anarquismoy revolucion, pp. 245-7; Semprun-
Maura, Revolucion y contrarrevolucion en Cataluna, pp. 249-50, 273; Durgan, ‘Trotsky, the POUM
and the Spanish Revolution’, p. 59.
83 Defensive resistance was the intent behind all the POUM’s May manifestos: see the material
cited by Semprun-Maura, Revoluciony contrarrevolucion en Cataluna, pp. 275-6.
 ,

The Barcelona May days 275

resistancebehind a Revolutionary Junta. 84 But Nin still sought to con-

vince the CNT of the need for unified resistance in Barcelona as a bar-
gaining counter in order to prevent the unleashing of a blanket repression
afterwards. 85
The CNT leadership was, however, reluctant to sanction even this
form of tactical resistance. The fact that it did not could certainly be
counted a serious strategic error, for it exposed not only the POUM but
also the CNT’s cadres to the full blast of state political repression from
7 May onwards. But it is vital to remember the intensity of the pressures
deriving from the war situation.
Since 18 July 1936 these had also reinforced the (pre-war) ‘politicising’
currents inside the anarcho-syndicalist movement led by the defacto gen-
eral secretary Horacio Prieto. 86
saw the CNT’s incorporation into
(Prieto
government as a logical progression of its role in the emergency structures
of committee power. 87 And he and others viewed the Council of Aragon
88
as performing classic governmental functions. ) The year 1937 also saw
a series of internal organisational changes in the CNT designed to erode
the old confederal autonomy - for many synonymous with the CNT -
and to centralise power in the hands of the national committee. 89 This
cracked down on the FAI’s direct-action groups, excluding them as such
from the organisation. A much tighter editorial control of the confederal
84 Cruells,Mayo sangriento, p. 70; Broue and Temime, The Revolution and the Civil War in Spain p. 284;
Semprun Maura, Revolucion y contrarrevolucion en Cataluha, pp. 281-2 - who wrongly dates the
manifesto as 6 May. According to Juan Andrade, Nin’s executive colleague, Spanish Trotskyists
(i.e. the diminutive group of Bolshevik-Leninists who belonged to the Fourth International) who

were present in Barcelona in May also agreed with the POUM’s assessment of the situation - in
spite of the triumphalism of their public manifestos: Fraser, Blood of Spain, p. 388 and Morrow,
Revolution and Counter-Revolution in Spain, pp. 143-4. For a generally more orthodox Trotskyist
assessment of the genealogy of revolutionary failure/the May events, see analyses by H. D.
Freund andj. Rous in ‘The Spanish Civil War. The View From the Left’, pp. 317-28, 345-402.
85 Testimonies of Solano and Andrade in Fraser, Blood of Spain, pp. 381-2; Bolloten, SCW, p. 435.
86
Prietohad resigned as general secretary in November 1936 after the CNT ministers had acccepted
the government’s departure from Madrid (see chapter 3 above), but he continued to advise the
acting general secretary, Mariano R. Vazquez.
87
Lorenzo, Los anarquistasy el poder, pp. 178—9; Prieto, El anarquismo espahol en la lucha politico, p. 11.
88
J. Gomez Casas, Los anarquistas en el Gobierno 1936-39 (Barcelona: Bruguera 1977), pp. 128, 148-
9; although the national committee were fundamentally unfavourable to the uncompromising
political direction taken by the Council and ultimately refused to defend it against government
intervention in summer 1937: Casanova, De la calle alfrente, p. 232.
89 Semprun-Maura, Revolucion y contrarrevolucion en Cataluha, pp. 242-52. For a sharp analysis of
internal renovation in the CNT, see J. Casanova, Anarquismo y revolucion, pp. 144-50 and De la

calle alfrente, p. 227. Casanova indicates a decrease in the two-way flow of initiatives inside the
CNT after 18 July 1936 even though the leadership cadres remained largely unchanged. This is

unsurprising as this internal battle over organisational forms in the CNT-FAI predates the war:
see chapter 1 above.
276 The Spanish Republic at war

press was implemented - over and above the norms of wartime censor-
90
ship. Bureaucratic controls were imposed. None of this occurred simply
because the CNT was collaborating with the Popular Front alliance, but
that collaboration accelerated the process of CNT centralisation, which
was, in some ways, a form of political modernisation. 91
In none of its armed uprisings across the Republican years since 1931
had the FAI ever won out against the forces of the state. In some ways it
seemed as if the Barcelona events were finally illuminating that pattern of
defeat. Certainly Garcia Oliver’s plea to his comrades on the barricades
‘not to cultivate the mystique of the dead hero’ seems haunted by such an
awareness. 92 The escalation from military coup to full-scale civil war had
widened the fault line in the CNT until by May 1937 the organisation
was itself divided by the barricades. The Barcelona May days in effect
constituted the CNT’s own ‘crisis of modernity’. 93
A transformation certainly occurred in the representation of the
CNT’s dead across the first ten months of the war. Gone by May 1937 was
the cult glorifying the fallen warrior and martyr, such as that which grew
up around the figure of the veteran anarchist leader Francisco Ascaso,
killed in the assault on Barcelona’s Atarazanas barracks in the July days
or that which suffused the collective narrative of the Iron Column. 94 Or,
supremely, the mythologising, quasi-religious aura and the exhortation to

90 S. Tavera and E. Ucelay da Cal, ‘Grupos de afinidad, disciplina belica y periodismo libertario
i93&- i 938’, Historia Contemporanea, 9 (1993), 167-90, esp. 177, 185.
91
These changes culminated in the decisions of the FAI plenum ofJuly 1937. Pardy this was about
legalising the FAI so it could participate in Republican political institutions. But the process of
change went deeper: Lorenzo, Los anarquistas espaholes, pp. 228-9; Peirats, La CNT en la revolucion
espahola, vol. 2, pp. 241-54; Jellinek, The Civil War in Spain, p. 571-2.
92 ‘No cultiveis en este momento el culto a los muertos.’ The speech, broadcast on the night of
4 May, is cited by many authors, including Cruells, Mayo sangriento, p. 63 and Semprun-Maura,
Revolucion y contrarrevolucion en Cataluha, p. 267.
93 As we have seen earlier, syndical reformers in the CNT had been driven, even before the 1930s,

by a desire to find some way out of the increasingly unequal (and therefore erosive) confrontations
between radical workers/anarcho-syndicalist cadres and the state. (For even in Spain the relative
increase in the technological/ operational sophistication of the security forces had opened up a
distance between them and their anarcho-syndicalist opponents in the social war.) The CNT
leadership’s public insistence that the May events had resulted from ‘foreign’ influence (see, for
example, Cattell, Communism and the Spanish Civil War, p. 145) also needs to be understood in
this context - as a means of deflecting attention and responsibility from these racking internal
conflicts (both within the CNT and between it and other Spanish organisations).
94 ‘Iron Column. Impassioned fighters for the Idea. Hearts of fire in the service of liberty. Visionaries,
whose hopes are set on a shining, humane tomorrow. Their flesh has suffered the prisons of
Reaction. They are the sons of those ones murdered by the Civil Guard with their “Ley de
Fugas” (shot in the back “while attempting to escape”). Martyrs, martyrs, martyrs . .Ascetics
.

with no divine master. Anarchists: victory is yours!’, from the Iron Column’s newspaper, Linea de
Fuego, 24 Oct. 1936, cited in Casanova, Anarquismoy revolucion, p. 113.
 The Barcelona May days 277

emulatory mourning95 evident at the funeral of Buenaventura Durruti,

who had died on the Madrid front in November 1936:

They embalmed the body, and put it on show, and even now one can look
through an opening into the tombs and see their leader sleeping under
glass They had brought him back from the Madrid front so that the An-
. . .

archists could look at his wounded body and decide by what treachery he had
been killed. It was too difficult for them to admit that he had been shot like any
ordinary man. 96

Durruti had been the comrade-in-arms of both Garcia Oliver and Ascaso
in the brutal Barcelona labour wars of the 1920s. The ‘three musketeers’
of popular legend, Durruti, Ascaso and Garcia Oliver had created the
most famous of all the anarchist direct-action groups, Los Solidarios, to
confront the hired gunmen of the monarchist state. But in spite of the
stories of fascist or communist snipers and internal betrayal, in Madrid
in 1936 there was neither conspiracy nor martyrdom: if Durruti’s end
symbolised anything, then it was the brutal happenstance of death in
war. 97
The CNT’s reportage on leaders, such as Domingo Ascaso (the brother
of Francisco), who died in the course of the May events in Barcelona,
was, in contrast, deliberately low key. 98 No longer were they martyrs to
the cause. The CNT’s supporters were now being exhorted to respect
other types of leader: the politician and office-holder rather than the
street fighter or radical egalitarian. But, in fact, even though Durruti
was perceived by many as the latter two, the call to emulate him was
already, by November 1936, an ambiguous one. The transformation of
his image represented the anarcho-syndicalist movement’s coming to
terms with the war. Garcia Oliver as minister also underwent a notable

95 ‘imitad al heroe del pueblo’ (‘emulate the hero of the people’), Abella, La uida cotidiana durante la
guerra civil, pp. 163-8.
96 Low and Brea, Red Spanish Notebook, pp. 215, 216.
97 For a resume of earlier theories on Durruti’s death, see Casanova, Anarquismoy revolucion, p. 251
and (more extensively but very unclearly) in Paz, Durruti en la revolucion espahola, pp. 689-714.
On Durruti’s death through the accidental discharge of a weapon at close quarters, the forensic
opinion of Dr Santamaria, Jefe de Sanidad of the Durruti Column (who carried out the post-
mortem), appears compelling. This is cited in J. Arnal, lo jui secretario de Durruti. Memorias de un
cura aragones en lasjilas anarquistas (Zaragoza: Mira Editores, 1995), p. 119. Even though the CNT
leader Federica Montseny was still referring to the unexplained nature of Durruti’s demise in
her (extremely bland) memoirs of 1987, Arnal (p. 201) claims that she had verbally confirmed
the truth to a Spanish journalist well before their publication. See also testimony of Clemente
Cuyas, in ‘Asi murio Durruti’, El Pais, 11 July 1993. Paz’s ostensibly extensive analysis, Durruti en

la revolucion espanola , pp. 715-24 sheds litde fight.

98 Cf.
Casanova, Anarquismoy revolucion, p. 252.
27 8 The Spanish Republic at war

reconstruction. He, unlike Francisco Ascaso and Durruti, had outlived

the time when ‘heroes’ could credibly take on the state single-handed."
Anarchist iconography serves thus as a cultural ‘barometer’ registering
100
the expansion of state political authority .

Yet we cannot simply point to the CNT’s leaders as the conscious

‘authors’ of these changes. Certainly for some the war had reinforced
pre-existing views in favour of modifying anarcho-syndicalist practice to
allow the incorporation of the CNT within parliamentary politics. But
for many more - although they did not consciously moot it, still less ar-
ticulate it in public - the war’s overwhelming practical imperatives had
greatly problematised ideological resistance to centralised forms of or-
ganisation. Yet most of these resources remained in liberal hands. This,
plus the limited capacity of CNT organisational forms to integrate and
centralise, saw the force of attraction exerted by the liberal state over
anarcho-syndicalist leaders increase as the war itself escalated. The very
real needs of the war saw both CNT and FAI leaders increasingly
effort
incorporated into the governing machinery of the liberal state, leaving
isolated and uncomprehending sectors of their own cadres and social
base whose daily experience led them to continue to resist its encroach-
ment.
The Barcelona had no organisational means of coordinating
radicals
the fight back from the CNT’s grass-roots committees - decapitated by
political developments inside the Generalitat. Above all, the historic rad-
ical anarchist hostility to the CNT
formation of industrial unions in the
would now come back to haunt them. In the spring of 1937 the CNT was
101
still in the early stages of transforming its unions into industrial ones.
If such structures had existed, then the battle for industrial production
which had raged across the months since the military coup
in the city
might have been harder for the government to win. Their lack points up
the flaw in explanations of anarchist failure that posit a binary choice:
‘the revolution or the war’. But radical anarchists could not ‘make the
revolution’ for the same reasons
that they could not take the lead in wag-
ing the war: they lacked the requisite structures in which to coordinate it.

99 Consider also the fact that the CNT

national committee would end by expelling the Council of
Aragon’s leader, the veteran radical Joaquin Ascaso (cousin of Francisco and Domingo), from
the organisation in September 1938: Casanova, De la calle alfrente, p. 233.
100
We should also note that in April 1938 the Republic would posthumously confer on Durruti the
rank of Lieutenant Colonel: Paz, Durruti en la revolucion espanola, p. 726.
It was February 1937 before industrial union structures even existed in the CNT in some crucial
101

war industries - such as metalurgia, Vega et al., ‘Socializacion y hechos de mayo’, p. 100 (n. 6);
Fraser, Blood of Spain, p. 544 (n. 3).
 The Barcelona May days 279

The CNT-FAI’s identification with localist popular resistance needs,

thus, to be problematised rather more than it has been. Such an identi-
fication undoubtedly indicates the anarcho-syndicalists’ superior under-
standing of the dynamics of popular mobilisation - of its basis in local
imperatives and community goals (whether in terms of a factory, urban
neighbourhood or village) and of its value as a liberating, socially and
culturally transformative process - as compared to the Popular Frontist
concentration on end result (mobilisation for the war effort). But CNT
radicals’ identification withcommunity rather than state and their heroi-
sation of the popular defence obscure serious ideological and strategic
contradictions: the internal division in the anarcho-syndicalist movement
threatening since 1931 had by May 1937 opened into an abyss.
By 5 May the tide had begun to turn against the resisters. The
Telefonica workers surrendered to the besieging police, and those be-
hind the barricades, bereft of orders from their leaders - other than one
102 -
urging them to ‘return to work’ remained confined to the defen-
sive as their positions crumbled. Beyond Barcelona too, police and pro-
government emboldened by the anarcho-syndicalist stalemate in
forces,
the capital, made concerted assaults on CNT and POUM premises and
CNT-held telephone exchange buildings in a number of towns across
Catalonia, including, most notably, Tarragona and Tortosa. 103 The strik-
ing similarity of this police action in both timing and pattern of attack -
-
suggests a common set of orders emanating from government authorities
104
in Barcelona.
Midday on 5 May saw the formation of an emergency four-man cab-
inet of Esquerra/Rabassaire, CNT and UGT representatives which ex-
cluded all those who had previously occupied ministerial posts. 105 The
absence of any representative with a named PSUC affiliation might be
interpreted as an attempt at cosmetic conciliation. But the new cabinet’s
lineon public order was as uncompromising as that expressed by Com-
panys on the eve of the conflict on 3 May. The central government in
Valencia confirmed its assumption of Catalan public order shortly after
the proclamation of the new Generalitat. The urgency was increased by

102
Broadcast by radio and ‘signed’ by the local union federations of both the CNT and UGT:
Semprun-Maura, Revolution y contrarrevolucion en Cataluna, p. 277.
103
Cruells, Mayo sangriento, pp. 86-90; B. Bolloten summarises a number of contemporary press
and other sources, SCW, pp. 452, 875 (n. 19); Semprun-Maura, Revolution y contrarrevolucion en
Cataluna, pp. 284, 287-9; Souchy, The May Days, pp. 95-100.
104
Semprun-Maura, Revolution y contrarrevolucion, p. 287; Cruells, Mayo sangriento, p. 90.
105
Bolloten, SCW, p. 441. This comprised Esquerra (Marti Feced), CNT (Mas), UGT (Sese) and
Rabassaires (Pou): Cruells, Mayo sangriento, p. 67.
28 o The Spanish Republic at war

the fact that President Azana, traumatised by his experience of being

trapped by the street fighting in the Catalan parliament building in cen-
106
tral Barcelona, was threatening to resign. This had to be avoided at all
costs since would have obliterated the Republic’s democratic credibility
it

at the very moment when

it was intensifying diplomatic efforts in an at-

tempt to procure the lifting of Non-Intervention. Azana was desperate -

as indicated by his suggestion to his close friend Prieto, the navy and air
minister, that he should order the Republican airforce to bomb clear the
short route from his residence to the port. This would have been like
dousing the May fires with petrol. But it took Prieto and other leading
republican figures to spell out the implications of the president’s impris-
onment Largo Caballero, whose personal dislike of Azana had been
to
fed by the preceding months of prickly interchange between the two.
Indeed, quite staggeringly, Largo did not see fit to communicate person-
ally with the President of the Republic throughout the whole of the May

It was Prieto who arranged, on his own ministerial responsibil-

107
crisis.

ity, the sending of two Republican warships to Barcelona. These docked

on 5 May, with orders to assist the Generalitat, but, first and foremost,
108
to evacuate the president. The central government also announced
the dispatch of militarised police units to Barcelona. But even as late as

5 May, Companys was hoping that a rapid, determined push on the part
of what were, for a short while at least, still Generalitat-controlled police
forces inside Barcelona might bring the situation sufficiently under con-
trol for Valencia to reconsider its decision. Unfortunately, the violence
on the streets exploded again, this time entirely shattering Companys’
coping strategy.
On the afternoon of 5 May, Antoni Sese, front-rank leader of the
CatalanUGT and new Generalitat minister, was shot dead outside a
CNT union building as he was being driven to assume his governmental
responsibilities. It is not possible to say for certain who killed Sese or why.
106
Zugazagoitia, Guerray vicisitudes, p. 268. The Catalan parliament building was situated in the old

Ciutadella arsenal: M. Vazquez Montalban, Barcelonas (London: Verso, 1992), p. 127; Azana’s
diary account of the May events is in the ‘Cuaderno de la Pobleta’ (20 May 1937), Obras com-
pletas vol. 4, pp.
, 575-88. The telegraphic recordings of Azana’s communications with Valencia
between 4 and 6 May are in the Servicio Historico Militar (Madrid) legajo 461.
107
S. Julia, ‘Presidente por ultima vez: Azana en la crisis de mayo de 1937’, in A. Alted, A. Egido

and M. F. Mancebo (eds.), Manuel Azana: pensamiento y accion (Madrid: Alianza Universidad,
1996), pp. 249-50. (Although the telephone lines were down, the telegraph system functioned
throughout.)
108
Bolloten,SCW, pp. 448-9; Cruells, Mayo sangriento, p. 71. Prieto was disparaging about Azana’s
preference for days of dread to a few moments’ resolution (i.e. making the short trip from the
palace to the port along the Paseo de Francia): Zugazagoitia, Guerray vicisitudes pp. 268-9. ,

Azana finally left for Valencia by air on 7 May: Vidarte, Todosfuimos culpables, vol. 2, p. 666.
 The Barcelona May days 28 i

Of all the hypotheses which have circulated, the least likely would still
seem to be that he was killed, in a kind of doomsday scenario, by someone
connected to the ultra-centralist current in the PSUC in order to force
a central government crack-down on Catalan autonomy. Even if one
accepts the rather too conspiratorial notion that Sese’s dissident com-
munist past (he had been a member of the BOC in the pre-war period)
made him ‘expendable’ in the eyes of centralist PSUC sectors, such a
strategy still entailed enormous
which could have sent the situation
risks

in Barcelona spinning completely out of control. But whatever the source

of the bullet which killed Sese - and it could, conceivably, have been a
109 -
stray, like the one which killed Domingo Ascaso the circumstances
of Sese’s death were sufficiently ambiguous, and the atmosphere of re-
ciprocal mistrust generated by days of inter-organisational bloodletting
so absolute, for the accusation that he was shot by an anarchist sniper to
110
be believed .

Probably more than any other single incident, it was Sese’s death which
precipitated the political intervention of the central government. In a
context where the continuation of violent street confrontation signalled
the continuing jeopardy of state authority, the symbolic significance of a
minister’s on either the Catalan cabinet or Valencia.
demise was not lost

The Generalitat was now wide open to the charge that it had failed to
contain a rising tide of disorder which was threatening the Republic’s
very capacity to resist militarily. Valencia’s appointee to the public or-
111
der portfolio, Colonel Antonio Escobar ,
was seriously wounded when
shot at on his arrival in Barcelona. And production in Barcelona’s war

109
Cruells, Mayo sangriento, p. 69.
110
Ibid., pp. (in particular, what Cruells terms ‘the atmosphere of personal extermination’
67-9
which affected all the participants, p. 67); Peirats, La CJYT en la revolucion espanola, vol. 2, p. 151;
Semprun Maura, Revolucion y contrarrevolucion en Cataluna, p. 276; Bolloten, SCW, p. 453, who
quotes PSUC leader Serra Pamies’ 1944 testimony (p. 875, n. 29) on Sese’s political past. (Sese
had stood as a BOC candidate for the Cortes in 1931 - polling some 500 votes.) See also Ucelay
da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, pp. 305-6, 308, and p. 311
(on Sese himself) - and see also chapter 1 above. Anarchist commentator Agustin Souchy also
refers to the possibility of a PSUC assassination in his contemporary (1937) account, published
in Buenos Aires. But mainstream nationalism still held sway in the PSUC (as Bolloten attests,
SCW, p. 448). The involvement of a PSUC ‘fringe’ seems unlikely - although it cannot be
ruled out entirely. Virtually all parties and organisations had their ‘uncontrollables’ - see the
discussion later in this chapter.
111
A cavalry colonel,Escobar had been head of the Civil Guard in pre-war Barcelona and thus
A conservative of profound Catholic faith (appearing
pivotal in the defeat of the military rising.
as Colonel Ximenez in Malraux’s L’Espoir), he would end the war as Commander-in-Chief of
the Army of Extremadura. He supported the coup which displaced the Negrin government (see
chapter 7 below) and was shot in February 1940 by the Franco regime: J. L. Olaizola, La guerra
del General Escobar {Barcelona: Planeta, 1983).
282 The Spanish Republic at war

industrieshad been disrupted by the generalised work stoppage since

4 May. Valencia’s take-over of the public order and defence briefs in
Catalonia meant that the Generalitat lost precisely those functions —
deriving from the 1932 statute - which were most highly charged
in
terms of nationalist identity Companys personally authorised the hand-
112
over, but in the circumstances he had little real choice. It meant the loss

of Catalan control over police and army in the region, but this still left the
Generalitat’s political-administrative and, most importantly, its economic
control intact. Behind Valencia’s partial political intervention there lay
the tacit threat of the complete suspension of the Catalan Statute -
something Companys obviously wanted to avoid at all costs.
would lie in store as the Generalitat sought to defend its
Bitter battles
‘statutory’ sphere of economic influence against a central government
increasingly desperate to centralise economic power as the impact of
Non-Intervention provoked mounting material crisis by late 1937 ." 3 But
bitter though this later jurisdictional conflict was between ‘Madrid’ and
‘Barcelona’, it should not obscure the fact that in May 1937 Companys’
primary objective remained the restoration of liberal political and eco-
nomic control in Catalonia. The 5,000 central government troops and
police who arrived in Barcelona late on Friday 7 May - soon to be rein-
forced by several thousand more 114 - enacted a repression which would
guarantee the very liberal order that the ERC, PSUC and Catalan mid-
dle classes had sought to reconstruct and defend since July 1936. 115 It
was only once this had been secured that Companys began to criticise
central government infringement of Generalitat powers.
With the arrival of police detachments from Valencia, the May days
were effectively over. State repression was, however, only just beginning.
All of the CNT leaders’ painstaking efforts during 6 and 7 May to secure a

112
Mayo sangriento, p. 74; Bolloten, SCW, pp. 437, 451 (also
Cruells, see nn. 15 and 16, p. 874); Broue
and Temime, The Revolution and the Civil War in Spain p. 285.
,

113
See chapter 6 below.
114
Solidarida Obrera,May 1937; Bolloten, citing this and other contemporary press reports, calcu-
9
thousand arrived in total over a few days: SCW, p. 460 (inc. n. 63).
lates that twelve
115
The fact that the new arrivals’ conquering cry was the leitmotiv of the left - jU.H.P! (jUnios
Hermanos Proletarios! (Workers Unite!)) - was an example of the classic process whereby the
legitimising discourse of ‘the revolution’ is assimilated all the better to consolidate a conserva-
tive order. Much same purpose was served by the fact that Valencia’s incoming nominee
the
as Barcelona police chief was Lieutenant Colonel Emilio Torres Iglesias, the former chief of
the anarchist militia column, Tierra y Libertad (Land and Freedom): Cruells, Mayo sangriento,
pp. 80-1; Semprun-Maura, Revoluciony contrarrevolucion en Cataluha, p. 285; Broue and Temime,
The Revolution and the Civil War in Spain, p. 285. Comparative examples of this process might
include the Liberal discourse of post-unification Italy or the French Radical Party’s use of the
symbols of 1789.
 )

The Barcelona May days 283

peace with guarantees’ came to nought. Indeed, even as they negotiated

c

in Barcelona, government troops en route to the city were participating

in acts of violent reprisal which dismembered the anarcho-syndicalist
movement. Symbolic acts - such the burning of the confederal flag -
were rapidly followed by the burning of CNT premises, and soon a
veritable wave of paseos was unleashed in which many cenetistas would be
6
killed."
Companys’ promise would be neither victors nor van-
that there
quished rang hollow as Republican prisons began to fill up with those
held in ‘preventive detention’ [prisioneros gubernativos and thus with little
hope of a trial. 117 On the streets, Valencian troops and Catalan police
behaved like occupiers, routinely demanding identity papers, tearing
up any CNT union cards they found and humiliating their owners.
118

But it was scant revenge, given the dominant tenor of regret among the
police and pro-government forces in general that the barricades had
come down before they had finished the job." 9 It is difficult not to hear
echoing in these sentiments the old attitudes of official Barcelona to the
120
‘rabble’. But the Republican state had to tread a fine line between
punishment and the needs of wartime mobilisation.
Barcelona’s proletariat may have been the beating heart of the barri-
cades, but it was also the crucial centre of the Republic’s war industry.
Factory production was gradually starting up again after 7 May, and
assuring it against further disruption was vital, not only in view of the
external arms embargo but also with the mounting rebel threat to war
production in the north. Thus, while punishment for the May days was

116
Cruells, Mayo sangriento, pp. 86-90; Peirats, La CNT en la revolucion espanola, vol. 2, pp. 157—61;
Semprun-Maura, Revoluciony contrarrevolucion en Catalufia, pp. 284, 287-9; Souchy, The May Days,
p. 98; Sole i Sabate andj. Villarroya i Font, La repressio a la reraguarda de Catalunya, vol. 1, pp. 2 12-16,
esp. pp. 213-14; Bolloten, SCW, p. 452, p. 875 (n. 19).
117
By July 1937 the CNT estimated that 800 of its members were in gaol in Barcelona alone:
Morrow, Revolution and Counter-Revolution in Spain, p.144; E. Goldman, ‘Political Persecution
in Republican Spain’, Spain and the World, 10 Dec. 1937, also cited in Souchy, The May Days,
pp. 104-7; Peirats, La CNT en la revolucion espanola, vol. 2, pp. 263-4; Sole i Sabate and Villarroya
i Font, La repressio a la reraguarda de Catalunya, vol.
1, pp. 2 17-24, 260 and pp. 279-86 on Republican

work camps.
118
Cruells, Mayo sangriento, p. 83; Bolloten, SCW, p. 456. This form of disciplining was fairly indis-
criminate, however, since the Generalitat had decreed compulsory unionisation at the end of
!
936 -

119
Cruells, Mayo sangriento, p. 79.
120
Cf. Comorera’s pre-war comments (as USC leader) on anarchism as the ideology of ‘sub-human’
and ‘degenerate’ individuals and ‘underworld parasites’, Ealham, ‘Policing the Recession’,
p. 354. See also the comments of exiled conservative Catalan nationalists that the Francoist
‘clean-up’ of the radical left in Catalonia in 1939 had saved them a task: R. Abella, Finales de
igjg. Barcelona cambia de piel (Barcelona:Planeta, 1992), p. 50.
enero,
284 The Spanish Republic at war

essential in order to guarantee future labour discipline, this had to be an

exemplary punishment which did not direcdy victimise the CNT since
its cadres still had the potential to disrupt production.
While the Generalitat took the opportunity discreetly to annul its pre-
121
vious decree measures recognising socialised control in industry, the
results of Republican government realpolitik were twofold. First, there was
conciliation with the CNT leadership, whose political subordination in-
creased as divisions in anarcho-syndicalist ranks grew more acute in the
wake of the May debacle. The CNT would never have more than an
emblematic government presence after May 1937 and in fact never took
up their seats in the Generalitat — at root because they could not come
122
to terms with their new political marginality therein. Second, there
was a political scapegoating of the POUM. (Indeed, the CNT leader-
ship’s muted response to this is an important indication of its acceptance
of Popular Frontism in the service of the war effort.
123
)
The POUM’s
124
public (and much published )
identification with those resisting on
the barricades, combined with its relative political marginality, made
it the ideal target for the symbolic function required by the Republican
state pour encourager les autres. On 16 June, in the wake of the fall of
Bilbao, the industrial powerhouse of the Republican north, the POUM
executive committee was arrested.
The fact that the POUM was also the target in internecine wars in
the communist movement facilitated the Republican government’s ob-
jectives by increasing the party’s isolation. But it is important to stress
that these two processes of targeting the POUM had separate agen-
das and, moreover, ones which rapidly came into conflict with each
other. The Republican authorities needed to make an example of the
POUM by bringing the full weight of liberal law and order to bear on
its leaders.
125
But the very basis of liberal legitimacy - the constitution-
ality of the state - was daily being violated in the latter weeks of June

as the Spanish Communist Party, Comintern representatives and some

Soviet police personnel set up private interrogation centres (checas) on
Republican state territory - but beyond the control of its constitutional

121
Morrow, Revolution and Counter-Revolution in Spain ,
p. 62 For the consequences of this on the ground
.

see M. Vilanova, ‘L'Escala y Beuda’, Historiayfuente oral 3 (1990), 53. As regards any assessment
,

of output levels in collectivised industry, there are no meaningful terms of comparison, first,
because Catalonia had not previously produced war materiel and second, because it was being
produced during the war in extraordinarily difficult economic and logistical circumstances.
122 123
Bolloten, SCW, pp. 494-7. Casanova, Anarquismoy revolucion, pp. 251-2.
124
The POUM press, La Batalla, was suspended at the beginning of June.
125
Members of the POUM executive would stand trial in October 1938: see chapter 6 below.
 The Barcelona May days 2 85

authorities - wherein they assaulted and assassinated anti-stalinist dissi-

dents with virtual impunity.
The intention here is not to suggest that Republican order was some-
how less implacable (although it was less lethal, at least in the short
126
term ). Rather it is to indicate that Comintern activity was directly
5

challenging the ‘monopoly of legitimate violence on which state author-

ity depended. This is very well illustrated in the story told by POUM
executive member Juan Andrade of his transfer from Madrid to the state
gaol in Valencia. The justice minister, Manuel Irujo of the PNV, sent a
detachment of Assault Guards whose main purpose was the surveillance
not of the POUM prisoners but of the accompanying communist police-
men in order to ensure that Andrade and his colleagues reached their
127
destination safely.
Andrade’s account raises a crucial further question, however. What
were the respective roles of the Comintern and the Spanish Communist
Party in the repression of the POUM? 128 While the ideological attack
on anti-stalinist dissent was prepared in Moscow, there were simply
not enough Comintern functionaries in Republican Spain to have car-
ried out - or even supervised -a systematic political repression of the
POUM. 129 The priorities and attention of the Soviet leadership were
until 1939 largely absorbed by domestic policy - a focus which was
massively reinforced by their own purges. 130 Undoubtedly Comintern

126
Later, imprisoned members of the POUM were caught between communist revenge and
Francoist punishment as Francoist troops bore down on Barcelona in 1939: J. Gorkin, Canibales
politicos. Hitler y Stalin en Espaha (Mexico, 1941); G. Regler, The Owl of Minerva (London: Rupert
Hart-Davis, 1959), pp. 324-5; Fraser, Blood of Spain, p. 389 (n. 1); Thomas, The Spanish Civil War,
p. 876. On POUM
and other left prisoners, see Goldman, ‘Political Persecution in Republican
Spain’, pp. 105-7; Broue and Temime, The Revolution and the Civil War in Spain, p. 315.
127
Fraser, Blood of Spain, p. 389; on the detentions see also A. Suarez, El proceso contra el POUM
(n.p.: Ruedo Iberico, 1974), pp. 83-7.
128
The POUM’s own criticisms were in 1937 focused on the sectarianism of the PCE, for example
in their letter to Republican president Azana in December 1937: copy in Azana’s personal
archive, AB {apartado 7, caja 137, carpeta 9).
129
As we already know from E. H. Carr’s study, The Twilight of the Comintern, it was a creaky,
ramshackle affair. The war in Spain severely tested its really quite meagre organisational and
personnel resources. As far as Soviet personnel in general were concerned, although we do not
have exact figures, current research estimates only a few dozen, and an absolute maximum of
thirty, high-ranking functionaries for the entire war period (i.e. certainly far fewer at any one time)

in Republican Spain. And while there was a relatively greater presence of Red Army personnel
(c. 3,000 and between 600 and 800 at any one time), they were overwhelmingly absorbed by

military functions (which includes those performed by civil engineers and interpreters). I am
grateful to Frank Schauff for this information.
130
G. Roberts, ‘Soviet Foreign Policy and the Spanish Civil War 1936-1939’, in C. Leitz and D. J.
Dunthorn (eds.), Spain in an International Context 1936-1959 (Oxford/New York: Berghahn Books,
1999), P- 9b Schauff, ‘Hitler and Stalin in the Spanish Civil War’, pp. 19-20.
 ,

286 The Spanish Republic at war

perceptions of the events in Barcelona were coloured by the rarefied

atmosphere In the Soviet Union. (This was borne to Spain by Stepanov,
who had arrived in the wake of the military debacle at Malaga (in

early February 1937) seeking explanations for it in terms of ‘spies and

131
traitors’. )
But, once again, it is important to remember that, whether in
respect of Malaga or the Barcelona events, Stepanov’s accusations found
a reply, in so far as they did at all, only because they chimed with the
desperate domestic circumstances of the Republic and, given these, the
very real and reasonable fear in a civil war of ‘the enemy within’. More-
over, hiscontinuing tirades in 1937 against ‘trotskyism and its helpers’
can themselves also be read as an indication of the Comintern’s frustra-
tion that the multiple and fissiparous political life of the Republic was
not in fact reducible to the ‘redemptive’ two-dimensional prescriptions
to which its reports and advice tended.
In the aftermath of the May days, moreover, Comintern personnel in
Spain seem notably to have targeted their aggression on foreign dissidents
(often themselves exiles). Why this was so lies beyond the scope of this
132

book. However, as epoques have guiding mind-sets, and inter-war

all

Europe’s derived heavily from forms of social darwinism, it does not seem
too speculative to suggest that the imperative within Stalinism to ‘cleanse’
the heterodox from the Comintern was akin in many respects to other
contemporaneous pathological hatreds of ambiguity and otherness, in
which we could also include - albeit in a minor category - the moral
panics of official Barcelona faced with the urban ‘rabble’. 133
It is also vital to ask whether the initially semi-autonomous police

action against the POUM obeyed rationales other than the Comintern’s.
The order to arrest the POUM originated inside the police force, not
in the cabinet. 134 The violence and unconstitutional methods used by
131
Elorza and Bizcarrondo, Queridos camaradas pp. 340-1, 387, 395.
132
Cf.Jellinek, The Civil War in Spain, p. 337.
133
Cf. P. and A. Abramson, Mosako roto, p. 206. For this very reason it also becomes problematic to
write the history of the May days and their political ramifications exclusively from the perspective
of Comintern sources. A recent example of the resulting distortion is in chapters 7 and 9 of
Antonio Elorza and Marta Bizcarrondo, Queridos camaradas. What the ‘enemy within’ represented
in rebel Spain was a contaminant in a moral/ social Darwinian sense: see chapter 2 above. There
are echoes of this at times in locally produced Republican propaganda (against ‘rightists’ and
‘hidden enemies’). And although this did not perform the same macro-political function in
Republican Spain, it would be useful to explore some of the parallels.
134 POUM youth leader Wilebaldo Solano referred explicidy to ‘the police coup of 16 June 1937’:
Bolloten, SCW, p. 891 (n. 84); Zugazagoitia (then interior minister), Guerray vkisitudes, pp. 291-4;
M. Irujo comments on an La CNT en la revolution espahola,
‘arbitrary police order’, cited in Peirats,
vol. 3, p. 234; Suarez, El proceso contra el POUM, pp. 83, 10 1 -4; Alba, ‘De los tribunales populares
al Tribunal Especial’, Justicia en guerra, p. 232. Cf. Azana, diary entry 18 Oct. 1937, Obras completas,

vol. 4, p. 828. Note in particular Azana’s comment that what was happening in the police was
worse than Nin’s disappearance itself.
 The Barcelona May days 287

the police after the May events are still routinely attributed to their
Communist Party membership. However, many of those involved were
‘new’ communists, but ‘old’ policemen (and, sometimes, long-serving
professional army officers, as in the case of the then Director General of
Security, Colonel Antonio Ortega, with whom the order for the arrests
orginated. 135 ) So the violence raises far wider questions about the abiding
failure of the Republic since 1931 to reform (i.e. demilitarise) police
culture and practice. The democratic Republic was barely five years old
when war exploded. But the Spanish security forces, whether army
the
or police, had a much longer tradition of taking matters into their own
hands and acting unconstitutionally. 136 Moreover, the onslaught against
the POUM came after the exposure ofa network of Falangist activity and
fifth column sabotage. To expect the ‘old’ police or military mentality

to have drawn a distinction between different sorts of anti-governmental

rebellion in wartime is unrealistic — especially when the left was, anyway,
a target with which many policemen were more comfortable. If we look
again at what was happening, a much older and more deeply ingrained
political culture than Stalinism begins to be apparent behind police zeal.

Additionally, as in other aspects of the post-May repression, there was

also a strong clash of regional cultures - for these were Madrid’s security
137
forces repressing the Catalan left.

The notorious case of the kidnapping and murder of POUM general

secretary Andreu Nin remains a conundrum. Nin was separated off from

135
Ortega was a long-serving professional army officer of moderate republican sympathies who
had joined the PCE in the war: Zugazagoitia, Guerra y vicisitudes, p. 292. Ricardo Burrillo, the
Barcelona police chief who signed the order for Nin’s detention, was also an army officer and
Assault Guard commander (Suarez, Elproceso contra el POUM, p. 83) and another wartime recruit
to the PCE. He was reputed to have been involved in the pre-war assassination of Calvo Sotelo
in July 1936; his career bespoke a certain ruthlessness and he was clearly not averse to the use of
political violence. But none of these attributes were suddenly acquired simply because Burillo
had joined the Communist Party: V Alba and M. Ardevol (eds.), Elproceso delP.O. UM. (Barcelona:
Lerna, 1989), p. 73. Burillo ’s military career went into decline in 1938 and he ended the war,
like a number of other professional officers who had been wartime communists, opposed to the

PCE. See biographical sketch in C. Zaragoza, Ejercito populary militares de la Republica (Barcelona:
Planeta, 1983), pp. 232-5 and entries for both Burillo and Ortega in Alpert, El ejercito republicano,

,
PP- 364, 381.
13
i
Even the Soviet intelligence chief, Alexander Orlov, would himself comment on this, cited in
Costello and Tsarev, Deadly Illusions, p. 300 and P. and A. Abramson, Mosaico roto, p. 190. This is
also borne out by some of the testimonies gathered in Alba and Ardevol, Elproceso del P. 0 UM., for
.

example the circumstances of POUM leaderJordi Arquer’s temporary release and the exchange
between him and Jaume Aiguader (then Esquerra minister in the Negrin cabinet), pp. 539-40;
also Gorkin, Elproceso deMoscu en Barcelona, p. 236. For more on the unconstitutional behaviour
of the police, see the discussion about the militarisation of Republican society at war in chapter
6 below.
137
For a suggestive reference to their hostility to the speaking of Catalan, see Suarez, El proceso
contra el POUM, p. 103.
288 The Spanish Republic at war

the rest of the POUM executive. His personal history may provide the
138
explanatiorfhere. Nin had lived in the Soviet Union throughout the
1 92 os. He had
been an executive member of the Red International of
Labour Unions. Then in 1926 he had joined the Left Opposition, acting
for a time as Trotsky’s secretary. All of this bound him inextricably into
the inner circle of the Bolshevik old guard in a way that the rest of
the POUM leadership was not. But the implication here is that Soviet
operatives were responsible for his murder. A massive amount has been
claimed and conjectured about this. But the fact remains that we still

do not know exactly what happened to Nin or who was involved. To

date nothing in the Comintern archives has shed any further light on the
affair. The NKVD material that has so far surfaced is less than conclusive,
while the NKVD archives themselves are not on open access.
A Catalan Television documentary on Nin’s assassination (
Operacio
Nikolai) made in the early 1990s claimed to offer definitive proof that
it was orchestrated by Alexander Orlov, the NKVD chief in Spain. 139
But in spite of its claims, the documentary raised more questions than
it answered - in particular about what clearly emerges as the key role

of the Spanish police service in Nin’s detention, torture and killing, al-
though this is never directly commented on in the programme. 140 On
24 July 1937 Orlov sent a coded report to Moscow which has been
interpreted as referring to his involvement in the Nin assassination —
although the details remain obscure. 141 Certainly we know that evidence
was forged in a vain attempt to link Nin, via the Falange, to Franco and
Nazi Germany. Orlov may have been involved, and he would certainly
have known about the forgery. He was also involved in - or at least knew
about - the liquidation of foreign dissidents, such as the Austrian social-
ist Kurt Landau. 142 But Orlov, having fled to the USA in August 1938
to avoid being purged, always denied all personal knowledge of Nin’s
assassination. In the 1950s he claimed that it had been carried out by a
138
For a brief political biography, see Durgan, ‘Trotsky, the POUM and the Spanish Revolution’,
p.69 (n. 5).
139 M. Dolors Genoves, ‘Operacio Nikolai’, Televisio de Catalunya, 1992.
140
The video also has a transcript - see pp. 15-18 for information on police involvement. The
Republican interior minister at the time, the socialist Julian Zugazagoitia, also referred to police
responsibility - although he emphasised the problem of post- 18 July recruitment to the security
Alba and Ardevol, Elproceso del P.O.U.M. p. 549. For more on the police’s role in the Nin
forces, ,

caseand the repression of the POUM, see chapter 6 below.

141
The report is cited in Costello and Tsarev, Deadly Illusions p. 291. But see the comments on this
,

book’s particular bias in Gazur, Secret Assignment. The FBI’s KGB General ,
p. xvii.
142
Orlov’s report to Moscow of 25 August 1937, cited in Costello and Tsarev, Deadly Illusions,
p. 286; Gorkin, El proceso de Moscu en Barcelona, pp. 220-1; Thomas, The Spanish Civil War,
p. 706.
 ,

The Barcelona May days 289

special Soviet flying terror squad unconnected to the in Spain. NKVD

This served the dual purpose of delivering the line his US hosts wanted
to hear at the height of the Cold War and avoiding incriminating himself
(Orlov was then seeking the right to permanent US residence and fear-

ful of anything that might prejudice the decision, especially in view of

some hostile questioning over his role in Spain from immigration service
officials.
143
)
But then in 1968, in response to a questionnaire put to him
by the historian Stanley Payne, Orlov retracted his earlier explanation

claiming that what had happened to Nin was all the work of Spanish
communists. 144 There was no obvious political motive for the change
of line. So why did Orlov revise his account? He died in 1973 without
making any further clarification. None of these unknowns rule out the
involvement of Orlov and/or other Soviet personnel in Nin’s death. 145
Indeed, there can be little doubt that Orlov, as intelligence chief, would
have been privy to the details. 146 But it is quite conceivable that Orlov
was assisting the Spanish communists in an enterprise that also served
his general purposes in, as he saw it, cleaning up security risks on the
Republican home front. After all, Orlov was empowered to take spe-
cific initiatives of his own without consulting Moscow.
147
With so many
unanswered questions about the role of the other Spanish communist
parties - PCE and PSUC - in the POUM repression, and the separate,
though overlapping, issue of ‘new communists’ and ‘old’ police culture,
as historians we should take care not to accept too readily the idea of sole
Soviet authorship. 148 Indeed, given the sheer density and multiplicity of
the political conflicts in the wartime Republic, its very simplicity should
give us some pause for thought.
From the pattern of communist intervention in the arrest and re-arrest
of Andrade and his POUM executive colleagues, it could be construed
that the object was to ensure that the party executive was subject to
formal criminal proceedings. They were originally arrested in Barcelona

143
Costello and Tsarev, Deadly Illusions, pp. 352, 356. Marcelino Pascua also pointed to the evident
unreliability of Orlov’s testimony given the context in which it was produced: M Pascua to
A. Vinas, 13 Feb. 1977, AHN/MP, 8 (13).
caja
144
Bolloten, SCW, p. 509. The now been published in English in Forumfur
Orlov questionnaire has
osteuropaische Ideen und 4 (2000), 238-43; see also Gazur, Secret Assignment. The FBTs
feitgeschichte,
KGB General pp. 337-8, 340, 341-5.
145
Orlov’s own musings on the role of PCE leader Jesus Hernandez also raise the knotty question
of whether any putative Comintern involvement in the Nin affair need have been synonymous
with a Soviet government directive: Gazur, Secret Assignment. The FBI’s KGB General, pp. 344-5.
146
Cf. G. Brook-Shepherd’s preface, ibid., p. xiii.
147
Costello and Tsarev, Deadly Illusions, p. 272.
148
See chapter 6 below for the rise of police and intelligence service influence in 1938 - as the
Republican home front became increasingly militarised.
2go The Spanish Republic at war

by communist police from Madrid and taken to police cells in Valencia

(thereafter being transferred to the state gaol). 149
Had the objective been
liquidation, it seems unlikely that Valencia would have been their desti-
nation - would seem to bear out
as the case of Nin, arrested separately,
(although the killing of Nin may ultimately have resulted from the failure
of his gaolers to extract a ‘confession’ to POUM espionage). When in-
ternational pressure procured the release of Andrade and his colleagues,
they were instantly re-arrested by communist police and this time taken
Madrid. As police action was now clearly occurring without
to a checa in
any constitutional check, their lives were indeed at risk. But as a result
of government pressure they were transferred several weeks later (at the
end of July) to state custody in Valencia prison. 150
Especially after the disappearance of Nin it was vital to the Republic’s
international credibility as a democracy that due process was followed in
the detention of the POUM
executive and that its members’ safety was
guaranteed. Negrin was no less concerned here than Prieto, Irujo and
Zugazagoitia (in the interior ministry). (Indeed, provided the POUM
leaders could be shielded from irregular police action, they were rather
safer inside state prison than they would have been on the streets — given
how high political tempers were running in some quarters.) Negrin dif-

fered from his cabinet colleagues, however, over how far they should take
the investigation of Nin’s disappearance. 151 Given it was a fait accompli ,

Negrin was determined to limit the damage that revelations of PCE and
Soviet involvement would do to the political coalition underpinning the
Republican war effort. But this did not mean that Negrin was simply
sanctioning what the PCE wanted. After all, the prime minister had
effectively sacked Ortega. 152 Equally, Comintern reports of this time ex-
pressed unhappiness at the government’s obstruction of the ‘war’ against
the POUM. 153 No doubt Negrin’s own political antipathy to the POUM
after May 1937 - which owed nothing to the PCE — made it easier for

149
Suarez, El proceso contra el POUM, p. 84.
150
The Suarez account also makes it clear, however, that, prior to their (second) departure for
Valencia, the POUM executive members had already been transferred from checa to state gaol
in Madrid - presumably also because of government pressure: El proceso contra el POUM, p. 85.
Andrade’s account is in Fraser, Blood of Spain, pp. 387-9. Republican state prisons also held
foreign dissidents as well as Spaniards: Morrow, Revolution and Counter-Revolution in Spain, p. 145;
E. Goldman, ‘Political Persecution in Republican Spain’, Spain and the World, 10 December 1937,
also cited in Souchy, The May Days, pp. 104-7.
151
Prieto, Convulsiones de Espaha, vol. 2, p. 117.
152
Zugazagoitia, Guerra y
vicisitudes de los espaholes, pp. 292-3.
153
Elorza and Bizcarrondo, Queridos camaradas, pp. 378-9.
 ,

The Barcelona May days 29

him to deal with the ethical implications of his decision over Nin. Negrin
had few hesitations about bringing the POUM to trial because, in the
middle of a war, its leadership had publicly (in its press) approved re-
bellion against the Republican state. His rationale around the POUM
was essentially no different from the one that he had followed in 1932
when he demanded that the full force of the law be brought against
the leaders of the anti- Republican military rebellion. But precisely be-
cause this was Negrin’s reasoning, precisely because of the supreme im-
portance which he had always accorded Republican state authority, it
was imperative that the judicial enquiry and trial of the affirm POUM
that same authority and the constitutional legal principles on which it
rested.
What was happening to the POUM clearly served the party political
interests of both the PCE and the PSUC. However, the party leaderships
were less interested in the repression of individual POUM dissidents
perse- indeed, there is some indication that the PCE, like the Republican
government, was wary of a backlash. 154 Rather, the thrust of the two
Spanish communist parties’ post-May days attacks on the POUM was a
strategic offensive designed to procure the party’s total exclusion from the
political life of the Republic at war.do not know enough about We still

the dynamic of this wartime communist movement

conflict inside the
in Spain. 155
Two things, however, are clear. First, the PSUC was not
a bystander in the process. 156 Its leaders would subsequently stress the
role of the Spanish Communist Party in POUM repression, arguing
that in the small world of Catalan communism, where organisational
loyalties had been extremely and where all the leaders
fluid until 1935
knew each other would have been impossible for anyone
personally, it

in the PSUC to believe that the POUM leaders were fascist agents (as
was the Comintern’s justificatory line). 157 This is no doubt true — and
certainly the communist policemen sent to arrest the POUM leaders
came from Madrid. 158 However, there were plenty of other tensions and
154
Ibid., p. 378; Fraser, Blood of Spain, p. 388.
155
Casanova, Anarquismoy revolucion, p. 247 . It is clear that the PCE mounted a well-organised anti-
POUM propaganda campaign which involved high-pressure tactics and intimidation: Gorldn,
El proceso de Moscu en Barcelona p. 250; Alba and Ardevol, El proceso del P.O.U.M., p. 541.
156
Cf. Suarez, El proceso contra el POUM ,
p. 88.
157
See PSUC leader Pere Ardiaca’s testimony in Fraser, Blood of Spain, p. 390. leader POUM
Andrade’s comments distancing both PSUC and PCE from responsibility for the repression
should be seen in the context of the post-war tendency of POUM
survivors to construct the
Comintern as the monolithic author of their political downfall: Fraser, Blood of Spain, p. 388.
158
Suarez, El proceso contra el POUM, p. 101.
292 The Spanish Republic at war

disagreements which could have provoked PSUC violence against the

POUM during the May Days. 159

In the pre-war Republic many acute intra-organisational conflicts on

the left had been played out violently. The coming of the war did not
wipe out these memories or the patterns of dispute which had originated
the conflicts. Indeed, as these frequently arose over issues of political in-
fluence, clienteleand membership rivalries, the circumstances produced
by rebellion and war, if anything, intensified such clashes in the Repub-
lican zone - hence the emblematic importance of the transfuga (political
160
renegade). These clashes occurred, as we have seen, between cenetistas
and ugetistas between socialists and communists and between the rival
,

branches of Catalan communism. Part political, part organisational, part

personal, these disputes also led to physical violence against individual
poumistas. But while this constituted in part ‘communist’ violence, it can-
not accurately be subsumed under the term ‘stalinist’, in the sense that it

was not a response to a Comintern game plan. Once the street fighting
had erupted in Barcelona, it precipitated a quantity of bloodletting on
all sides. The CNT, UGT, PSUC and POUM, as well as other lesser

players, were all involved as the ‘ghosts’ of decades of labour wars and
161
political infighting stalked the streets and meeting rooms of Barcelona.

159
Not the least of which were intense intellectual jealousies in
what was a very small world.
The most bitter personal animosities members who had joined
were those between the BOC
the POUM in 1935 and those who had joined the PCC (soon the PSUC). Indeed, the personal
hatreds between POUM and ex-PC C people were in general the most intense. We need to think
back to the process by which both the POUM and the PSUC were formed (see chapter 1 above).
As Enric Ucelay da Cal has pointed out, the disputes between the two parties would be made
more complex because - in spite of how they are frequendy portrayed - both the POUM and the
PSUC contained social democratic and ‘neo-bolshevik’ components, ‘Socialistas y comunistas
en Cataluna durante la guerra civil’, p. 311 . The traces of these disputes even appear embedded
in the narrative of POUM leader Julian Gorkin’s own memoir, El proceso de Moscu en Barcelona :

see, forexample, pp. 227, 246 - although, of course, the author himself interprets these details
as proof of something else entirely.
160
In Catalonia people usually passed from the CNT to the sphere of left parliamentary politics
(UGT and communist parties). This sort of transfuga was a fairly common phenomenon on the
inter-war Spanish left: examples include Roldan Cortada, Rodriguez Salas (see Bolloten, SC W,
p. 417) and RafaelVidiella (Sese’s replacement). Andreu Nin’s own political trajectory went
from left Catalanism through the PSOE and the CNT onto Bolshevik-Leninist ranks. Anarchist
commentators have always been at pains to point out that the CNT had no institutional history of
taking reprisals against transjugas (seej. Peirats, in Souchy et al., The May Days p. 20). But it must ,

have exacerbated other intra-left organisational tensions. As chapter 1 above also indicates,
there was, additionally, a great deal of ‘traffic’ up to 1935 socialist and
between the various
communist organisations in Catalonia. Reflecting the pattern elsewhere in Spain, some leading
members of the Esquerra also passed to the PSUC in summer and autumn 1936: Ucelay da
Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 311.
161
Cruells, Mayo sangriento, pp. 62 69; Ucelay da Cal, ‘Socialistas comunistas en Cataluna durante
,
y
la guerra civil’, p. 313.
 The Barcelona May days 293

Even the ritualistic reciprocal tearing up of union and party cards in the

days after the May streetfighting could also be seen as a point where
162
‘top-down’ state repression met this clientelist intra-party conflict.

The recruitment of ‘opportunists’ was another constant and wide-

spread accusation levelled by each of the Republic’s political groups
against the rest since the start of the conflict. Socialists blamed re-
163
publicans and communists; republicans blamed the communists, and
everyone blamed the CNT, who, in turn, blamed the communists and
republicans.
164
The most numerous - namely the anarcho-syndicalists
and the communists - were both accused by leading socialists on var-
ious occasions of violent outrages against their membership which the
denouncers publicly identified as forming part of a brutal battle for party-
105
political/ organisational advantage within the Republican home front.
The CNT’s (militia) recruitment from among prison populations freed
166
at the start of the war was not easily forgotten.
But by early 1937,
as the socialists were suffering a degree of organisational eclipse from
which the PCE was benefiting, there were also various denunciations of
a similar type levelled against communist office-holders - for example a
scandal involving the detention of PSOE militants in Murcia led to the
107
dismissal of the civil governor.
would have been extremely difficult, given the war, for those joining
It

Republican political organisations to have been screened - even if the

162
Spain and the World press report, 22 Sept. 1937, in Souchy et al., The May Days, p. 100; Cruells,
Mayo sangriento, p. 82 See also Azana’s acute remarks about inter-party competition for members,
.

Azana, Obras completas, vol. 4, p. 593.

163
Graham, Socialism and War, passim.
164
Cf. CNT’s jaundiced view of the political ‘promiscuity’ of Left Republicans in Aragon: ‘son del
ultimo que viene (they follow whoever’s in political fashion): with the Republic in 1931; with
the right and Gil Robles during the black biennium (1933-5); after the military rising with the
CNT and now, once again, with the government’, letter from CNT union (Teruel province),
November 1937, cited in Casanova, Anarquismoy revolucion, p. 223.
165
In so far as the PSOE (though not the UGT) remained mostly outside this argument, it was
because the war had supposed a ‘membership crisis’ of sorts for a variety of reasons. See chapter
3 above for a discussion of these. (Our knowledge of this crisis does not come from membership
figures- which for the war are hard to come by (especially for the PSOE) - and, in any case,
difficult to interpret in such an exceptional period. Rather, we know of this crisis from the
consistent, across the board discussion of it by a variety of socialist party and union leaders: see
Graham, Socialism and War, passim.)
166
Ibid., p. 63; Vidarte, Todosfuimos culpabks, vol. 2, p. 656; G. Moron, Polltica de ayery polltica de
mahana (Mexico: n.p., 1942), pp. 77-8; Broue and Temime, The Revolution and the Civil War in

Spain, p. 273.
1 7
Graham, Socialism and War, p. 63; Morrow, Revolution and Counter-Revolution in Spain, pp. 73-5.
Although Morrow alludes to it, I purposely do not cite here the famous polemic involving
Cazorla and the Madrid Defence Council which erupted in April 1937. As I explain in chapter
3 above, this was a significandy more complex conflict which developed from the perilous and
fraught front-line position of the capital city.
294 The Spanish Republic at war

will had existed. But infighting was inevitable anyway in a context of ac-
celerated mass political mobilisation. The war had massively inflated the
reciprocal currency value of political membership in a society in which
deep-rooted traditions of political nepotism and ‘fixing’ ( enchufismo) com-
bined with low levels of political education.
168
A ‘penumbra’ certainly
existed at the edges of the most successful and competitive organisations
where political clientele met criminal fraternity. Precisely what relation-
ship these elements had to the political activities of the communist and
anarcho-syndicalist organisations, however, is rather harder to assess.
The input of these and other ‘uncontrollables’ is, by definition, an
imponderable. But once meltdown had occurred in Barcelona on
4 May, the ensuing confusion in the region offered opportunities for the
‘resolution’ of all manner of scores - at least some of which would not
have been directly related to the big political issues at stake. State control
was significantly greater than in July 1936, but it seems reasonable to as-
sume would still have been elements of lumpen activity under
that there
the blanket of May. As we have already seen, paseos were notorious for
providing a useful cover for acts of personal as well as political revenge
and for outright criminal activity. 169
There was some evidence that ‘uncontrollables’ on the fringes of the
CNT may have been responsible for the killings of the leading Italian
anarchist Camillo Berneri and his secretary Francesco Barbieri, whose
bodies were found on the streets of central Barcelona, near the Generali-
170
tat building, during the night of 5-6 May. Their deaths have frequently
been attributed to Comintern activity, but largely speculatively, on the
basis of the political context in which they occurred.
171
A search was
made of Berneri’s flat on 4 May by two men ‘wearing red armbands’
who carried away documents. The CNT press report of 11 May which
broke the news of the double assassination was censored to prevent its
identifying the searchers as PSUC policemen. But we do not really know

168
The very ambiguity of ‘communist police action’ as discussed here is itself indicative of the still

relatively weak nature of state-derived professional identities as against those of political parties
whose force of attraction still depended at least in part on clientelist practices (i.e. parties were
still viewed as offering access to direct material benefits, career advancement etc.).
169
See chapter 2 above for these problems in relation to the July days of 1936; also chapter 3 on the
Madrid Defence Council, public order and the occurrence of kidnapping and extortion ‘scams’
involving its employees.
170
Cruells, Mayo sangriento, p. 69; Semprun-Maura, Revoluciony contrarrevolucion en Cataluha pp. ,
279-
81.
171
The report in the CNT press, Solidaridad Obrera, on 11 May 1937 (see Peirats, La CNT en la

revolucion espahola, vol. 2, pp. 148-50) linked the killings to the disappearance of the Russian
dissident communist Marc Rhein: Broue and Temime, The Revolution and the Civil War in Spain,

p. 305 285-6); Souchy etal., The May Days, pp. 40-2, 105.
(also pp.
 . ,

The Barcelona May days 295

whether the larger group who came to arrest Berneri and Barbieri on
5 May
were also PSUC affiliates, or, indeed, even police at all.
While some CNT sources held to the theory of PSUC police involve-
ment, 172 the CNT’s official report on the events, published in June 1937,
accused pro-Fascist elements of the radical Catalan nationalist party,
Estat Catala, of colluding in the killings with agents of the Italian se-
cret police (OVRA). There is a great deal of circumstantial evidence
173
to recommend this explanation. However, other intelligence which
came to leading members of the PSOE in government also suggested
that Berneri and Barbieri may have been killed on the orders of Angel
Galarza, interior minister in Largo Caballero’s central government, in
order to prevent Berneri going public on Galarza’s involvement in the
embezzlement of public funds. 174 Galarza, apart from being a fairly in-
competent minister, was a late-comer to the PSOE who had a very
dubious reputation among his colleagues as something of an oppor-
tunist and sharp operator reminiscent of the ‘old world’ of monarchist
politics. 175 According to this intelligence report, Galarza’s ‘fixer’ was an

Italian anarchist called Gigi-Bibi who had already employed other Ital-
ian anarchist contacts in Barcelona to secure some of the incriminating
documents in Berneri’s possession. Although none of this is conclusive,
Galarza could easily have ordered a police search and then taken other
measures accordingly. Indeed, his ‘anarchist’ contact could conceivably
have been an OVRA agent himself. Although the OVRA’s activity in
Barcelona was relatively marginal, we do know that its agents kept
Berneri under close surveillance and it is quite possible that, to this

172
Cf. A. Souchy, Los sucesos de Barcelona (Valencia: Ebro, 1937, transl. as The Tragic Week in May a
publication of the CNT-FAI’s external information service which Souchy directed) and Souchy’s
own essay in TheMay Days (1987), p. 42.
173
See Bolloten, SCW, pp. 453, 875-7 f° r a resume of the available literature and current state of the
enquiry on Berneri’s death. Souchy also implies the possible removal by those searching Berneri’s
flat on 5 May of Berneri’s manuscript dealing with Mussolini’s expansionist Mediterranean
policy: The May Days, p. 42
174
Copy of (anonymous) 24 Nov. 1937, found in the correspondence of Luis
report, dated
Araquistain, political lieutenant of Largo Caballero and his ambassador in Paris until the cabinet
changes of May 1937. See Howson, Armsfor Spain, pp. 225-7.
175
Certainly there were complaints in the PSOE about the dubious characters sometimes to be
glimpsed ‘in the wings’ of Galarza’s ministerial suite: Graham, Socialism and War, pp. 63-4, 99;
War in Spain, p. 558. The
Vidarte, Todosfuimos culpables,vo\. 2, pp. 673, 655, 862; Jellinek, The Civil
‘old politics’ represented by Galarza would also be integrated to the new Francoist order. For,
in the rebel zone, although there was never any crisis of political order, networks of corruption
also existed. The point about the Republican zone is that although these old behaviours were
still present, they were also beginning to come up against other political values - such as those

of the ‘public-service’ state - which carried the transformative potential for new professional
identities and a different political culture.
296 The Spanish Republic at war

end, they may have infiltrated foreign anarchist groups operating in the
176
city.

Certainly there were particular aspects of CNT practice which might

be said to have facilitated infiltration - whether by
Italian or Spanish
fascists 177 or by those involved in straightforwardly criminal pursuits.

However, the fact that the wartime ‘penumbra’ spread wider than the
CNT-FAI suggests that this picture is best explained in relation to the
generally destabilising impact of coup and war. These had produced a
‘wild new frontier’ environment which saw gangsterism and local feuds,
political or otherwise, waged in the space still unfilled by reconstructing
state power. 178 It was this that probably explains the intensity of the post-
May days violence in the smaller towns of the Barcelona area. 179 For onto
state repression -
in the form of the police-led attacks on Telefonicas
and CNT/POUM premises - were grafted other sorts of score-settling
in more opaque, but still often very violent, local conflicts. This was the
180
picture not only in Catalonia but also in neighbouring Aragon.
Although many imponderables remain over the May days, it is at least
clear that the events cannot be reduced to a Cold War parable of an alien
Stalinism which ‘injected’ conflict into Spanish Republican politics. The
Comintern’s ‘clean-up’ of dissident communists in Barcelona in May
and June 1937, ethically unattractive though it was, constituted but one
element in a bigger picture. Nor can it even explain everything that
happened to the POUM.
In the wake of the May events, the POUM would be made a polit-
ical example in a way no other Republican group was. But how, why

and (crucially) when that happened all make for a far more complex story
than the one habitually laid down in the existing martyrologies of POUM
demise - as we shall see.
181
Moreover, if we always compose the party’s
history backwards from the May days - and in particular from the atro-
cious murder of Nin - then we shall never understand the political and
176
See the work of C. Rama (ed.), Camillo Berneri: Guerra de closes en Espana 1936-1937 (Barcelona:
Tusquets, 1977) and Fascismoy anarquismo en la Espana contemporanea (Barcelona: Bruguera, 1979).
See also S. G. Pugliese, Carlo Rosselli: Socialist Heretic and Antifascist Exile (Cambridge, Mass.:
Harvard University Press, 1999).
177
In contrast to Madrid, however, the indigenous fifth column in Barcelona was, as yet, marginal. It
would only become a force of any significance in the grim days of 1938 when the Republic faced
an utterly bleak international horizon and was on the edge of material collapse. See chapters 6
and 7 below.
178
Cazorla himself points this out: see minutes of the Madrid Defence Council reproduced in
Arostegui and Mariinez, La Junta de Defensa de Madrid, p. 446.
179 180
Cf. Cruells, Mayo sangriento, p. 90. Casanova, Anarquismo y revolucion, pp. 253-63.
181
For an analysis of the proscription of the POUM in the context of the Republic’s deteriorating
wartime situation in 1938, see chapter 6 below.
 The Barcelona May days 297

cultural complexity that made the POUM what it was - with everything
that ‘complexity’ signifies in terms of internal tensions. In this respect, we
should remember too that the POUM had had its own experience of vi-
182
olent internecine conflict earlier in the war. This stemmed from a par-
ticular genealogy across the 1930s which had brought the POUM to en-
capsulate within major political division between radical politics
itself the
183
and liberal political reform now rending the wartime left as a whole.
Finally, we can conclude that the May days were an urban rebellion,

directed against state power and reflecting the particular configuration

of CNT strength - and weakness - in Catalonia. This is confirmed by the
fact that the last act of the May days took place, not in Catalonia, but in
neighbouring Aragon - the agrarian counterpart of ‘red’ Barcelona. The
city’s had been dismantled, but Aragon’s institutional
physical barricades
barricades against central statepower still remained, in the form of the
governing Council of Aragon, backed by the armed strength of CNT and
POUM cadres on the eastern (Aragon) front. The central government’s
appointment on 5 May of General Pozas as commander of the Eastern
Army, as well as military head of the Catalan region, set the stage for
the subsequent military offensive in Aragon which would take place in
August.
What would open the way for Aragon to be brought fully into the orbit
of the central Republican government, however, was the
change crucial
in the constellation of that government - a change by the also detonated
May day events in Barcelona. The PCE’s discontent with Largo’s running
of the war ministry had already reached critical mass before May. In all
respects, the party’s criticism reflected the collectively held view of Largo
in the cabinet: he was inept, devoid of his own ideas, and made policy
with a cabal of military and civilian supporters of questionable abilities
and (in some cases)perhaps dubious loyalty too; he neither consulted
the cabinet nor convoked the High War Council (Consejo Superior de
Guerra). 184 It had taken Largo until the beginning of March to get the
Generalitat to agree to submit the Catalan militia to the control of the
Valencia war By mid April the PCE had been given the go-
ministry. 185
ahead by the Comintern to find a way of separating Largo from the war
186
brief. But this did not mean removing him as prime minister. Stalin
182
See the suggestive testimony of Paul Thalmann to B. Bolloten, SCW, p. 860 (n. 12).
183
Cf.Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 311.
184
Azana, Obras completas vol. 4, pp. 591-2, 594-5; Elorza and Bizcarrondo, Queridos camaradas,
,

pp. 340-1, citing Stepanov’s Comintern report of 17 March 1937.

185
Salas Larrazabal, Historia del ejercito popular, vol. 4, pp. 1042-5; Bolloten, SCIY, p. 419.
186
Elorza and Bizcarrondo, Queridos camaradas p. 341.
,
298 The Spanish Republic at war

saw Largo asperforming a vital political role - and one in which he could
not easily be replaced. It was the symbolism of Largo that guaranteed
for the Republican alliance sectors of the labour movement and political
left — especially in the CNT — who might not otherwise support the war
187
effort but who were crucial to it. Largo’s premiership was, however,
already extremely problematic, given that he was now estranged from
virtually the entire cabinet. What most concerned republicans and par-
liamentary socialists - and perhaps above all President Azana - was the
war’s crucial international-diplomatic dimension. None were unaware
of Largo Caballero’s domestic usages. Nevertheless, both republicans
and socialists were desperate for a prime minister with a greater intel-
lectual grasp of the vital importance of international diplomacy to the
188
Republic’s fate. They wanted someone ‘fluent’ in such milieux who
could plead the Republic’s case, indeed who, in his own person, encap-
sulated the liberal democratic principles being defended in arms. For
the socialists too, therewas also the ongoing friction with Largo and his
supporters over the liaison committee policy with the PCE. For those so-
cialists who identified with the PSOE leadership, the Caballeristas were

not only ignoring party discipline in voicing their opposition, but were
also undermining a policy whose primary rationale was the practical
support of the war effort and the government.
Once the May day protests had exploded in Barcelona, the issue of
public order rapidly focused minds in the Valencia cabinet. The assassi-
nation of Generalitat minister Sese on 5 May increased the republicans’
resolve to find a firmer hand for the helm: someone who was prepared
to crack down even if it meant unpopularity with some sectors of organ-
ised labour and the left. 189 The PCE leadership was also clear about this

187
The same conclusions on Stalin’s position here have also been reached independendy by Frank
SchaufF, in ‘Sowjetunion, Kommunistische Internationale und Spanischer Biirgerkrieg 1936-
1939’, doctoral thesis, University of Cologne, 2000 and in his unpublished research paper
‘Failure in Emergency: The Spanish Civil War and the Dissolution of the Comintern’, p. 27. It
would be Largo’s own refusal to separate the two ministerial portfolios which made inevitable
his departure from the cabinet in May.
188
Cf. ‘One of the main reasons [Largo] had to be replaced was because of his evident and notorious
incapacity as prime minister, at least during such a war which was complex and exhausting not
only because of its internal dimension but also because of the powerful influence of certain
great powers whether they intervened militarily or held back from the conflict — there was a
complex diplomatic configuration in which precious, and irrecuperable, time was lost both in the military and
(my italics). Undated comments in the personal archive of PSOE member and
political spheres'

Republican ambassador M. Pascua, legajo 5, carpeta 6 (doc. 8 in series although unnumbered).

189
Irujo, in the name 01 the PNV’s parliamentary group, called for a new prime minister who could
‘impose constitutional order and the rule of law on the home front and put an end once and
for all to the uncontrollables, to committees and to other kinds of violence’: M. Irujo, cited in
H. Raguer, La polvoray el incienso. La iglesiay la guerra civil espahola (1936-39) (Barcelona: Ediciones
Peninsula, 2001), p. 325.
 The Barcelona May days 299

and, by 7 May, the basis of an anti-Largo cabinet alliance had formed

between republicans and communists.
190
The PSOE national executive
was no less a part of this alliance. But it kept a lower profile - something
which reflected Prieto’s preference for behind-the-scenes manoeuvring,
but also the party leadership’s not inconsiderable collective fears that too
obvious an attack on Largo would exacerbate the internal divisions in
the socialist organisation - something they were keen to avoid at all costs
since it would further advantage PCE recruitment. The republicans also
remained cautious about a full frontal attack on Largo. The May events
were themselves a reminder that excluding him from the cabinet itself
might provoke further destabilising popular protests in the streets and
workplaces of other Republican cities. The conundrum preoccupying
republicans, communists and the PSOE was just how powerful Largo
Caballero really was.
On Thursday 13 May, during an extremely acrimonious evening cab-
inet meeting, the crisis finally broke. The PCE had tabled a severe criti-

cism of Largo’s war and public-order policy. When the premier refused to
accept any of it, the two communist ministers walked out of the meeting.
Largo was for continuing the meeting, but Indalecio Prieto reminded
all

him PCE’s withdrawal constituted a government crisis and that

that the
the president needed to be informed immediately. At this point, Largo
seems to have been looking to Azana to back his formation of a new cab-
inet entirely excluding the PCE. Largo was also much exercised by the
communists’ demand - raised as part of their broader public-order con-
cerns - that the POUM be declared an illegal party after its involvement
in the May day events. Although the prime minister felt no sympathy
for the POUM’s political stance, he was reluctant to bear the personal
responsibility for state action against any organisation of the left. Partly
this was an ethical concern, but partly it reflected his anxiety over the
damage which would inevitably be done to his reputation as leader of
the UGT if he were to become publicly associated with the repression of
the POUM. Largo’s concern that the courts should opine on the POUM
before the government acted was constitutionally impeccable. But the
Republic was at war, fighting for its life, and the events in Barcelona could
easily have capsized Republican resistance. Moreover, that Largo could
have believed that it was feasible to exclude the PCE from the cabinet
is the clearest possible indication of his limited political understanding.
Neither the international nor domestic situation nor his own political
isolation in the cabinet permitted it.

190
M. Azana, Obras completas, vol. 4, pp. 591-2; Graham, Socialism and War, p. 91.
300 The Spanish Republic at war

Largo was up against President Azana’s own intense — if still

also
dissembled - dislike of him. Azana had been opposed to his appointment
as premier back in September 1936 and had not changed his opinion
since - finding Largo obtuse and obdurate. The president had wisely kept
his own council on this matter (Largo certainly had little idea how much
he was disliked), but now Azana glimpsed the opportunity to acquire
a prime minister on his own wave length. His desire so to do had, of
course, been galvanised by the fact that Largo had ignored him entirely
during his anguished days of virtual imprisonment in Barcelona. At the
same time, however, Azana was determined that Largo’s removal should
not be seen as a presidential caprice. He knew that this would only
bolster the outgoing premier’s credibility. Largo’s removal had to come
as a consequence of his rejection by the Republican political alliance -
including the parties that represented working-class constituencies.
verbal sound and fury on 14 May from Largo’s unconditional
The
backers in the UGT
executive as well as threats of mass protestCNT
action momentarily fazed the PSOE - as it did some republicans,
including the Republic’s vice-president, Martinez Barrio. 191 By the end
of that same day, however, had become clear that Largo was not going
it

to negotiate with the PCE. Instead he was hoping simply to put off the
crisis by disappearing to conduct military operations in Extremadura.

Their success, so Largo hoped, would be enough to silence the PCE

and end the crisis in his favour. Given the general lack of confidence in
Largo as war minister, Azana viewed this scenario with not much less
scepticism than he did the Caballeristas’ totally impracticable scheme
tofoment an anti-Francoist rebellion among the Rif tribes of Spanish
Morocco. 192 The Extremaduran offensive - never in fact implemented -
was intended to take the pressure off the north, while also cutting Franco’s
troop supply lines in the south. Whether it could ever have achieved the

191
Azana, Obras completas, vol. 596-7.
4, pp.
192
See ibid., pp. 589, 594 for cogent views on the Moroccan plan (and a shrewd assessment
his
of Largo’s naive belief in popular uprising as a mystical event). As has been demonstrated,
there were structural political reasons why the proto-nationalists with whom the Caballeristas
de Baraibar) were in contact (and to whom they were paying substantial
(particularly Carlos
sums of money) could not deliver what they promised: R. M. de Madariaga, ‘The Intervention of
Moroccan Troops in the Spanish Civil War: A Reconsideration’, European History Quarterly, 22
(Jan. 1992). Prietodebunked the Moroccan plans along similar lines: see Guerray Revolucion vol. 3,
,

p. 80 (n. 3). There

is information on de Baraibar’s role as go-between in Araquistain’s papers,

legajo 25, B30. Pascua also points out that Largo and de Baraibar were financing the Moroccan

plan without telling Negrin in the treasury. Pascua (who would take charge of the Republic’s
Paris embassy in June 1938) also cites reports critical of Largo’s Moroccan scheme from the
Spanish consul general in Rabat (French zone), 12, 28 May and 23 June 1937, AHN/MP, caja
2 (14.4).
 , ,

The Barcelona May days 301

latter objective - not least because of the Republic’s lack

is questionable
of an offensive capability. In the event, the Republic would opt instead
to activate the eastern (Aragon) front to relieve pressure on the north. It
is certainly true that the PCE dragged its feet over Extremadura. But so
did Prieto. General Miaja, who had had his own severe conflicts with
the prime minister, also procrastinated over the dispatch of troops from
Madrid to participate in the operation. There was a fear of uncovering
the Madrid front - where Franco’s best troops were concentrated -
and a shared conviction that Largo was militarily incompetent. Both
sentiments existed in their own right rather than merely as foils for other
personal or political agendas. The strategic value of the Extremaduran
offensive was also somewhat dubious given that the rebels had access to
Portuguese roads, air bases and telephone facilities. 193 Moreover, the fact

that no evidence of any detailed military dispositions for the operation

was ever found in the war ministry suggests that the reservations of
those opposing Largo were well founded. 194
But if the outcome of the May cabinet crisis can be said to have had
an architect, it was the PSOE’s Indalecio Prieto. 195 The sheer uncon-
stitutionality of Largo’s ad hoc postponement of the cabinet crisis pro-
voked him into action - doubtless strongly encouraged by Azana, who
was particularly exercised by such things given what little remained of
constitutional etiquette during wartime with parliament closed. Azana,
unlike many of his republican colleagues, was unworried by the noise
from Largo’s backers because he believed that they overestimated their
leader’s strength in the IJGT. 196 (Largo’s base there had, anyway, been
undermined by the split in the national executive - several members
of which now backed Prieto and the PCE. 197 ) Thus, on the evening of
14 May it was the PSOE which forced wide open the political crisis by
withdrawing its appointees from the cabinet. Prieto himself remained, as

193 Araquistain would make great claims for the Extremaduran offensive in his pamphlet El comu-
nismoy la guerra de Espaha published in France in 1939 (and reproduced in L. Araquistain,
first

Sobre laguerray en la emigration (Madrid: Espasa-Calpe, 1983), p. 218). Some of these claims were
subsequently endorsed by Francoist military historians - see R. Salas Larrazabal, La historia del

ejercito popular vol. 1, pp. 1076-83. The plan of operations also appears in Colonel Jose Manuel
Martinez Bande’s La ofensiva sobre Segovia y la batalla de Brunete (Madrid: San Martin, 1972),
pp. 237-40. But such historians also had a clear political agenda. With this in mind, Salas’
sudden belief in the offensive capacity of the Republican army is, to say the least, remarkable.
194 Prieto and an (unnamed) officer of the Republican general staff, both interviewed by G. Jackson,
The Spanish Republic and the Civil War, p. 372, n. 13.
195 The key link between Azana and Prieto throughout the May crisis is the republican former
premier, Jose Giral. A close friend of Azana’s, Giral had also relied heavily on Prieto’s advice
during his premiership at the start of the war - see chapter 2 above.
196 197
Azana, Obras completas, vol. 4, p. 596. Graham, Socialism and War, pp. 168-70.
 ,

302 The Spanish Republic at war

ever, in the wings, sending his fellow PSOE ministers, Juan Negrin and
the veteran socialist Anastasio de Gracia (labour minister), to inform
Largo of the party’s decision. 198
Largo’s manner of confronting the situation amply demonstrates his
political limitations. Lirst, a fit of pique, then an attempt to ignore all

his opponents entirely - but this without offering any further incentive
to the CNT to close ranks with his supporters in the UGT. The cabinet
which Largo proposed to Azana on 17 May was an instrument designed
to punish the PSOE
executive and, above all, Prieto. The plans to do
by some months, however. Already in early March
so predated the crisis
Araquistain had tabled proposals to Largo which were similar in their
overall design (of marginalising the Caballeristas’ socialist rivals in the
cabinet) to those of 17 May. 1
" In the latter the PSOE would have been
reduced to 2 ministries out of 14. Negrin remained in the treasury. But
Prieto was relegated combined ministry of trade, industry and agri-
to a
200
culture. (The PCE, however, retained two posts — although swapping
agriculture for labour.) But there was even worse to come. In spite of
the PCE’s key demand that Largo relinquish the war portfolio - tacitly
seconded by the PSOE and the republicans - Largo had augmented his
responsibilities here to oversee a new ministry absorbing Prieto’s former
navy and airforce brief. To cap it all, the CNT was thrown mere scraps —
the now politically marginal post of justice, as well as that of health. As
the CNT lost no time in pointing out, this showed a fine lack of grati-
tude (among other things) on the premier’s part, since it put them on an
equal footing with the PCE had initiated the crisis. 201
- the party that
Immediately the CNT made it known that it would not participate in
any cabinet in which it did not retain control of the trade and indus-
try portfolios. Finally, the unwieldiness of the cabinet - against which
Azana had expressly warned — indicated Largo’s stubborn adherence to
an antiquated political formula whereby peripheral portfolios (and non-
were allocated proportionally to all groups in the Popular
portfolio posts)
Front coalition - a practice which generated a monster entity entirely

198
Largo Caballero, Mis recuerdos p. 205; Graham, Socialism and War, p. 97.
199
Graham, and War, pp. 92-3. Prieto was allocated public works.
Socialism The other PSOE
minister was, as before, Anastasio de Gracia in the labour post.
200
Azana, Obras completas, vol. 4, p. 601. A full list of Largo’s proposed cabinet
is in Guerra
y revolucion
en Espana, vol. 3, p. 81. The had two ministries (public works and propaganda),
Left Republicans
as did the PSOE and PCE. Republican Union had one (communications/merchant navy) and
the Basque and Catalan nationalists (PNV and Esquerra) had a minister without portfolio
apiece.
201
There was extensive coverage of the CNT’s reactions in the press. A collection of related cuttings
is also to be found in the Archivo Historico Nacional (SGC), Salamanca, serie Bilbao, carpeta
39.
 The Barcelona May days 303

inappropriate to the tight communication and cooperation required in

wartime.
But the most striking feature of Largo’s proposed ‘solution’ to the
cabinet crisis was his absolute refusal to relinquish personal control of the

war ministry. He maintained this position even though he knew that it was
the one non-negotiable issue with the PGE and that neither the PSOE nor
the republicans would accept a cabinet without communist participation.
Moreover, Largo was holding out in almost complete political isolation.
Behind him he had only his old retainers on the, now divided, UGT
executive. Largo had boxed himself into a political corner. As a result,
the inevitable came to pass: Largo resigned on 17 May. The question
which immediately arises, however, is why, if Largo set such store by the
war ministry, did he not fight for it by making a political deal with the
CNT - his only potential allies for such a ministerial combination? One
answer would be that Largo and his supporters were poor strategists.
But that is stretching credibility when the CNT option was so glaringly
obvious.
In fact, Largo’s inability to take the only route out of the cul-de-sac in
which he found himself obeyed a much older political rationale: that of
the entrenched organisational values of the Caballerista union bureau-
202
cracy. By the time the May crisis broke, Largo and his union old guard
were engaged in a battle with the PCE for control of the UGT whose
origins lay in the pre-war period. By spring 1937 it was also clear that the
united youth organisation (JSU) had exited the socialist orbit entirely.
This organisational conflict entirely shaped Largo’s understanding of
what was at stake during the May crisis. Araquistain had warned him on
the eve of the cabinet crisis that socialist-communist rapprochement -
and maybe even a merger - was inevitable given the international con-
and that setting his face against it absolutely would guarantee
figuration,
his own eclipse as Prieto’s star rose. 203 But the political values underpin-
ning Largo’s worldview made it impossible for him to accept that the
war effort constituted an overriding imperative, requiring certain politi-

cal alliances — such as the one the PSOE executive was actively pursuing
with the PCE. Largo saw himself as a lone champion of socialist (and
above all UGT) organisational integrity Prieto was consorting with the
‘enemy’. In Largo’s eyes, this alone was enough to debar him from the
war ministry - possession of which was pivotal to political ascendancy in
202
For more on these, see chapters i and 3 above.
203
Araquistain to Largo Caballero, 2 May 1937 in AH-26-36 (Madrid); Nenni, Laguerra de Espana,
pp. 48-9.
304 The Spanish Republic at war

the wartime Republic. It was this same ‘logic’ which prevented Largo
from making concessions to the CNT - an even older enemy in his long
war of organisational position on behalf of the UGT. 204 Hence his total
lack of response to the CNT, which, already by the early weeks of March,
was calling for a pre-emptive union alliance to head off the hostile party
forces in the cabinet. That Largo behaved thus in May 1937 tells us
something else too about his unspoken political assumptions — although
it is something deeply counter-intuitive, and thus difficult to grasp for any

observer operating with historical hindsight. For socialist organisational

integrity to have been more important to Largo than winning the war
(for this is the logical conclusion of his stance), then, at some level, he has

to have assumed that, even if the Republic lost, there would still be some
point to that ‘integrity’. In other words, Largo was scripting Francoism
be a rerun of the Primo de Rivera dictatorship of the
as if it were going to
1920s. According to this scenario, even if no democratic regime existed in
post-civil-war Spain, there would still surely be a space for a ‘responsible’
trade union — namely the UGT. Nor was Largo the only socialist leader
with union ties who would think in this way. 205 Moreover, such tacit
assumptions would become more prevalent through 1938, feeding mis-
guided ‘pro-negotiation’ currents as the Republic’s military position grew
206
more fragile.

In the last analysis, the May cabinet crisis had exposed both Cabal-

leristas and the CNT to the consequences of their political contradictions

and their inability to lay the basis of a strategic alliance over the period
1934-6. 207B Y x 937 Largo Caballero, as titular head of the ‘socialist
found himself without any solid political base. 208 But even then,
left’,

he was unwilling to offer political concessions to the CNT On its own,

the anarcho-syndicalist movement lacked, as we have seen, the organi-
sational capacity to articulate a political alternative to the Popular Front.
In fact the UGT had some similar defects. 209 But, in any case, the

204
Julia, ‘Presidente por ultima vez: Azana en de mayo de 1937’, p. 251; although it seems
la crisis
to me that Julia’s analysis exaggerates the magnitude and coherence of the ‘union threat’ to
the Republican alliance of political parties (see p. 249). On this see Graham, Socialism and War,
PP- 92-3-
205
Cf. the views of Julian Besteiro in 1939, Obras, vol. 3, pp. 435-7, cited in Preston, Comrades,
pp. 187-8. See also chapter 7 below.
206 207
See chapters 6 and 7 below. See chapter 1 above.
208
H. Graham, ‘The Erosion of the Socialist Left 1934-1937’, in Lannon and Preston (eds.), Elites
and Power in Twentieth Century Spain, pp. 127—51.
209
Itwas not designed to allow easy or regular communication ‘horizontally’ between the inter-
mediate levels of the various industrial federations - as opposed to communication at the apex,
between the national executives of each federation.
 The Barcelona May days 305

radical minority in the UGT to whom the CNT might have appealed
210
for support had never occupied the primary leadership positions in
the union hierarchy. This remained in the hands of cautious bureaucrats
whose spirit was epitomised by Largo himself. Isolated and faced by the
united opposition of his parliamentary socialist, communist and repub-
lican cabinet colleagues, Largo was obliged to resign the premiership.
He had been ousted not by a stalinist plot, as he and his supporters
would later insist, but by the entire Republican alliance. As the PSOE’s
general secretary Ramon Lamoneda would later pointedly comment:
‘Caballero always claimed to have been “kicked out by the Communists”,
which was in part true, since everyone kicked him out, from Azana
to Martinez Barrio.
5211
And the CNT stood by while this happened -
alienated by Largo’s political parsimony - a bitter reminder of the old pre-
war antagonisms. The fact that the Caballerista power base was so badly
eroded by 1937 facilitated Largo’s ejection from the cabinet. But the com-
mon purpose of republicans, socialists and communists was crucial to its
realisation.
President Azana had achieved But if Largo’s removal came
his wish.
through the collective decision of the Republican alliance, the president
was singularly more personally proactive in the selection of PSOE fi-
nance minister Juan Negrin as the new premier. That said, the range
of individuals fulfilling the minimum criteria from which the president
had to choose was not very great. Neither a republican nor a communist
was feasible: the former because he would have lacked credibility inside
Spain, the latter because not only would it have been deeply divisive
inside the Republic, but it would also have destroyed its international
campaign to secure the lifting of Non-Intervention. The new prime
minister had perforce to be a socialist. But he could not be a Caballerista
since none of them would never have accepted the subordination of the
trade unions which had long been the objective of republicans, commu-
nists and parliamentary socialists alike - redoubled after the May events

in Barcelona. Nor, in any case, would it have been easy to find someone
of the requisite calibre in Caballerista ranks. (Indeed, only Araquistain
appeared prime ministrable. But Azana would never have appointed the

210
The aspirations of a radicalised rural base after the February 1936 Popular Front elections saw
some collaboration between the CNT and sectors of the UGT landworkers’ federation, the
FNTT, in the south: Payne, Spain’s First Democracy, pp. 301-3. This political constellation would
reappear in late 1937 as part of the Caballeristas’ ongoing war for control of the UGT: Graham,
Socialism and War, pp. 198-218.
211
R. Lamoneda, ‘El secreto del anticomunismo’, unpublished post-war writings, ARLF-166-40,
p. 6 (Fundacion Pablo Iglesias, Madrid).
306 The Spanish Republic at war

man he saw as^the architect of the May 1936 crisis which had kept Prieto
212
out of the premiership at such a critical juncture. )
By a process of
elimination, therefore, the new premier had to be a parliamentary so-
cialist. The expected choice was Prieto himself- for his grasp of political
realities both domestic and international, his strategic intelligence, min-
isterial experience and sheer drive, as well as his pivotal role inside the
Spanish socialist movement and close friendship with the president. But,

as Azana himself indicated, Prieto was needed for that most crucial of
cabinet responsibilities: the war ministry. 213 Indeed, the consolidation
of land army, airforce and navy in a single ministry, thus rectifying the
anomaly pending since 4 September 1936, had not been the least of
the goals of the cabinet front against Largo in May. With all his faults,

Prieto was truly irreplaceable in the war ministry. The premiership, on

the other hand, required someone of a less emotionally volatile nature
than Prieto. needed someone who had good - or at least neu-
It also
tral - relations with all sectors of the Republican alliance. But Prieto

was at loggerheads with an important sector of his own movement. Nor

were his relations with the CNT good. Hence Azana’s preference for
214
‘the calm energy’ of Negrin. His republican convictions would not
recommend him to the CNT, but at least he was (as yet) a neutral figure
in their eyes, unlike Prieto. Likewise, Negrin, although clearly identified
with the PSOE executive and Prieto, had not been prominent in the in-
ternecine socialist conflict. Moreover,Azana and Prieto were in perfect
agreement that the multi-lingual and cosmopolitan Negrin, with a good
network of contacts in Europe, was the Republic’s international politician
215
par excellence .

But in spite of Azana’s very clear statement that he himself chose

Negrin, as well as the evident rationale behind the appointment, a great
deal of mythology has grown up suggesting that Negrin was the Commu-
nist Party’s candidate or, even more fantastically, that he was imposed
by the Soviet Union. Much of mythology derives from particular
this
and contentious prime minister
interpretations of Negrin’s policies as
in the later stages of the war and not least from the voluminous and

212
See chapter i above.
213
Negrin would himself be both premier and war minister later in the war. But that was the result
of the political crisis in the PSOE which saw Prieto, as the only other feasible incumbent, depart
the cabinet (and, in effect, active political life). See chapter 6 below.
214
Azana, Obras completas, vol. 4, p. 602.
2,5
For a biographical sketch of Negrin, see H. Graham, ‘War, Modernity and Reform: The Pre-
miership of Juan Negrin’, in Preston and Mackenzie (eds.), The Republic Besieged, pp. 163—6.
 The Barcelona May days 3°7
216
vehemently anti-communist post-war writings of Indalecio Prieto. But
the fact remains that in May 1937 Negrin was as much Prieto’s choice
as he was Azana’s and for precisely the same reasons: his ability to ar-
ticulate the Republican alliance internally, his republican convictions,
and international fluency And while it is
singular political intelligence
true that Negrin’s profilewas conducive to Stalin’s political objectives
in supporting the Republic, it makes no sense to claim this as the driv-
ing force behind his premiership: first, because of the role played by
Azana and Prieto, second, because, as we now know, Stalin had been
far from adverse to Largo’s remaining as premier and third, because,
in any case, neither the Soviet Union nor the PCE was in a position
to ‘impose’ a premier on the Spanish Republic in May 1937. Negrin
had a good working relationship with Stashevsky, the Soviet commercial
attache, with whom he was necessarily in frequent contact in his ca-
pacity as treasury minister. But Negrin’s relations with Soviet personnel
were no different from or indeed closer than Prieto’s. In fact, socialists

as diverse as Araquistainand PSOE executive member Manuel Albar

repeatedly commented on the excellence of Prieto’s relations with both
Soviet diplomatic personnel and the PCE. 217 Negrin was aware and
concerned far earlier than Prieto about the negative impact of inter-
party competition within Republican state bodies, as a result of which
he debarred members of the PCE from the reorganised Carabineros. It
was also Prieto who suggested in spring 1937 that the PSOE accept the
PCE’s proposal of merger for the sake of the war effort. 218 Negrin was
opposed (as, in the event, was the PSOE national executive - with the
exception of Jeronimo Bugeda). Had relations not been good between
Prieto and the PCE, the party would scarcely have joined the rest of the
cabinet in strongly backing him for the war portfolio, which it considered
a far more pivotal post than the premiership - as the clash with Largo
indicates.

216
From Comoy por que sail del ministerio de defensa nacional. Intrigas de los rusos en Espana (Paris, 1939)
to the anthology Convulsiones de Espana (3 vols., Mexico City, 1967-9) passing through a torrent
of similarly directed spoken and printed words in the intervening years (until Prieto’s death in
1962). Some were republished inside Francoist Spain in ‘explanatory’ editions courtesy of the
regime’s policemen-ideologues: see Toy Moscu (Madrid, 1955) (prologue, commentary and notes
by Mauricio Carlavilla).
217
Graham, Socialism and War, pp. 132-3 (see also n. 29 for Lamoneda’s comments to the same
effect).
218
Ibid., p. 132; Vidarte, Todosjuimos culpables, vol. 2, pp. 620-1; Nenni, La guerra de Espana, pp. 48-9
(n. 16), in which he also cites a characteristically pithy encapsulation by Lamoneda of Largo
Caballero’s apparent volte-face on socialist-communist unification.
308 The Spanish Republic at war

The new Negrin government emerged in mid May 1937 just as the
rebel offensive against Vizcaya was reaching its height. The then six-
week-old assault on the Republican Basque Country demonstrated the
profound change in how the rebels were fighting the war. Franco, al-
though still obsessed with taking Madrid, had been persuaded by his
German advisers in the wake of the Guadalajara debacle of the need to
expand his armed forces through mass conscription. An expanded army,
however, required equipping. Hence the campaign against the Repub-
lic in the industrial north acquired added urgency. The large army put

together by Mola was backed by air support from the small but well-
equipped Condor Legion and by Italian air units - both under German
command but ultimately at Franco’s orders. The air attacks began with
Durango on 31 March and reached their crescendo on 26 April when
the small town of Guernica, the symbolic seat of Basque nationalism,
was annihilated by three hours of saturation bombing. (It had no anti-
aircraft defences.) Franco’s key strategic target in the attack was not a
military one, however, but rather morale: Guernica was intended to kill
the Basque appetite for resistance. And in an important sense it achieved
this, along with the vast toll of human destruction. 219 Two days later, Mola
publicly linked the fate of Guernica with that of Bilbao, declaring ‘we
shall raze Bilbao to the ground’. 220 Franco himself was more concerned
to acquire its industrial capacity than to see it destroyed as a source of
moral pollution, as Mola sought. Nevertheless, in other respects Franco
was no less committed to the ‘redemption’ of the Basque region, as of
everywhere else in the Republican zone, through violence and the mass
physical elimination of his opponents. 221 Throughout April the rebels
inflicted increasing artillery and aerial bombardment against which the
defenders - virtually devoid of aircraft - could do little. The terror this
produced increased political divisions within the ranks of Vizcaya’s
defenders, ensuring the gradual collapse of resistance.
In March, as the Catalan cabinet crisis had raged over the public-order
question and the disarming of the worker patrols, the Basque president,
Aguirre, had had his own smaller-scale show-down with the CNT in
Bilbao. The ostensible issue was the ongoing dispute over production
219
From a market-day population of some ten thousand people, the Basque government estimated
the death toll at 1,645 with a further
889 injured.
220
Bilbao was first bombed on 31 August 1936 and repeatedly in September, including with leaflets
threatening a brutal bombing campaign and other reprisals if the Basques resisted: Tunon de
Lara, ‘Algunos problemas historiograficos de la guerra civil en Euskadi’, p. 134.
221
Preston, Franco, pp. 240-2, 245.
 The Barcelona May days 309
222
of the CNT’s press. On
24 March, the CNT’s print workshop was
expropriated by the Basque government and the CNT’s regional com-
mittee arrested temporarily as a securin’ measure. Basque nationalist
troops were also sent to surround garrisoned CNT troops, whom they
had been told were about to rise against the government. 223 The build-
up of tension in Catalonia had brought calls from some sectors of the
PNV for the disarming of the CNT’s battalions or their decanting into
The Basque Communist Party - to which the gov-
disciplinary brigades.
7

ernment had ceded the CNT’s printing premises - seconded these calls.
But Bilbao was not Barcelona and Aguirre was almost certainly more
concerned to discipline a marginal and (as he saw it) troublesome po-
litical force in the knowledge of the coming rebel offensive. But CNT

morale was damaged. It opted to produce its paper in Santander, where

a more liberal censorship operated. 224 And for all it was a lesser political
force in the north, CNT support was a necessary component of Basque
military resistance.
Faced by a seriously worsening military situation in the wake of the
bombing of Guernica, the Basque government formally declared the
creation of an Army of Euskadi. 225 Then, as the Italians crossed the ria
de Guernica and tensions escalated in Barcelona, Aguirre assumed
its supreme command on 1 May. This was communicated to Largo

Caballero on 5 May when the street war in Catalonia was already a

fact. The horror with which some sectors of the PNV contemplated

the unfolding of events in Catalonia - which they saw as confirming all

their doubts about the Republic - further impelled the party’s ongoing
attempts to seek a separate mediated peace with the rebels. But these
attempts proved fruidess would all subsequent ones. 226
- as
The emergence of the Negrin government on 17 May went some way
tow ards stabilising the fears of Aguirre and the PNV The new premier,
7

for all he represented mando unico also symbolised the \ictory of ,

Republican normality over the ‘dangerous classes’ of Barcelona. On the

positive side too was the appointment of Prieto to a new unified defence
ministry’. This, as w ell
appointment of Colonel Rojo, the architect
T
as the
of Madrid’s defence, as the new
chief of the general staff, overcame at
least some of Aguirre’s reservations about accepting military instructions

222
Chiapuso, El gobiemo vasco y
los anarquistas, pp. 117, i3off.
223 224
Ibid., pp. 141 -8. Ibid., p. 194.
225
By decree published in the Diario Oficial del Pais Vasco, 26 April 1937.
226
De Meer, El Partido Xacional Vasco, pp. 410—550.
 ,

3 10 The Spanish Republic at war

from the centre. Prietoappointed General Mariano Gamir Ulibarri to

command the newly formed ‘Army Corps of the Basque region’ (Cuerpo
del Ejercito del PaisVascongado ). 227 Although the avoidance of the word
‘Euskadi’ is to be noted, this measure nevertheless recognised that there
was no such thing as an overarching Army of the North’. Henceforward
Llano de la Encomienda would command the forces in Santander and
Asturias, while Gamir - to whom Aguirre handed over command -
oversaw those of Vizcaya. Relations were improving now that Vizcaya
was the object of serious strategic attention from the Republican chief of
staff. Gamir arrived to take up his command at the end of May. The con-

version of the Basque battalions into Mixed Brigades got underway and
Aguirre’s former reluctance to accept political commissars was over-
come. In spite of the Basque units’ retention of their specific political
228
identities, their performance improved But it was probably already
.

too late. Moreover, there was an ongoing and increasingly anguished

defend Bilbao.
conflict over the lack of aircraft to
Although this dearth of air support provoked from the isolated,
besieged Basques angry accusations of treason and would subsequently
give rise to a bitter polemic over responsibility for the fall of Vizcaya
between Largo, Prieto and Aguirre 229 the fact was that the Republic
,

did not have the requisite planes to defend the north against the rebels’
plentiful German and Italian air support. (Indeed, the German aircraft
in the north were the best ones deployed throughout the entire war.) The
situation was compounded by a variety of other factors: inadequately
functioning communication channels, some practical incompetence
and competing needs (as well as rivalries) on the centre front. But the
main problem remained the lack of both planes and pilots. While Largo
had passed the responsibility on to his military command, Prieto made
desperate personal attempts to find more air cover. But often the planes
procured never made it to their destination. Non-Intervention meant
that some were impounded in France while others crashed, turned back
or went off course because of bad weather or technical insufficiencies.
Modifications were made to increase flight capacity in order to allow

227
On Gamir, see appendix on army officers in Alpert, El ejercito republicano p. 371.
228
This may have been attributable in significant part to the withdrawal of Aguirre from military
command responsibilities: Thomas, Spanish Civil War, p. 687. Certainly the other components
of the Vizcaya and northern Popular Fronts were unimpressed by Aguirre’s abilities in this area.
229
J. A. Aguirre, Informe del Presidente Aguirre algobierno de la Republica sobre los hechos que determinaron el
derrumbamiento delJrente del Norte (Bilbao, 1978), especially, pp. 353-62 (for exchange of telegrams
between Largo, Prieto and Irujo). The loss of the north was also grist to the mill of the internal
battle in the PSOE: Graham, Socialism and War, pp. 134, 279 (n. 35).
 1

The Barcelona May days 31

direct flights from the central front to Vizcaya, thus avoiding the perils
of the stop-over in French territory. But even so, it is unlikely that many
arrived in the north. 230
Under such conditions, Bilbao’s defenders stood little chance. They
had no means of inhibiting the technique of ground attack from the
air being developed by Franco’s Condor Legion strategists. During the
second half of May, the rebels closed the siege ring around Bilbao. During
the first days of June, Condor Legion planes and artillery bombarded
the ‘Iron Ring’ - the double ring of trenches cut into the hills outside the
city. Even apart from the fact that the structural plans had earlier been
betrayed to the rebels, these fortifications were in any case inadequate.
Uncamouflaged and lacking in depth, they had been built by engineers
and architects with purely civilian experience. Moreover, the defenders
were up against the most powerful guns ever deployed by the Germans
in Spain. Anti-aircraft guns - superfluous in view of the Republic’s lack
of planes — were also deployed as light field artillery. The trenches were
pulverised by tracer shells and the defence ring breached on 12 June.
With the memory of Durango and Guernica still fresh, the Basque
government made a strategic decision, in consultation with the mili-
tary command, to evacuate Bilbao and continue the fight from other
231
fronts. But this was also a key political decision driven by the PNV’s
desire to protect Basque capital and property. This was further demon-
strated by Aguirre’s absolute refusal to allow the implementation of a
scorched-earth policy. Prieto’s orders to destroy heavy industrial plant
were ignored and gudaris posted to protect such installations (armaments
and explosives factories, steel-making plants, shipyards and heavy engi-
neering works) in case other groups, most notably the Basque CNT, tried
232
to implement Prieto’s instructions. For a week Asturian, Santander

230
The question of aircraft numbers in the defence of Vizcaya is a fraught one with complex and
fragmentary sources that do not agree: Tunon de Lara, ‘Algunos problemas historiograficos de
la guerra civil en Euskadi’, pp. 141 —3. While Howson does not provide a specific breakdown
for the Basque campaign, his overall assessment/ figures confirm the dearth: Howson, Arms for
Spain, pp. 141-2, 209-10, 212-13, 234-5, 255-7, 302-3. (Cf. Tunon de Lara’s (unreferenced)
estimate of a rebel air strength tenfold the Republic’s and of one hundred planes sent from the
central (Madrid) front of which only one-third arrived.) It must be borne in mind also that the
Republic could not afford to uncover the central front.
231
The rebels had feared a repeat of the siege of Madrid in Bilbao: Preston, Franco, p. 280.
232
Zugazagoitia, Guerra y vicisitudes de los espanoles, p. 312; Ghiapuso, El gobierno vasco y los anarquistas,
pp. 215-17; Ambou, Los comunistas en la resistencia nacional republicana, p. 147. (Plant was later
destroyed in Santander, but Ambou remarks that much was not in Asturias - if for different
reasons, Ambou, p. 236); Tunon de Lara, Algunos problemas historiograficos de la guerra civil

en Euskadi’, p. 143 (the Basque police ( ertzainas) also guarded churches, other religious buildings
and prisons).
312 The Spanish Republic at war

and Basque unks retreated westward - and thence by boat to France, or

direcdy to Santander and Gijon to carry on the fight. On 19 June the
rebels entered Bilbao without opposition.
The loss of Vizcaya redoubled Negrin’s determination to impose a
uniform Republican order on the rest of the zone. First in the firing
line was Aragon. Whereas Largo Caballero had implictly legitimised
the Council of Aragon by engaging in a discussion of its sphere of in-
fluence and competences, the highly centralist Negrin simply ignored
it, refusing to answer any requests for ‘clarification of status’. Although

the Council had agreed in June to abolish its public-order section (in
order to demonstrate an acceptance of government authority here), it
was accused by the Popular Front parties of not fulfilling this promise.
The Negrin government’s first real ‘communication’ with the Council
was to send in troops in August to reinstate the machinery of municipal
government.
The central government’s target here was not rural collectivisation/^/*
se, but the dissolution of the Council’s political authority and the destruc-
tion of the organisational sinews of CNT power in the region: ‘the moral
and material needs of the war imperiously demand the concentration of
authority in the hands of the state’. 233 Raison de guerre was here reinforc-
ing the government’s underlying beliefs - that neither political power
nor the deployment of significant economic resource should be in union
hands and that privatised agriculture should be afforded a greater de-
gree of material protection. 234 Hundreds of cenetistas were imprisoned. 235
In allowing the departure of reluctant collectivists, the dissolution did,
however, constitute an important factor in the growing crisis of the col-
lectives. But it is important to realise that, just as the Council’s political
crisis (sparked by the June public-order question) was partly internal, 236

so too was the anti-collectivist polemic. When the central government

sent in the army, this activated and potentialised the Council’s political
and social opponents inside Aragon.

233 Text of dissolution decree, cited in Fraser, Blood of Spain p. 390. Fraser himself has made the
,

point eloquently that much Aragonese collectivisation continued to function after the enforced
dissolution of the Council: P. Broue, R. Fraser and P. Vilar, Metodologia historica de la guerray la
revolucion espanolas (Barcelona: Fontamara, 1982), p. 125.
234 Cf. the CLUEA dispute in Valencia. The Negrin government clashed with the unions over a
political principle as much as over their ability to deliver a centralised operation and the foreign
exchange derived therefrom: Bosch Sanchez, Ugetistasy libertarios, pp. 382—3.
235 Many of the Council’s political office holders as well as those accused of crimes against private
property were still in prison when Franco began his final offensive against Aragon in March
1938: Casanova, De la calle al/rente, p. 233.
236
Casanova, Anarquismoy revolucion , pp. 258, 264—71.
 3

The Barcelona May days 31

Once government troops had been sent into Aragon, the destabilisa-
tion of CNT authority created local political opportunities for others. 237
After the harvest had safely been gathered in, the Left Republicans en-
dorsed the Council’s dissolution - in accordance with Azana’s great
wish - when the Aragonese Popular Front met in Barbastro. Having
lost out to the PCE where Aragonese peasant smallholders were con-
cerned, 238 Left Republicans now competed to acquire the CNT’s clien-
tele in the contest for political influence within the region’s new municipal
structures. For its part, the PCE was able to maintain a political influence
via both thearmy and the post-Council influx of Ministry of Agriculture
personnel to register the collectives and assess the status and needs of
production in the region. 239 In the latter respect, this process mirrored
what had occurred earlier and on a bigger scale in the Valencia region
as small farmers and more affluent constituencies had from autumn
1936 looked to reconstructing state agencies like the Institute of Agrar-
ian Reform (as well as to the PCE’s Provincial Peasant Federation (FPC))
as a means of mobilising against what they perceived as the threat of
collectivised economic forms. 240 But whatever the specific colour of the
politics, in Republican eastern Aragon, unlike more developed Valencia,

it was the war itself that had produced the beginnings of mass politics

proper.
The real nature of the Council’s social base remains difficult to ascer-

tain since arms were used to build it and also to dismantle it - and in
both cases those arms came from outside Aragon. The net effect of the
Council’s economic intervention had been redistributive. But both its
quasi-governmental functions and the element of implicit if not explicit
duress involved in some of the collectivisation process made the Coun-
cil controversial both within the CNT and beyond. In the last analysis,

it was born in, and of, an emergency, and operated in such difficult

237 On post-Council politics, see ibid., pp. 264-97; Fraser, Blood of Spain, pp. 390-4.
238
Casanova, Anarquismoy revolucion, p. 221.
239 The Agricuture Ministry’s Institute of Agrarian Reform (IRA) required registration as a condi-
tion of legalisation. But this process would be obstructed by the activation of the military front
in late 1937 and ended by the rebels’ final offensive in Aragon in March 1938.
240
In Valencia it was the consolidation of government power which activated the regional con-
flict over collectivisation: Bosch Sanchez, JJgetistasy libertarios, pp. 39-49. The promulgation of

normalising legislation from August 1936 onwards - and especially the decree of 7 October so
favourable to smallholders and tenant farmers - provided anti-collectivist social constituencies
with ammunition. Many of the peasants recruited by the PCE’s Provincial Peasant Federation
(FPC) had before the war been attached to Catholic unions linked to the Valencian Regional
Right (DRV) - a component of the mass Catholic party CEDA. By the end of 1936 there were
explosive social and political conflicts between ‘individualists’ and ‘collectivists’ in villages across
the Valencia region: Bosch Sanchez, Ugetistasy libertarios, p. 122.
3 14 The Spanish Republic at war

conditions and*was so short-lived that it is virtually impossible properly

241
to measure it against the revolutionary schemas it enshrined . What
is clear, however, is that the Republicangovernment used the evidence
of local conflicts in Aragon - both of a class nature and otherwise —
242
to justify its own intervention . Thus, in thename of the war effort,
republicans, parliamentary socialists and communists together brought
the last independent stronghold of regional particularism in Republican
Spain under the control of the central state.

But there was also another crucial factor which explains the precise
timing of the Aragonese action: the need to activate the eastern front
in an attempt to relieve the pressure of all-out rebel offensive against
beleaguered Asturias - now all that remained of the Republican north.
Since the start of the war, the rebels had kept only a bare minimum of
troops on the Aragon front. As a result, the Republicans concentrated
on simply maintaining their own lines. (This, aside from any political
agenda, was the key to the poverty of arms among the militia. The
Republic simply could not afford to divert scarce weaponry to a super-
fluous front - and for the first year of the war, that status was defined
by the content of rebel war policy.) But if the Republic now took the of-
fensive and attacked in the east, then Franco would be forced to divert
resources from the north. This, however, required the prior full military
integration of CNT and POUM militia forces. But these forces had only
begun to accept the authority of the Valencia war ministry in March 1937
and had a distinctly over-optimistic sense of their military capacity pre-
cisely because it had never been tested. The move into Aragon in August

1937 was thus a means of belting and bracing the militarisation of the
front. 243
A scarcely less important objective was to increase agricultural produc-
tion in Aragon. With central political control in place, it was hoped that
this could also be used to mitigate the acute problem of feeding the pop-
ulous Republican cities - and, above all, Barcelona with its huge refugee
population, now
being further increased by the influx from Asturias.
Implicit here was the assumption that collectivisation had diminished
productivity by its inefficiency and demoralisation of smallholding con-
stituencies. In fact, the balance sheet for collectivised wartime production

241 242
Casanova, Anarquismoy revolution, pp. 319-20. Ibid., pp. 258-60.
243 Alpert,
El ejercito republican, pp. 82-3. The militarisation of all Catalan-originated militia in
easternAragon was completed by the end of April 1937 — except for the POUM’s 29th division,
which was reorganised as a result of the political action against the POUM in summer 1937:
Casanova, Anarquismoy revolution, p. 114.
 The Barcelona May days 3*5

in Aragon was far from obviously negative in 1936-7. 244 But eastern
Aragon had never contained territory of sufficient yield for it to serve as
an adequate emergency granary for urban Republican Spain - irrespec-
tive of the mode of production therein employed. 245 Moreover, as the
war came to the eastern front, so the conditions of production deterio-
rated along with the supply of manpower and availability of transport.
-
In the end, it was the war causing economic dislocation and the shrink-
age of markets - rather than political opponents or state action which
was the greatest single eroder of collectivised agriculture as it was of
the privatised variety. In Aragon, as elsewhere, war under conditions of
Non-Interventionist embargo would erode the material fabric and the
psychological resilience of collectives and collectivists as it would all other
facets of the Republican home front.

244 See Ministry of Agriculture figures for comparative regional productions in 1936 and 1937,
reproduced in Thomas, Spanish Civil War, p. 559; Bosch Sanchez, Ugetistasy libertarios, pp. 378—9.
Production in 1936—7 went down in both Catalonia and Valencia - where private peasant
agriculture was more widespread.
245 In the pre-war period, the three provinces of Aragon had in any case only provided 7 .6 per cent
of Spain’s total cereal crop. Over half of that 7.6 per cent had come from Zaragoza province -
of which the most productive sector was in rebel hands: Fraser, Blood of Spain , p. 348.
 CHAPTER 6

Negrin’s war on three fronts

A long time, even for heroes ... 1

When Negrin took over as prime minister in May 1937 his primary objec-
tive was to procure the lifting of Non-Intervention as the greatest obstacle
to the Republic’s active prosecution of the war. Within a month, the loss
of industrial Vizcaya had increased his urgency. Negrin was the ideal can-
didate to carry the Republican war effort onto this crucial third ‘front’ in
Europe’s diplomatic arena - while also simultaneously maintaining the
military front and the home-front mobilisation that underwrote it. But
the start of his premiership coincided with the end of Blum’s in France
and Baldwin’s in Britain. Neither development favoured the Republic.
But, as always, it was Britain’s position on Non-Intervention that held
the key to its predicament.
The majority opinion in the British cabinet was certainly one of pas-
sive pro-Francoism. 2 But the crux of the problem for the Republic was
the strategic role played by Non-Intervention within Britain’s overall
foreign policy. From the end of May 1937, under new prime minister
Neville Chamberlain, the cabinet was steering hard for rapprochement
with Italy. Britain’s remained the defence
pivotal foreign policy goal
of its and a modus vivendi with Italy in the
extensive imperial interests,
Mediterranean was judged the best means of securing it. To this end,

1
Milton Wolff, Another Hill (Urbana/Chicago: University of Illinois Press, 1994), p. 1 19.
2
E. Moradiellos, ‘The Gentle General: The Official British Perception of General Franco during
the Spanish Civil War’, in Preston and Mackenzie, The Republic Besieged, pp. 1 -19. When Madrid
had been besieged in autumn 1936 there had been a favourable swell of opinion in the British
cabinet for granting the rebels belligerent rights. The foreign secretary,Anthony Eden, had
overcome this only with difficulty. By the time of the blockade of Bilbao in late March and early
April 1937, however, Eden had less difficulty in containing the staunchly pro-rebel First Lord of
the Admiralty, Samuel Hoare, when he called for foodstuffs en route to the Basques to be formally
classified as war material (although, in fact, the rebels were treating food as suchanyway and
blockading the port as had been granted belligerent rights.) Eden himself wanted the Royal
if they

Navy to escort British merchant ships through to Bilbao: Alpert, A New International History of the
Spanish Civil War, p. 122.

3 l6
 Negrin’s war on three fronts 3 17

Non-Intervention offered a useful diplomatic framework in which to sta-

bilise relations with Italy while attempting the sought-after alliance. The

massive increase in Axis aid to Franco after March 1937 would cause
some anxiety. But British policy makers remained confident that the
power of sterling and, failing that, of the Royal Navy to blockade would
ensure a ‘friendly’ rebel Spain even in the worst-case scenario. Given this
reasoning, then, for the majority of the British cabinet the machinery of
Non-Intervention was a valuable resource - irrespective of Italy’s per-
sistent flouting of it. When Italian action was seen as directly damaging
British interests, such as in the sinking of merchant shipping by Italian
submarines in 1937, steps would be taken to resolve matters beyond the
framework of Non-Intervention. 3 Britain’s persistent diplomatic utilitar-
ianism where Non-Intervention was concerned would, thus, prevent the
consolidation of the Republic’s defence possibilities in 1937.
From early on in the war both the Italians and Germans had continu-
ally attacked shipping bound for Republican ports, even though they had

no authority so to do. 4 As most aid came by sea, by 1937 these attacks

were, from the start, a major threat to the Republic’s ability to sustain
a military defence. Once the Komsomol was lost on 14 December 1936,
the Soviet Union - whose merchant marine was small and whose vessels
were dispersed across far-flung seas (the Caspian, Aral, Arctic/White,
Baltic and Black Seas and the Ocean) - required the Repub-
Pacific
lic to find other sources of transport for war materiel it procured on the

Spaniards’ behalf. 5 The Republic had begun the war with around three-
6
quarters of the Spanish merchant fleet. But it suffered heavy losses,
which meant that it could not compensate for the lack of Soviet or other

3 The Nyon Conference of September 1937 was convened to discuss the problem of ‘unknown’
submarines interfering with British and other shipping. Italy was invited to join even though it
was an open secret that the submarines in question were Italian: Alpert, A New International History
ofthe Spanish Civil War, pp. 142-3. After Nyon, the attacks on British shipping more or less stopped.
4 Only belligerent rights permitted the intercepting of merchant shipping on the high seas.
5 Howson, Arms for Spain, pp. 130-4 (the Komsomol was unlikely itself to have been carrying arms);
D. T. Cattell, Communism and Spanish Civil War, p. 77. Howson gives comparative details on
the

size and fleet tonnage, pp. 133-4. Recourse to an atlas also reminds one of how the immensity of
Russia made inordinately complex the task of fleet coordination across all the separate seas.
6
The Republic nationalised those merchant ships in its ports at the start of the war. But vessels
at sea were often permanently lost to it, as were those docked in Germany, Italy and the USA,
where the rebels and their supporters mounted legal challenges over ownership. There is a lack
of clarity still over the size of the Spanish merchant marine. See F. and S. Moreno de Alboran
y de Reyna, La guerra silenciosay silenciada. Historia de la campaha naval durante la guerra de 1936-39
(4 vols., Madrid: Graficas Lormo, 1998) and R. Gonzalez Echegaray, La marina mercantey el trafico
maritimo en la guerra civil (Madrid: San Martin, 1977). It was in the region of 880 vessels, of which
the Republic had 660 and the rebels 220. I am grateful to Gerald Howson for his help with the
technical and quantitative material relating to naval strength.
318 The Spanish Republic at war

vessels. Enemy and — to a lesser extent — difficulties relating to

action
inexperienced crew and the lack of a proper command structure eroded
Republican shipping. By contrast, the rebels managed to preserve most
of their smaller share of merchant vessels. But, more importantly, the
German government sent ships for their use which sailed under flags
of convenience (in particular the Panamanian) and were thus beyond
the reach of the any case very short-lived) Non-Intervention naval
(in

patrols of Spanish coasts between April and June 1937. The Republic
had neither the resources nor the contacts to do likewise - at least not
on anything like the same scale. In addition, vessels bringing war materiel
to the rebels from Italy were openly escorted by naval warships.
By the late summer of 1937 rebel and Axis naval aggression had vir-

Mediterranean as a supply route for the Republic.

tually sealed off the
Access via the French frontier thus became vital to its survival. The
Radical-led government in France that had replaced Blum’s in June
was continuing to operate a policy of ‘relaxed Non-Intervention’ (‘Non-
Intervention relachee’). 7 Nevertheless, this border policy was erratic. And
even when the aid did pass, it constituted ‘the discreet smuggling ... of
small quantities of war material and of civil aircraft in ones and twos’ —
8
which was very far from offering a basis to turn the war around. For
as long as Non-Intervention existed, the Republic was, in reality, con-
demned hand-to-mouth existence.
to a
But in spite of the enormous, gathering question mark over the fea-
sibility of maintaining - still less augmenting - war materiel Negrin ,

realised from the start that the political-diplomatic climate made it im-
perative for the Republic to move onto the offensive. It needed a clear
military victory to demonstrate that it could hold its own (if not more)
and that Franco was not heading inexorably for victory. Unlike President
Azana, who from spring 1937 was convinced that the most that could be
achieved was a negotiated end to the war via international mediation,
Negrin would throughout 1937 act upon his belief that the Republican
war effort was a going concern. 9 If Non-Intervention could be ended
and foreign volunteers withdrawn, then it might even be possible for the

7 Blum remained as vice-premier (until January 1938), as did Cot in the air ministry.
8
Howson, Armsfor Spain , p. 233.
9 For the differences of emphasis between Azana and Negrin in 1937, see R. Miralles, ‘Paz hu-
manitaria y mediacion internacional: Azana en la guerra’, in A. Alted et al. (eds.), Manuel Azana:
pensamientoy accion (Madrid: Alianza, 1996), pp. 263-8 and R. Miralles Juan Negrin: al frente de la
de la Republica (1937-1939)’, Historia Contemporanea, 15 (1996). These differences
polltica exterior
would begin have serious political consequences from the military crisis of spring 1938 when
to
the Republican zone was split into two - see below in this chapter.
 .

Negrin’s war on threefronts 319

Republic to win. But Negrin understood too that any negotiated settle-
ment would also demand a resilient Republican war machine: because
only from a position of strength could he force the rebels and their
backers to accept something other than unconditional surrender. Partly
with this in mind the Republic mounted the Brunete offensive in July
on the western Madrid front. Probably the bloodiest single battle of the
10
war, fought in the extreme heat (with temperatures of over 100 degrees
in the shade), this bid to raise the siege of the capital was doomed to
failure by the Republic’s lack of offensive capacity - to which hold-ups
of material at the French frontier at a crucial moment contributed.
Even when aid did arrive in Republican Spain, the fundamental point
to grasp is that its very intermittence as well as its variable quantity and
quality (often unascertainable in advance, frequently different from the
specifications and always so modey as to be the despair of the command 11 )
made it virtually impossible for the Republic to plan very far ahead
ahead or to sustain such actions as it undertook. The year 1937 did
see some improvement in terms of the stabilisation of military fronts
and the building up of an indigenous arms industry in Catalonia. But
domestic arms production came nowhere near to meeting the needs of
12
the army. For as long as Non-Interventionist embargo existed, there
could therefore be no overall Republican stabilisation. 13 Indeed, not even
the lifting of Non-Intervention would have put the Republic on an equal
footing with the rebels. Neither Britain nor France nor the Soviet Union -
re-armers all - would have offered the kind of integrated state-backed
military aid such as the rebels continued to be assured of from the Axis,
and But ending Non-Intervention would, nevertheless,
especially Italy.
have given the Republic access to sufficient sources of arms properly to
fight an offensive war. 14
During the summer and autumn of 1937 the full weight of the
Republic’s diplomatic offensive to reverse Non-Intervention was focused
10
In terms of the proportion of casualties to overall participants, on both sides.
11 **
Howson, Armsfor Spain , pp. 250-1, note
12
Once the improvised Catalan war industry was up and running (by autumn 1936), it produced
6 million cartridges per month until the end of the war in Catalonia. But Mussolini could
authorise dispatch of 10 million in a single month and, overall, provided 319 million for the
rebels: P. Preston, ‘Italy and Spain in Civil War and World War’, in S. Balfour and P. Preston

(eds.), Spain and the Great Powers (London: Roudedge, 1999), p. 173. Catalan industry (which by
spring 1937 was effectively the only Republican war industry) could muster litde over half of this
number. Borja de Riquer, unpublished paper on the Catalan war effort, Escuela de Verano, San
Lorenzo del Escorial (Madrid), August 2000. See alsoj. M. Bricall, Politico econdmica, pp. 56-72.
13
Cf. Cattell, Communism and the Spanish Civil War, pp. 76-7.
14
A. Vinas, ‘Las relaciones hispano-francesas, el gobierno Daladier y la crisis de Munich’, in
Espaholesy franceses en la primera mitad del siglo XX (Madrid: CEH-CSIC, 1986), p. 179.
320 The Spanish Republic at war

on France. 15 Anxieties French government and diplomatic corps

in the
over defence vulnerability were increasing in direct proportion to escalat-
ing Axis aid to the rebels and Italian influence in the Mediterranean. At
the same time, Britain’s unilateral policy of rapprochement to Mussolini
made France feel doubly isolated. As a protest, in July the French
cabinet suspended the service of international observers on its land

frontier withSpain and exerted diplomatic pressure on Britain to end

Non-Intervention. A diplomatic window of opportunity for the Spanish
Republic seemed to be opening. French pressure on Britain led to a pro-
posal to Italy for the withdrawal of all foreign volunteers from both rebel
and Republican zones in return for recognition of limited belligerent
16
rights.

The Soviet Union was also inclined to consider such an exchange. The
outbreak of the Sino-Japanese War in early July, and its rapid escalation,
was the first of several developments which, by 1938, would displace
Spain from its privileged position in Soviet foreign policy. 17 Already by
mid 1937 Stalin may well have been coming to believe that impossible
odds were being stacked against a Republican victory. 18 But its continued
resistance remained, nevertheless, a significant component in his strategy.

Axis energies were, for the time being, tied down in Spain, and Nazi
Germany thus deterred from any attack in the east. Moreover, even Non-
Intervention itself - which the Soviet Union had consistently opposed
once it became clear that it was not an efficient means of curbing aid to

15
This initiative was carried forward by Negrin himself, his foreign minister Giral and also by
the Republican ambassador to Britain, Pablo de Azcarate, who transferred his attention from
London to Paris: E. Moradiellos, ‘Una mision casi imposible: la embajada de Pablo de Azcarate
en Londres durante la Guerra Civil (1936-1939)’, Historia Contemporanea, 15 (1996), 135!!
16
E. Moradiellos, Laperfidia de Albion. Elgobierno britanicoy la guerra civil espanola (Madrid: Siglo XXI,
1996), p. 184.
17
Although had always been a relative privileging - given the primacy of domestic policy for
this
the Soviet leadership up to 1939: Roberts, ‘Soviet Foreign Policy and the Spanish Civil War’,
in Leitz and Dunthorn (eds.), Spain in an International Context (1936-1959)
P- 9 1 Soviet foreign -

policy seems to have been made informally rather than in cabinet or politburo meetings. The
Soviet bodies involved in Spain (defence, foreign affairs and Comintern) did not liaise with each
other but were all individually subject to general political guidance by ‘a group of functionaries
around Stalin who came
together in Moscow from time to time to discuss foreign policy issues’.
This policy-making group included Molotov, Ezhov, Voroshilov, Litvinov and Dimitrov: Schauff,
‘Hitler and Stalin in the Spanish Civil War’, p. 18.
18
Although this was not a clear position, cf. the questions being asked in an internal Comintern pol-
icydocument dated 9 September 1937: ‘[h]ow can we achieve the break-up of the bloc currently
standing against Republican Spain, and how can we achieve a change in the policies of the demo-
cratic countries? Which kind of concessions can we make to England and France concessions . . .

compatible with the existence of the Spanish Republic, in order to change their behaviour? We
do not have any clear thought in this matter’, Schauff, ‘Hitler and Stalin in the Spanish Civil
War’, pp. 13, 14.
 JVegrin’s war on threefronts 32

the rebels - had its diplomatic uses. By maintaining a tension in British

relations with the Axis (noth withstanding Chamberlain’s bid for Italian
rapprochement) it impeded any putative four-power concert against the
USSR - the possibility of which Stalin continued to fear. 19 But irrespec-
tive of Spanish Republican resistance, the Soviet Union could rule out
neither four-power concert nor German aggression definitively. Provi-
sion for such an eventuality had to be made. So while Negrin continued to
gear his diplomacy to protecting the Soviet lifeline, what Stalin probably
had in his sights was ensuring the Republic’s sufficient resistance capacity
while at the same time reducing the call it made upon war material that
was domestically produced in Soviet factories. Production levels in heavy
industry for the period 1937-9 would be badly affected by the purges and
by industrial reorganisation (a proportion, but not all, of which resulted
from the purges). In some cases heavy industrial production in the Soviet
Union did not reach 50 percent of the projected targets. 20 In an attempt
to reduce Republican demand, by October 1937 Stalin was prepared to
accept the British proposal of conceding limited belligerent rights to the
contenders in Spain in return for an agreement on the withdrawal of
foreign troops (even though Soviet diplomats thought it madness that
Britain was prepared to concede such rights to Franco, given the potential

19
Hence the Soviet Union’s hostile reaction at the end of May 1937 to Prieto’s wild-card suggestion
that, in response to the German shelling of Almeria (a virtually undefended port) the Republic
should declare war on the aggressor in order either to force German withdrawal or to provoke
a European war: Alpert, A New International History of the Spanish Civil War , p. 14 1; Bolloten, SCW,
-
PP- 574 5 -

20
Ex-NKVD chief Orlov apparently claimed that a reconfiguration of Soviet policy was in any
case made explicit to relevant Soviet personnel as early as summer 1937: Payne, The Spanish
Revolution p. 274; D. T. Cattell, Soviet Diplomacy
,
and the Spanish Civil War (Berkeley, Calif.: University
of California Press, 1957), p.115. (We still have no direct documentary evidence from NKVD
sources.) E. H. Carr suggests a rather less abrupt or explicit shift, The Comintern and the Spanish Civil
War (London: Macmillan, 1984), p. 51 (n. 22). Certainly the information we have on Soviet arms
indicates the latter (Howson, Arms for Spain appendix 3, pp. 278-303). Frank Schauff suggests
,

that, in terms of overall domestic production, the proportion of Soviet war material directed to
Spain between 1937 and 1939 was high - especially for planes (in the light of Howson’s figures).
Soviet propaganda itself obviously played a major role in shaping the Republicans’ belief that the
Soviet Union was stronger than it was, but holding back. As the logic of this analysis indicates,
there is absolutely no sense in Burnett Bolloten’s wild suppositions that the Soviet Union was
exporting industrial goods and plant machinery from Republican Spain (see Bolloten, SCW,
p. 581, n. 32 and endnote text p. 905). In extrapolating backwards from an eastern European
scenario post 1945, Bolloten is leaving out of account Soviet policy priorities up to 1939. But
even if these had not existed and Republican Spain had been producing something sufficiendy
industrially advanced to be of interest (which it was not), there were neither ships to carry it nor
a viable route to take. The Soviet Union’s imports from Spain were not great and, in the main,
consisted of citrus fruit. But, again, once the Mediterranean became impassable (in August 1937)
even this was highly problematic. ‘Even in the sphere of foodstuffs the Soviet Union had to supply
the Spanish Republic’, Schauff, ‘Hider and Stalin in the Spanish Civil War’, pp. 14, 16.
322 The Spanish Republic at war

impact on its merchant shipping). The granting of belligerent rights

would not have resolved the Republic’s arms procurement problem,
as the British plans proposed the maintenance of Non-Intervention
controls. But it would, crucially, have allowed the Republic properly to
defend its war materiel en route to Spain. Less wastage through sinkings
and impoundings would clearly have been in Stalin’s interests, and there
were sufficient Republican warships to serve as escorts in Mediterranean
21
waters.
But the volunteer withdrawal proposals were stalled endlessly. Neither
Mussolini nor Hitler was prepared to decrease assistance to the rebels.
Indeed, quite the opposite: what had underlain their withdrawal from
the Naval Patrol in June 1937 was the realisation that there could be
no certainty of a rebel victory unless they injected more aid. The with-
drawal of foreign combatants from Spain would have seriously damaged
Franco’s fighting capacity, but the Republic’s hardly at all by mid 1937.
As it was, however, Britain accepted the Italian insistence that the pro-
posals for troop withdrawal should be discussed in the Non-Intervention
committee rather than by the League of Nations. This was the equivalent
of guaranteeing a massive delay. The Soviet Union had thus to maintain
a certain level of aid. France, on the other hand, confounded by Britain’s
acquiescence, continued to resist pressure to close its border.
It is unlikely, however, that the League of Nations route would have
yielded more. (Indeed, Azana’s apparent faith in the League’s ability to
oversee a viable troop withdrawal scheme sits oddly with his scepticism
over Negrin’s attempts to draw France away from British influence.)
Negrin’s resolution of 18 September 1937 to the League that foreign
troops should be withdrawn from Spain, or else Non-Intervention lifted,
was duly approved. But the weight of absentions, including from the key
European players, ensured that it remained just a resolution. Negrin con-
tinued to work on the French as the weak link in Non-Intervention. But
little was achieved. Border restrictions remained, and the noises over a

French military expedition against Italian-held Majorca subsided along

with French resolve. 22 Inevitably, Franco continued to procrastinate
over the British plans for troop withdrawal. With the help of Germany
and Italy, seemingly endless objections were raised. The debate would
21
Franco’s navy had more cruisers and submarines, the Republic’s more destroyers: Gero to
Dimitrov, 19 Nov. 1938, Radosh, Habek and Sevostianov, Spain, Betrayed. The Soviet Union in

the Spanish Civil War

505 (doc. 80).
, p.
22
Negrin’s request for French officers to reinforce the Republican army remained as a French blind
eye to the recruitment of reserve officers: Miralles, ‘Juan Negrin al frente de la politica exterior
de la Republica’, p. 151; Azana, ‘Cuaderno de la Pobleta’ (27 September 1937), Obras completas,
vol. 4, p. 806.
 Negrin’s war on threefronts 323

effectivelybe frozen until the spring of 1938. The measure of the Re-
public’s diplomatic failure was that it could achieve neither movement
on troop withdrawal nor the lifting of Non-Intervention.
Meanwhile, the fall of the remainder of the Republican north threat-
ened. The Brunete offensive of July 1937 had been intended to take
pressure off the north. But though it retarded the fall of Santander (as
Franco had to divert troops 23 ), it could not prevent collapse. Santander
was taken by the rebels on 26 August, whereupon they opened their final
northern offensive against Asturias. The Republic tried a further military
diversion by activating the eastern (Aragon) front. But the underlying
pattern of Brunete was repeated: a temporary success when the Republi-
can forces took the towns of Quinto, and then Belchite on 6 September, 24
then a failure to maintain the momentum. The capture of the two towns
entirely exhausted their resources. As usual the Republican forces were
gravely handicapped by their lack of hardware, or at least appropriate
hardware. It is also the case that the knowledge of lack of materiel had a psy-
chologically inhibiting effect on the Republican command. 25 But there
were also other sorts of troop and logistical insufficiencies in Aragon: on
a front that had been inactive for months, intelligence as obvious as the
26
location of enemy fortifications was still unavailable. By 19 October
Gijon had fallen, and with it what remained of Asturias was rapidly
lost.

There were other factors too complicating the Republic’s prosecution

of the war. Not the least of these was the abiding tension between the
civil and military authorities that stemmed back breach of trust
to the
caused by the military coup itself. The enduring suspicion of the profes-
sional military acted as an insidious and massively destructive pressure,
coming as it did on top of the eternal batde against inadequate resources
and undertrained personnel. This was one of the crucial ways in which
the military rebellion would go on working, indelibly, against Republic
long after its immediate organisational effects had been contained. 27 It

23 Both rebel and Republican lines were always thinly manned outside the immediate batde areas -
quite simply because the human resources available did not afford more.
24
The towns were taken street by street and house by house and involved bitter hand-to-hand
fighting which is described in Milton Wolff’s autobiographical novel, Another Hill pp. 69-86. ,

The journalist Herbert Matthews described parapets of corpses and characterised Belchite after
its capture as ‘a fetid mass of wreckage’, Two Wars and More to Come (New York: Carrick S. Evans,

1938), PP- 301-10; see also de la Mora, In Place of Splendor pp. 336-7.
,

25 Rojo to Prieto, 28 September 1937, letter reproduced in J. I. Martinez Paricio (ed.), Los papeles
del general Rojo (Madrid: Espasa Calpe, 1989) (documentary section is not paginated).
26
Jackson, The Spanish Republic and the Civil War, p. 398.
27
The trial of Asensio, Villalba and other officers for their responsibilities in the loss of Malaga
(see chapter 3 above) also needs to be understood in this context. The case had been under
324 The Spanish Republic at war

underpinned -Negrin’s opposidon to declaring a full state of war in Re-

publican territory. Prieto in the war ministry was coming to favour this
by late 1937, and the chief of staff, General Rojo, deemed it essential. 28
But the prime minister stood firm. In spite of the organisational — and
disciplinary - advantages that would have accrued in other respects, he
would repeatedly refuse to declare a state of war until the very end, when
the collapse of Catalonia was imminent. 29
In the face of military reverses and increasing diplomatic isolation,
it was vital to bolster mass political and economic mobilisation on the
Republican home front in order to sustain the defensive effort at the core
of Negrin’s strategy. For this reason the PCE remained as crucial in 1937
to Negrin’s domestic policy on the civilian front as it was to the organi-
sation of the Republic’s military defence. Not only did the party entirely
endorse Negrin’s drive for war centralisation, but it also offered a dy-
namic instrument through which to achieve the necessary psychological
mobilisation of the population, as well as its engagement in war work.
(This had also been the raison d’etre of the liaison committees established
with the PSOE in the spring.) The particular talent of the PCE lay in its
practical application of multiple discourses of mobilisation. The effects of
acutely uneven development meant that Republican society remained,
inevitably, deeply fragmented during the war. The PCE’s technique of
adapting the register of the message to the cultural/ educational level and
expectations of its differing audiences 30 was thus far more effective on the
ground than the PSOE’s somewhat austere and demanding appeal to a
civic socialism, and certainly more appropriate than Negrin’s ‘universal’
language of constitutional liberalism, transferred, without much modifi-
cation, from his interventions in European diplomatic fora. 31The PCE
was thus at the core of multiple forms of Popular Frontist political and
cultural mobilisation during 1937 which criss-crossed urban and rural

investigation since February 1937. They were brought to trial in October, as the north was
and given prison sentences: Nadal, Guerra civil en Malaga pp. 417-18. But the case was
falling, ,

subsequently reconsidered and dismissed in July 1938. For the role of Negrin (who supported
Asensio) see Zugazagoitia, Guerray vicisitudes, pp. 453-4.
28
Azana, Obras 785; Los papeles del General Rojo pp. 91-2, 97-8.
completas, vol. 4, p. ,

29 A state of war was declared on 23 January 1939. Negrin’s earlier reluctance also stemmed from
his evaluation of the military personnel available. To war safely would have
declare a state of
required a more plentiful supply of senior officers of Rojo’s calibre. But he knew that they were
relatively few. In September 1937 Negrin had lamented Miaja’s lack of talent - alarmed at the
prospect of his wielding such control in the Madrid area: Azana, Obras completas, vol. 4, p. 767.
30 Graham, Socialism and War, p. 119; Largo Caballero, Mis recuerdos, p. 227.
31
The text of Negrin’s wartime speeches is scattered in press and archive. A number feature in the
accompanying documents volume of Alvarez, Negrin, Personalidad historica, vol. 2 But it would be.

very useful to have an edited volume of Negrin’s speeches.

 JVegrin’s war on threefronts 325

worlds and different constituencies of Spaniard in terms of class, region,

gender and age. In this sense, the PCE could, then, lay some claim to
have constituted itself during the war as the most successful Republi-
can party of all. For a time at least, it delivered, both within itself and
the organisations of the wartime Popular Front, the counter-hegemonic,
inter-class alliance that republicans and socialists had sought, but mainly
not achieved, since the birth of the Republic in 1931. And for all that
socialists and republicans criticised the PCE for its ‘populism’, this was
a utilitarian populism vital to the Republic’s survival in wartime. 32 The
PCE could not, of course, solve the significant ideological and policy con-
tradictions inherent in Popular Frontism in Spain.But it was precisely
thePCE’s discipline, deriving from its democratic centralist structure,
which made it strong enough to bear them — without sustaining the or-
ganisational fragmentation that so debilitated thePSOE. This we could
term the functionality of democratic centralism.
But democratic centralism did not obviate significant internal stresses
in the PCE. The party’s rapid growth and the influx of a lot of people
without political experience took its toll on efficiency and coherence,

as was noted by Comintern adviser Palmiro Togliatti, who arrived in

Spain in July 1937. 33 Moreover, the fact that the PCE was incorporating
disparate constituencies within itself meant that competing political and
economic agendas tended to resurface inside both the party and Popu-
lar - especially as the Republic’s material conditions
Front organisations
eroded during 1937 and into 1938. One example of this, which we shall
consider was the increasing tension in the PSUC between Cata-
later,

lanists and centralisers. Another important area of wear and tear on the
PCE related to Popular Frontist agrarian policy.
As we have seen, in areas like Catalonia, Aragon and Valencia, where
considerable social resistance to collectivisation existed among small-
holders and tenant farmers, the PCE (and the PSUC) had sought to
channel that opinion. 34 From the agriculture ministry too the PCE’s
Vicente Uribe had promoted measures - from the decree of 7 October

32
Casanova, Anarquismoy revolution, p. 22 1 cites Aragonese republicans complaining how the PCE’s
political line can’t be sincere ‘unless apostasy is the order of the day (consigna)’; for PSOE

criticisms, seeGraham, Socialism and War, p. 119.

33 Carr, The Comintern and the Spanish Civil War, pp. 52-3; P. Togliatti, Escritos sobre la guerra de Espaha
(Barcelona: Critica, 1980), p. 9 - introduction by P Spriano. On Togliatti in Spain, see also
his Opere 1933-1944, vol. 4 (Rome: Editori Riuniti, 1979) and A. Agosti, Togliatti (Turin: UTET,
r
99 6), pp. 225-43.
34 This also happened in a more low-key way in the Republican south - for example via the PCE’s
championing of co-operative forms over collectives: see L. Garrido Gonzalez, Colectividades agrarias
en Andalutia: Jaen (1931 -1939) (Madrid: Siglo XXI, 1979), pp. 57ft.
326 The Spanish Republic at war

1936 onwards - designed to protect such sectors in order to ensure their

productivity for the war effort. This accorded with the respect for liberal
individualism and property rights at the heart of the Popular Front. But
for the PGE the essential goal by 1937 was maximising production -
and no doubt, in Comintern minds, the Soviet Union’s experience of
war was
civil Republican Spain, however, that meant
influential here. In
that collectivists as well as smallholders had to be kept on board the war
effort. 35 (As we have seen, reining in Aragon had been more about defus-

ing the power of the Council and activating the military front than it had
been about hostility to collectivisation.) As a result, the agricultural min-
istry, via its Institute of Agrarian Reform (IRA), oversaw the codification
of the collectives during 1937 — many of which were in the Republican
south. 36 This was a means of extending state control, to be sure. But
neither was it hostile to agrarian collectivisation per seT1 That would not
have been politic - especially given the context of pre-war collectivist
cultureand traditions in Castilla La Mancha and the rural south-east -
where communists too were sometimes involved in these initiatives. 38
When the issue of private-property rights resurfaced in the course of the
IRA’s work, however, the Republic chose to postpone discussion of this
as a principle. The collectivists were deemed to have usufruct. But the
instances of former owners’ suits being favourably adjudicated by the
IRA in 1937 and 1938 created a sense of instability.
39

It is well known that the tensions between collectivists and small farm-
ers led to a number of violent clashes in Republican territory. But it

is difficult to see how any policy option could have avoided this, given
the degree of political and social fragmentation obtaining in the Re-
public. The PCE made certain choices both for party-political reasons
and because of its understanding of war needs. But the conflicts it had
to mediate as a result were already inherent in the post-coup political
situationand social breakdown of forces.
As the war went on, however, the focus of conflict in the agrarian
sphere would cease to be so specifically collectivism versus individualism.

35 Ronald Fraser explains this dilemma well in Blood of Spain p. 370. ,

36 The figures suggesting a 25 per cent increase in the number of collectives between mid 1937 and
late 1938 (even after the loss of Aragon) point to a furious process of registration: ibid., p. 393
and p. 373 (n. 1).
(n. 1)
37 R. Fraser summarises IRA aid to registered collectives: ibid., p.373 (n. 1). See also Instituto de
Reforma Agraria, La Politica del Frente Popular en agricultura (Madrid/Valencia, 1937) and V Uribe,
La politica agraria del Partido Comunista speech given in Valencia 4 July 1937 (Barcelona, 1937).
,

38 Garrido Gonzalez, Colectividades agrarias en Andaluda: Jam (1931-1939)', N. Rodrigo Gonzalez, Las
colectividades agrarias en Castilla la Mancha (Toledo, 1985).
39 Adelante,
29 May 1937; Bolloten, SOW, p. 238.
 Negrui’s war on threefronts 327

By the second half of 1937, the extension of price controls brought rural
producers of all sorts government bodies. Then, as
into conflict with
the conditions of production deteriorated further and the effects of the
war began to bite in Catalonia (with an acute subsistence crisis in urban

areas), tensions would erupt around the issue of military requisition from
rural producers. Given that the PCE’s overriding priority was the war,
then, as it grew more difficult to provision the home front as well as the
army, it was obliged to back state regulation not only for the collectives
(in terms of wages and conditions) but also for individual farmers. By

the end of 1937 the effects of prioritising war needs were beginning to
be the major cause of peasant alienation and urban-rural tension. 40 And
the collectives too — whether in Aragon or the Republican south — were in
the end eroded as much by the deteriorating conditions of war as by any
specifically anti-collectivist action on the part of PCE or government.
The very fact of the PCE’s rapid political ascendancy had also in-
stantly exacerbated the historic rivalry and longstanding organisational
tensions with the PSOE/UGT. The evolution of this conflict would be
further complicated by the internal di\ision in the socialist movement,
the ramifications of the PCE’s links with the Comintern and the cumu-
latively erosive impact of the war. But it is crucial to appreciate that the
breakdown in relations between the two movements sustaining the Re-
publican polity at war came as a result of the interaction of these factors
rather than as the consequence of a purportedly monolithic Stalinist-
inspired sectarianism emanating from the PCE.
The alliance between the PCE and the parliamentary PSOE forged
early in 1937 had been based on a common perception of the importance
of the Popular Front and of the need for joint party action behind the war
effort. As we have seen, the parliamentary socialists also intended it to fa-
cilitate the marginalisation of Largo Caballero and his supporters, whom
they judged to be responsible for the damaging divisions in the move-
ment since 1934 (and now, with the war, unaffordable). But the was PCE
also increasingly embroiled in an organisational battle with Caballerista
socialists over influence in the UGT. 41 This w as a continuation of the
pre-war which Largo and his veteran ugetista leadership were as
‘battle’ in

determined to absorb the upstart young communist organisation (as they

viewed it) as the PCE was to steal a march ov er Largo. 42 But the outbreak
of the war had giv en the PCE certain advantages. This, in turn, inten-
sified the party ’s ambitions - and in particular those of the Comintern

40 For more on 41
this see below in this chapter. Graham. Socialism and War, passim.
42 See chapter i above.
328 The Spanish Republic at war

delegate, Codovilla, who, in what became branded very much as his own
initiative, pushed for the creation of a single proletarian party (the partido
unico). But, as we have also seen, this brought Codovilla up against
Comintern instructions from Stalin not to force the issue and to be mind-
ful of the plurality of the Popular Front in Spain. For the PCE to have
behaved otherwise would have undermined Stalin’s prime objective: to
sustain Republican resistance. But after the military reverses of early

1937 and with immense pressure still on Madrid, Stalin had accepted the
,

opinion emanating from the socialists and republicans in the cabinet that
Largo was an obstacle to the implementation of appropriate war policy.
Once Largo Caballero had exited from government in May 1937,
however, he began to construct the reasons for his departure entirely
according to the old rationale he knew best: that of socialist-communist
organisational rivalry. He had been ejected from government as the result
of a communist conspiracy. The Caballeristas saw themselves confirmed
in their suspicions by the increasing rapprochement of Prieto’s socialists
to the PCE and by Prieto’s own ‘wild card’ suggestion, in the summer of

1937, that the primacy of war probably required the fusion of the PSOE
and the PCE. Even though this was instantly vetoed by the PSOE exec-
utive, it fed the flames of the Caballeristas’ own escalating propaganda
war against the PCE. The tension was further increased on the ground
by what the socialists saw as the arrogant proselytism and sectarianism
of an expanding Communist Party. What was happening here was also
the result of more than one factor, however.
Certainly the sectarianism inherent in the PCE’s self-perception as
the revolutionary vanguard party had been stimulated by its wartime
successes. But it is rather too simple to explain the PCE’s aggressive
proselytising practice across 1937 exclusively in terms of ideology. Cer-
tainly there were those in the PCE(and the Comintern) who believed
that the May days had pointed up the dangers of Popular Frontist plural-
ism and that it was time to revert to an worker bloc led
earlier policy of a
by a ‘single party’ of socialists and communists, inevitably under the po-
litical tutelage of the latter. 43 But these tensions were, as yet, submerged

beneath disciplined adherence to the official line. At the same time, how-
ever, what wartime success had also meant to the PCE was a mass influx
of new members, usually with low levels of political education. As we
have seen, the circumstances of the civil war had made it important for
people to join political parties and organisations for all sorts of practical

43 Elorza and Bizcarrondo, Queridos camaradas, p. 451.

 Negrin’s war on threefronts 329

reasons - protection, advancement, sometimes even to ensure access to

food. These new communists also contributed to shaping the wartime
party by infusing communist sectarianism with old-fashioned clientelism
and the politics of patronage. 44 By this I mean the politics that typified
pre-Republican Spain. Clienteles served political parties in return for
protection and favours. The coming of the Republic had initiated a pro-
cess of change, with the PSOE symbolising a new ethos of public service.
But the change was its infancy when the war erupted. Political cul-
still in
ture and people’s mentalities at local/village level were still, for the most
part, framed by an older understanding of politics. The pervasive culture
left as much as the right. Many of the new
of clientelism influenced the
PCE saw the party in these terms and were deeply hostile
recruits to the
to members of the PSOE as potential rivals for job preferments or other
favours. But all political groups operating in the Republic were affected to
some extent by clientelist assumptions. They fed internecine conflict and
the left’s continuing obsession with transfugas. Indeed, Largo Caballero’s
own input in the conflict with the PCE was in part constructed via his
immense anger at finding that the old-guard UGT leadership had, for
the first its members’ allegiance within the
time ever, a serious rival for
socialist camp Largo would rewrite the script of the May cabinet
itself.

crisis in the light of what was, by late 1937, an increasingly bitter and

rarefied batde with the PCE inside the UGT. He began to see himself as
the great anti-communist warrior, defending the organisational integrity
(and, not coincidentally, his own control) of the UGT.
Fortunately for the broader needs of the war, Prieto and the PSOE
executive held firm to its policy of inter-party liaison with the PCE. But
there were ominous signs as some local and provincial party sections,
hostile to the new communists’ thrusting ambition, challenged the au-
thority of the PSOE executive and began to withdraw from the liaison
committees. 45 As their own relations with local PCE organisations dete-
riorated, so Largo’s increasingly public anti-communism began to chime
with their own experience. His political demise, in retrospect, was acquir-
ing a whole new,and powerful, meaning. Once again, Largo Caballero’s
influence would be based on a symbolism that hid a far more complex
and contradictory reality. But this time, unlike in September 1936, his
symbolic power would hinder, not help, the beleaguered Republic.

44 Until we have a thoroughgoing social history of the PCE in the 1930s, this must remain a working
hypothesis. But the circumstantial evidence, even thus far, is compelling. This facet of the PCE

is an important, albeit implicit, theme in Graham, Socialism and War.

45 Ibid.,
p. 80.
33 ° The Spanish Republic at war

The containment of this potentially very damaging situation between

the PSOE and the PGE was assisted by the fact that Stalin reined in
Codovilla. In September 1937 he was recalled to Moscow and Togliatti
took over his role as Comintern delegate, with Stepanov remaining as
deputy and Gero continuing in Barcelona. 46 Togliatti
was keen to en-
sure good relations with the PSOE in order to strengthen the Popular
Front and thus Republican resistance. He thus sought to stabilise the
liaison committee initiative and contain the PCE’s more zealous advo-
cates of communist vanguardism and the ‘single party’ (such as Dolores
Ibarruri). 47 For these tactical reasons, but also perhaps because of his
own political understanding of Popular Frontism, Togliatti was critical
too of the Comintern’s idea that the PCE should press for new elections
tobe held in the Republican zone. 48 In fact, the proposal drew no real
support and thus died on the drawing board in the autumn of 1937.
Negrin replicated his response to the calls for a single party: he pro-
crastinated amiably. The PSOE - although some
executive did likewise
criticism was expressed of the shortsightedness of the proposal in terms
of the negative international diplomatic reaction it would provoke -
crucially in Britain. Socialists and republicans also believed that the coup
had made the renovation of the parliamentary assembly desirable. But
they also knew that it would have to wait until the war was over. This
was for practical reasons, but also because the legitimacy of the electoral
results would depend on the balloting of the entire Spanish population.
Even the PCE’s own leaders saw that their identification with the elec-
toral initiative would be counter-productive as it would isolate and thus
weaken the party. 49
But nor is it as clear as has been suggested that Stalin had taken
up the idea of elections with the intention of promoting the political
46 These changes followed Togliatti’s September report, which was very critical of Codovilla’s
highly dictatorial style. On this see also Falcon, Asalto a pp. 144, 150-1, 160. Togliatti’s
los cielos,

report is in Escritos sobre laguerra de Espana, pp. 143-50 and Carr, The Comintern and the Spanish Civil
War pp. 94-8.
,

47 Togliatti wrote to Moscow in November about what he saw as the worrying extent of Codovilla’s
sectarian influence on Ibarruri: Togliatti, Opere 1933-1944, pp. 288, 291 (also in Togliatti, Escritos
sobre laguerra de Espana, pp. 149, 164); Elorza and Bizcarrondo, Queridos camaradas, p. 417 and for a
summary of Togliatti’s goals and influence, Rees, ‘The Highpoint of Comintern Influence? The
Communist Party and the Civil War in Spain’, in Rees and Thorpe (eds.), International Communism
and the Communist International 1919-43, pp. 157-8. There are conflicting views of Togliatti’s own
leadership But he was certainly significantly less interventionist than Codovilla: Schauff,
style.

‘Failure in Emergency: The Spanish Civil War and the Dissolution of the Comintern’, p. 28;
Carrillo, Memorias, p. 263.
48
Togliatti’s report of 30 August 1937 cited in Elorza and Bizcarrondo, Queridos camaradas, p. 401.
In full in Togliatti, Escritos sobre laguerra de Espana, pp. 126—42.
49 Elorza and Bizcarrondo, Queridos camaradas, pp. 404-5.
 Negrin’s war on three fronts 331

career of the PCE - still less a fully fledged ‘popular democracy’ in

50 There seems
Republican Spain. be a strong element of post- 1945 to
hindsight operating in such assessmentsand a concomitant tendency to
downplay the contingencies of the time in shaping Soviet policy. 51 War
needs remained the determining factor in Stalin’s view of the Republic.
The autumn and winter of 1937 saw the final loss of the north and hard
fighting in Aragon which exposed the continuing weak points in the
Republican army. And all the time the enemy was getting stronger. The
Axis had made clear its determination to arm and supply the rebels unto
total victory. The fact that it was at this point that Stalin thought of looking
towards the consolidation of the PCE’s organisational protagonism in the
Republican polity - and, above all, towards maintaining its high profile
in the army - strongly suggests that the goal was to throw the PCE
into the breach to ensure that the Republic would be able to go on
resisting.

Although party-political interests played a part, this imperative of re-

sistance also underlay the PCE’s confrontation with Prieto in the defence
ministry over his desire to downsize the Political Commissariat in the au-
tumn of 1937 52 Prieto’s October reforms were intent upon reducing both
,

the numbers of commissars and their authority - especially at the lower

levels of the army. In effect, Prieto was seeking to depoliticise the army
and, as he saw it, to reinforce the hierarchy of command53 (although
in fact it was envisaged that commissars would go on existing at the
higher levels — in brigades, divisions and armies). Prieto’s measures were
intended to have the same ‘normalising’ functions as those which, for ex-
ample, prevented soldiers from participating in political demonstrations

50 See, for example, ibid., pp. 384-405 (esp. p. 402).

51
The very structure of Elorza and Bizcarrondo’s Queridos camaradas encourages this since it sepa-
rates the discussion of Comintern ideological influence/political objectives in Republican Spain
from its war (i.e. Republican resistance capacity) on Soviet/PCE
discussion of the impact of the
policy - as if the former were somehow fixed and independent of the latter. But the balance of
current research indicates precisely the ‘complex, contradictory and uncertain’ nature of Soviet
policy in Spain. It was ‘driven as much by circumstances as anything else’, Roberts, ‘Soviet
Foreign Policy and the Spanish Civil War’, Kremlin never sought to
p. 83. Note also that the
integrate the various Soviet bodies involved in Spain (defence, foreign affairs, Comintern) -
which also suggests contingency over a monolithic agenda: Rees, ‘The Highpoint of Communist
Influence? The Communist Party and the Civil War in Spain’, p. 160; Gazur, Secret Assignment.
The FBI’s KGB General, p. 334.
52 A. Elorza and M. Bicarrondo (
emphasise Prieto’s anti-communism
Queridos camaradas, p. 408)
over his ‘defeatism’ as the Comintern opposition to him as minister in late
main motive for
1937. But this would seem to introduce a false distinction. As the Comintern interpreted the
situation by late 1937, the PCE’s pro-active military presence was the only reliable guarantee of
the Republic’s continued resistance.
53 Broue and Temime, The War
Revolution and the Civil in Spain, p. 314; Alpert, El ejercito republicano,

pp. 184-9.
332 The Spanish Republic at war

and increased army pay differentials (5 October 1937). It is also worth

reinforcing the point that political affiliation of any sort was considered

undesirable, not just communist affiliation. There is no doubt that many

professional officers were keen to see such a normalisation. But in their
enthusiasm to return to what they knew they tended to forget that the
Republic was not fighting a war in ‘normal’ conditions. And it was pre-
made the role of the commissar as crucial in 1937-8
cisely this fact that
as had been, in a different way, in 1936.
it

In waging their own ‘war’ to look after the physical and psychological
welfare of their men, the commissars were the Republic’s best guarantee
of morale and discipline in the face of the permanent, cumulative and
highly erosive material inferiority and the structural handicaps under
which it was forced to fight. 54 In fact the kind of political education
commissars pursued was frequently low grade and very general — for
example explaining the Republic’s war of national independence to what
were, by the second half of 1937, units of recruits rather than volunteers.
From the start of the war it had been the PCE that had recognised the
importance of the Commissariat, contributing its best men. But Prieto
continued to be exclusively exercised by the party-political, including
clientelist, PCE’s strong influence in the Commissariat.
aspects of the
By and repeatedly challenging new PCE nominations,
‘uprooting’ it

however, he can only have undermined the institution’s efficiency. He

clearly thought that a price worth paying in order to achieve a better
party-political balance inside the Commissariat. And indeed this helped
to damp down PSOE, CNT and PCE on the home
the tensions between
front. But whetherwas a price worth paying in terms of the army’s
it

cohesion and efficiency is more uncertain.

Nor can Prieto’s campaign be seen purely in terms of his desire to
professionalise Republican institutions. Negrin yielded to no one in
prizing the independence of state authority. 55 But without a victory in
the war he knew that there would be no Republican state. And, as Negrin
himself recognised, PCE commissars, like PCE army commanders, were
almost invariably good, and frequently the best the Republic possessed.
Prieto was normally no less given to intelligent pragmatism than Negrin.
Except, that is, in matters where he sensed his own political reputation

54 De la Mora, In Place of Spendor pp. 352-3.

,

55 I omit entirely from the discussion here the vexed question of how Soviet military advisers related
to the command structures in the Republican army. We simply do not as yet have the means
of corroborating or disproving the claims about Soviet autonomy made in some of the memoir
literature.
 Negrin’s war on threefronts
333

to be endangered. Prieto’s ‘war against the communists’ in 1937-8 had

a clear political meaning. However, it was not the meaning that he
subsequently came to invest it with in his voluminous post-war writings.

The seamless narrative of Prieto as the great anti-communist warrior

opposing the Soviet Union’s bid to satellitise the Republic through
the PCE was a post hoc construction, increasingly determined by the
exigencies of Republican exile politics in an environment of intensifying
Cold War antagonisms. 56 (In the latter half of the 1940s the objective of
exile Republican politics was to persuade the Allied powers to intervene
against Franco. Prieto consciously played the anti-communist card as a
means of constructing a viable political platform from which to attract
western support for a Republican option to replace Franco.The role
came naturally to Prieto because he could tap his quite genuine
and intensely visceral dislike of the PCE and he doubdess ended by
believing that there was an absolute continuity between his wartime
sentiments and post-war discourse.) By contrast, Prieto’s civil wartime
anti-communism in some ways recalls that of his great rival and antago-
nist, Fargo Caballero, in the UGT. Both were fuelled in part by bruised

ego and personal political rancour, and both were politically myopic
in that they undermined the Republic’s ability to resist. 57
In spite of the conflict over political commissars, the Popular Front
was held together by the joint efforts of Togliatti on the one side and
the PSOE executive on the other. In particular it fell to the PSOE’s
general secretary, Ramon Eamoneda, repeatedly to mitigate the damage
done by the party’s high-profile, ‘historic’ leaders. 58 Togliatti’s task was
facilitated by the abiding difficulties of communication between Moscow
and Spain. The habitual delays had meant that, effectively from the start
of the war, the communists in Spain were making their own decisions -
even if these subsequendy had to be defended in terms of the broad
Comintern line. 59 This situation created a certain latitude for Togliatti
to continue stressing the Popular Front as a pluralistic alliance to protect

56 As Lamoneda shrewdly observed, Prieto’s anti-communism was a strategy for exile: Lamoneda,
‘El secreto del anticomunismo’, unpublished notes (FPI) cited in Graham, Socialism and War ,

P- 144 -

57 It is interesting to contrast the tenor and content of Prieto’s notorious ‘anti-communist’ speech
of August 1938 to the PSOE National Committee (as well as the extremely negative reactions it
evinced) with his later, far more grandiloquent anti-communist expositions which, in an entirely
met with general approval in PSOE and republican circles. On this
different political context,
see Graham, Socialism and War, pp. 142—4.
58 For Lamoneda, see ibid., pp. 157-8.
59 Rees, ‘The Highpoint of Communist Influence? The Communist Pam and the Civil War in
Spain’, pp. 145, 148-9, 150-2.
334 The Spanish Republic at war

and foment a new democratic politics in Spain. 60 His conception chimed

in important ways with republican and socialist ideas about how the
war would necessitate the reshaping of the Republican polity. For all
components of the Republican alliance, this had to be reinvented as
something apart from the discredited oligarchic order that had backed
the military rebellion. The historically very weak central state was being
restored as the central instrument of economic modernisation. But for
this to work (and in the immediate term for the war to be fought), it had

to be underpinned by an anti-hegemonic alliance reaching across social

classes (such as republicans and socialists had been seeking since 1931).
It was this vital Spanish inflection of the Popular Front strategy that gave

it a specific utility and resonance in the civil war.

If the Republic’s military situation had stabilised rather than deteri-

orated after late 1937, then the Popular Front might have evolved and
been strengthened. Although this is a counterfactual hypothesis, it seems
reasonable, and indeed important, to make the point - not least because
seldom is it made clear the degree to which the military reverses and
defeats of 1938 were themselves a major cause of the Spanish Popular
Front’s progressive erosion and eventual disintegration. Indeed, what
was often perceived - and certainly reported in hindsight - as the in-
creased political stridency of the PCE leadership in 1938 was at least
in part generated by its desire to hold the line on resistance after the
military situation began steeply to decline into the spring of 1938 and
the rebel armies began their drive against Aragon. In many ways too
this strategy had grown apart from the Comintern’s to become grafted

as a domestic PCE agenda driven by the party’s own ambitions and inter-
ests. This would be evident in the party leadership’s reluctance to accept

Comintern instructions in February-March 1938 that the PCE ministers

should leave the cabinet in the hope of improving the Republic’s stand-
ing with the British and French establishments, thus facilitating Stalin’s
61
long-sought-after diplomatic front against Hitler.
It fell to the PCE, as the party most single-mindedly engaged with
maintaining the war effort, to back Negrin unconditionally in his

60
Togliatti was a leading member of the Comintern rather than merely its functionary. His ex-
perience in Spain would shape his strategic and conceptual thinking around the key questions
new type’ which, in turn, would later
of anti-fascist resistance, revolution and ‘democracy of a
inform the policies of other European communist parties and especially his own PCI.
61
Elorza and Bizcarrondo analyse the interchanges between Comintern and PCE in Queridos
camaradas, pp. 410—16. Stalin’s strategy here reinforces the contingency of Soviet policy in Spain.
It was always adapting to changing circumstances. There was no rigid ideological blueprint - in
spite of what Elorza and Bizcarrondo finally seem to suggest on p. 420.
 Negrin’s war on threefronts 335

increasingly hard line on political centralisation to ensure the concen-

economic resources in central government hands. The
tration of scarce
batde with the Council of Aragon had partly been over control of for-
eign exchange. In Valencia too, the Republican treasury’s ongoing batde
to centralise control of citrus exports led to increasing (and ultimately
armed police) confrontations across 1937 with the unions’ collection and
export agency, CLUEA - a conflict which was closely bound up with that
62
between pro- and anti-collectivist currents in the region. In Catalonia,
the drive for economic centralisation also crystallised tensions between
the PCE and the PSUC as well as driving a wedge between centralisers
and Catalanists inside the PSUC. Both conflicts would intensify in the
course of the damaging confrontation looming between Negrln and the
Generalitat.
The crisis of May
1937 in Barcelona had seen the Generalitat obliged
to cede control of public order to the central government. From then
on, relations between the two deteriorated apace. In many ways it was
a ‘dialogue of the deaf’. Each side imputed the worst kind of political
manipulation to the other, thus blocking any real discussion of the per-
spectives and needs of each. The loss of public order, the touchstone of
autonomy, had understandably traumatised Companys and the
political
Esquerra. 63 Catalan nationalists were also aggrieved at what they saw
(justifiably) as a lack of recognition from Valencia of the achievements
of the improvised Catalan war industries. 64 The central government,
desperate for arms and munitions (in particular because of the loss of
material through Italy’s torpedoing of ships) was, understandably, always
demanding more. Rather less justifiably, however, the centralist preju-
dices of the governing majority led it to conceive the ‘solution’ in terms
that were practically simplistic and politically highly damaging. Appar-
ently all that was required to achieve higher production was greater
62
For the CLUEA dispute, see Bosch Sanchez, Ugetistas y libertarios, pp. 117-23, 336-40;
W. Bernecker, Colectividades y revolution social: el anarquismo en la guerra civil espanola 1936—1939
(Barcelona: Critica, 1982), pp. 123-6; Abella, La vida cotidiana durante la guerra civil, p. 82.
63 Companys criticised the order dissolving the Council of Aragon as a veiled attack on Catalan
autonomy; Prieto expressed extremely harsh judgements of Companys (‘Companys is mad as in
5
“should be confined to an asy lum” ), Tarradellas and Comorera: Azana, Obras completas, vol. 4,
p. 760 and p. 802 (La Pobleta entries for 31 Aug. and 20 Sept. 1937 respectively).
64 An improvisation memorably evoked by the historian Ernie Ucelay da Cal as ‘una titanica
chapuza’. The most famous example of this ‘dialogue of the deaf’ in regard to military' production
is in the famous exchange of letters betw een Prieto and Companys in December 1937. Prieto

complains about what he still needs and hasn’t received, while Companys paints a picture of
the staggering amount that the Catalans have achieved in such inauspicious circumstances: De
Companys a Indalecio Prieto. Documentation sobre las industrias de guerra en Cataluna (Buenos Aires, 1939).
See also Azana, Apuntes de memoria p. 82. ,
336 The Spanish Republic at war

political will whence the necessity of sweeping out half-hearted Catalan

managers from the factories and installing government agents (i.e. engi-
neers and civil servants) in their stead. Negrin’s case for nationalisation
to increase war efficiency was strongly inflected by his Spanish national-
ism. 65 But if the prime minister gave a determined lead here, he found
firm backing as much from the PSOE as from the PCE. For the socialists
were as centralist as President Azana’s Madrid-based republicans. While
the Republican President too gave indirect expression to his bitterness
66
towards Companys and the Generalitat, it was Prieto who expressed
himself violently in full cabinet meetings on the subject of Catalan ‘self-

ishness’. A month government’s October 1937 move

after the central
to Barcelona, with the express intention of exerting greater economic
began with
control over the region, the process of factory intervention
67
piecemeal take-overs. After some resistance from the Catalans, it would
conclude nearly a year later in August 193 8 when the government as-
sumed direct control of the war industries in their entirety — thus
region’s
provoking the cabinet crisis that would see the departure of both Catalan
and Basque nationalist ministers. 68
The huge influx of central government personnel and civil servants
to Barcelona in the autumn of 1937 stretched the already meagre supply
of food and space in the refugee-packed city to breaking point. It also
led to spiralling political tensions. The Spanish-speaking incomers requi-
work accommodation. At the same time, with an
sitioned buildings for
international public in mind, Negrinwas projecting an image of the
Republican war effort as a ‘Second War of Independence’. (This was

65 Azana, Obras completas, vol. 4, p. 701 (La Pobleta, entry for 29 July 1937); p. 745 (entry for 23 Aug.
1937), P- 8 2 3 (entry for 14 October 1937); E. Lister, Nuestra guerra (Paris: Coleccion Ebro, 1966),
pp. 244-5. See also Besteiro’s comments that the vision of (a unified) Spanish history in Negrin’s
speech of 18 June 1938 was so reactionary that it could have been given by a fascist or a Carlist.
Besteiro to Pascua, 10 July 1938 AHN/MP caja 2 (16).
66
This is apparent in Azana’s fictionalised dialogue about the war, La velada en Benicarlo. See also
various diary entries, for example Obras completas, vol. 4, p. 699 (28 July 1937); pp. 707-8 (29 July
r
937)-
67
These factories were placed under the control of the Catalan branch of the Ministry of Defence’s
subsecretariat of armaments (first created in June 1937), A. Monjo and C. Vega, ‘La clase obrera
durante la guerra civil: una historia silenciada’, Historiay fuente oral, 3 (1990), 88, n 13. For the
intervention and ensuing tensions, see also Bricall, Politico economica de pp. 291-3. In
la Generalitat,

its public explanation of themove to Barcelona, the central government emphasised the need
for a good dialogue make autonomy work. But, between the lines, the message was
in order to
clear in the reference to everyone putting the war effort first. Text of broadcast speech by interior
minister J. Zugazagoitia, 28 Oct. 1937, in AHN/MP, caja 6 (18).
b8
For a time the Generalitat refused to allow government representatives into the factories: Fraser,
Blood of Spain,
p. 227 (n. 1). By 1938 it would even seem that the government was transferring
Catalan police and even firefighters out of the region: Fraser, Blood of Spain, p. 383 (n. 1).
 jVegrin’s war on threefronts 337

after the ‘First

5
of 1808-14, when sectors of the Spanish population (most
mythically, the popular classes of Madrid) had offered armed resistance
to the new Napoleonic order.) But precisely because 1808 had subse-
quendy been constructed as the espaholista myth of patriotic resistance
par excellence , this made it difficult for Catalanists to assimilate. 69 The
same sort of problem supervened in the PCE s propaganda production.
5

For all its abstract Popular Frontist discourse on national rights, the party
demonstrated complete lack of understanding of the need to make
its
5

a cultural ‘translation in Catalonia of the material it produced for the

rest of Republican territory: This was - strategically speaking - rather
shortsighted. After all, Catalonia was the one region of Spain where
the inter-class alliance —
which Popular Front aimed - had already
at 7

assumed a real form by the spring of 1936. But PCE my opia should
scarcely surprise - given that the official Spanish Communist Party had
never had any real presence in Catalonia prior to the civil war. 7 ° For
the same reason, it was hard for the PCE’s leaders to understand the
complex skein of allegiances and rivalries on the Catalan left. It was an-
other world entirely - which they sought in vain to fathom by deploying
7

a Manichean schema in part derived from the two-dimensional ideo-

logical categories in vogue with the Comintern, thus further straining
relations with the PSUC.' Inside the PSUC too there was a widen-
1

ing divide between Catalanist social democrats and communists as the

5
former backed Company’s growing criticism of the Negrin gov ernment
while the communist sector supported its centralisation measures. 72
5
Companys criticisms of the Negrin gov ernment in the Catalan par-
liament found a more strident echo in the Catalanist press. There was
a call for an end to the publication of newspapers in Castilian as the
5
‘language of the coloniser The editor was subject to judicial action. 73
.

69 None of these observations is intended to imply any particular assessment of how the Catalan
experience of 1808-14 (i.e. in terms of collaboration and resistance) compares with other areas of
Spain. Everywhere pro- and anti-French positions were complex and contradictory. In Catalonia,
were various popular rebellions against the French, and the pressures of
as elsew here, there
occupation rapidly eroded ajrancesado pro-French) support. The different reactions in 1937-8
had more to do with post hoc nationalist constructions of 1808-14 (both Spanish espaholista and
Catalanist . Nationalist historians in Catalonia also chose - logically - to concentrate on instances
of specifically Catalan rebellion against the Spanish crown.
70 See chapter 1 above.
71
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil: p. 314.
72
Togliatti, Escritos sohre la guerra de Espaha. pp. 180-1, 247-8: E. Gero to Dimitrov. 19 Nov. 1938,
cited in Radosh. Habek and Sevostianov. Spain Betrayed The Soviet Union in the Spanish Civil War.
Elorza and Bizcarrondo. Queridos camaradas, p. 427.
p. 51 1;
73 Azaha, Obras completas voL
4, p. 701.
- which also indicates harsh PSOE press criticism ofwartime
Catalanism.
338 The Spanish Republic at war

But in some senses the central government had itself driven up the tem-
perature by its own political heavy-handedness. In essence, a majority in
the cabinet, starting with Negrin, felt that there was far less need to pla-
cate the Esquerra now that the GNT had been domesticated. 74 And this
sense no doubt also underwrote Azana’s own support for new elections in
Catalonia. But this hostility towards all Catalanist claims and objections
was unwise. For quite apart from the CNT’s status, the long-drawn-out
jurisdictional dispute with the Generalitat sapped both government and
regional resources and seriously eroded Catalan morale in the process.
The Republic could afford none of these things.
Precisely because of these gathering problems, the middle of August
had seen a prohibition on all political street demonstrations in Barcelona
aimed at damping down the myriad political discontents and suppressing
overt manifestations of war weariness and other forms of defeatism. But
this inevitably meant increasing press censorship too. Catalan liberals, as

well as libertarians, 75 criticised this as evidence of the erosion of ‘political

liberty’ in the Republican zone under the Negrin government. Political
sensibilities were understandably bruised by Negrin’s high-handed cen-
tralism. Nevertheless, in many crucial respects Negrin was as genuinely
a liberal by conviction as any of his critics in the Esquerra.
The prime minister was actively pursuing domestic policies designed
to consolidate a liberal market-based economy and a parliamentary
polity in Republican Spain. 76 In this ‘normalising’ vein I place initia-
tives such as that of the Caja de Reparaciones, set up by Negrin under

treasury auspices. 77 This body was responsible for supervising the eco-
nomic restitution of expropriated property. 78 Liaison between it and the
Tribunal de Responsabilidades Civiles set up (in the Ministry ofJustice)
to investigate complicity with the military rebellion was, in theory, in-
tended to increase the authority of the state by making it the arbiter of
restitution. But in fact, the Tribunal, unlike the Caja, was never very
active. Negrin was also concerned to limit dismissals of staff from state

74 Jellinek, The Civil War in Spain p. 571.

,

75 Tavera and Ucelay da Cal, ‘Grupos de afinidad, disciplina belica y periodismo libertario’,

pp. 186-7.
76 ‘We are fighting for all Spaniards . . . The Republican government has no enemies except those
who will not accept the rule of law’, Negrin’s broadcast speech, 22 Oct. 1937, reproduced in
Servicio Espahol de Informacion (Valencia), 24 Oct. 1937, copy in AHN/MP
caja 1 (15).
77 Although Negrin’s close collaborator, the left republican Francisco Mendez Aspe, would formally
take over the treasury portfolio in April 1938, in fact Negrin himself continued to be the main
architect of the Republic’s economic policy until the end of the war.
78 Del Rosal, Justicia en guerra, pp. 239-45. See also G. Sanchez Recio, La Republica contra los rebeldes
y los desafectos. La represion economica durante la guerra civil (Alicante: Universidad de Alicante, 1991).
 Negrin's war on threefronts
339
employment to those against whom actual instances of military rebellion,
treason, espionage or flight of capital could be proved. 79 Late 1937 and
early 1938 would also see the public return to the Loyalist zone of some
eminent politicians from the centre-right of the Republic’s pre-war spec-
trum - including the ex-ministers Miguel Maura and Manuel Portela
80
Valladares.
Negrin’s appointment as prime minister had also seen an acceleration
of the process of normalising the judicial process as the Popular Courts
were incorporated into the established framework of ‘ordinary’ justice
81
(the Audencias Provinciales). While other modifications of the judicial
process under Negrin would prove controversial, as we shall see, this
measure nevertheless very probably helped diminish one set of prob-
lems connected with undue political influence of a clientelistic variety
82
as well as outright corruption in some cases. In terms of guaranteeing
constitutional rights, Negrin supported his interior and justice ministers,
respectively the PSOE’s Julian Zugazagoitia and the Basque National-
ist Manuel de Irujo, in their attempts to consolidate the normalisation
of the Republican police, judicial process, courts and prison system. In
the latter, length of professional service rather than political affiliation
became the criteria for the promotion of staff - especially senior ones.
There also occurred an unpublicised release of priests who were in prison
solely because they were priests.

By the summer
of 1937 private Catholic worship was effectively
permitted by the Republican authorities, although churches remained
preceded Negrin’s appointment as prime
closed. This de facto tolerance
minister.But until Negrin there was no active and sustained cabinet
support for Basque Nationalist minister Irujo’s bid to initiate a grad-
From May 1937 until the fall of
ual process of religious normalisation. 83
Catalonia inJanuary-February 1939 Negrin never ceased to support and

79 Cf. decree of 25 December 1938 (published in the Gaceta de la Republica). Part of this decree is

cited in Ortiz Heras, Violencia Politico en la II Republica, p. 73 (n. 13).

80
Portela attended the Cortes meeting held at the end of September/beginning of October 1937:
see Azana’s La Pobleta diary entry for 30 September 1937, Obras completas, vol. 4, pp. 81 1, 807.
81
Sanchez Recio, Justiciay guerra en Espana, p. 96.
82
R. Gullon, ‘Justice et guerre civile: souvenir d’un procureur’, in Serrano, Madrid 1936—1939]
P.Nenni, La guerra de Espana (1958; Mexico: Ediciones Era, 1975), p. 53 (although Nenni erro-
neously implies that the Popular Courts were entirely replaced by the Special Tribunals against
espionage, treason and defeatism (created 22 June). These were in fact additional. See below in
thischapter for more on the range of new courts in 1937-8.
83 For Irujo’s January 1937 proposals, rejected by the cabinet as impracticable, see M. de Irujo, Un
vasco en el ministerio dejusticia, memorias (3 vols., Buenos Aires: Editorial Vasca Ekin, 1976-9), vol. 1,

pp. 125-7. For Irujo’s statement of policy intent as part of the new Negrin cabinet in May 1937,
see de Lizarra, Los vascos y la Republica espanola, pp. 173-87.
340 The Spanish Republic at war

promote initiatives to this end. It made eminent sense in public-order

terms since public worship could be monitored, unlike religious cere-
monies in private homes, which could be used as a cover for subversive
activities - or, at the very least, to hold collections for Socorro Blanco

(White Aid) in support of rebel sympathisers. But, at heart, Negrin was

motivated by conviction here rather than political expediency. The ab-
sence of religious liberty was a fundamental ethical and political flaw
in a Republic whose legitimacy rested, so he believed, upon its con-
84
stitutionality. Liberal republican ideas were shaping the Republican
economy and polity, but they had to be extended out into Republican
society Progress was slow — but not because Negrin lacked the political
will. Rather extreme caution and discretion in how, when and where

religious ceremonial should be reintroduced were the inevitable conse-

quence of the maelstrom unleashed by the military coup and, as part
of that, of the Spanish Catholic Church hierarchy’s overwhelming sup-
port for, indeed legitimation of, the rebels. This had created a dense and
fraught atmosphere in Republican territory which no amount of polit-
ical will from the top could conjure overnight. If the reintroduction of

religious freedoms was to be successful, then it had to be taken one step

at a time. On this point at least the Catalans saw eye to eye with Negrin.
Ironically, the Christian democrats of Unio Democratica found it much
easier to agree with the prime minister on matters of religious policy
than with their fellow Catholics in the PNV 85
By mid 1938 things would begin to move. The Negrin government
approved a series of communications between Cardinal Vidal i Barra-
quer (Bishop of Tarragona, resident in Rome), his emissary, Santiago
Rial (Vicar General of Tarragona) and the Vatican in the hope of nor-
malising the (never broken) diplomatic relations between the Republic
and the Holy See 86 and, most importantly, of obtaining its support for
87
the discreet re-establishment of public worship in Catalonia. Although
cautious, the Vatican was not unwilling. The opposition came rather
from inside the Spanish - and Catalan - hierarchy. But the government

84 Negrin had never been remotely anti-clerical. For his family background and political beliefs see
H. Graham, ‘War, Modernity and Reform: The Premiership ofjuan Negrin 1937-9’, in Preston
and Mackenzie, The Republic Besieged, pp. 163-96 and H. Raguer, ‘La politica religiosa de Negrin
( 937~ 939)’>
I I unpublished article, pp. 7-8 and (same author) La polvoray el incienso pp. 325-30,
,

_
347) 354 8- See also chapter 3 above.
85 H. Raguer, ‘La politica religiosa de Negrin’, pp. 6, 7 and in letter to the author, 9 Aug. 1999.
86
A useful summary is in Lannon, Privilege, Persecution and Prophecy, pp. 207-8.
87
Raguer, ‘La politica religiosa de Negrin’, pp. 2—5; La polvoray el incienso, pp. 331-58.
 Negrin’s war on threefronts
34 1

pressed on. In October 1938 the burial of a Basque war hero took place in
Barcelona with full public religious ceremonial and several Republican
government ministers in formal attendance. This statement of the
Republic’s commitment to a pluralismwhich included public Catholi-
cism was still was followed up almost
not an easy one to make. But it

immediately by the creation of a specific body charged with overseeing

the implementation of public worship (the Comisariado de Cultos). It was
headed by an old colleague of Negrin’s, Jesus Maria Bellido i Golferichs,
like the premier a professor of physiology. With the collaboration of Unio

Democratica, and the tacit support of the Vatican, this had very nearly
achieved its aim in Tarragona province when the military collapse of
88
Catalonia supervened.
Negrin’s domestic policies also played a notably important role in his
conception of the third, diplomatic front. Projecting a decisive image of
the Republic’s liberal credentials was, he knew, an important weapon in
on Non-Intervention.
the battie to convince Britain to modify its position
But, again, it would be wrong prime minister’s strategy
to interpret the
here as ‘opportunistic’ as so many commentators have done. The poli-
cies Negrin attempted to pursue inside Republican Spain during 1937
and into 1938 — religious and otherwise — were of a piece with the polit-
ical, economic and cultural options he had vehemendy defended since

18 July 1936 and, in their fundamentals, those he had defended since he

had first entered politics in the Republican Cortes of 1931. What would
steadily consume his attempts to develop the liberal political practice of
the Republic - above all in its judicial practice - was the exceptional
pressure of the war, mounting inexorably throughout 1938. 89 Moreover,
in assessing what happened one must also remember the newness of
the democratic system in Spain. The formal superstructure itself only
dated back to 1931. But any kind of broader democratic culture was very
much work in very early progress given the enduring consequences of
underdevelopment - high levels of illiteracy and the concomitant lack of
basic education. Indeed, with the possible exception of Catalonia, most

88
Raguer, ‘La politica religiosa de Negrin’, pp. 7-8. As Hilari Raguer’s article makes clear, the
Vatican was anxious that Catholic practice should be reinstated in Republican territory by means
other than the military offensive of the victorious Francoist armies.
89 Negrin, in his speech of 30 September 1938 to the Cortes, effectively recognised how the ‘virtual
state of war’ (i.e. quasi-martial law) had inevitably eroded some aspects of the constitutional order
builtup after the July days in 1936. But he urged his critics to compare the current constitutional
order with past times or with other polities at war - in effect not to judge the Republican state
while it was in a ‘state of emergency’. Negrin’s speech is reproduced in Alvarez, Negrin, personalidad
historica, vol. 2 - see particularly pp. 74-5, 71-2.
342 The Spanish Republic at war

middling sectors - even if reasonably educated - still operated with a

highly clientelist understanding of politics. 90
Moreover, in that the Spanish war was a civil war, there was the ad-

ditional serious problem of a fifth column or ‘enemy within’. The fifth

column in Republican Spain covered a range of people and activities
that expanded as the war went on. Most crucially, it involved passing
intelligence to the rebels and engagement in a range of activities which
either handicapped the Republican war effort (for example production
of counterfeit ration books, work certificates and dispensations from mil-
itary service) or demoralised the civilian population (spreading rumours
of shortages, negative military reports etc.). Apart from the ‘autonomous’
pro-rebel groups and individuals thus engaged, there were also by 1937
articulated networks, including a Falangist one. They received financial
support from wealthy rebel supporters in exile or the opposing zone.
Most dangerous of all were those in direct contact with international es-
pionage and in particular the Gestapo and OVRA (Italian secret police),
both of which were active inside Republican Spain. In Madrid - where
the fifth column was most active - it was often linked to sympathetic
diplomatic corps via which it acquired safe houses, facilities and assured
means of radio communication with the rebels. 91
Such ‘invisible’ internal enemy activity posed a unique problem for the
Republic. 92 In the beginning it was because state collapse had allowed
such individuals and groups to act with impunity. Thereafter it was be-
cause, the Republic’s critics notwithstanding, a democratic polity would
always be relatively constrained in the methods it could employ against
the fifth column. This was true even though, as the war went on, fifth
column activities posed an ever greater threat to an economically be-
sieged and internationally isolated Republic which, short of everything,
had to face up to incessant offensive military action from the enemy.
The rebels never had to operate under these kinds of pressures, which

90 In such a context, the hugely erosive impact of the war and the pressing needs of post-war
reconstruction would, by the very end of 1938, lead Negrin to posit some necessary curbing
of the highly fragmented party politics of 1931—6: see Marchenko to Voroshilov (reporting
conversation with Negrin of 10 Dec. 1938) in Radosh, Habek and Sevostianov, Spain Betrayed.
The Soviet Union in the Spanish Civil War (doc. 79), p. 499. But precisely what Negrin thought and
meant by cannot be read ‘transparendy’ from a Comintern report.
this
91
On the column, see J. Cervera, Madrid en guerra. La ciudad clandestina igg 6-iggg (Madrid:
fifth

Alianza, 1998) and M. Uribarri, La quinta columna espahola, vol. 1 (Havana: n.p., 1943). See also
discussion in chapter 3 above.
92
enemy spies could not be easily identified by cultural markers
‘Invisible’ in the sense that these
such as language, accent, manners or customs.
 \egrin's war on threefronts 343

ineritabh magnified internal political and social tensions. These were

precisely the factorswhich ‘armed* the fifth column so powerfully in
Republican Spain. The opportunities for home-front sabotage would
increase as time went on. And with an external enemy that w as always
winning, alw ays advancing, fears about the enemy within doubly dan-
gerous because hidden produced an ever heightening state of anxiety
and tension - especially after the material and psychological crises in-
duced by the splitting of Republican territory in April 1938 - wiiich itself
contributed significantly to justifying more repressive and eventually
quasi-militarised judicial practices. 93
If one takes into account these extreme circumstances, then, it is re-
markable that the general historical opinion of the Republic's internal
political and judicial evolution from mid 1937 should be quite so hostile.
Even historical commentators wiio see themselves as broadly sympa-
thetic to the Republic implicitly apply to it standards that no extant
democracy has yet attained in wartime. In modern times, w ar has al-
ways eroded constitutional practices. In the Second World War even
'mature* democracies such as the USA and Britain modified judicial
procedures and curtailed the constitutional rights of their citizens some
of wiiom were interned in the name of the w ar.- But in Republican
Spain democracy w as only five y ears old witen polity and society were
plunged into a gruelling war and a civil w ar at that — something which is
always more brutal. Ine\itabh the norms ofsuch a young democratic cul-
ture could not have permeated all sectors of the state or society Yet if one .

compares Republican police or judicial practices with the constitutional

realities of European democracies like Britain and France during the First

World War - a more appropriate historical comparison in this respect

than the 1939-45 conflict - then the untenability and sheer anachronism
of many of the criticisms of the Republic's constitutional flaw s become

93 CL Hidalgo de Cisneros’ outburst of firustraoon ai the potential cost of guaranteeing civil ifoerces
tor the enemy: de la Mora. In Place sf Spiexdm. p. 344. The effects of the dkison of Republican
dik chapter But k is worth noting here that the existing Literature
territory are discussed later in
ignores the scale of the resulting chaosand dislocation- There is a vivid oral account by one
American International Brigades Irving Goff. who. as a member of the Republican guerrilla
CMps.wa5sentwkhactJleague.BilAah0.10tr. mvain to stem the retreat ofRepublican soldiers
occasioned by the breakthrough of Francos armies to the Medkerranean. Taped interview with
Jim Carriger, courtesy of Peter Carrofl. The tapes are also lodged in the Bancrof Librarv.
University of California Berkeley
93 C£ Emergency Powers Bi! in Bri tain Me. 1(440 . Calder The .1fytk sf the B&tiz. 123. See also
A. W B. Simpson, fa the Highest Dcgm Qdkw: Deiemtxm zeztkmt Thai m \ 1krtmt Rntam Oxford:
Oxford University Press. 1944 .
 y

344 The Spanish Republic at war

apparent. 95 For they assess it not in the real context of its time, place and
culture, but against some idea of Republican perfection. 96
It was the need to respond to the very real problem of espionage

and sabotage in Republican Spain that led Prieto to consolidate and

strengthen the state’s intelligence services. On
9 August 1937 he an-
nounced the creation of the SIM (Servicio de Investigation Militar). 97
Responsible directly to him as defence minister, the SIM functioned pri-
marily as a military police force. But given its counter-espionage brief, it

was also responsible for gathering political intelligence on - and there-

- the home front. The vital importance of such functions in
fore policing
wartime had been brought home by the May days and reinforced by the
events surrounding the detention of the POUM leadership in mid June
1937 . A vulnerable Republic needed its own intelligence instrument with
incisive powers to separate rumour and hearsay from hard evidence of
treasonous activities. In view of the scandal surrounding the disappear-
ance of POUM leader Andreu Nin, 98 Prieto’s concern about extraneous
‘friendly’ politicalagendas (namely the Soviet Union’s) was scarcely sur-
prising. 99But as we shall see later, it is important to separate the specific
political anxieties Prieto and other ministers may have had as a result of
the Nin affair in 1937 from the dynamic of SIM expansion during 1938,
which was overwhelmingly a response to the needs of the war.
Nevertheless, the circumstances of Nin’s disappearance in June 1937
had provoked an international outcry. Negrin was swamped by telegrams
and letters of protest from abroad. As well as being a foreign-policy and

95 See, for example, B. F. Martin, The Hypocrisy ofJustice in the Belle Epoque (Baton Rouge: Louisiana
State University Press, 1984) - a useful study of the quite acute limits on defacto (as opposed to
de jure) political rights under the Third Republic (an example on p. 234); Simpson, In the Highest
Degree Odious. See also later in this chapter for the discussion of death sentences passed by the
Republic’s special courts.
96 To which the enduring myth of the European and American left - ‘the last great cause’ - has
doubtless inadvertently contributed.
97 Decree published in the Diario Oficial del Ministerio de Defensa Nacional, Bolloten, SCW,
p. 547. The draft text stipulating the SIM’s remit was personally composed by Prieto, Como
por que sail del ministerio de defensa nacional, p. 77. Clearly this did not mean that the Republic had
lacked intelligence-gathering instruments before this, or indeed that the SIM had a monopoly
on intelligence work after its creation: D. Pastor Petit, Espionaje. Espaha
1936-1939 (Barcelona:
Bruguera, 1977), p. 112; Payne, The Spanish Revolution, pp. 346-7. In the aftermath of the coup of
18 July 1936, the intelligence services had presented the Republican government with the same
problem as the rest of the state apparatus: how to tell who was loyal among a personnel largely
inherited from the monarchy.
98 For the Nin kidnap and assassination, see chapter 5 above.
99 See Orlov’s comments on this as part of his questionnaire response to the historian Stanley Payne
in 1968, in Forum fur osteuropaische Ideen und feitgeschichte,
4 (2000), 229-50 (Orlov’s comments,
pp. 245—6). These are also cited in Bolloten, SCW, pp. 546—7.
 Negrin’s war on threefronts
345

it also undermined the fundaments of the liberal

public-relations disaster,
state to which the prime minister was committed. Police action was,
100
as we have already discussed, occurring beyond ministerial control .

The challenge facing Negrin was to reimpose civilian political authority

and to deal with the underlying problem of security - in no small part
related to the increasing fears concerning espionage and fifth column
activity generated by Republican political and military vulnerability. It
was for both these reasons that Negrin introduced in late June a special
court for espionage and treason (the Tribunal Especial de Espionaje y
101
Alta Traicion ), whose first task was to regain control of the process of
investigating the charges against the POUM.
There seems litde doubt that Soviet representatives - whether from the
intelligence services or the Comintern remains unclear - helped forge
a link (quite literally) to try to implicate the POUM in other ongoing
102
discoveries of espionage activity against the Republic . It is crucial to
bear in mind this context of real espionage, in which the rumours against
the POUM became embedded, to appreciate why the accusations were,
103
up to a point, ‘successful ’. The imperative of war is a violent one,
not least because it renders most sorts of ambiguity ‘unsafe’. Real and
present danger invested the anti-POUM rumours with a power inside
104
Republican Spain that they would not otherwise have had But the .

105
forged evidence against the party leadership was very poor was not . It

taken seriously by anyone in the Republican judiciary or polity who saw

it. Notably disparaging were Irujo and the Esquerra’s Jaume Miravitlles,

106
Generalitat propaganda minister In the small circle of the Catalan
.

100
See chapter 5 above.
101
Gaceta de la Republican 23 June 1937. Text cited in Alba and Ardevol, El proceso del P.O.U.M.,
pp. 529-35. These courts were presided over by five judges - three civil and two military. All
were ministerial appointees - respectively from justice (two civil), interior (one civil), defence
(made nominations to the justice ministry for the two military positions).
102
See chapter 5 above.
103
Sole i Sabate and Villaroya i Font, La repressio a la reraguarda de Catalunya (1936—1939), vol. 1,

pp. 243-58; Cervera, Madrid en guerra. La

ciudad clandestina, pp. 283-337. Cervera deals with a
number of fifth column networks, including the one used by Orlov to forge evidence against the
POUM (p. 303). Some of the author’s further deductions regarding Nin’s death are, however,
rather dogmatic - given the current state of our knowledge.
104
Azana, Obras completas, vol. 4, p. 692 (La Pobleta); cf. the efficacy of Franco’s rumour mongering
here too: Cattell, Communism and the Spanish Civil War pp. 146-7. ,

105
J. Gorkin, El proceso de Moscu en Barcelona (Barcelona: Ayma, 1974), p. 163; see Bolloten, SCW,
pp. 513-14 for a resume.
106
Bolloten, SCW, p. 509, p. 890 (n. 58). Miravitlles’ disbelief in I. Suarez, El proceso contra el POUM
(n.p.: Ruedo Iberico, 1974), p. 172; J. Miravitlles, Episodis de la guerra civil espanyola , p. 189; Peirats,
La CNT en la revolucion espahola, vol. 3, p. 238. Miravitlles had once been a member POUM
himself: Low and Brea, Red Spanish Notebook, p. 202.
 1 ,

346 The Spanish Republic at war

political left, tightly knit notwithstanding its organisational divisions, it

is inconceivable that anyone believed the accusations. It is improbable

too that a politician of Negrin’s sophistication did not also appreciate
the denseness of the political web being spun around Nin’s disappear-
ance, although the prime minister had no specific information: indeed,
the very lack of it was precisely the measure of the
crisis this supposed for

Republican state authority Negrin’s comment to both President Azana

and Zugazagoitia that ‘anything was possible’ is reasonably astute and,
taken in its full context, need not have related only to his voiced thoughts
on putative Gestapo involvement. 107
But in spite of Nin’s fate and the detention by the Republican authori-
ties of other POUM executive members, the party continued to have an

organisational existence of sorts right through until the spring of 1938.

The small sections of the POUM in Valencia and Castellon continued
to participate in municipal politics, retaining their city councillors until

December 1937. 108 In its Catalan heartland, the POUM’s party-political

life was more precarious. Nevertheless, a new executive was appointed

to replace the imprisoned one. 109 This picture alerts us to the error of
seeing the POUM’s post-May
1937 experience as part of a seamless
whole of imported and monolithic stalinist persecution. The specificity
of the Catalan situation should also return our attention to the internal
dynamic of longstanding rivalry and animosity on the Catalan left and
thus to themany unanswered questions about the PSUC’s part in POUM
repression. Instead, the Anglo-American historiography — following
somewhat uncritically the Catalan POUM’s memoir material 110 - has

107
Azana, Obras completes, vol. 4, p. 692 Zugazagoitia, Guerray vicisitudes, pp. 292-3 PSOE executive
;
.

member Juan-Simeon Vidarte wrote that after the war Negin told him that he believed ‘the
communists’ had killed Nin: Todos fuimos culpables, vol. 2, p. 729; also Azana, Obras completes,
vol. 4, p. 638.
108
See V Alba and S. Schwartz, Spanish Marxism vs. Soviet Communism (New Brunswick, n.j.:
Transaction Books, 1988) for the POUM’s municipal activities after May 1937. Supporting
contemporary press sources are also cited by Bolloten, SCW, p. 890 (n. 79). An order to dissolve
the POUM was issued by the examining magistrate in December 1937: Alba and Ardevol, El
proceso del P. 0 U.M. p. 47
.
,
This would take effect in the wake of the guilty verdict at the October
.

1938 trial: Elorza and Bizcarrondo, Queridos camaradas, p. 380.

109
Bolloten, SCW, p. 513, p. 891 (n. 84).
110
In particular those by Julian Gorkin and Victor Alba, which tend repeatedly to reduce complex
national political conflicts to the ‘hidden hand of Moscow’. It is interesting to contrast this with
the party’s own criticisms in a letter to Azana in December 1937 when the focus is entirely
on the PCE’s own sectarianism: POUM to Azana, Archivo particular de Manuel Azana (AB),
apartado 7, caja 137, carpeta 9. Gorkin and Alba’s own bete noire, Orlov, in his reply to Payne’s
1968 questionnaire, shrewdly remarked on the evident political interest, in a post-war era of
Cold War ascendancy, of reducing the POUM repression to an ‘affair of the Soviets’ (Orlov in
Forum far osteuropaische Ideen und £'eitgeschichte 4 (2000), 239). Moreover, the Valencian POUM,
 JVegrin’s war on threefronts 347

tended instead to conflate the June 1937 detention and the October
1938 trial. But it was only in 1938 that the party was obliged to oper-
ate semi-clandestinely.
111
The new POUM executive was imprisoned in
April 1938, when the splitting of the Republican zone sparked a major
military and political crisis. The party’s fate would therefore ultimately
be tied up with the much larger process of militarisation happening un-
der the impact of the war, as the rebels stepped up their offensive and
as the Republic’s diplomatic isolation became total and its home-front
112
conditions desperate.
As the sense of siege increased towards the end of 1937 - with the
of the north and
final loss the diplomatic impasse over Non-Intervention
and troop withdrawal - Negrin sought the institution of special courts to
tighten up prosecution of ‘espionage, treason and defeatism’." 3 These
emergency courts (Tribunales Especiales de Guardia), like the already
functioning ones for espionage and high treason, would apply a highly
summary procedure reminiscent of military courts which thus meant
the suspension of normal constitutional guarantees for the defence of
the accused." 4 By way of safeguard, death penalties imposed by the
emergency courts had to be ratified by the cabinet. This infringement
of the independence of the judiciary was criticised at the time, as it has
been subsequently by historians, as unconstitutional - precisely because

while itself critical of the PCE’s behaviour, was also hostile to the Catalan POUM’s political
position. Years later, in 1973, Joaquin Maurin himself - whose own very different experience
of the war meant that he was never again able to tune in to the POUM
leadership’s post-Nin
wavelength - would write to Victor Alba in terms critical of the party’s wartime stance: Elorza,
‘La estrategia del POUM
en la guerra civil’, in La II Republic. Una esperanzaJrustrada (Actas del
congreso Valencia Capital de la Republica) (Valencia: Edicions Alfons el Magnanim, 1987),
pp. 133-6.
111
Bolloten, SCW, p. 513; Alba and Ardevol, El proceso del P.O.U.M., pp. 135, 145; Elorza, ‘La
estrategia del POUM en la guerra civil’, p. 135. Orlov’s own intelligence reports for spring 1938
refer to ongoing surveillance: Costello and Tsarev, Deadly Illusions, p. 364.
112
Cf. Alba, ‘De los tribunales populares al Tribunal Especial’, Justicia en guerra, p. 232. The author
not only ignores all of these factors, but also, unaccountably, appears to think it reasonable to
compare Republican democracy in the 1930s with late-twentieth-century western models. By
the same token of course, Alba’s assessment thereby implicitly disregards the democratic deficits
in liberal parliamentary systems today. Conversely, aspects ofJulian Gorkin’s own account of
his experience of the Republican judicial system sit oddly with some of his political conclusions:
see Gorkin, El proceso de Moscu en Barcelona, pp. 2 10-13.
113
Gaceta de la Republica, 1 Dec. 1937.
114
In the espionage and treason courts two out of the five judges had to be military ones: see
text of foundational decree, cited in Alba and Ardevol, El proceso del RO.U.M., pp. 530-1;
A. Gonzalez Quintana, ‘La justicia militar en la Espana republicana’, Justicia en guerra, p. 186.
The Emergency Courts were presided over by a professional judge but composed of military
and police officials, and conviction was always based on the confession of the accused who had
no counsel: Sole i Sabate and Villaroya i Font, La repressio a la reraguarda de Catalunya (1936-1939),
vol. 1, pp. 268-76, particularly p. 269.
348 The Spanish Republic at war

115
the Republic had not declared a state of war ( estado de guerra). But this

assessment ignores the extraordinary fraughtness of such a step for the

Republican government.
The civil war had begun with a military rebellion against the con-
stituted civilian power. The rebels had themselves declared a state of
war precisely because, according to existing public-order legislation, this
conferred upon the military supreme political authorityfor the duration II6 .

It is, then, entirely unsurprising that Negrin should have

been reluctant
to risk any of his military commanders ‘confusing’ the limits of military
authority once again. Julian Casanova provides the key to what was hap-
pening in 1938 when he refers to summary justice being implemented
117
in a ‘virtual state of war’. Negrin was beginning to operate as if a state
of war had been declared — but without declaring one dejure — for fear
that even graver unconstitutional acts might supervene. Under Negrin’s
system, there was summary justice, but at least it was controlled by the
118
civilian political authority, represented by the cabinet.
However, the implementation of the emergency courts provoked the
resignation in December 1937 of the Basque Nationalist Minister of
Justice, Irujo.
119
He could not accept them - even given the exceptional
120
context of the war. (We should also note that it was Irujo, along with
Prieto, who had consistently voted against the implementation of the
death penalties imposed by Republican courts - the rest of the cabinet
usually voting affirmatively.) Irujo’s under-secretary, Mariano Anso, took

115
The Republic was still governed under the preliminary ‘state of alarm’ {estado de alarma) that had
continued, virtually unbroken, since February 1936: Gonzalez Quintana, ‘La justicia militar en
la Espana republicana’, Justicia enguerra, p. 187.
116
The 1933 Public Order Act passed by the Azana government had not changed the legal provi-
sions which allowed a state of war (or siege) to be declared by the military (without consultation) -
even though this infringed the Republican constitution. Nor did the Republic demilitarise the police. In
July 1936 the rebels had simply issued a ‘Bando declaratorio - del estado de guerra’. Franco
would rescind these provisions by decree in 1948. On this see M. Ballbe, Orden publico y militarismo
en laEspana constitucional, pp. 13, 361-2. Ballbe also indicates that the provisions under the 1933
Public Order Act were even more favourable to the military than under that of 1870. The
r
933 Act envisaged a war council exclusively of military representatives while the 1870 law had
specified military and civilians.
117
J. Casanova, ‘Rebelion y revolution’, in Julia et al., Victimas de la guerra civil pp. 161-2. ,

118
Negrin’s justification in his speech to the Cortes (30 September 1938), reproduced in Alvarez,
Negrin, personalidad historica, vol. 2 pp. 71-2. See also his public reply to the Generalitat’s criticisms
,

in La Vanguardia, 28 April 1938 (‘Justicia dura, pero justicia’), a section of which is cited in Sole
i Sabate and Villaroya i Font, La repressio a la reraguarda de Catalunya, vol.
1, pp. 270-1.
119
Anso, To fui ministro de Negrin, pp. 208-9.
120
M. Irujo, Un vasco en el ministerio dejusticia, memorias (3 vols., Buenos Aires: Vasca Ekin, 1976-9),
vol. pp. 83, 87; A. de Lizarra, Los vascosy la Republica espahola (Buenos Aires: Vasca Ekin, 1944),
1,

pp. 188-90; Bolloten, SCW, p. 514. Bolloten’s insistence that the police courts were communist-
driven is nowhere corroborated - not even by his own account/the material he cites.
 ,

Negrin’s war on threefronts

349

over in the justice ministry until the cabinet reshuffle of April 1938. But
Negrin personally assumed responsibility for seeing the emergency court
proposals through the cabinet. To his mind, the defence of Republican
democracy required exceptional measures - in wartime even more than
back in 1932 when he had argued for the execution of General Sanjurjo,
the leader of the abortive military coup of September. Irujo was never
reconciled to this view, but he did, nevertheless, return as a minister with-
out portfolio, persuaded by the Basque premier, Jose Antonio Aguirre,
that the PNV’s cause would be better served by maintaining a presence
in the cabinet.
Criticism of the emergency courts’ unconstitutionality also came from
Catalan quarters. While these too had a clear ethical base, there was also a
political ambiguity. When the Generalitat complained about the erosion
of ‘political liberty’ it primarily meant the erosion of regional autonomy
and of the political rights of nationalists. The Esquerra had not been
averse to the central state dealing harshly with rebellious sectors of the
libertarian movement or the POUM in Barcelona in May 1937. (Nin’s
disappearance was, of course, of a different order - although here again
it could conveniendy be categorised under ‘the depredations wrought

by Madrid’. 121 ) The constitutionality of police action was an issue - but

in particular because of the implications it might have for centralising po-
litical control. The Generalitat’s main complaint against the SIM was

that it amounted to a Spanish imposition, which breached the spirit

122
of the autonomy statute. SIM personnel were mainly non-Catalans
drafted into the region. Their lack of ‘local knowledge’ may have made
their actions more swingeing on occasion - although, as a military police
force, nowhere was the SIM noted for its lightness of touch. Nevertheless,
there was an additional cultural dimension to the SIM’s unpopularity in
Catalonia 123 - a point we have already noted in relation to earlier
(Madrid) police action in the detention of the POUM’s Catalan lead-
ership. By the time the POUM trial came about in
October 1938, the
Catalan nationalists’ worst fears about political centralisation would have
been realised. Even so, the eventual sanction of the Republican courts
121
See Azana’s bitter comments here: ‘It comes as no surprise that [Companys and his supporters
in the Generalitat]have exploited the Nin case. That Companys affects moral indignation and
assumes the role of constitutional champion after everything that has happened in Catalonia
under his presidency is intolerably cynical’, La Pobleta diary entry for 28 July 1937, Obras
completas, vol. 4, p. 699; see also Azana, Apuntes de memoria p.78.
122
The SIM in Catalonia nevertheless had its own separate command structure: D. Pastor Petit,

Espionaje.Espana 1936-1939 (Barcelona: Bruguera, 1977 ), p. no; Payne, The Spanish Revolution,
p. 346 (citing Manuel Uribarri (not Ulibarri), who headed the SIM for a period during 1938).
123
Sole i Sabate and Villaroya i Font, La repressio a la reraguarda de Catalunya, vol. 1, p. 260.
350 The Spanish Republic at war

against the POUM - for rebellion against the constitutional order 124 -
was entirely consistent with the model of social and political order to
which the Generalitat subscribed and which it had been keen to see re-
established in May 1937 And, as the outcome of May had demonstrated,
.

the Republican state was engaged upon reconstructing economic, social

and cultural hierarchies appropriate to such an order. 125
Accordingly, as we have seen, the Negrin government encouraged
the return of some leading political conservatives and restored to them
control of their property. Indeed, more generally too, the burden of eco-
nomic decisions by Republican bodies was in favour of restoring property
rights to former owners provided that no active complicity in the mil-
itary rebellion could be proved. Nor were those of more conservative
political persuasion targeted by the post-Popular Tribunal Republican
authorities for the merefact of their political opinions, previous allegiances,
cultural values or economic
Republican practice was an attempt to
class.

be consistent with the liberal principles enshrined in President Manuel

Azana’s speeches and which also informed Negrin’s policies: the rights of
individuals over those of social groups were to be guaranteed by due con-
stitutional process. The Republic had overcome the arbitrary power of
the committees and patrols: illegal detentions and paseoswere now rare. 126
At the same time, there was an increase over 1937—9 m the numbers of
people detained in prison and in work camps (where those detained per-
formed war work on fortifications and the like). 127 The CNT denounced
the economic and political conservatism of the ‘new Republic’, claim-
ing that the increase was made up of social and political constituencies
identified with itself. In other words, the Republican legal system was be-
ing used to discipline the left and ensure that it accepted the regime’s
political and economic authority. (In this, of course, the agenda was
not essentially different from 193 1-3.) But it had been the CNT’s own
Garcia Oliver who had created the work camps in December 1936 in

124
Alba and Ardevol, El proceso del P.O.U.M., pp. 479-90, esp. pp. 485-7. The court specifically
found there to be no case to answer in terms of treason or espionage.
125
See Raguer, Lapolvoray el incienso, pp. 328-9. For the Republic’s imposition of hierarchies - apart
from the obvious examples of the nationalisation of war industries and the construction of the
New Republican Army, consider also the spring 1937 decree excluding militiawomen from the
fronts and the fact that May 1937 saw an end to the publication of anonymous soldiers’ poems
(the famous romanceros) in favour of rather more elite focused literary production. This change
is apparent, for example, from Mono Azul, the literary weekly published under the auspices of

the Alliance of Anti-fascist Intellectuals - a body epitomising the Popular Front ethos.
126
See Irujo cited in Raguer, Lapolvoray el incienso, p. 329.
127
Sole i Sabate and Villaroya i Font, La repressid a la reraguarda de Catalunya, vol. 1,p. 283; pp. 281-6
includes some extracts from testimonies; Julia et al., Victimas de la guerra, pp. 256-7.
 )

JVegrin’s war on threefronts

35 1

his capacity as justice minister. The increase in the scale of their opera-
tions has to be understood most immediately in the context of the war
and the deterioration in the Republic’s position. It seems probable also
that the restitution of a liberal Republican order in law, economy and
society would have had a impact on the sociological
qualitative profile
of the Republican prison population over the war. But the very fact of
war, especially a civil war, complicates the prison picture and makes this
128
difficult to assess.

As far as the overall erosion of popular morale on the Republican home

front is concerned, however, we should be careful of exaggerating the im-
pact ofjudicial questions in a society whose experience of constitutional
rights freely exercised was still quite limited. Far more important sources
of the erosion of morale among large sectors of the population were
hunger, shortage and the increasingly frequent severe air raids on big
urban centres against which the Republic, short of fighter planes, could
offer people little protection. These problems became most acute in
Catalonia. December 1937 saw the mass bombing of Barcelona. In
March 1938 came a second saturation Italian bombing wave, more in-
tense than anything inflicted on Madrid. In its effects - and probably
in the intent too — the damage was overwhelmingly civilian. Delayed-
fuse and lateral-force bombs were used - both designed to cause max-
imum civilian casualties. Diplomatic protests proved futile. 129 Wealthy
districts were also bombed - including the Ritz and Hotel Majestic,

where foreign correspondents were hit. But the bombing hit hardest in
refugee-crammed areas. As the rebel armies advanced, so an unending
flood of refugees poured into Catalonia. By late 1937 the food situa-
tion was dire. Among the urban poor, hunger bordered on starvation -
as indicated in Quaker reports of wartime relief work, which include a
wrenching account of conditions in one anarchist ‘ragged school’ in the
centre of Barcelona. 130 For this besieged and bombed urban population,

128
For example, the impact of black-market activities or low-grade ‘fifth column’ offences. P. Pages
i Blanch, La guerra civil espanyola a Catalunya (igj 6—igjg), 2nd edn (Barcelona, makes a
1997)
start and bears out the libertarian /left testimonial literature. See also P. Pages i Blanch, La preso
Model de Barcelona. Historia d’un centre penitenciari en temps de guerra (igg 6-igjg (Barcelona: L’ Abadia
de Montserrat, 1996), pp. 303-404 and C. Canellas et al., Historia de la preso Model de Barcelona
(Lleida: Pages, 2000).
129
After the Nyon conference in September 1937, Italy modified its tactics. The rebels were given
Italian submarines and Italian aircraft on Majorca flew with Spanish markings. From then on,
bombing of all Republican ports and cargo ships thus bound could be carried out with virtual
impunity for the rest of the war: Alpert, A New International History of the Spanish Civil War, p. 145.
130
Report dated 1 November 1937, FSC/R/Sp box 1 (file 3) and 14 December 1937, FSC/R/Sp
box 5.
352 The Spanish Republic at war

increasingly ground down by hunger and shortage, the situation was

psychologically even harder to bear in the absence of any sign of relief
or improvement on the international horizon.
Integral war weariness was, then, also part of the cumulative, collateral
effect of Non-Intervention. Not only had this made it impossible for the
Republic to sustain an offensive war, but it had also undercut its ability
to provide adequate staple supplies to sustain the home front and com-
pensate for the severe dislocation of the national economy. Maintaining
resistance under conditions of Non-Intervention obliged the straitened
Republic to spend virtually all its resources on war materiel and, there-

after, to concentrate its scarce staple supplies on the army. Prieto had

always said that the winning side would be the one with the health-
ier home front. Ultimately, increasing hardship and hunger were what
undermined the Republic’s legitimacy on that home front.
This was also true away from the towns and cities. Hunger was not
by any means a monopoly of urban dwellers. Rural populations, as pro-
ducers (whether collectivist or individual) more often had direct access
to food. But harvest difficulties or being on the wrong side of village
power relations could cancel out any advantage here, 131 as did the abid-
ing problem of all wars, namely that the army began to live off the land.
Across the entire Republican zone, rural populations wearied of the
army’s ‘parasitism’, which compounded the other effects of economic
dislocation.Complaints once made against the militia were now levelled
against army units. In Aragon, where many of these retained autonomous
supply functions, 132 there were criticisms of soldiers helping themselves
to poultry and other produce. ‘Informal’ requisition often seemed little
more than pillage. The CNT press warned against ‘banditry’ - while the
PCE, which owed its political entree to Aragonese politics to the prob-
lem of militia depredations, sought to impose tougher military discipline.
How successful it was in ameliorating this particular problem remains
unclear. Certainly there was a deterioration in the situation during the
Republican military retreat through Aragon to Catalonia in the spring of
1938 when numerous violent conflicts between civilians and soldiers were
131
Ucelay da Cal, ‘Cataluna durante la guerra’, pp. 348-9. The author makes a number of
important points about the complex question of which groups went hungriest in Republican
Spain during the war. However, he is rather optimistic about the fate of urban workers. Many
sectors of the urban poor would not have had access to the union facilities he describes. In any
case, the Generalitat’s economic liberalisation had seen the abolition of the food committees
responsible for provisioning poor neighbourhoods in 1936, and it had reduced the efficacy of
initiatives like consumer co-operatives too (see chapter above). Nor were these or the communal
4
kitchens provided by unions immune to the growing food crisis of 1938.
132
Casanova, Anarquismoy revolution, p. in.
 Negrin’s war on threefronts
353

reported. 133 Moreover, in Aragon as elsewhere, the PCE’s stringent dis-

ciplinary line in all matters would frequendy be interpreted by the CNT
as anti-libertarianism. The CNT regularly defended young deserters, or
peasants discontented with the government’s requisition of their crops —
which meant, in consequence, heightened intra-organisational political
tensions. 134 These were increasing more generally too by 1938 as civilians
became alienated by other manifestations of militarisation such as the
increased billeting of troops in houses and school buildings. 135 One can
see this weariness too in the rising tide of general complaints directed
against military personnel - for example the careless driving of military
vehicles, some places had led to accidents and even fatalities. 136
which in
The general attrition attendant on wartime economic dislocations
came to affect smallholders, tenant farmers and those in collectivised
agriculture alike. They were already subject to unpopular war taxes. A
shortage of labour and material (seed and fertilisers) as well as worsening
conditions of production eroded both yield and producer morale. 137
Collectivists and smallholders alike would also come to criticise gov-
ernment measures such as the brigadas de cultivo (agricultural brigades)
under the control of the UGT and the CNT landworkers’ federations.
These were workers drafted in from urban centres to replace the agri-
cultural labour force lost to conscription or other forms of war work. In
a sense, one can appreciate the complaint levelled by Ricardo Zabalza,
leader of the UGT land federation (FNTT), that it would have been
more effective to create general brigades for fortification work, leaving
work their own land. By the same token, such
the rural population to
an exchange between rural and urban environments might, in different
circumstances, have had the potential to forge a greater sense of na-
tional political cohesion for the Republic. But the seriously deteriorating
conditions ruled out any potential for such a ‘Republicanisation’ or ‘na-
tionalisation’ of differing political and social constituencies. Instead, the
rural world turned in upon itself. Collectives as well as smallholders were
quite ready to hoard food. 138 While this provoked urban complaints, a
>33 For example, soldiers taking bread from a Catalan village shot dead a CNT municipal officer
attempting to stop them: Julia et al., Victimas de la guerra civil, pp. 253^
134
See Elorza and Bizcarrondo, Queridos camaradas, p. 407 for an example of how intra-
organisational tensions were now bound up with economic tensions and war weariness - espe-
cially around issues of compulsory government purchase of crops.
135
This was already a problem in Aragon by late 1937: Azana, Obras completas, vol. 4, p. 812.
136
Simeon Riera, Entre la rebelioy la tradicio, pp. 233-4.
137
For the exhausting conditions of production (including under fire) and its attrition of the southern
collectives, see Garrido Gonzalez, Colectividades agrarias en Andaluda, pp. 95-100.
138
See Fraser, Blood of Spain, p. 371 (n. 1).
354 The Spanish Republic at war

greater factor in urban shortage was the chronic lack of transport to

deliver foodstuffs to consumers. 139 By 1938 the Popular Front would be
undermined by these intractable practical problems as much as, if not
more than, it was by specific political sectarianisms. But the language of
the latter would often be used to give voice to this burden of war weari-
ness and practical despair - a shift to which one needs to be increasingly
atuned as 1938 progresses. The harsher discipline in evidence on both
the military and home fronts would further contribute to this alienation
— although its very intensification was also a desperate response to an
increasingly impossible material situation.
Increasing frustration at the impasse the Republic faced internation-
ally, as well as the attritional tension of maintaining a war effort under
such conditions, inevitably began to deepen the political divisions in
Republican ranks. The CNT’s general secretary, Mariano R. Vazquez,
continued to support the government while remaining deeply suspicious
of the PCE. Many in the movement were much more disaffected, how-
ever, and as the military situation further deteriorated their voices would
become louder. Most worryingly of all for Negrin at this stage was the
accelerating strife in the PSOE/UGT. If he had to contend with CNT
discontent and the low-intensity war with the Generalitat, the very least
Negrin needed was to be able to rely on his own party. 140 But the Cabal-
leristas’ increasingly public dissidence had reached red alert by the autumn

of 1937. Their battle with the PCE for organisational influence in the
UGT was regularly presented as a high moral crusade against ruthless
political opportunists. This inevitably found a response among all those -
in the socialist movement and beyond — who had lost out in the bruising
organisational struggle unleashed by the war. Largo was also able to tap
into a current of incipient anti-Soviet sentiment that was beginning to ap-
141
pear. This too had its source in war weariness and mounting frustration
atRepublican impasse. In such circumstances, anti-Sovietism was the in-
evitable counterpoint of the popular mythologisation of the Soviet Union
in late 1936. The hopes then had been unrealistic, and the resulting

139 This is a constant reprise in Quaker relief reports.

140
As the veteran socialist Gabriel Moron expressed it, ‘with his own party, [Negrin] felt as if he
were communing with the Void’, Politico, de ayery politico de manana (Mexico: n. p., 1942), p. 109;

see also Buckley, Life and Death of the Spanish Republic , p. 401.
141
The government, alarmed by the likely destabilising effects should this escalate, put up discreet
posters in Valencia and Barcelona to request that care be exercised in public pronouncements:
Jackson, The Spanish Republic and the Civil War , p. 405. In Moscow, Pascua also tried to limit the
diplomatic damage caused by the Caballeristas’ increasingly incendiary press editorials: see
AHN/MP, caja 2 (2) for Pascua’s complaints to Negrin (letter of 28 Nov. 1937) and examples of
his diplomatic notelets to Stalin.
 JVegrin’s war on threefronts
355

disappointment - as Non-Intervention strangled arms deliveries and

Stalin began to reassess his position — was commensurately intense. The
fact that the PCE had wrapped itself in the mantle of Soviet strength -
so effectively projected by its own propaganda - meant that it too would
feel the full force of disappointed expectations.
But if Largo was the chief beneficiary here, he overstepped the mark
when he insisted on his right to air his political grievances with the PCE
publicly. The Communist Party was a crucial element in the Popular
Front alliance sustaining the war effort. The Caballeristas’ public and
vociferous anti-communism constituted wrecking tactics of the worst
sort. It was this that galvanised the PSOE
effectively under
executive -
Prieto’s direction —
remove Largo and his supporters from all posi-
to
tions of authority in the party. By i October 1937 they had also success-
fully evicted Largo from the UGT leadership.
142
With a war on, neither
Negrin nor indeed the PSOE leadership could afford to give Largo his
head - however much respect he still commanded as a ‘historic’ leader
of the movement. The incendiary quality of the high-profile personal
animosities involved in this internal socialist dispute should not obscure
the fact that what was at issue in the Caballeristas’ irresponsible politick-
ing behind the lines was the very stability of the Republican home front
in this most difficult of wars.
At the front itself, morale was relatively better for rather longer during
143
193 7- The political commissars had played an important role in curb-
ing desertions and generally serving as a positive force for discipline —
this not only via elementary political education but also by dispensing

practical health care and general education. Looking after the welfare
and interests of their men meant keeping up fighting morale — something
that was doubly crucial in circumstances of such material deprivation.
Prieto’s attempts in late 1937 to normalise the army had, as we have seen,
weakened the commissariat - especially at the lower levels of the army.
Indeed, one could argue that was precisely this erosion that intensified
it

the need for the brutal discipline imposed by the Republican military

142
For events in the party and union see Graham, Socialism and War, pp. 112-31, 167-89 and a
brief summary in Jackson, The Spanish Republic and the Civil War, p. 406. Jackson tends, however,
to overemphasise Negrin’s direct intervention in the proceedings. Prieto was the architect. (It
was Prieto’s (then) political lieutenant, Ramon Gonzalez Pena, the Asturian miners’ leader and
(significantly)a hero of the October 1934 rising, who led the ousting of Largo from the UGT
executive: see Graham, Socialism and War, p. 178.)
143
Although there were incidents - for example the rebellion in a unit being transferred from the
Madrid front to Aragon in August 1937: Azafia (La Pobleta, 31 Aug), Obras completas, vol. 4,
p. 760.
356 The Spanish Republic at war

police (SIM)*once the Republic’s military situation became critical in

1938 - a topic to which we shall return later. 144

On
15 December 1937 the Republic launched a major offensive on
the Aragon front in a further bid to seize both the military and diplo-
matic initiative. After the fall of the north, Franco was concentrating
his forces to storm Madrid - planned for the eighteenth. Teruel was
thus another attempt at a diversion. But the fact that the Aragonese
city was already some two-thirds surrounded by existing Republican
lines, plus the element of surprise, meant a much better chance to follow

through an offensive campaign. Republican morale was good, in spite of

winter weather as bitter as the fighting — which replicated the building-
to-building, hand-to-hand combat of Madrid and Belchite. With rebel
reinforcements rushed from the Madrid zone, it was touch and go. But
the Republicans took Teruel on 7 January. They held it until late Febru-
ary, with International Brigade troops drafted in late January to stiffen

the resistance. But the cost in casualties was huge as they were battered
by heavy air and artillery attacks. Nor did the Republicans have the re-
serves to counter-attack outside the city. In the end the sheer weight of
enemy back-up forced the Republicans to evacuate Teruel before they
were encircled. They had held on so long for reasons of prestige and
morale. But the cost of doing this exhausted the meagre resources of the
Republican army, while, in reality, its courageous performance caused
not a ripple on the international diplomatic scene. 145 Teruel was not a
defeat because of logistical insufficiency or lack of training or discipline,
it was a rout occasioned by material lack - of arms, artillery and, most
ominously, of aircraft.
The Republic’s chances of forcing international diplomatic move-
ment, or France to break ranks with Britain over Non-Intervention,
were fading. In spite of the January breakthrough in Teruel, Chautemps
formally closed the French frontier, making the passage of war material
even more uncertain. In reality, very little aid at all reached the Re-
public during the entire, crucial year from June 1937 to June 1938 - for
the most part because of the extraordinary difficulty of getting it there.
The key to the Republic’s military starvation lay in its enemies’ success-
ful quarantining of its Mediterranean ports. This left only the lengthier

144
This is not intended to imply that the commissariat ceased to be important. It was still in-
strumental in maintaining discipline during the horrendous conditions of the Ebro batde: see
Alpert, El ejercito republicano, p. 203. On both the SIM and the Ebro see later in this chapter.
145
The desperate retreat from Teruel involved the Republic’s forces fighting their way back through
enemy lines. In the process they abandoned a quantity of military hardware the Republic could
ill afford to lose.
 Negrin’s war on threefronts
357

routes via the Baltic Sea and the Atlantic - used by the Soviet Union
146
from December 1937 until summer
But rebel control of the
193 8.
Straits of Gibraltar still meant
was no direct access by sea to
that there
the Mediterranean. The odyssey of Republican war material thus con-
cluded in an always problematic land transportation across France from
the Atlantic to the Mediterranean coast where, from January 1938, it hit
Chautemps’s reclosed border. French policy-making circles had been in
at least several minds over what to do about Spain. But by early 1938
the defence imperative was pulling the Chau temps government towards
seeking its own accord with Fascist Italy. While Negrin perceived this
danger, he chose to continue trying to exert a countervailing influence in
the belief that the mounting tension in Europe would demonstrate the
relatively small worth of such an accord and thus force France to break
ranks with Britain.
Unfortunately for Negrin, however, developments in Britain would
further weaken his diplomatic gambit. Italy had continued to work on
deepening the rift between Chamberlain and his foreign secretary, Eden,
who was increasingly concerned by the implications of massive Axis aid
to the rebels.Chamberlain had himself once hoped that negotiations with
Italy might yield a quid pro quo that would reduce its troop and armament
commitment to Franco. But the nazification of Austria in February 1938
had made him doubly anxious to secure an Anglo-Italian agreement.
Mainstream cabinet opinion was backing him to do this irrespective of
whether concessions could be exacted from Mussolini over Spain. Eden
continued to hold out against any accord with Italy or recognition of its
Abyssinian empire until Mussolini had agreed to cease attacks against
merchant shipping and civilian targets in Spain as well as to end CTV
troop dispatch there. But Eden’s views isolated him in the cabinet, thus
prompting his resignation on 19 February. 147 By mid April an Anglo-
Italian settlement would be a reality even though Ciano had made it
clear that Italian forces would not leave Spain short of a rebel victory.
Non-Intervention inspectors had been in Spain counting foreign troops
since February 1938. But Franco would continue to obstruct the planned
withdrawal through to June and beyond - even though he was only ever
prepared to sanction it at a minimal level of a few thousand Italians. The
Republic could thus bring an end neither to Non-Intervention nor to the
underwriting of the rebel war effort by Axis might - evident in the massed
ranks of planes, tanks, trucks and CTV troops that had confronted the

146
Howson, Armsfor Spain , pp. 235-6.
147
Alpert, A New International History , pp. 143, 152-4.
358 The Spanish Republic at war

Republican army at Teruel. By 9 March 1938 Franco’s offensive to take

control of Aragon was underway Both Quinto and Belchite fell within
the first twenty-four hours of the rebels’ rapid advance.
The aftershock of Anschluss (Hitler’s occupation of Austria) brought
about the temporary reopening of the French frontier by Chautemps,
whose government then gave way in mid March 1938 to a new one under
Blum. 148 But by this stage the rebel army was heading for the Valencian
coast. When it reached it, the Republic zone would be split in two. Negrln
urgendy petitioned the Blum cabinet on 12-14 March
for military aid

1938, since such a division of territory posed an enormous challenge

to Republican military viability and responding to it would require the
urgent and large-scale reorganisation of the Republic’s armies. But the
French government showed little resolve — faced with British opposition
and also its own warnings of a possible Axis reaction.
Chief-of-Staff’s
No doubt residual political ambivalence towards the Spanish Republic
in military and diplomatic circles further inhibited French action, which,
in the end, was limited to a defensive measure: the stationing of more
troops on the Franco-Catalan frontier.
Although some stockpiled material did start to pass across the now
open border, it is unlikely that it did so on the grand scale often implied. 149
In any case, there seems to have been a fall-off in Soviet-procured aid after
December 1937. 150 Certainly this was the impression in France by May
1938 - which only added to the government’s reluctance to aid the
Republic. 151 The Soviet Union did intervene in spring 1938 to facili-
tate the passage of one small arms consignment from Czechoslovakia.
But it was a drop in the ocean. In other circumstances, Czechoslovakia,
as the world’s largest exporter of armaments, would have been an ob-
vious source for the Republic from the start. The Soviet Union could
have acted as the ostensible purchaser to procure the necessary exporta-
tion documentation - was also a signatory to Non-
for Czechoslovakia
Intervention. But the coolness of the Czech government towards the So-
viet Union, in spite of their mutual assistance pact of 1935, had ruled this
out. (As a result, the Republican ambassador in Prague, Luis Jimenez de
Asua, was obliged to do business with the usual array of dubious and cor-
rupt ‘contacts’ who embroiled him in the sadly familiar round of daylight

148
Although the decision towas taken by the outgoing Chautemps cabinet.
open it
149 Alpert, A New International History,
p. 155; Moradiellos, La peifidia de Albion, p. 332.
150
Howson, Armsfor Spain, pp. 239-41 and pp. 302-3.
151
Republican ambassador Pascua also derived the same impression from his Soviet opposite
number in Paris in May 1938: Vinas, ‘Las relaciones hispano-francesas, el gobierno Daladier y
la crisis de Munich’, pp. 176, 183.
 JVegrin’s war on three fronts 359

robbery. 152 ) But, in any case, by 1938 Spain was fading as the focal point
of Soviet foreign policy Escalating Japanese aggression (in the form
of further expansion in China), Anschluss and, by late March, the loom-
ing, increasingly all-consuming Czech crisis itself would all stand in
Negrin’s way. 153 This was so especially because, in spite of cumulative
German aggression, Britain was still absolutely opposed to Stalin’s pro-
posal of a grand alliance against Hider, made at the League (of Nations)
on 18 March.
Meanwhile, as rebel troops rapidly overcame Republican resistance
in Aragon 154 and Barcelona was subject to savage bombing, the pres-
sure on Negrin was mounting. President Azana was looking for a way
of opening up the question of mediation. A current of pro-mediation
republican opinion was circulating around him, swelled by the adher-
ence of Catalan and Basque nationalists as well as some cenetistas and
members of the PSOE. War minister Prieto, in profoundly pessimistic
mood, freely expressed his belief that the army was demoralised by the
enemy’s overwhelming military superiority, that the Republic had there-
fore lost and that it should sue unequivocally for peace. Britain’s im-
movability clinched matters for Prieto. However, he had no illusions
about the rebels, who were unlikely to negotiate over what they believed
they could take by military action. So it is difficult to conclude other
than that Prieto was at least prepared to risk that the conflict would
end entirely on Franco’s terms. 155 Picking up on this, the French gov-
ernment made an offer of mediation via its ambassador, Labonne. 156
The offer was discussed by the cabinet. But it was rejected in the face
of vehement opposition from Negrin, who argued that the Axis would
not allow it to prosper and that even attempting it would damage the
Republican cause at home and abroad at a critical moment when it
152
Howson, Armsfor Spain, pp. 153-63,144.
153
Roberts, ‘Soviet Foreign Policy and the Spanish Civil War 1936-1939’, pp. 90-1. It is as well
to remember that almost all the waymajor crises affecting the Soviet Union’s western
through the
defences (Anschluss in March 1938, Munich in September and Hitler’s Prague coup of March
1939), 1416 Soviet Union was also fighting a relatively large-scale war with Japan in the east
(along the frontier of Mongolia and Manchuria): H. Ragsdale, ‘Soviet Military Preparations
and Policy in the Munich Crisis: New Evidence’, Jahrbiicher fur Geschichte Osteuropas, 47 (1999),
216 (n. 31).
154 Thomas,
For a brief summary of the rebel’s drive through Aragon, see Spanish Civil War,

pp. 798-802. Once again, as at Teruel, the rebels’ air superiority was crucial to their success.
155
Note, for example, Prieto’s scepticism when, in response to the March military crisis, the army
and airforce chiefs of staff, Rojo and Hidalgo de Cisneros, proposed offering themselves to Franco
in a bid to facilitate peace and mitigate its conditions for the rest of the army: Zugazagoitia,
Guerray vicisitudes, pp. 383-4.
156
Zugazagoitia, Guerray vicisitudes, pp. 384-5; Miralles, ‘Paz humanitaria y mediacion interna-
cional: Azana en la guerra’, pp. 268-70.
360 The Spanish Republic at war

could not afford to appear weak. It was this that broke open the po-
litical crisis inside the Republican cabinet. At issue were the purpose
and ethics of resistance as well as the principle of prime ministerial
authority.
On 16 March the PCE and CNT leaderships orchestrated a street
demonstration against ‘the treasonous ministers’ near the Pedralbes
Palace (Azana’s residence in the Barcelona suburbs) to coincide with the
cabinet meeting taking place there. One week later, in a blatant breach
of ministerial solidarity, not to mention press censorship, the communist
education minister, Jesus Hernandez, writing under a pseudonym, also
attacked Prieto in the press. 157 Given Negrin’s view of the press’s cru-
cial function in maintaining morale, he can scarcely have approved, if

indeed he had advance knowledge. The premier, however, did know in

advance about the demonstration of the sixteenth and justified it to the
cabinet. 158 Indeed, he may even actively have suggested it to the PCE
as a way of consolidating his own political position against the threat
of Azana’s resignation and ‘unconditional surrender’ apparently taking
shape against him in the cabinet. 159
But danger was - and would remain - inchoate.
in fact, the cabinet
Negrin’s policy of maximum resistance was approved in spite of repub-
lican reservations. For these did not (nor would they ever) amount to an
alternative policy. The republicans wanted mediation, but had no con-
crete strategy for how it might be achieved. Having weathered the cab-
inet storm, then, on 27 March, an astounded Negrin heard the French
ambassador, Labonne, enquire of him whether he, like Prieto, also con-
sidered the war to be lost. Until that point Negrin had been clear about
keeping Prieto in the war ministry in spite of the Teruel defeat and the
mounting pressure from the PCE to remove him. But the military situa-
tion was critical. Having seen the French frontier reopened in the wake of
Anschluss, Negrin could now brook nothing that undermined his efforts
to get the French to send arms. The thought of the impact of Prieto’s
tragic queen performance in front of Labonne - reprised in front of the

157
Article(s) by Juan Ventura in La Vanguardia and Frente Rojo 23 March 1938. La Vanguardia was
,

more or lessthe Negrin government’s official press medium by this stage. (Its editor was the
socialist deputy, Fernando Vazquez Ocana: see his memoir, Pasiony muerte de la segunda Republica
espafiola, p. 78.) This was not the first press attack on Prieto, however. See, for example, Dolores

Ibarruri’s savage speech of 27 February 1938 to the PSUC, Ibarruri, En la lucha, p. 249.
158
Elorza and Bizcarrondo, Queridos camaradas, p. 413.
159
Vidarte, Todos jiiimos culpables, vol. 2, pp. 823-35, 820-1; Zugazagoitia, Guerra y vicisitudes,
pp. 389-91; Anso, To jui ministro de Negrin, p. 214; Togliatti, Escritos sobre la guerra de Espana,
p. 193. The demonstration was an attempt to evoke the spontaneous popular disapprobation
of 18 July 1936 in Madrid when Martinez Barrio sought to treat with General Mola.
 Negrin’s war on three fronts 361

entire cabinet on 29 March — was finally just too much. 160 The prime
minister went back on his arrangement with the PSOE executive. He
161
still wanted Prieto in the cabinet, but not as defence minister. But
Prieto, pardy from conviction, pardy because he felt his own political
reputation to be on the line, insisted on the war portfolio as the price of
his cabinet participation. His fatal obstinacy, not devoid of egoism, was
reminiscent of his great rival and antagonist, Largo Caballero, almost a
162
year previously.
meant that, in the end, Negrin had no choice
Prieto’s intransigence
but to leavehim out of the cabinet that was reconstituted on 5 April.
For Negrin, resistance — as the Republic’s only viable option - dictated
seamless public optimism from all Republican ministers. In the end,
the real difference between Negrin and Prieto was not their intellectual
grasp of the situation, but their subjective response to it. Negrin drew
strength from adversity while Prieto seemed to cave in before the bleak-
ness. Negrin honed down his energies to a single fierce point directed at
the main strategic objective: sustaining resistance. To this end, he em-
ployed a useful psychological device: he avoided considering the whole.
Instead he concerned himself with what was practically necessary to keep
resistance on track in the immediate term - namely maintaining an army
in the field, supplied and fed. 163 He did this not because he believed the

160
Zugazagoitia, Guerra yuicisitudes, pp. 395-6; Negrin’s own account is in Epistolario Prieto y Negrin,

pp. 23-4 and speech of 1 Aug. 1945, Palacio de Bellas Artes, Mexico, reported in Nouedades,
in
6 Aug. 1945 - a copy of which is lodged in AHN/MP, caja 14 (12). In this speech Negrin says
that removing Prieto was one of the hardest decisions of his life. But he had to do it because the
military situation was verging on utter disaster, with the front broken and Franco’s armies only
an hour’s march from Barcelona. Burnett Bolloten also cites Negrin’s public denial that Prieto’s
removal was the result of Comintern or PCE pressure: ‘on the graves of our war dead . there
. .

is not a word of truth in it’, SCW, p. 581 quoting from Documentos politicos para la historia de la

republica espanola vol. 1 (Mexico City" Coleccion Malaga, 1945), p. 21. Given Negrin’s personal
,

value system, that was a very powerful - and far from formulaic - oath.
161
Negrin, Epistolario Prieto y Negrin p. 55 (n. 7); see Zugazagoitia, Guerra y uicisitudes, pp. 402-4 for
,

his own and the PSOE executive’s attempts (in the persons of Lamoneda and Albar) to mediate
between Prieto and Negrin.
162
In particular Prieto’s behaviour regarding the ambassadorship in Mexico: Zugazagoitia, Guerra
y uicisitudes, pp. 409-10. Prieto was prepared to go in order to undertake negotiations with the

Mexican gov ernment about the receipt of Republican refugees - but only ifthis was an offical cabinet
mandate. As Negrin pointed out, he could not agree to this because the cabinet was a sound box.
In a matter of days the credibility' of the Republic’s resistance policy would be in tatters. See
also Negrin in Epistolario Prieto y Negrin, pp. 53-4 and Graham, Socialism and War, pp. 142—3.
163
Epistolario Prieto y Negrin, pp. 26-8; Irujo, Un uasco en el ministerio de justicia, vol. 1, pp. 82-3;
Moron, Politico de ayer, pp. 83, 86; Zugazagoitia, Guerra y uicisitudes, pp. 421-2. Even hostile
witnesses record this conscious attempt by Negrin to bolster morale: for example see comments
(‘entusiasmo ficticio’, ‘balones de oxfgeno’) by Colonel Cuevas, appointed Director General of
Security in the interior ministry in April 1938, Recuerdos de la Guerra de Espaha (Montauban: n.p.,
1940), p. 58. For an astute summary of the essence of Negrin’s wartime strategy, see Anso, Tojui
ministro de Negrin, p. 215.
 .

362 The Spanish Republic at war

Republic could win militarily. Franco’s destruction of a large quantity of

the Republic’s best troops at Teruel had put paid to whatever remained of
those hopes. But Negrin did believe that the international situation had
to break and that if the Republic could only hold on militarily, then that
would save it. Franco could be forced to negotiate,
seismic political shift
but he would never do so if the Republic’s will to resist were seen, even
momentarily, to waver. Negrin was well aware ofjust how grim material
conditions were for many people on the home front. But as this was part
of the ‘whole’ that he could not significandy mitigate, he chose just to let

This was also true for the closely related problem of the increas-
it lie.

ingly harsh disciplinary codes operating on both ‘fronts’ — military and

civilian. In the circumstances it was impossible to investigate individual
cases of injustice without unravelling the very fabric of ‘iron resistance’
(‘resistencia a ultranza’ - ‘con pan o sin pan’). 164 But both politically
and socially this would become Negrin’s Achilles’s heel. The departure
of Prieto intensified the internal tensions in the socialist movement to
a point just short of open crisis. 165 As an inevitable part of this it also
further incapacitated the functioning of the liaison committees between
the PSOE and the PCE, also rapidly being ground down by their mem-
bers’ sheer physical and mental exhaustion in a war where all the exits
were blocked. In addition, Negrin was now facing the estrangement of
the sector inside the CNT, identified with the general secretary, Mariano
Vazquez, that had until then backed his resistance policy.
The new cabinet line-up was ultimately determined by Negrin’s need
to have unconditional commitment to the war from his ministers. Thus
the republican Giral was replaced in the foreign ministry by Alvarez del
166
Vayo, reoccupying the post he had lost back in May 1937 The removal
of Giral, a very close friend of Azana’s, was also intended to obstruct the
president’s tendency to conduct informal diplomacy on his own account.
This had always been anathema to Negrin’s constitutional sensibilities,
but now the backdoor diplomacy was also absolutely incompatible with
167
his own policy strategy.

164
Cf. Zugazagoitia, Guerra y
vicisitudes, pp. 420, 421-2, 429-30.
165
Graham, and War pp. 136-63.
Socialism ,

166
Although Negrin took personal charge of high-level European diplomacy. Nevertheless, he was
clear about del Vayo’s advantages over Giral: for all that foreign diplomats might wonder about

del Vayo’s grip on reality, he would at least never provide them with defeatist words for then-
governments to use against the Republic: Zugazagoitia, Guerra y vicisitudes, pp. 405-6, 428—9,
430.
1 7 In late October 1936 Azana had, on his own authority, sent Bosch Gimpera to London to
sound out the British over mediation. The veteran socialist leader Julian Besteiro had fulfilled
 Negrin’s war on threefronts 363

The need to support the war effort had also been the crucial factor
deciding the PCE’s continuation in the cabinet — in the form of a sole
minister — Vicente Uribe in agriculture. In mid February Stalin had sent
notice that the party should withdraw because of the negative effect its
168
presence was having internationally. This indicates that Stalin had
not yet entirely given up hope of shifting Britain’s position on collective
security had to come at the expense of the Republic —
— even if this

whose resistance Stalin must by now have considered to be strictly time-

limited. Nevertheless, the Soviet leader had absolutely no reason to want
to hasten this process and particularly not after Anschluss in March. So
the PCE’s leaders, who, understandably, wanted to remain in the cabinet,
were able to argue both truthfully and effectively that, without them, the
policy of resistance itself would be significantly weakened 169 because
Negrin would find it harder to neutralise the pactist caucus gathering
around the Republican president, Manuel Azana. I7 °
Jesus Hernandez’s departure was, nevertheless, more or less inevitable
after the clash with Prieto. Hernandez would become Commissar Gen-
eral of the Army. The CNT’s Segundo Blanco replaced him in ed-
ucation, although his presence in the cabinet did little to counteract
the CNT’s growing alienation from the government. Both the home
office and justice ministry had appointees who were less experienced
than their predecessors but unconditionally loyal: respectively, the socia-

Paulino Gomez
lists (who replaced Julian Zugazagoitia) and Ramon
171

Gonzalez Pena, the ‘historic’ leader of the Asturian miners’ union (who
replaced Irujo). Formerly closely identified with Prieto, Gonzalez Pena
had been engaged across 1937 in the organisational struggle to remove
control of the UGT executive from Largo Caballero. Negrin himself as-
sumed responsibility for the war ministry. Irujo remained in the cabinet
as minister without portfolio, while Giral rejoined it in the same capa-

city. The crisis had thus been managed, allowing Negrin, along with his
a similar function when he represented the Republic at the coronation of George VI in May
1937 Miralles, ‘Paz humanitaria y mediation inter nacionaT, pp. 257-8 and S. Julia, ‘Presidente
:

por ultima vez’, in Alted et al., Manuel Azana: pensamiento y accion, pp. 255-6. For Azana’s own
prediction of the coming clash over the question of mediation, see his ‘La Pobleta’ diary entry
for 7 August 1937, Obras completas, vol. 4, p. 716.
168
Elorza and Bizcarrondo, Queridos camaradas, p. 410.
169
See ibid., pp. 413-16 for a reconstruction of these interchanges.
170
Guerra y revolucion en Espana, vol. 4, pp. 75-6; Bolloten, SCW, p. 582. Nor was the PCE entirely
confident about Negrin’s commitment to resistance: Togliatti, Escritos sobre la guerra de Espana,
pp. 200, 231; Graham, Socialism and War, p. 138.
171
Gomez had done a good job where he had been put in charge of public order - a
in Catalonia,
responsibility incumbent on the central government since the May days: Zugazagoitia, Guerra
y vicisitudes, p. 293.
364 The Spanish Republic at war

chiefs of staffi (where there was also continuity in Rojo and Hidalgo de
Cisneros) to concentrate on maximising the possibilities for military re-
sistance.The PSOE journalist, Zugazagoitia, relieved of the home office
portfolio at his own request, remained close to Negrin, offering impor-
tant support in his secretarial role at the war ministry. 172 Indeed, even
Prieto would remain at Negrin’s disposition for a time. 173 But republi-
can and nationalist disaffection and the widening rift in the PSOE were
inevitably making serious inroads into Negrin’s political support base.
In the first days of April 1938 the northern wing of the rebels’ advance
into Aragon took the city of Lleida and then the important power sta-
tion at Tremp, temporarily blacking out Barcelona and decreasing its
industrial output thereafter. Meanwhile, the rebels’ central units drove
down the Ebro valley to the Castellon-Valencian coast. 174 On 15 April
General Alonso Vega reached it at Vinaroz and split Catalonia from the
centre-south zone. On the same day Britain signed the Anglo-Italian
naval agreement and continued to pressure France to close the frontier -
even though Britain’s own merchant ships were still being sunk by Franco.
As always, the assumption held — as it would most damagingly of all in
the escalating Czech crisis - that there was no Axis (and more particu-
larly no German) bluff to be called. Britain’s persistent overestimation of
Germany’s rearmament levels eroded any belief in the strategic purpose
of diplomatic resistance to its expansionist aims.

172
On Julian Zugazagoitia ’s important working relationship and friendship with Negrin - which
endured until virtually the end of the war - see Zugazagoitia ’s own memoirs, Guerra y vicisitudes,

pp. 403-4, 407-8, 410, 411, 427-8, 434. Zugazagoitia’s memoirs, first published as Historia de la
guerra en Espaha in Buenos Aires, 1940, are probably the best high political memoirs ever written
on the civil war. As a friend of both Negrin and Prieto, Zugazagoitia tried immensely hard to
prevent the clash between the two in 1938. As a veteran socialist, he was discreetly critical of
what he saw as Negrin’s excessive tolerance of PCE sectarianism: Graham, Socialism and War,
p. 280 (n. 54) and letter to M. Pascua (Paris), 25 April 1938, AHN/MP, caja 2 (bis)
139, p.
16 (Embajada de Paris). But what finally strained the personal relations between the two was
Negrin’s increasingly erratic manner of working (a personality trait exacerbated as the wartime
pressure on him mounted), which exasperated the punctilious and methodical Zugazagoitia. He
repeatedly complains (but frequently more in sorrow than in anger) that Negrin is impossible to
help. This crucial tension is already evident from the published memoirs, but it becomes much
clearer when one consults his (sharp and witty) letters to Pascua, for example 11 and 26 April
1938, 5 May, 17 and 20 June and 6 Oct. 1938, AHN/MP, caja 2 (bis) 16 (Embajada de Paris).
Nevertheless, Zugazagoitia felt compelled to withdraw from Prieto’s orbit because of his ‘war’
against Negrin. Julian Zugazagoitia was detained by the Gestapo in France in 1940. He was
handed back to Franco and executed. Memorial articles in El Socialista (Mexico D.F.), 1 Feb.
1942.
173
For the post hoc ideological construction of the April cabinet crisis, see Graham, Socialism and
War, pp. 136-7.
174 For a useful map of the rebel advance, see Thomas, Spanish Civil War, p. 799; also Jackson, The
Spanish Republic and the Civil War, p. 410.
 JVegrin’s war on threefronts 365

Meanwhile, as Franco had more and more Italian aid and troops as
well as US trucks and oil, the ruptured Republic faced a massive crisis on
both its military and home fronts. Indeed, militarily, the war might even
have been over at this point. In the period immediately after the splitting
of the zones, the Republic’s defence situation was the worst it would

ever be any point

at in the entire war. There was no continuous front
between Vinaroz and Barcelona. If the rebels had gone straight on for
Barcelona at that point, then they could not have been stopped. 175 But,
to the astonishment of the Republic’s political leaders and its military
high command, Franco turned away from Catalonia, instead diverting
his troops south for a major attack on Valencia. 176
Communication between the two Republican zones was extremely dif-
ficult. Radio contact was uncertain and intermittent. Axis submarines

bombed sea traffic, effectively putting Valencia out of reach of Barcelona,

except by aircraft. But these had a limited capacity and, though relatively
more reliable, they were still subject to enemy attack. Food for Catalonia
had therefore to come from France. But the increasingly fraught politics
of the border made it a precarious source. Catalonia, with its massive
number of refugees, suffered acute food shortages. After the definitive
closure of the French frontier in June 1938 a subsistence crisis of major
proportions loomed. 177 But all over Republican territory, subsistence
and deteriorating material conditions fed an acute sense of vulner-
crisis

abilty, isolationand danger. 178

The survival of the Republic beyond April 1938 depended on the
rapid reorganisation of its armies and the equally rapid political gal-
vanisation of its home front. These were the imperatives that made the
PCE indispensable to Negrin if he was to be able maintain his resistance
strategy with a view to forcing Franco to the negotiating table. 179 The
prime minister and the party promoted in parallel similar ideas about
the war’s nature and objectives. Under the banner of Union Nacional

175
J. Negrin, speech, Palacio de Bellas Artes, i Aug. 1945, Mexico, AHN/MP, caja 14 (12); Azana,
Obras completas, vol. 4, p. 537.
176
Azana, Obras completas, vol. 3, p. 537; V Rojo, /Alerta los pueblos! (Buenos Aires, 1939; Barcelona:
Ariel, 1974), pp. 40, 46-50, 54-5.
177
This is very clear from Quaker relief work sources. Miscellaneous field reports for 1938 in
FSC/R/Sp/box 1 (file 4); /box 2 (files 3 and 4); /box 4 (field reports, vol. 2: (D. Ricart) report
on Catalonia in 1938).
178
Zugazagoitia commented to Pascua on this too, observing that Negrin’s speeches of exalted
resistance were failing to connect with this public mood, letter of 20 June 1938, AHN/MP caja
2 (2) 16.
179
To manage the territorial division, the PCE created a double party leadership - the main one
in Barcelona included Diaz, Ibarruri, Uribe, Delicado and Anton, while in Madrid, Checa,
Hernandez, Giorla, Mije and Dieguez functioned as a deputy leadership.
366 The Spanish Republic at war

(national unity) they reinforced the civil war as primarily one against for-
eign colonisation and for the right to build a national polity in which
Spaniards of different classes and ideas could participate. 180 The same
liberal constitutionalist nationalism underpinned Negrin’s war aims -
181
the famous ‘Thirteen Points - published on May.
5
This most moder- 1

ate of political programmes was intended for international consumption

and as a blueprint for mediation — while, inside Spain, ‘Union NacionaP
was designed to attract conservative and Catholic elements in both zones
who were uneasy at the increasing influence of Germany and Italy. In line
with this policy, the leadership of the united socialist-communist youth
organisation (JSU) proposed extending membership to Catholic youth.
The move was denounced by many former young socialists who, already
worried by the implications of Popular Frontism for the left, saw it as a
5 182
further attempt to ‘decaffeinate the youth movement ideologically.
5
But if ‘Union Nacional accelerated the break-up of the JSU, nor was
its espousal an easy step for the PCE. It went beyond the widely under-

stood definition of the Popular Front and was tantamount to ‘stop [ping]
being Communists altogether, at least until the war was won 183 The
5
.

PCE leaders toed the line. But that was far from meaning that all were
convinced by the subordination of a specifically class-based discourse
and worker interests to ‘national unity 184 Nor were the PCE’s leaders
5
.

blind to the dangers involved for the party in submerging its identity in
the war effort, given the political climate of mid 1938 with popular war
weariness increasing apace. It meant mortgaging the party’s credibility

180
For example, text of Negrin’s Madrid speech of i8June 1938, ‘jEspana para los espanoles!’

(published in pamphlet form, FPI); see also J. Diaz, ‘Union Nacional de todos los espanoles’,

30 March 1938, reproduced in Tres anos de lucha vol. 3, pp. 128-32.

Frente Rojo, ,

181
These were published in many places and languages. The points are listed by Zugazagoitia,
who assisted in their production, Guerra y vicisitudes , pp. 430-2 and also in Guerra y revolucion en
Espaha ,
vol. 4, pp. 88-9. The exact origin of the idea for the Thirteen Points remains unclear. But
Stepanov’s claim that the PCE produced the entire document, word-perfect, does not ring true:
Elorza and Bizcarrondo, Queridos camaradas, p. 418. For an English resume of the Thirteen Points,
see Thomas, Spanish Civil War, p. 820 and analysis in Jackson, The Spanish Republic and the Civil
War, p. 452. The Thirteen Points also implicitly indicated the withdrawal of the International
Brigades.
182
For the dispute in the JSU, see Graham, Socialism and War, pp. 69-74, 112-16, 226, 229 and ‘The
socialist youth in the JSU: the experience of organizational unity 1936-8’, in Blinkhorn, Spain
in Conflict 193 1 -1939, pp. 83-102.
183
These shrewd words are those of the journalist Vincent Sheean, who heard Ibarruri’s speech
to the PCE central committee on 23 May 1938, five weeks after the territorial split: V Sheean,
Not Peace Sword (New York: Doubleday, Doran & Co., 1939), pp. 185 — 8.
but a i

184
Togliatti had already censured a February 1938 speech of Dolores Ibarruri’s precisely because
it implicitly called for a return to a class-based politics for the wartime PCE: Elorza and Bizcar-
rondo, Queridos camaradas, p. 417. Jose Diaz’s address ‘Union Nacional de todos los espanoles’,
in Tres anos de lucha, also alludes to this tension inside the party.
 Negrin’s war on threefronts 367

and reputadon to a victory whose possibility seemed remote even to the

most disciplined amongst them. 185 Indeed, even the Comintern saw this
186
danger. But given that maintaining Republican resistance was also in
Soviet defence interests, the PCE was being impelled inexorably towards
serving as Negrin’s indispensable support - as ‘el partido de la guerra’
(the war Rather than the
party). PCE using Negrin, then, it was the
prime minister who found in the PCE the only instrument capable of
sustaining his resistance policy and holding things together politically
187
and militarily. It would prove capable of doing so through the second
half of 1938 precisely because of its iron discipline. All other Republican
political organisations were now, under the immense material strains
imposed by the war and through sheer human exhaustion, either hostile
to Negrin’s policy of holding out or else (more commonly) badly divided
by the question. But the PCE would, in the end, pay a huge price for
its discipline. The very fact that it was identified with holding the line

meant that it would become increasingly isolated. It was the bete noire of
many, with a variety of longstanding political grievances (much inten-
sified precisely because of everyone’s increasing sense of helplessness).

But the PCE would also become the target of a rising tide of other, far
more amorphous, much less overdy ‘political’ (but no less real) discontent
which, in the end, signified, quite simply, the cumulative desperation and
desolation of a war-weary population at the end of its tether.
In spite of Blum’s resignation on 8 April, the new French premier,
Daladier, kept the borderopen in spite of British pressure to close it. 188
However, this did not bode as well as first appeared for the Republic.
The open border was merely a bargaining device to try to negotiate
with Franco the resumption of pyrites exports to French factories. The
most important figure in the new French cabinet was the foreign min-
ister, Georges Bonnet, from the most conservative sector of the Radical

Party. Insular and pragmatic, he was no less wedded to an Italian

185
In October 1937 Dolores Ibarruri had agreed with Azana that the division of Catalonia from
the rest of the Republic would mean defeat: Azana’s diary entry for 13 October 1937, Obras
completas, vol. 4, p. 820.
186
Stepanov, cited in Elorza and Bizcarrondo, Queridos camaradas, p. 421 and see also pp. 417-18.
187
Prieto was in a sense also suggesting this when he remarked, off the cuff, just after the Pedralbes
demonstration ‘against capitulation’ that he was convinced that Negrin had asked the PCE to
organise it: Zugazagoitia, Guerra y vicisitudes, p. 391. See also Luis Araquistain, cited in Cattell,
Communism and the Spanish Civil War, p. 233 (n. 11); Buckley, Life and Death of the Spanish Republic,

p.402 and Negrin’s own comments in the course of the cabinet crisis of spring 1938, ‘for me
today there is only one significant distinction between political parties - whether they want to
continue fighting or whether they want to surrender’, Anso, To Jui ministro de Negrin, p. 215.
188
For the politics of the French border, see Moradiellos, La perfidia de Albion, pp. 272-85.
368 The Spanish Republic at war

rapprochement than was Chamberlain, and even more to the instru-

mental use of appeasement. For Bonnet, whether it was Spain or eastern
Europe, political guarantees for ‘small’ nations against Germany were
an ethical-political luxury too far.
On 11 May the League of Nations also rejected the Spanish demand
for an end to Non-Intervention presented by Republican foreign minis-
ter Alvarez del Vayo. Only Spain and the Soviet Union voted in favour.
Negrin’s intention here had been to procure an injection of military aid
to bolster the military reorganisation underway after the territorial divi-
sion of the Republic. 189 But now Republican military resistance and the
pursuit of a diplomatic breakthrough were both tacitly aimed at forcing
a mediated settlement with guarantees. Negrin was actively pursuing
190
this. As its sine qua non was effective Republican resistance, however,
the prime minister could not make this dual policy explicit. With his eyes
on both the enemy high command and his international political and
diplomatic audience, the only viable public watchword Negrin could
give to the Republican population was ‘iron resistance’.
On 13 June the French border was finally reclosed. Franco had insisted
upon this before he would settle the matter of pyrites imports that so
exercised the Daladier government. Non-Intervention relachee was over.
But it was even worse for the Spanish Republic. Franco was also insisting
on the return of gold reserves that it had lodged in France. As a result
these funds, deposited in the Mont de Marsan, would effectively be
frozen from July 1938 - even though nothing definitive had yet been
agreed about the bilateral withdrawal of volunteers. 191
By this stage too the Republic’s gold deposits were perilously close
to being exhausted. A year previously Negrin had warned that the war
would end for the Republic the day the last gold peseta was spent. He
189
See also Negrin’s representations to the British government in this regard, ibid., p. 269.
190
Zugazagoitia, Guerra y vicisitudes, pp. 429-30, 433; Azana, Apuntes de memoria, p. 82. There is
also a cryptic confidential PCE report for March 1938 referring to Negrin’s travels incognito
to Paris, PCE XVIII (frame 224). Apart from oblique references such as these,
archive, film
we do not know very much about Negrin’s secret diplomacy. He never spoke of it even
still

to his closest collaborators. For example, around the time of Negrin’s secret talks with the
Francoist emissary, the Duque de Alba, in September 1938, his secretary and colleague
Julian Zugazagoitia remarked that the prime minister seemed preoccupied, but all he ever
spoke of was ‘frivolous things like women and cabaret’, Zugazagoitia to M. Pascua, 6 Oct.
1938, AHN/MP, caja 2 (bis) 16 (Embajada de Paris). There is little in the Quai d’Orsay or the
PRO to enlighten us further. Another potential source for Negrin’s secret diplomacy would be
the archives of the French interior ministry - which I have not investigated to date. I am also
grateful for the advice of Professor Santos Julia and Dr Enrique Moradiellos concerning their
researches in, respectively, the Quai d’Orsay and PRO.
191
A. Vinas, ‘Gold, the Soviet Union and the Spanish Civil War’, European Studies Review 9 (1) ,

(Jan. 1979), 120-1.

 Negrin’s war on threefronts 369

knew would be over unless he could secure

that Republican resistance
-
credits from the Soviet Union since nowhere else would furnish them. In
July Negrin sent his former ambassador to the Soviet Union, Marcelino
Pascua (from spring 1938 ambassador in Paris), back to Moscow with
192
the request. Stalin agreed to make a $60 million loan available to the
Republic. This was in addition to the $70 million agreed the previous
February. 193 But this second loan was made when there was virtually no
194 Without the July credit the Republican war effort could
gold to back it.
not have survived through the second half of 1938. 195 The fact that these
creditswere conceded tells us something crucial about Soviet policy
towards the Republic by mid 1938. With the division of its territory,
the closure of the French frontier and the escalating Czech crisis, it
seems clear that the Soviet cupola no longer believed that the Republic
could win. 196 But Stalin was still prepared to invest in prolonging its
resistance. 197
June would also see the crystallisation of the internal political divi-
sions in the Republic over the feasibility of pursuing a mediated peace.
Opinion had become more evident
hostile to Negrin’s resistance policy
among republicans after the Labonne soundings in March. But Azana,
galvanised by the Republican defeat at Teruel and increasingly alienated
from Negrin, was beginning to look beyond his own republican circle
to disaffected socialist leaders in search of an alternative government
constellation.
When
Negrin was appointed prime minister, Azana had been opti-
mistic that herewas the man who would work for the mediated peace to
which he was increasingly committed. 198 But the two would come to be
increasingly at odds. Azana was certainly no less capable than Negrin

192
Ostensibly Pascua’s return to Moscow was for the official farewell ceremony to him as ambas-
sador: Negrin to Pascua, AHN/MP, caja 14 (17), July 1938.
193
A. Vinas, ‘The Financing of the Spanish Civil War’, in Preston(ed.), Revolution and War in Spain,

pp.271-2 and ‘Gold, the Soviet Union and the Spanish Civil War’, pp. 118-19. Correspondence
on Soviet credits between Pascua and Negrin 22 June 1937 to 11 March 1938, AHN/MP, caja
2, carpeta 2.
194 Vinas, El oro de Moscu, p. 413. M. Pascua to A. Vinas, 13 Feb. 1977, AHN/MP, caja 8 (13). Angel
Vinas’ work, El oro de Moscu (and the more technical volume from which it was derived, El oro
espahol en la guerra our main source of information on Republican wartime financing.
civil), is

But, as Vinas indicates, still know nothing about how the July 1938 credit was implemented.
we
195
Vinas, ‘Gold, the Soviet Union and the Spanish Civil War’, pp. 120-2.
196
Vinas, El oro de Moscu, pp. 406-8.
197
Reports by Soviet diplomatic personnel in early 1938 certainly suggest a serious intention still to
make Republican defence viable: see report of 25 Feb. from Soviet plenipotentiary in Britain to
Voroshilov, cited in Radosh, Habek and Sevostianov, Spain Betrayed. The Soviet Union in the Spanish
Civil War, pp. 427-8.
198
Julia, ‘
Presidente por ultima vez: Azana en la crisis de mayo de 1937’, pp. 255—6.
37 ° The Spanish Republic at war

of imagining the enormous distance between the cultural and political

worldviews of Republican and rebel leaderships. " But since mid 1937
1

Azana had been consumed by one overriding objective: the desire to

stop the descent into internal warfare in the Republic - which was in
considerable measure a product of the impossible pressure under which
it was being placed. 200
For his part, Negrin could keep his finger on the pulse of international
affairs and, most crucially, gain some sense of what was happening inside
the Nazi cupola, through the confidential reports he received (until the
Munich crisis of September 1938) from Jimenez de Asua, the Republican
ambassador in Prague. 201 Negrin knew that Franco would not negotiate
if he thought he could win outright. Negrin also understood that oth-

ers would not necessarily share his conviction and that, although what
drove him was a powerful sense of patriotism, he could not expect to
be popular. 202 Believing that only the projection of iron resistance could
bring Franco to the negotiating table, he remained fearlessly focused
on removing anyone who threatened this. Hence Negrin’s dismissal of
Azana’s brother-in-law, Gipriano Rivas Cherif, from his post as Republic
203
consul general in Geneva in May 1938. Partly at Azana’s instigation,
Rivas Cherif had been involved in some extremely indiscreet (indeed
public and entirely non-viable) soundings over mediation involving a
number of South American countries.
The Rivas Cherif saga was the catalyst in the collapse of relations
between president and prime May, the estrangement be-
minister. After
came personal as well as political. 204
Negrin saw Azana as behaving
unconstitutionally and wrong-headedly and damaging the Republic at

199
Cf. Negrin’sremarks in 1938 to the French military attache, Colonel Morel, concerning the
complete mind-set (‘otro modo de pensar’) separating the two sides, cited in J. Marichal, El
intelectualy la politica (Madrid: CSIC, 1990), p. 100. Azana recounts many incidents which gave
him similar pause for thought, for example his diary entry for 19 July 1937 on rebel killings in
Teruel, Obras completas, vol. 4, pp. 685-6.
200
See Azana’s diary entry for 29 June 1937, Obras completas, vol. 4, p. 638, where Prieto provides
a resonant echo of these fears: have no choice but to hang on until it all falls apart. Or
‘Me
until we start attacking is how I have always believed that this will end’.
one another, which
For Azana’s (war-long) attempts at mediation, see Miralles, ‘Paz humanitaria y mediation
internacional: Azana en la guerra’, pp. 257-76; Azana, Obras completas vol. 4, pp. 588, 655-6,
,

833; Guerray revolucion en Espana, vol. 3, pp. 179-84; Bolloten, SCW, p. 904 (n. 54).
201
These dispatches can be consulted in the Archivo de Barcelona (Archivo Reservado and Archivo
de la SIDE (Section de Information diplomatica especial)), in the archive of the foreign affairs
ministry in Madrid.
202
This can be glimpsed in his comment about Talleyrand to Zugazagoitia, Guerray vicisitudes,

p. 429.
203
398-400, 427; Bolloten, SCW, p. 926
Ibid., pp. (n. 48).
204
Zugazagoitia, Guerray vicisitudes, pp. 426-7.
 ,

JVegrin’s war on three fronts

37

a time when (in the military crisis following the territorial split of April
205
1938) it could not afford to appear weak. Azana saw Negrin as in-
creasingly distant and, wrongly, assumed this to be political arrogance.
5
Negrin’s ‘distance was more the result of his reducing support base
and the increasing burden that he bore personally. 206 (We must also
remember that, given that Negrin rather than Alvarez del Vayo was the
Republic’s effective foreign minister and diplomatic negotiator-in-chief,
then no one knew better than he how unpromising the environment for
negotiation was - something which itself added to the pressure.) Negrin’s
distance was not arrogance, self-importance or disdain but rather a reac-
tion born of an increasingly acute sense of his own responsibilities. This
was coupled with an ever more uncompromising sense of the real politi-
cal and ethical priorities of the situation. As he memorably remarked to
one socialist colleague: T’m no more important than any other Spanish
citizen - except when I represent those who are giving their lives for the
Republic.’ 207
This context explains much about Negrin’s immense anger when he
learned of the political intrigues of the mediation camp in Barcelona
Azana had entertained vain hopes of Prieto taking over the
in June.
government in May. 208 Now he was making overtures to the veteran
PSOE leader, Julian Besteiro, to bring him out of his self-imposed in-
ternal exile of municipal service in Madrid. Besteiro was openly hostile
to Negrin and by this point amenable to fifth column Falangist voices
intent upon persuading him that he should take over as prime minister in
order to facilitate ‘peace negotiations’. 209 At the time, Negrin was away
from Barcelona on a tour of the military fronts in the centre zone where
Franco’s troops were battling in the Levante, their sights set on Castellon.
Cutting short his tour, Negrin returned to Barcelona to issue his famous
public denunciation of the ‘charca politica’ (political mudhole). ‘If the

205
But see also Azana for his views of Negrin’s constitutionalism, Obras completes, vol. 4 (Pedralbes,
3 M ay 1938), p- 878.
206
Negrin did have a set of socialist ministerial secretaries in the cabinet, but this was insuffi-
cient in the circumstances: Graham, Socialism and War, p. 158; Zugazagoitia, Guerra y vicisitudes

pp. 388, 417, 433-4; Azana, Obras completes, vol. 4 (Pedralbes, 3 May 1938), p. 879. It is
little wonder that he sometimes needed to escape to the cinema, about which Azana would

complain.
207
‘Yo soy cualquiera - menos cuando represento a los que mueren por la Republica’, F. Vazquez
Ocana, cited in Alvarez, Negrin. Personalidad historica, vol. 2, p. 252. Fernando Vazquez Ocana
had been a PSOE deputy for Cordoba. During the war he was the editor of La Vanguardia and
head of Negrin’s press office.
208
Azana, Obras completas, vol. 4, p. 884; Zugazagoitia, Guerray vicisitudes, p. 427; Graham, Socialism
and War, p.144.
209
Preston, Comrades, pp. 180-1.
 ,

372 The Spanish Republic at war

people and tl^e troops only knew the sort of things that were going on,
5210
they would sweep all of us politicians away.
In fact, Azana’s plans would never prosper. The disapprobation of
Negrin’s policy in republican and socialist circles never amounted to
more than currents of opinion. But Prieto and, perhaps even more,
Besteiro were incensed by the prime minister’s words, which they inter-
preted as a personal attack. The episode reveals how the resolution of
the spring cabinet crisis had left Negrin more vulnerable to his political
enemies as he became increasingly isolated. By June, stresses inside the
new UGT leadership were also being felt. Its general secretary, the vet-
eran miners’ leader and justice minister, Ramon Gonzalez Pena, who
had been a major player in defeating the Caballeristas in late 1937, thus
returning the UGT to full support of the government, was now backing
a new PSOE executive under Besteiro to explore mediation.
211

Moreover, the CNT was now internally split over the issue of continued
resistance. Already in March its former secretary general, Horacio Prieto,
had publicly declared that the war was lost and it was time to sue for
peace. The CNT justified this stance in terms of its political hostility
212

to the PCE, and doubtless their fierce rivalry was a salient factor here.
But underlying the was the shattering experience of
political invective
defeat at Teruel and the subsequent rebel conquest of Aragon. Ending
the war was promoted as a means of reducing hardship and suffering.
It was even seen, quite startlingly, as an opportunity for rebuilding the

CNT — something which suggests a syndicalist myopia akin to Largo

Caballero’s unspoken assumptions that once the PCE was dealt with,
Francoism might not prove too bad for the union movement. Apart
from the practical feasibility of resistance, however, some libertarian
sectors, grouped around the FAI’s peninsular committee, were also now
beginning openly to question its political purpose. The commitment to
liberal economics enshrined in Negrin’s Thirteen Points crystallised their
opposition to the CNT national committee’s continuing support for the
prime minister. 213

210
The Times 21 June 1938; Zugazagoitia, Guerra y vicisitudes, pp. 443-4. On Negrin’s alienation
from his own cabinet as a result, see Gero to Dimitrov, 19 Nov. 1938 (doc. 80) in Radosh, Habek
and Sevostianov, Spain Betrayed. The Soviet Union in the Spanish Civil War, p. 506.
211
Togliatti to Dimitrov, 19 June 1938, cited in Elorza and Bizcarrondo, Queridos camaradas, p. 420;
cf. Graham, and War, pp. 213-18.
Socialism
212
CNT national plenum, Barcelona, March 1938, Lorenzo, Los anarquistas espaholesy elpoder, p. 255.
Peirats, La CNT en la revolucion espahola, vol. 3, p. 99 (see pp. 83-99 for the course of the liber-
213

tarians’ internal crisis in spring 1938). There was also internal disagreement over the attitude
to take towards the government’s creation of a religious body, the Comisariado de Cultos, with
 \egrin's war on threefronts
373

In June 1938. the Italian CTY participated in attacks on Valencia

which were successfully blocked by Republican fortifications. Soon, how-
ever. the Republic would go on the offensive against the CTY when, on

25 July, the Popular Army launched its surprise attack across the River
Ebro. In immediate tactical terms this was designed to force Franco's
troops back, thus relie\ing the pressure on Valencia. But. even more im-
portandy, Negrin was gambling on the fact that by throwing everything
into the offensive he could comince his audience of Great Powers -
and Britain in particular — that the Republic would not easily or quickly
be beaten and that they should exert pressure to bring Franco to the
negotiating table to end the war.
With the element of surprise in its favour, the Republican offensive
had considerable initial success - particularly because, as we have already
noted, the fronts in Spain were always sparsely manned. But Franco, un-
like the Republic, could readily call up trained reserves. The Republican

troops dug in and endured. Battie was waged for over one hundred days.
Acute blockages at the French border told on the Republic, robbing its
actions of staying power. 214 Franco, calling urgendy on his backers, threw
everything into repelling the offensive. Most crucially, he would by the
end have ovenv helming air superiority. 215 The Ebro witnessed massive
air batdes. Nothing like this had ever been seen before, nor would it
be again until the Batde of Britain. Republican communications were
bombed to oblivion and. as so many international brigader memoirs tes-
tify, the troops were blasted off the bare and rocky hillsides by the sheer

force of the incendiary material launched.

For all the Republican troops conditions worsened dramatically as the
rebels' vasdy superior bombing capacity began to tell. The ine\itable re-
sponse to terror from the skies was terror on the ground. Harsh military
be understood in the context of the scale of the military
discipline has to
challenge importance to the Republic's continued resistance. 216
and its

Moreover, while there were batde-hardened troops at the Ebro, including

international brigaders. there were also many new conscripts — barely

responsibility for ensuring that religious liberty was defacto as well as dejure. In the end the CXT
approved it. Lorenzo. Los anarquistas espamotesj d poder pp. 259-60.
.

214
Howson. Armsfor Spain. p. 241.
The Franeoist array had 225 bombers to the Republicans' 50 although they were more e\enly
matched on fighter aircraft - 150 Republican to 175 on Franco's side E. CastelL L. Falco.:

X. Hernandez. O. Junqueras. J. C. Luque andj. Santacana. La batalla de VEhre Historia, paisaige

. .

patrimom Barcelona; Pdrtie. 1999 p-177- C£ Modesto's comments to Constancia de la Mora.

.

In Place of Splendor, p. 369.

216
It is interesting to see this point made unequivocally in D. T. Cattell's 1965 study. Communism
and the Spanish Cidl W ar. p. 209.
374 The Spanish Republic at war

trained peasant boys of seventeen and eighteen, or men in their forties,

whose notion of military cohesion was slight and who were poorly able
to cope with the terrible conditions of massive enemy onslaught. Had
the Republic been able to institute general conscription earlier, then the
worst effects of this situation might have been mitigated. But, as it was,
serious disciplinary problems and, notably, a high level of desertion in
the midst of a military emergency made brutal penalties unavoidable as
a means of enforcing obedience. Indeed, such discipline could be life-
saving and was, at any rate, indispensable to the overall maintenance
of frontline resistance — as in any regular war. 217 The fact that many of
the officers administering these tough disciplinary measures were com-
munists certainly further excited sectarian political conflicts - especially
with the CNT. And there is no doubt that military ‘justice’ could be
very unjust in individual cases —
again, as happens in any regular war.
Sometimes too sectarian motives impelled communist commanders. But
it would be a distortion to portray what was going on here uniquely in

terms of the political sectarianism of 1936. By the second half of 1938 —

and probably earlier - the CNT’s denunciations of ‘communist’ violence
against its militants usually referred to punishments meted out to indi-
218
vidual soldiers who had gone absent without leave from the front.
While this was entirely understandable behaviour at a human level, in
the military circumstances obtaining by 1938 it could not be sanctioned
without risking the viability of Republican defence.
Partly there was a clash of cultures here that then translated into a
clash of politics. New conscripts with very rudimentary training did not
understand the iron discipline of the front. For this very reason they
often gravitated to the CNT’s orbit, where the political culture was more
conducive to them. The rural backgrounds of the recruits were very
whose ‘anti-militarism’
often similar to those of the volunteers of 1936-7,
had reflected their desire to be able to shift between the military front
and their agricultural tasks. This culture clash would also echo during
the POUM trial in October 1938. The communist political commissar,
Ignacio Mantecon, appearing as a witness, would denounce the fact
that troops on the Aragon front had fraternised with the enemy in no
man’s land. 219 Mantecon’s account acquired an overt political gloss. He

217
Wolff, Another Hill, pp. 120-1.
218
See, for example, Peirats, LaCNT en la revolucion espafiola, vol. 3, pp. 199 (although the author
himself constructs what was happening at the front entirely and utterly as evidence of the PCE’s
sectarianism).
219
Alba and Ardevol, El proceso del P.O.U.M. pp. 222-3; Wolff, Another Hill, pp. 67-9.
,
 Negrin’s war on threefronts
375

accused the POUM of fomenting treasonous behaviour, while also, by

implication, discrediting the CNT. But it is important to understand that
he was referring back to a period in early 1937 before the Aragon front
had been activated and to young men who had had little or no experience
of warfare or military discipline. The absence of death too had, until then,
made possible those transient understandings between local boys.
But the divided Republic of 1938 was a world away from
territorially
Aragon in spring 1937. Behind the lines too there were increasing signs
of disaffection and distress. Angry mothers demonstrated against the
5

mobilisation of their teenage sons in the infamous ‘quinta del biberon

220
(baby call-up). The sense of siege and isolation deriving from the
and war under impossible odds created a reciprocally
territorial split
The fear of the ‘enemy within - for the fifth column
destructive dynamic.
5

was indeed more active and more dangerous now precisely because
the Republic was weaker — and the crack-down on deserters and draft
dodgers ( emboscados), to whom the whiff of treason now also attached led
to the increasing militarisation of the home front. It fell to the SIM, as
the military police, to trackdown deserters and those evading military
service. In so was imposing military discipline on unwilling
doing it

people. This explains the increasing opprobrium attaching to the service

and its increasing clashes with many sectors of the civilian population,
including libertarians.
The predominance of Spanish communists in the SIM reflected the
already established preference for the PCE of Spanish military and police
personnel (obviously the most useful to the SIM). This is also often
taken as synonymous — without the case ever being explicitly argued —
with the supposed heavy influence of Soviet security advisers in the
SIM. 221 But the main thrust of Russian advice was technical — on new
counter-espionage techniques - rather than political. Indeed, the terror
inside the Soviet Union between 1936 and 1938 wrought such
itself

acute destruction on Soviet and Comintern operational structures out-

222
side that efficient control of anything was probably out of the question.
Moreover, where the Spanish security forces were concerned, there were
220
From February 1938 the Republic was calling up reservists, a process culminating in the in-
corporation of those due only for military service in 1940 and 1941 - hence ‘biberon’, which
literally means a baby’s feeding bottle.
221
Most obviously in Radosh, Habek and Sevostianov, Spain Betrayed. The Soviet Union in the Spanish
Civil War p. 476. See also Bolloten, SCW, pp. 601 -6. But the evidence Bolloten presents is, once
,

again, ambiguous. See also Orlov in Forum Jur osteuropaische Ideen und geitgeschichte, pp. 246-8.
222
Schauff, ‘Hitler and Stalin in the Spanish Civil War’, p. 19. Radosh, Habek and Sevostianov,
Spain Betrayed. The Soviet Union in the Spanish Civil War, p. 496 (doc. 78) cite a brief note (10 Nov.

1938) from Marchenko (charge d’affaires in Republican Spain) to Litvinov. He relays Negrin’s
376 The Spanish Republic at war

several other %r more

erosive sources of political intrigue and conflict in
evidence. First, there was the underlying ‘old class war’ between the ‘old 5

police (if ‘new’ communists) and the left - especially libertarians. Then
there was the ongoing organisational competition between the PCE and
the PSOE that affected the SIM as it did most other Republican enti-
223
ties. Many saw it as being to their personal and career advantage to
have the party patronage of the PCE. This slotted into a clientelist under-
standing of politics held by many of the ‘new communists’ whom the war
had brought to accelerated prominence. The underlying organisational
competition in theSIM had also been further exacerbated by Prieto’s re-
peatedly unfortunate choices of individuals to head the service. 224 When
Paulino Gomez took over as minister of the interior in April 1938 he had
sought, with Negrin’s backing, to curb PCE clientelism in the SIM. 225
But here, as in other areas of the administration, this was a delicate pro-
cess — given how much the underdeveloped Republican state depended
on party and union personnel to sustain crucial wartime functions.
Finally, there is another dimension of this organisational competitiveness
that should not be overlooked. Given that the war had brought youth to
the political forefront in Republican Spain, it also seems highly probable
that there was a gendered dimension to the clashes between young com-
munist commanders and their libertarian counterparts. This was male
youth proving its macho credentials as much as its political metal. 226

remarks that the remaining NKVD

personnel ought to keep a low profile and maintain their
distance from the SIM. This desire for discretion relates to Negrin’s diplomatic strategy after
the withdrawal of the International Brigades, which was geared to achieving British and French
support to ensure a mediated end to the war. Negrin was also concerned not to inflame anti-
Soviet and anti-communist feeling, growing as a part of the general weariness and desperation
within Republican Spain. Radosh and his co-editors claim, unaccountably, that the Marchenko
report proves that the NKVD controlled the SIM. Even within the terms of their own argument,
it seems odd to be claiming this for November 1938, when Soviet infrastructural support for the

Republic was rapidly being wound down.

223
Graham, Socialism and War pp. 198-244.
,

224
Negrin, Epistolario Prietoy Negrin, p. 33. Ex-SIM director Manuel Uribarri’s own books [La quinta
columna espanola and El SIM de la Republica (both Havana: n.p., 1943)) were intended as a personal
defence. But their disingenuousness tends to confirm that he was not the right man for the
job. For a resume of Prieto’s appointments, see Bolloten, SCW, p. 601. See also here Orlov’s
comments, decades later, that he believed the Republic had genuinely lost out because Prieto
was determined to install his favourites in the SIM: Gazur, Secret Assignment. The FBPs KGB
General pp. 133-5. Orlov told Payne that it was as a result of this that he withdrew from SIM
,

affairs in October 1937 to concentrate on intelligence work related to the Francoist zone, Forum

Jiir osteuropaische Ideen und eitgeschichte , p. 248.

225
Negrin, Epistolario Prieto y Negrin, p. 33.
226
There is an oblique allusion in Peirats, La CNT en la revolucion espanola, vol.
3, p. 218. Santiago
Garces was twenty-two years old when he was appointed to acting head of the SIM in 1938:
Bolloten, SCW, pp. 601—2.
 ,

Negrin’s war on threefronts

377

The SIM had taken over the Republican work camps that now also
227
housed deserters and draft dodgers. Conditions in these varied. But
violenceand arbitrary treatment grew less controllable everywhere as
the Republic’s military position deteriorated in the chaotic conditions
following the rebel advance in Aragon and the rupture of the front in
228
April 1938. An increasingly harsh military discipline was exerted upon
deserters and draft dodgers thereafter. Prisoners taken at the front were
also sent to thework camps. Conditions worsened for all prisoners —
military and civilian. There were some deaths in the work camps -
mosdy from neglect. However, there was at least one instance in 1938
when the Republican authorities were forced to intervene against a work-
camp regime in which prisoners who were, for example, too sick to work
were being shot outright. 229 In mid August, as the Ebro battle raged,
the Republican government issued an amnesty designed to get trained
fighting men back from the camps and prisons to the front, where they
were desperately needed. 230
The SIM also now controlled all forms of the previously private party-
political prisons - which was to be expected, given its nature and function
as an instrument of the Republican state. The prisons it ran were not
checas as conjured in both Francoist and anti-stalinist left literature -
that is, they were not clandestine places of illegal detention. But the SIM
did carry out unconstitutional detentions on behalf of the Republican
state, and it also sometimes used torture to extract the confessions on

which convictions in the emergency courts were based. It was, neverthe-

less, also effective in dismantling spy and saboteur networks when the
231
Republic was at its most vulnerable.
There were also cases of SIM agents being implicated in the extra-
judicial killings of individuals accused of fifth column activities. Some-
times there was substance to these accusations and sometimes not. But
it is significant that killings like this - many of which were carried out

by ordinary civilians, not by policemen — coincided with moments of

227
See Fraser, Blood of Spain p. 179 for an early testimony.
228
This military context of panicked chaos of defeat and retreat saw all manner of brutality -
both by soldiers and against them. The International Brigades were also caught up in this: see
a summary in Thomas, Spanish Civil War, p. 801. This may have some bearing on the recent
appearance of dubiously ‘retrospective’ hypotheses about the brigaders having been subject to
a reign of terror from the day they set foot in Spain.
229
F. Vega Diaz, ‘El ultimo dla de Negrin en Espana’, Claves de Razon Practica, 22 (May 1992), 61.
230
Gaceta de la Republica, 18 August 1938.
231
On SIM efficiency, see Sole i Sabate and Villaroya, La repressio a la reraguarda de Catalunya (1936-
1939)i v°l- PP- 261-2. The midwife of the SIM was the war: ‘tecnica y terror al servicio
i,

judicial’, the authors’ own resume in Julia el al., Victimas de la guerra civil, p. 244.
378 The Spanish Republic at war

high military tension, when the Francoist forces were achieving major
breakthroughs. This suggests that those killed, often irrespective of any
particular actions, were killed for what they were seen to represent and,
above all, in lieu of that advancing enemy. Many forms of violence would be
unleashed in the maelstrom of military defeat and retreat in 1938, much
of it beyond anyone’s control.
This erosion of the Republic’s constitutional fabric was at odds with
Negrin’s political principles. But he had to accept it - as a part of the
‘whole’ he blotted out. This was the price of maintaining resistance. 232
Negrln, however, continued to see a clear distinction between violence
that was imposed by a constitutional state and arbitrary violence. While
he sought to end SIM torture, he would continue to defend, against
constant Catalan criticisms, the rigours of the special courts and their
recourse to the death penalty. 233 The death sentences provoked protests
from President Azana and bitter debate in the cabinet. 234 But Negrin’s
view would prevail from April through the summer. As the two armies
clashed at the Ebro, the cabinet considered a batch of capital sentences
imposed by the special courts. Negrin’s view was that, given the sacrifices
being made at the front — including now by soldiers who were virtually
children - then the home front had to be subject to a discipline worthy of
that sacrifice - in short, ‘this is war and it has to be for all of us’. Again he
faced bitter opposition, and although 58 of the 62 death penalties were
eventually approved, was by a majority vote, not cabinet unanimity. 235
it

To date we know that the Republic’s special wartime courts approved

(and the government endorsed) at least 173 executions. 236 Many more

232
See Negrin’s distressed but world-weary response ‘one more horror’ (‘una atrocidad mas’) on
hearing of the killing of prisoners in a work camp: Vega Diaz, ‘El ultimo dia de Negrin en
Espana’, p. 61.
233 Azana, Obras completes, vol. 4 (diary entry, Pedralbes, 3 May 1938), p. 878; although, in the
tinder-box atmosphere of summer 1938, it seems improbable that Negrin would have thought
that simply by issuing an order to Garces (the head of the SIM) he could eradicate all instances
of torture in the cells. His denial to the journalist Henry Buckley in 1939 needs to be seen in
the context of the publicity ‘war’ for British and French support: Thomas, Spanish Civil War,
p. 669 (n. 5).
234 Sole Sabate and Villaroya, La
i repressio a la reraguarda de Catalunya (1936-1939), vol. 1, pp. 259-
68; Julia et al., Vktimas de la guerra civil, pp. 248-9; Azana (diary entry for 22 April 1938), Obras
completes, vol. 4, p. 875 (‘Little more than a week after I spoke publicly of forgiveness, they throw

58 executions at me’).
235 Sole i Sabate and Villaroya, La repressio a la reraguarda de Catalunya, vol. 1, pp. 274-5; de Irujo, Un
vasco en el ministerio dejusticia, vol. 1, (1) p. 83.
236
Sole i Sabate and Villaroya, La a la reraguarda de Catalunya, vol. 1, pp. 268-76; Sanchez
repressio

Recio, Justiciay Guerra en Espana. Los Tribunals Populares (1936-1939), pp. 168-75, a l so PP- 68-9,
70. To date, we have no figures for death sentences carried out (as opposed to those passed)
by special courts outside Catalonia. But as Sanchez Recio’s analysis indicates, sentence review
 Negrin’s war on threefronts 379

death sentences were commuted, while the majority of people brought

before the special courts were either acquitted or received other kinds of
penalties (prison sentences and fines of varying severity). In view of the
civil war context, and the fact that both soldiers and civilians came before
these special courts, the picture we have does not sustain an apocalyptic
reading of Republican state violence. As Negrin himself observed in his
public reply to his critics: ‘The law is harsh, but we have to act thus against
those on the home front whose actions either directly endanger soldiers’
lives or undermine their heroic sacrifices. The law is harsh, but its terms

are public: those who choose to break the law know, before they do so,
what the penalties are and what they are risking. It’s a harsh law, but no
more harsh (and, in practice, less so) than those which have operated in
other countries that have found themselves in circumstances similar to
ours.’ 237 Not everyone, however, agreed with Negrin’s equation of home
and military front discipline, even in the context of an increasingly bitter
civil war and the desperate military straits of April 1938 and after. The

Generalitat commissioned a legal investigation that in July 1938 reported

in harshly critical terms on the constitutional irregularities inherent in
the practice of militarised justice in Catalonia.
By now the ‘persistent Catalanism’ that Negrin saw underlying the
Generalitat ’s criticisms was also a real bone of contention. As we have
already seen, the prime minister was clear about his priority in the war:
he was fighting As he had commented to Zugazagoitia in the
‘for Spain’.

summer of 1938, he would prefer to hand over to Franco than see Spain
dismembered. All he asked first was that Franco cut his ties with the
Axis. 238 But Catalan disaffection over the militarisation (and thus cen-
tralisation) ofjustice remained acute. It was to be one of the major causes
of the cabinet crisis of mid August 1938 which saw the final departure

(by the special court for espionage and treason and, ultimately, by the cabinet) attenuated the
practice of the special courts as a whole. Moreov er, the special courts in some Republican areas
(Cartagena and Castellon, for example) were notably more lenient in their general sentencing
policy.
237 La Vanguardia, 28 April 1938. Part of Negrin’s Sabate and Villaroya, La
text is cited by Sole i

repressio a la reraguarda de Catalunya vol.

, 1, pp. 270-1. The fact that the Republic’s special tribunals
tried both soldiers and civilians complicates the comparison with wartime justice in other cases -
such as, for example, Britain or France during the 1914-18 war. But such comparisons would

be unsatisfactory anyway. The appropriate comparators are necessarily other European ciml
wars. Such comparative work is in its infancy, but see J. Casanova, ‘Civil Wars, Revolutions
and Counterrevolutions in Finland, Spain and Greece (191&-1949): A Comparative Analysis’,
International Journal ofPolitics, Culture and Society 13 (3) (2000), 515-37 and S. N. Kalyvas, The Logic
,

of Violence War (Madrid: Fundacionjuan March, 2000).

in Civil
238
J. Zugazagoitia, Guerra y vicisitudes, p. 454 and see epigraph to chapter 5.
380 The Spanish Republic at war

of both the Catalan and Basque nationalists from the government. 239
The cabinet crisis also brought a conclusion of sorts to the abiding in-
ternal tensions in the PSUC. Once Companys had effectively become
the leader of the opposition to Negrin’s government - with Prieto’s de-
parture from the cabinet in April -
the PSUC’s ability to compete with
the Esquerra on its own political ground was increasingly blocked. The
PSUC could break out of this impasse, but only by destroying the internal
equilibrium between Catalanists and communists, that is, by ceasing to
‘emulate’ the Esquerra, as it had been doing since July 1936, and opting
instead unambiguously to back Negrin’s policy of centralisation. This
was effectively what happened in August when the Esquerra minister
was replaced by one from the PSUC. 240
But if the PSUC could offer some support to Negrin at cabinet level,
developments inside the much more crucial PSOE were making the
party ever less available as a prime ministerial instrument with which
to resist capitulationary currents. In the wake of Prieto’s exit from the
government, the PSOE executive showed its disapprobation by refusing
either to convoke the national PSOE-PCE liaison committee or to issue
a joint declaration of support for the government. The Socialist Party’s
inner turmoil also explains why
was in April that Negrin abandoned
it

his principle of a PSOE monopoly in the Carabineros by appointing a

PSUC chief. 241 In terms of government personnel, the PSOE was still
providing Negrin’s major support. But war weariness at the grass roots
was making the party in the centre-south zone ever less prepared to
endure the friction of organisational rivalry with the PCE. Many of the
inter-party liaison committees were by the second half of 1938 on the
verge of collapse. Indeed, some had already been suspended. 242
Press censorship was another source of conflict. Even though this was
in PSOE hands, some leading party members (including Zugazagoitia’s
successor in the interior ministry, Paulino Gomez) complained that
Negrin was letting the PCE get away with more than the PSOE. 243
Certainly there were notorious occasions where the PCE flouted censor-
ship: for example the broadside against Prieto at the time of the March
1938 cabinet crisis or, later, as we shall see, the circulation of leaflets

239 In the persons of Irujo andjaume Aiguader.

240
Ucelay da Cal, ‘Socialistas y comunistas en Cataluna durante la guerra civil’, p. 315; although
in fact the conflicts inside the PSUC continued: P. Togliatti, report of 2 May 1939, Escritos sobre
1

la guerra de Espana, pp. 248-9.

241 242
Bolloten, SCW, pp. 609-10. Graham, Socialism and War, pp. 145-7, ^o, 224.
243 Ibid.,
pp. 158-9; Togliatti, report of 31 May 1939, Escritos sobre la guerra de Espana, pp. 232-3.
 Negrin’s war on threefronts 381

defaming the POUM in the run-up to the trial of its executive commit-
tee in October. But this evasion of censorship points not to Negrin’s own
partisanship but to the fissures in a state that was obliged to rely heavily
on clientelist party formations. 244 After all, the Caballeristas were also
freely engaging in similar sorts of partisan propaganda activity, albeit di-

rected at other targets. Negrin was a strong believer in the pre-eminent

importance of good morale on the home front. The fact that the PSOE
was his own party too probably made him especially impatient with the
continual factional sniping from the Caballeristas’ remaining provincial
party presses — where the steady stream of pamphlet production by the
dissidents belied the fact that there was a war on.
The PSOE general secretary, Lamoneda, was keen to contain these
tensions between Negrin and the party since, like Negrin, he too un-
derstood that the war effort demanded PCE and collaboration with the
was no alternative to Republican resistance. More-
that, thus far, there
over, Lamoneda and his fellow executive members also realised that any
explosion would damage the PSOE first and foremost. In a desperate
attempt to pull the party together Lamoneda tried to reincorporate in
the PSOE its estranged ‘historic’ leaders - Prieto, Largo Caballero and
Besteiro - August national committee, taking place on the eve
at the
of the movement’s fiftieth anniversary in 1938. 245 Lamoneda’s efforts
were probably doomed from the start. But it was Prieto’s extraordinarily
politically inopportune speech to the floor that administered the coup de
grace. The veteran leader made no substantive political points. Indeed,
he left before the agenda was debated. But the damage was done. Along
with Prieto, a number of other leading PSOE members withdrew from
the life of the party. These included Jimenez de Asua, the Republican
ambassador in Prague, and Fernando de los Rios, who was ambassador
in Washington. 246 Nevertheless, Prieto’s speech was immediately criti-
cised - even by friends - as the massive piece of spleen venting it was. 247
It seemed as if Prieto had decided, like Largo before him, that the war

was no longer worth the political candle. If the Republic could not win,
then it was time to stop sacrificing the PSOE.
The national committee meeting was taking place during difficult days
on the Ebro. Rebel reinforcements were in the process of containing the
Republican assault south of the river. In the wake of this, Negrin sought

244 See Azana, diary entry for 23 Aug. 1937, Obras completes, vol. 4, p. 745.
245 Preston, Comrades, 246
pp. 181-2. Graham, Socialism and War, p. 153.
24/ The episode is analysed in detail ibid., pp. 142—5; see also pp. 147-9.
382 The Spanish Republic at war

to clamp down on socialist dissidence in the interests of the war. But

this triggered the beginning of political meltdown - particularly in the
248
JSIJ. To the increasingly vociferous Caballerista wing were now added
the voices of many more from the pre-war Socialist Youth who were hos-
tile to the extremes of ‘union nacionaf, which they saw as negating the
meaning of the civil war and stripping the youth organisation of its so-
cialist identity. The political violence exploding within several provincial
youth federations - most notably Albacete and Alicante - was couched in
the language of the Caballerista crusade against the PCE. But across the
gruelling months of the Ebro battle, from July to November, the unrav-
elling in theJSU was driven by a more profound disillusion: the growing
awareness that no amount of political discipline or self-sacrifice on the
young socialists’ part could improve the Republic’s military situation or
make up for its lack of everything needed to wage an effective war.
The battle of the Ebro epitomised the war of attrition against the
Republic, whose superhuman, long-drawn-out resistance was bought at
terrible cost to the best troops of its army. Even in its vastly improved
form, this army could not withstand the might of Axis supply. Both
sides suffered heavy losses of troops and equipment. 249 But while for
the Republic these were irreplaceable, the rebels could always bring up
armed and supplied reinforcements. To do this, however, Franco was
reliant as never before on the massive deployment of Axis material - and
especially on German artillery and aircraft. In November, in order to
secure the wherewithal to drive the Republicans back across the river,
Franco finally submitted to Nazi Germany’s demands for mining con-
cessions in Spain. For Negrin this was conclusive evidence that he was
fighting the war to preserve national sovereignty and prevent Spain being
turned into an economic satellite of the Reich.

248
Ibid., pp. 223-31.
249 On the Ebro see Castell et al
. ,
La batalla de I’Ebre. Histdria, paisatge, patrimoni. Jesus Castillo
Domenech has argued cogendy that the stricdy military outcome of the Ebro battle was more
ambiguous than often claimed and that it is only subsequent political events that fix our view of

Republican defeat. His key points are: (1) the Republican army of the Ebro successfully blocked
Franco’s attack on Valencia; (2) Franco was forced (by Republican strategy) to engage in a war
of position that limited the use he could make of his superior strength (in terms of war materiel );
(3) Republican losses (of men and material) were no greater than Francoist losses; (4) the Ebro
offensive allowed the Republic to maintain its resistance beyond 1938 (the core of Negrin’s
strategy). In a different international political environment, that might have been enough to save
it. Archives of the Abraham Lincoln Brigade (ALBA) website,
30 June 2000. Conventional
military histories in Servicio Historico Militar (Col. J. M. Martinez Bande), La batalla del Ebro
(Madrid: Editorial San Martin, 1988); R. Ballester, La batalla del Ebro (Barcelona: Bruguera,
1974)-
 .

JVegrin’s war on three fronts 383

Butwas not events in Spain that turned Republican retreat into

it

certain defeat - it was what happened in far away Czechoslovakia. As

the Ebro battle raged, the crisis broke over Nazi Germany’s claims to the
Sudetenland. When Britain and France capitulated to Hitler at Munich
at the end of September, they signed away not only Czech independence
but Spanish Republican democracy too. Had Britain stood firm over
Czech guarantees, then the horizon would not have closed down so
dramatically for the Republic. 250 Crucially, right up to the British and
French capitulation, Stalin was, in spite of all the difficulties, prepared to
mobilise Russian forces and, to this end, was in September sending Soviet
planes to Czechoslovakia. 251 An agreement with Romania to allow the
252
passage of Soviet troops was also in the But after the Munich offing.

meeting - to which the Soviet Union was not invited - Stalin drew his
conclusions about the remoteness of any grand alliance against Nazi
Germany.
Soviet foreign policy remained fluid, however. At the end of August,
Stalin accepted Negrin’s proposal to withdraw what remained of the
International Brigades. 253 It would not affect the Republic’s defensive
capacity and could help counteract British hostility. Military advisers had
alsomainly been withdrawn by this point. But Comintern personnel re-
mained. Credits too had been supplied to the Republic. Nor were future
consignments of Soviet-procured military aid ruled out. There was no
sudden cutting of ties with the Republic — even after Munich - because
its resistance was still valuable. But given the Republic’s diplomatic iso-

lation and the enormous difficulty now of delivering war material to it,

itseems clear that Stalin had ceased to see Republican resistance as an

instrument for achieving any putative anti-Nazi alliance with the democ-
racies. If this were now to be realised, it would be sealed over the grave
of Spanish democracy.
The Munich debacle, coming in the midst of the Ebro battle, had a
devastating political effect inside Republican Spain. This was particularly
true in the centre-south zone, bounded by hostile territory and the sea.

250
M. Alpert, A New International History, p. 167
251
H. Ragsdale, ‘Soviet Military Preparations’, pp. 210-26, esp. 220-1.
252
221-3; although the poor state of Romanian road and
Ibid., pp. rail links would have seriously
hindered land transportation (Ragsdale, pp. 219-20).
2 53
Voroshilov to Dimitrov and Manuilsky, 29 Aug. 1938, cited in Radosh, Habek and Sevostianov,
Spain Betrayed. The Soviet Union in the Spanish Civil War, p. 469; Elorza and Bizcarrondo, Queridos
camaradas, pp. 421-2; International Solidarity with the Spanish Republic 1936-1939 (Moscow: Progress
Publishers, 1975), p. 328; J. L. Alcofar Nassaes, Los asesores sovieticos en la guerra civil espahola

(Barcelona: Dopesa, 1971), p. 150.

384 The Spanish Republic at war

There, since ihe spring of 1938, isolation had been feeding a sense of
abandonment and desolation, increased by what were now very wretched
material conditions and mounting hunger for most people. Acutely aware
of this situation, Negrin had secret talks in Zurich in the second week
of September with Franco’s emissary, the Duque de Alba. But these
had only confirmed Franco’s uncompromising desire for unconditional
surrender. Negrin saw no choice but to continue Republican resistance to
force Franco to change his mind - or, rather, to oblige others to change it
for him. The withdrawal of the International Brigades 254 — which began
in October - had likewise been agreed to raise the diplomatic stakes,
thereby increasing the pressure on Franco. 255
It was only now too, in October, against the calvary of the Ebro, with

the Republic’s political options totally closed and with Negrin increas-
ingly fearful of internal collapse that the trial of the POUM’s leaders
went ahead in Barcelona. (After Munich there was
no longer any clearly
point in worrying about adverse British reactions to the trial.) For Negrin
the POUM trial constituted iron in the soul: a symbolic punishment de-
signed to cauterise the real fifth column. (Members of the fifth column
network in which there had been a vain attempt to implicate the POUM
had already been convicted.) Many have supposed that the PCE’s influ-
ence was crucial here in persuading Negrin to hold the trial. The party
was certainly baying for POUM blood. But just as important here were
Negrin’s consistently held political views and his own perception of the
needs of the fronts. He interpreted the perspective of the ordinary com-
batants through the prism - indeed the prejudices - of his own vehement
centralism and statism.
There had been lengthy legal wrangles between prosecution and de-
fence council over whether the POUM case should be tried by a mil-
itary or civil court. Even so, the preparation of the prosecution’s case
seems basically to have been concluded by June 1938 - in the run-up
to the Ebro. 256 (So it had taken approximately the same amount of

254 The withdrawal was announced by Negrin at the League of Nations on 2 September: Moradiel-
1

los,La perfidia de Albion, pp. 315-16. General Rojo had already made it clear to the prime minis-
ter that this would not damage the Republic’s fighting capacity. Rojo to Negrin, 9 September

1938, reproduced in Martinez Paricio, Los papeles del general Rojo (in (unpaginated) documents
section).
255 Moradiellos, La perfidia de Albion, p. 330. The Brigades’ farewell parade took place in Barcelona
at the end of October. On the rebel side, there was only a token withdrawal of 10,000 Italians
(CTV) in order to activate the Anglo-Italian agreement of 16 November. The bulk of the CTV -
some 35,000 troops - remained until the end of the war.
256
Gorkin, El proceso de Moscu en Barcelona, p. 234.
 -

Negrin’s war on threefronts 385

time as the judicial investigation for the trial of General Asensio and
members of the Republican military high command for responsibilities
in the collapse of Malaga in February 1937 257 ) But the POUM trial
took place at a far more critical juncture for the Republic politically
and militarily In spite of this pressure, however, and the fact that the
PCE had done its best to intimidate the defence council and to harden
public opinion against the POUM, the trial followed constitutional pro-
was not in any sense a ‘Moscow show trial - in spite of
258 5
cedures. It

the title of the memoirs published in 1974 by POUM defendant Julian

Gorkin. 259 (The third and final Moscow trial had taken place in March
1938.) Moreover, given that the five judges in the POUM trial were state
appointees (as decreed for the courts of espionage and high treason), we
could deduce that the Republic’s young constitutionalism was far health-
ier than might reasonably have been expected in the dreadful autumn of

1938-
The POUM leaders were found not guilty of treason or espionage.
But some of them were convicted of rebellion against the Republican
260
state. It has been claimed that Negrin demanded that the court im-

pose the death penalty — on the understanding that the cabinet would
commute it. The prime minister would appear to have considered this
261
option in October 1937. Moroever, as we have noted before, it was
consistent with his demand that the death penalty be applied to General
Sanjurjo, who had led the military rebellion of August 1932 against the
Republic. Of course, the POUM
leaders had endorsed and joined an
already occurring rebellion rather than planning or leading their own.
But for Negrin there was also the added irritant of the party’s perceived
Catalanism (in spite of Andreu Nin’s best efforts here). Most of all in
Negrin’s calculations, however, there was the acute military situation of
late 1938. There is nevertheless something problematic in the speculation

over Negrin’s call for the death penalty: namely the manner in which it

257 See chapter 3 above.

258
The anti-stalinist left outside Spain had feared it would not - see, for example, the French left
socialist Marceau Pivert’s visit to Pascua on this matter: M. Pascua (in telegram) to J. Alvarez
del Vayo, 13 July 1938, AHN/MP caja 1 (21). Pascua suggested publishing the trial proceedings
to allay fears in France.
259 El proceso de Moscu en Barcelona is in fact an expanded edition of Gorkin’s earlier work, Canibales
(Mexico: 1941). On Gorkin, see H. R. Southworth, ‘Julian Gorkin, Burnett Bolloten and
politicos

the Spanish Civil War’, in Preston and Mackenzie, The Republic Besieged pp. 261-310. ,

2 6°
The sen tence reproduced in Alba and Ardevol, El Proceso del P.O.U.M. and also in Suarez, El
is

proceso contra el POUM pp. 202-9.

,

261
A. Elorza and M. Bizcarrondo cite the somewhat telegrammatic notes made by the PCE
ministers of a cabinet meeting held in late October 1937, Queridos camaradas p. 379. ,
386 The Spanish Republic at war

is said to have, occurred. The public prosecutor in the trial, Jose Gomis,
was very close to the government. Yet he did not demand the death
penalty for any of the POUM defendants. Instead, hearsay has Negrin
attempting to intervene at the eleventh and three-quarter hour of the
262
trial to The oddity of this does not, of
influence the presiding judge.
course, prove that it may be that finally Negrin, de-
never happened. It

prived of the support of his own party and much of his own government,
had himself succumbed to the panic-induced siege mentality that was
gradually taking over in the Republican zone.
The denunciation or persecution of those deemed to be ‘enemies
within’ was now also fuelled by popular anxieties over the war. As
the post-Ebro retreat of Republican forces began in the second half
of November - and with it a huge movement of civilians - the vacuum
of authority on the ground and the ensuing chaos led to many violent
263
incidents. Soldiers took food and intimidated or killed civilians who
tried to stop them. Retreating soldiers were sometimes shot by other
military personnel who saw them as deserters. Civilians too were at
times caught up fatally in this maelstrom. Its most notorious victim was
the belligerently pro-rebel Bishop of Teruel, Anselmo Polanco, who had
been in Republican custody since the beginning of 1938. He would be
killed in unclear circumstances in February 1939 en route to the French
264
frontier.

Among Republican security forces too - the SIM included - it was

the war’s cumulative influence that explains their modus operandi. The
political culture of the Russian advisers in Republican Spain also had a
strong component of siege mentality, of course - deriving in part from the
extreme experience of the Russian Civil War and the subsequent inter-
national isolation of the Soviet Union. But they plugged into existing and
well-founded Republican fears as much as they shaped them. This was
as true in 1938 as it had been during the siege of Madrid in November
and December 193 6. 265 Comintern fears of the ‘enemy within’ in

262
Bolloten, SCW, pp. 518-19.
263
Julia et al., pp. 259-62 and see also pp. 252-3.
Victimas de laguerra civil,
264
Raguer, La polvoray pp. 237-9, 178. If indeed Polanco’s killers knew who he was,
el incienso,

then his quite bellicose political sidetaking is enough to account for what happened. To
ascribe his death to a form of atavistic and visceral popular anti-clericalism seems otiose
and, indeed, deeply misleading, Montero Moreno, Historia de la persecucion religiosa en Espaha,

pp. 424-7.
265 Habek
See Col. Sverchevsky (‘General Walter’) to Voroshilov, report of 2 Aug. 1938, in Radosh,
and Sevostianov, Spain Betrayed. The Soviet Union in the Spanish Civil War, pp. 477-87, which gives
 Negrin's war on three fronts 387

war-torn Spain were doubtless fuelled throughout by the atmosphere

given off by the Moscow But this was never simply transferred
trials.

‘neat’ or unmediated to Republican Spain — as if the latter were simply

a blank screen to be written on. Rather the real and increasing vulner-
ability of the Republic to fifth column sabotage blended with the fears,
ambitions and sectarianism of a populist, massified PCE. This was mag-
nified through the ‘sound box’ of the Comintern to become mutually
reinforcing.
In the end, to focus criticism on the SIM specifically - and especially
on a mythic, Sovietised one — is to miss the point that the SIM was
symptomatic of a broader problem. The harshness of its policing activi-
ties was part of how the imperatives of war had led to the militarisation

of the Republican home front in a desperate attempt to hold off the dis-
integration born of material inferiority and lack - simply to hold things
together and discipline the war effort. Increasingly harsh police action
and the erosion of constitutional and judicial norms were not evidence
of communist, still less Soviet, influence. They were evidence that the
war was consuming everything: individuals, parties (the PCE included)
and the very fabric of Republican democracy
For the Caballeristas and others with political grievances inside the
socialist organisations, and for many anarchists and republicans as well,
there now seemed no point in repressing their anger and dissent any
longer.There was also a growing sense that no peace terms could be
worse than what they were already enduring. Negrin, from a different
vantage point, knew But although the logic of his resistance strat-
better.

egy was impeccable, to come up hard against the opinion

it was soon
of influential members of his military high command, in particular that
of the chief of staff, General Rojo. Their belief that, after the massive
destruction of the Republican Army at the Ebro, any resistance capacity
was to be measured in months would give a tremendous boost to Negrin’s
political opponents. Precisely because of this, and given the disintegra-
tion inside the PSOE, Negrin looked increasingly to the PCE by the
end of 1938 to bolster resistance, thus further alienating many socialists.
Itwas Munich, then, which lit the long fuse on the time bomb primed

ample evidence of how the immense pressures of war against the odds were intensifying fears
of the ‘enemy within’. The language /categories employed in the report (by a Soviet officer to
his political masters) are the monolithic ones of the purges. But we have to read their meanings
in the light of the rising military crisis on the Ebro and the desperate war weariness of the
Republican home front.
3 88 The Spanish Republic at war

in centre-south Spain by hunger and demoralisation. It would explode

some months later in the shape of the Casado coup. 266

CONCLUSION
Isolated in Europe, the Republic was always dependent upon the full-

scale political and economic mobilisation of its home front to sustain

the war effort. Within the Popular Front it was the PCE that had come
closest to appreciating the importance of strategic propaganda work
and, thereby, went some considerable way to achieving mass home-front
mobilisation. As a result, the reconstituted wartime Popular Front was
able to go some distance over 1937 to articulating the political power and
state infrastructure crucial to a sustained defence against the military
rebels and Germany and Italy. But thereafter this would be undermined
by lack of time and financial and human resources. The Republic was
at full stretch simply administering the military and home fronts, as well
5
as sustaining its ‘war on the diplomatic front. There was neither energy
nor resources to devote to extending social reform (in health, welfare,

education) beyond what Largo’s October 1936 decrees had envisaged.

Ultimately, then, the way the war had
be financed (since very limited
to
credit was available from outside) prevented the Republic from building
a social contract with those fighting and dying for it. How, in the face of
increasing material shortage and cumulative military defeats, could the
Republic believably enact or sustain a wide mobilisation in the name of
a social order that did not yet exist? 267
Precisely because of the overwhelming material lack of everything in
Republican Spain by 1938 it is deeply problematic even to pose fur-
ther questions about how far its political shortcomings ‘explain’ its col-
lapse. T he Popular Front certainly failed to develop sufficient multiple
discourses and mobilising strategies. A more politically liberal, more
materially inclusive Republican project (than that of 193 1-3) remained
unrealised. But how could even the most sophisticated or modern of
political systems have sustained popular morale once it could not deliver

266
See chapter 7 below.
267
The national Republican community may, ironically, have been realised only in the army -
through the work of political commissars, education and welfare facilities and through the
solidarities forged under fire. This seems a reasonable working hypothesis for the period prior
to the desperate mobilisations, shortagesand (consequent) disciplinary brutality of 1938 that
eroded military morale. But as yet we lack any thoroughgoing study of the Republican
seriously
army that might address such questions.
 Negrin’s war on threefronts 389

to its population the basic wherewithal of daily life? By 1938 many were
actually starving. In the end, the political erosion of the Popular Front
project always comes back to the long term impact of Non-Interventionist
embargo. The ‘common cause’ of any war effort is always provisional,
268
conditional and potentially fragile. In the face of too great a mate-
rial disadvantage, no amount of ideological mobilisation can ever be
enough. 269

268
See Calder, The Myth of the Blitz p. 90.
,

269
See J. Barber and M. Harrison, The Soviet Home Front IQ4 / -1945 (Harlow: Longman, 1991),
p. 177.
 .

CHAPTER 7

The collapse of the Republican homefront

To on because there was no other choice,

fight
even was not possible, then to salvage
if winning

what we could - and at the very least our self respect.

Why go on resisting? Quite simply because we

knew what capitulation would mean.

The angel would like to stay, awaken the dead,

and make whole that which has been smashed.
But a storm is blowing from Paradise . .

This storm irresistibly propels him into the future

to which his back is turned, while the pile of debris
before him grows skyward.
This storm is what we call progress. 1

At the meeting of the Republican parliament held at the beginning of

October 1938 in Sant Cugat del Valles, Negrin requested and received an
unconditional vote of confidence from the assembled MPs. This meant
that a vote had been cast in favour of the prime minister’s strategy of
defensive resistance - resistance, that is, until a peace could be achieved
which offered guarantees That Negrin obtained this
for the defeated.
result in the Cortes, notwithstanding his vehement critics in both the
PSOE and republican ranks, highlights two crucial realities. The first
was that no one else was prepared to take political responsibility for
the Republic’s fate. Second, this was so because no one had - or ever
would have - any alternative political strategy to substitute for Negrin’s. 2
1
The first two quotations are from Negrin in the Epistolario Prieto
y Negrin, p. 37 (‘Seguir luchando,
porque no habia mas remedio para, si no se podia ganar, salvar lo que se pudiera o, al menos,
salvar el decoro’) and p. 44 (‘Resistir, ,:por que? Pues sencillamente porque sabiamos cual seria
el final de la capitulacion’), both in Negrin’s letter of 23 June 1939. The third is from Walter

Benjamin, Theses on the Philosophy ofHistory IX.

,

2
Graham, Socialism and War, pp. 156—7; S. Julia, ‘La doble derrota de Juan Negrin’, El Pals,
26 Feb. 1992.

390
 The collapse of the Republican home front 391

It was precisely a tacit awareness of this that explains why none of the
republican intrigues against Negrin ever solidified. 3
But even if there was no alternative, except unconditional surrender
by the Republic, unfortunately that did not mean that Negrin, even
equipped with his vote of confidence, necessarily had the wherewithal
to pursue his own policy In the wake of the Ebro retreat, the prime
minister would manage to negotiate further arms supplies on credit from
the Soviet Union. In connection with this, the patrician communist and
chief of the Republican Airforce, Ignacio Hidalgo de Cisneros, visited
Moscow in early December. 4 A not insignificant amount of war material
was dispatched - even though it fell short of Negrin’s shopping list. 5
But vital as was the promise of Soviet supplies, the domestic requirements
of Negrin’s defensive resistance were just as important. And here the
political backing of his PSOE-based government was vital. But Negrin
could no longer rely on this.

In the middle of November the Ebro retreat had begun. By the 18th
the last units of the Republican Army had withdrawn across the river
to its north bank. Knowledge of the magnitude of this military reverse
split political divisions inside the PSOE. At Azana’s invi-
wide open the
tation, Besteirocame from Madrid to Barcelona to talk about a peace
government. But both men knew that they lacked the positive politi-
cal backing - in the PSOE or beyond - to take such a step. Besteiro
also attended the extraordinary meeting of the PSOE national executive
committee on 15 November, called by general secretary Lamoneda in a
last-ditch attempt to solder leadership unity in the face of the Republic’s

3 For this reason, it is somewhat problematic to refer to the opposition to Negrin as a ‘party of
peace’ - as do Angel Bahamonde Magro and Javier Cervera Gil in their excellent study of the
end of the war, Asi termino la guerra de Espana (Madrid: Marcial Pons, 1999).
4 Cisneros’ memoirs exaggerate the significance of his trip (see Cambio de rumbo (Vitoria-Gasteiz:
Ikusager, 2001), pp. 543-49). As Marcelino Pascua, Republican ambassador in Moscow, indi-
cated, previous Soviet credits had always been negotiated by Negrin personally through formal
diplomatic channels, and there is no reason to suppose that it was any different in the last months
of 1938, Pascua to A. Vinas, 13 Feb. 1977, AHN/MP caja 8 (13). Vinas agrees, El oro de Moscu,

p. 419.Cisneros’ trip was a way of underlining the urgency of Negrin’s - independent - request.
Conceivably, it was also intended to function as a kind of ‘diplomatic semaphore’ to Britain and
France that the Republic intended to continue resisting. It may also have been designed to stay
the PCE’s diminishing belief in Negrin’s own commitment to resistance. For the text of the letter
from Negrin to Stalin carried by Cisneros, see Guerray revolucion en Espana vol. 4, pp. 198-200.
,

5 Howson, Arms for Spain p. 243: Pascua to Vinas, 13 Feb. 1977, AHN/MP caja 8 (13). As Pascua
,

indicates, we certainly cannot assume the accuracy of Hidalgo de Cisneros’ claim that a one
hundred million dollar loan was agreed, virtually over dinner ( Cambio de Rumbo p. 548). To date,
,

however, we lack access to Negrin’s own papers, where documentary evidence of the figure, and
how and when it was agreed, may be located.
392 The Spanish Republic at war

mounting military crisis. But Besteiro carefully insisted that his presence

did not mean he

accepted ‘membership’ of the party executive. Indeed,
he used the occasion to make a blistering indictment of Negrin’s war
policy and its consequences for the PSOE. His hysterical denunciation
of the communists replicated Francoist propaganda in both its vehe-
6
mence and its naivety. Virtually in the same breath, however, Besteiro
admitted that there was no alternative to an alliance with the PCE if the
Republic was to be kept alive. In short, he was acknowledging Negrin’s
dilemma as prime minister but, like Prieto before him, Besteiro refused
to share the political responsibility which should logically have derived
from his reasoning. All three veteran (
historico) socialist leaders — Largo
Caballero as well as Besteiro and Prieto - used the November meeting
to reiterate their withdrawal from party life - effective since the PSOE
national committee of August. Henceforward, the historicos and some of

their followers would actively be searching for a way to end the war.
Besteiro, like Largo, seemed to be assuming that a Franco dictatorship
would resemble that of Primo de Rivera in the 1920s, with its space for
collaboration with ‘responsible’ socialists. 7 Besteiro and others - both
inside and outside the socialist movement - were coming to believe, con-
sciously or otherwise, that all that stood between them and peace was
the blind will of the PCE as ‘the party of war’. Indeed, Besteiro would
seemingly be coming to believe the Francoist propaganda line that, by
handing over the PCE, the Republicans could ‘purify’ themselves and es-
tablish a basis for post-war reconciliation ‘between Spaniards’ (although
obviously not Spaniards who were communists). Otherwise, it is hard to
make sense of the role Besteiro would later play in the rebellion against
Negrin led by Colonel Segismundo Casado, the commander of the army
of the Centre, in March 1939.
Munich and the Ebro defeat had, then, radically destabilised Negrin’s
support base in the PSOE. 8 Precisely because of this, Negrin was obliged
to look to the PCE for ‘iron in the soul’ - even though this inevitably
further alienated many socialists. In early December Negrin vainly
floated the idea of dissolving the political parties into a single political

6
A verbatim account of Besteiro ’s tirade is to be found in the minutes of the 15 November meeting,

Fundacion Pablo Iglesias (Madrid), AH-20-5; Preston, Comrades pp. 182-4. ,

7 Besteiro would articulate this sentiment during the Casado events, Bahamonde Magro and
Cervera Gil, Asi termino la guerra de Espana, pp. 407-8. Nor was this a sentiment confined to
party leaders: see the preparations of the Ciudad Real PSOE in March 1939 to ‘re-establish the
fabric and content’ of their organisation, Adelante, 8, 9 March and Avarice 14, 17 March 1939, cited
in Graham, Socialism and War, p. 239.
8
See Lamoneda’s comment to Togliatti on 21 November 1938, cited in Elorza and Bizcarrondo,
Queridos camaradas, p. 425.
 The collapse of the Republican home front 393

coalition (‘Frente Nacional’ or ‘National Front

5

). He saw it as a means
of breaking the deadlock of factionalism and also freeing himself from
the international accusation that he was an instrument of the PCE.
But the Politburo was not keen 9 At heart, most of the Spanish com-
.

munist leaders were no happier than their socialist counterparts at the

prospect of a diminished party-organisational identity. (The PSOE lead-
ership, needless to say, opposed Frente Nacional root and branch.) The
communists were also wary of Negrin’s personal absorption of more
and more executive authority. Indeed, relations were deteriorating be-
tween premier and party — not least because of the suspicion, shared by
Togliatti, that Negrin’s coalition idea was also inspired by his desire to
break through to negotiations with opposition elements in the Francoist
10
camp .

However, it was not Negrin’s agenda that constituted the heart of the
PCE’s dilemma, but the tenor of feeling deep inside the PSOE and the
general war weariness beyond. Both had played their part in the fierce
response to the dismissal on io November of a socialist political commis-
sar, Fernando Pinuela, by the communist leader, Jesus Hernandez, then

head of the political commissariat for the centre-south army group 11 .

Relations between the PCE and the PSOE were so bad on the ground
that, to all intents and purposes, the Popular Front was extinct. That
being the case, if defensive resistance was to be maximised, one could
argue that it would have been in the best interests of the PCE as a party to
abolish the fiction of the Popular Front and take up a much more visible
level of governmental responsibility. But the fact was that the PCE was
simply not powerful enough to do this on its own initiative — even in
the absence of any other coherent Republican alliance or strategy. Nor,
even if it had been, would the Soviet Union have had any interest in a
communist ‘coup’.
Maintaining Republican resistance for as long as possible went on be-
ing important for the Soviet Union after Munich because it was a means
of staving off fascist aggression in the east that would be directed against
itself. This was why Stalin agreed to send further aid to the Republic

in late November - Union having more pressing

in spite of the Soviet
foreign-policy concerns than Spain by then, and in spite of the logis-
tical problems of transportation and the risks of the arms falling into

enemy hands. (This aid would reach the Republic by the second half of

9 Togliatti, report of 21 May 1939, Escritos sobre la guerra de Espafia, p. 237.

10
Elorza and Bizcarrondo, Queridos camaradas, p. 426.
11
Graham, Socialism and War, pp. 160-1, 227.
394 The Spanish Republic at war

12
January 1939* But it was already too late to make any difference. )

In fact, the Soviet leadership never entirely gave up on the possibility

of prolonging Republican resistance until this was definitively capsized
13
by the rebellion against Negrin in March. Nevertheless, Stalin knew
that Spanish resistance was time-limited. He was also contemplating
the potential of an anti-German alliance with Britain and France being
achieved - even after Republican defeat. For this strategy to remain feasi-
ble, however, there could be no departure in the interim from the Popular

Front line in Republican Spain. Hence the Comintern’s vehement re-

jection of Negrin’s Frente Nacional idea, which chimed with Togliatti’s
own reservations and those of the Politburo. 14 Precisely because
this was

the Comintern’s logic, makes little sense to interpret the POUM trial
it

of October 1938 as having been somehow imposed on the Republic by

external Soviet pressure.
The PCE thus continued trying to galvanise resistance from behind
the slogan of the Popular Front. But this was an immensely dangerous
strategy for the party. It meant that the PCE would inevitably be focusing
on itself all the intensifying political hatreds, frustrations and general war
weariness in the Republican zone without having any means of protect-
ing itself against those animosities. Ironically, this scenario was obliquely
forecast by the PSOE executive during the conflict over Pinuela’s dis-
missal. Indeed, it is a fear of what popular and populist ‘anti-communism’
(i.e. a mixture of war weariness, the peculiar isolation-inspired anxiety

permeating the centre-south zone, political disillusion of diverse origins

and cypto-Francoism) would do to the PSOE if it were unleashed that
explains why the party executive backed down from its confrontation
with Negrin over Pinuela’s reinstatement. 15 For the PCE, however, the
consequences of its increasing vulnerability would not be long in making
themselves felt.

12
The Soviet Union was particularly worried by Franco’s demand that France should hand over
the material to him. The Union insisted that the French government return the arms
Soviet
since they had been sent on credit and were therefore still not paid for: Howson, Armsfor Spain,
p.243.
13
Elorza and Bizcarrondo, Queridos camaradas, p. 433 cite a telegram sent from Moscow on
5 March 1939. It requested information on the state of Republican resistance and made clear that
continuing aid was still on offer if that resistance was holding and passage through France could
be assured. The fear of weaponry falling into enemy hands was now acute, hence Voroshilov’s
earlier cautious response of 16 in Radosh, Habek and Sevostianov, Spain
February 1939, cited
Betrayed. The Soviet Union in the Spanish Civil War, p. 512 (doc. 81).
14
Comintern telegram to Togliatti of 10 December 1938, cited in Elorza and Bizcarrondo, Queridos
camaradas, p. 427.
15
Graham, Socialism and War, p. 161.
 The collapse of the Republican homefront 395

Meanwhile, whatever the recommendations of Xegrin’s defensive re-

sistance strategy, sustaining it required not only material aid from the

Soviet Union and internal support from the PCE (in lieu of a func-
tioning Popular Front government) but also an army in fighting con-
dition and an international power prepared to bring about mediation.
But mediation would always remain beyond the reach ev en of Negrin’s
tremendous political will and intelligence. And in the wake of the Ebro
losses, it was far from certain whether the Republic's forces in Catalonia
1 *1

could withstand the coming enemy offensiv e — in terms of either mate-

rial or morale. Once the Ebro retreat had begun, and in the light of

the shattering knowledge of Munich, the feeling was widespread that

the war was in the process of ending. General Rojo sought to implement a
series of diversionary actions on other Republican fronts in the centre-
south zone in order to pre-empt Franco’s offensiv e and take the pressure
off Catalonia. But these were, for whatever reason, delayed by Miaja
and Matallana, the generals in charge of the army group of the centre.
Certainly the Ebro defeat had eroded Xegrin’s credibility in the eyes
of many professional army officers who were now beginning to reassess
the validity and viability of prolonging the resistance. 17 In the ev ent, the
various diversionary actions came too late and were too small to serv e
18
the purpose. As a result, Franco’s Catalan offensiv e, which had begun
on 23 December, accelerated rapidly from 3 January 1939 as the Repub-
lican forces rapidly ran out of reserv es, both human and material. On

15 January, Tarragona fell to the rebels, thus opening up a dir ect route
to Barcelona. Franco’s progress was facilitated by’ the fact that Madrid’s
military intelligence was riddled with spies. His forces had known about
the div ersionary strategies as soon as had those Republican military’ com-
7

manders charged with implementing them - and in some cases probably

ahead of those Republican commanders.
Republican forces in Catalonia continued to fight. But for the most
part they were fighting to delay Franco’s advance as they’ themselves
7

retreated.Moreov er, the bleakness of the international environment and

the chaos and panic of these times further hastened the collapse of civilian
morale in Catalonia. Since April 1938 urban Catalonia had been isolated

16
Some 220,000 men, compared to c. 500,000 in the centre-south armies. But of the forces in
Catalonia, only just over half (c. 140,000) were encadred in the effective, mobile Mixed Erigades.
17
Bahamonde Magro and Cervera GiL Asi termino la guerra de Espana. pp. 350-1.
18
M. Taguena Lacorte, Testimonio de dos guerras Mexico: Ediciones Oasis. 1973; 2nd edn. 1974).
pp. 264—5, 2 74~5- The reduced scale of the operations also seems to have been at the insistence
of Miaja and Matallana.
396 The Spanish Republic at war

from Valencian food. By the last quarter of 1938 it was facing what
today’s aid agencies would term a major humanitarian crisis. Starvation,
homelessness and threatening epidemic disease - all fed the rising tide
of war weariness and disillusion.
On the night of 21 January Rojo told Negrin that the Republican
front was broken in crucial places en route to Barcelona. The fall of the
capital was imminent. As the military crisis in Catalonia deepened, the
prime minister ordered the evacuation of state and government appara-
tus which began its slow journey of stages to the French frontier, amid a
great tide of soldiers and civilians. On 23 January, Negrin finally declared
a state of war across Republican Spain. The evacuation of Barcelona
marks the point at which, to all intents and purposes, central government
and Republican state machinery ceased to exist as operational appara-
tuses of power - even if they certainly went on existing as legal entities.
Republican forces covered their departure. But now there was no front,
and nothing to stop the advance units of enemy troops - Navarrese,
Italian and Moroccan — who entered the city unopposed on 26 Janury
*
939 -

The manner of Barcelona’s fall has generated a range of accusatory

myths and other suspiciously totalising ‘explanations’, in which the sub-
ject of blame ranges from Catalan nationalism through stalinist counter-
revolution to the depredations of the central Republican state. Catalan
nationalists’ commitment to the war had certainly been eroded by
Negrin’s heavy-handed centralism. This had also produced debilitat-
ing tensions within the PSUC leadership between Catalanists and com-
munists. So too the erosive daily struggle for material survival which
sapped the will to resist across 1938-9 could have been internalised by
CNT-related worker constituencies as the consequence of earlier revo-
lutionary failure. But it is difficult to extrapolate from this to ‘outcast’
or even working-class Barcelona as a whole and impossible to know if
even CNT-related sectors went on explaining things predominantly in
these terms throughout the war. What we do know, however, is that
Barcelona was a refugee-packed city blasted by bombs in 1938, starved
of food, and then of hope by the ‘bomb’ that was Munich. By the end
of 1938 Catalan industry was militarised to the teeth. But production
could not be sustained - not least because of the power shortages that
had followed Franco’s conquest of important hydroelectric resources in
the spring of 1938. Among troops as well as pro-Republican civilians
the knowledge of a totally blocked international horizon weighed like
 The collapse of the Republican homefront 397

lead. In short, the answer to whywas not

‘Madrid, November 1936’
replicated in ‘Barcelona, January 1939’ lies in the intervening years of
warfare and what they had cost both materially and psychologically. 19
Between ‘Madrid’ and ‘Barcelona’ lay a gulf of experience and knowl-
edge. The Republic had done and given everything. But everything was
clearly never going to be enough. Hence the growing sentiment: ‘let it
20
be over, however it ends, just let it be over now’.
Nor would the inevitability of closure ever be far from Negrln’s mind
in these February days. At midnight on 1 —2 February in the old castle at
Figueras he summoned the remnants of the Republican Cortes to meet
for the last time on Spanish soil. His three conditions for ending Repub-
lican resistance — a guarantee of Spain’s territorial integrity, a national
plebiscite to decide the country’s political future, and a guarantee of no
reprisals against the Republican population (with a right of evacuation
for all those themselves to be at serious risk) - had already in
who felt

fact been reduced to the last of these three. But the third condition, as
Negrin had already informed France and Britain, and as he would reiter-
ate many times over subsequent days, was not negotiable. The diplomats
were silent. But Franco’s vindictive reply to Negrin would soon ring loud
and clear in the scarifying terms of the Law of Political Responsibilities,
21
published on 16 February. Yet still France and Britain (and especially
the latter) kept Negrin to capitulate. He was caught
up the pressure for
in an impossible situation. On the one side stood an implacable military
and political enemy which would not cede what could be taken by force
and, anyway, saw the purging and punishment of Republican Spain as
integral to its new political project. On the other side, the British and
French political establishments had not the least expectation of influenc-
ing Franco - and probably little concern to do so either.
When Negrin accompanied Azana over the frontier to France on
5 February, the prime minister already knew from Rojo that what
19
Franco in Barcelona (United Editorial: London, 1939), pp. 5, 13 (this was an eyewitness report of
the fall of Barcelona, written by a Quaker relief worker, Muriel McDiarmid); R. Abella, Finales
de enero, 1939. Barcelona cambia de piel (Barcelona: Planeta, 1992), passim , but especially pp. 85-6,

87 -9 134-7
,
-

20
The feeling was widespread - as recalled, for example, by Eduardo Pons Prades, interviewed in
‘Victory and Defeat’, the final programme of the Granada Television series The Spanish Civil War
(first broadcast 1982).
21
For Negrin’s acute verdict on this, see his speech in the Palacio de Bellas Artes, Mexico, : August
1945: ‘in war without quarter, such as ours has been, either every crime is a
a savage civil

“common crime” or none are’ (referring to Franco’s dubious claim to make a distinction under
the law between political crimes and common crimes perpetrated during the war), Documentos
politicos para la historia de la Republica Espahola (Mexico: Coleccion Malaga, 1945), pp. 25-6.
398 The Spanish Republic at war

remained of Gatalonia was falling. 22 (Franco had taken Girona on the

4th, and the island of Minorca was in the process of a British-brokered
surrender.) Having signed the final orders for retreat on 8 February,
Negrin, accompanied by Rojo and the cabinet ministers Julio Alvarez
del Vayo, Vicente Uribe and Francisco Mendez Aspe, 23 saw the first
Republican units march across into France. Most of the remainder would
have crossed over by the 11th. 24 The end of Republican resistance in
Catalonia vastly increased the anxiety of the Soviet Union that mili-
tary aid dispatched to the Republic might enemy hands before
fall into
25
reaching the centre-south zone. General Rojo, who on 29 January had
confided to Azana and Negrin that only minimal, short-term defensive
resistance was possible, also signalled his effective withdrawal by remain-
26
ing in France. But Negrin, Alvarez del Vayo and Santiago Garces, the
head of the SIM, returned immediately to Spain on 8 February, flying
from Toulouse to Alicante in the centre-south zone.
Negrin’s January declaration of a state of war had effectively handed
power in the centre-south zone to the military - and specifically to Miaja
and Matallana as, respectively, chief of the centre-south army group and
chief of the general staff. Negrin had had no choice if he wanted to keep
the centre-south zone afloat. But in thus downgrading the authority
of civil governors and other civilian agents of the central government in

22
Azana was acompanied by his old friend Jose Giral, first prime minister of the Republic at
war and later foreign minister, and the Republican vice-president, Diego Martinez Barrio. The
Basque and Catalan premiers, Jose Antonio Aguirre and Luis Companys, left later the same
day.
23 Once Negrin had taken over the defence portfolio from Prieto in April 1938, Mendez Aspe,
Negrin’s deputy in the treasury, effectively assumed ministerial responsibility.
24 The bulk of the Republican units had come across by 11 February, covered by the 35th Division,
under the command of the very young Lieutenant Colonel Pedro Mateo Merino: see his memoirs,
la nuestra. Andanzas y reflexiones de un combatiente republicano (Madrid: Editorial
Por vuestra libertady
Disenso, 1986).The Thirty-Fifth International Division, to give its full title, had encompassed
the remainder of a number of International battalions, although by the time of the Ebro these
were largely manned by Spanish soldiers.
25 Even the Soviet aid that had arrived in the middle ofJanuary often remained in, or was returned
to, France for fear that it would fall into enemy hands, Howson, Arms For Spain, p. 244. The
planes that crossed the frontier in pieces were sent back because there were no longer functioning
aerodromes in Catalonia where they could be rebuilt.
26
Zugazagoitia, Guerra y vicisitudes, pp. 530-1. Zugazagoitia’s criticisms of Rojo’s refusal to obey
Negrin’s order to return to Spain were even harsher in private. Both he and Marcelino Pascua
(who described Rojo as having a ‘difficult’ or ‘prickly’ personality (‘espinoso caracter’)) were
appalled by what they saw as the evasion of personal responsibility, sustained by half-truths
at best, evident in Rojo’s book Alerta los pueblos! (1939), Zugazagoitia to Pascua, 12 Feb. 1940
j

and Pascua to Zugazagoitia, 27 March 1940, AHN/MP, caja 2 (bis) 16 (exilio). This, as well as
the charged and hyper-personalist politics of exile, no doubt explain Rojo’s notable hostility to
Zugazagoitia and his lukewarm response to the appeal against Zugazagoitia’s death sentence by
the Francoist courts.
 The collapse of the Republican homefront 399

what remained of Republican territory, he was laying the structural basis

for later military-led action, and ultimately rebellion, against himself.
But why did Negrin return from France at all? Few, if any, of his military
and political collaborators were expecting him too. 27 On io February
Negrin met Miaja and Matallana, at the Penon de Ifach, near Alicante.
On the succeeding two days he held cabinet meetings near Valencia and
then in Madrid. (All his ministers had followed him back to Spain, save
28
the republicans (who included Giral and Mendez Aspe). ) It is virtually
inconceivable by this stage that Negrin was still hoping against hope
for a diplomatic - or other - explosion on the wider European scene
29 Negrin
to save the Republic. was now focusing intently on one thing:
how to make Franco publicly guarantee that he would take no reprisals
against the defeated population. Prolonging resistance was about main-
taining pressure to this end. All of this chimes with comments Negrin
had made in 1938: ‘[njegotiate? But what about the poor soldier in
. . .

530
Medellin? By which he meant what would happen to ordinary sol-
diers on the farthest edges of the Republican front who could not get
out? Once Catalonia had fallen, prolonging centre-south zone resistance
was supremely about creating the dme and space to structure a staged
retreat, maintaining Republican control of air facilities and, above all,

of the key Mediterranean ports long enough to allow the evacuation

of those personnel most at risk of reprisals. Such a strategy would also
have allowed time to train and implant guerrilla units whose logic was
also evident to a prime minister who believed that coming — and
in the
inescapable - European conflagration, the Republic would still have
everything to play for.

It was - in which the PCE would neces-

precisely this strategic vision
sarily have a crucial role - that reinforced Negrin’s ethical rejection of the
one form of intra-organisational ‘politics’ now rising like scum to the sur-
face of the becalmed centre-south zone. This was an anti-communism
of diverse and contradictory origins that favoured ‘sacrificing’ the PCE

27
For memoir assessments see Cordon, Trayectoria, p. 470; Vega Diaz, ‘El ultimo dla de Negrin en
Espana’, pp. 61-3. Negrin himself commented in 1945, ‘If I hadn’t gone back at that point, I

would have no self-respect today, I simply could not have lived with myself (‘no hubiera podido
sobrevivir al asco de mi mismo’), speech in the Palacio de Bellas Artes, Mexico, 1 August 1945,
Docwnentos politicos para la historia de la Republica Espanola, p. 26.
28
Zugazagoitia, Guerra y vicisitudes, The returnees: Uribe (PCE), Moix (PSUC Tomas
pp. 542-4. ,

Bilbao (ANV), Gomez (PSOE), Gonzllez Pena (PSOE/UGT), Segundo Blanco (CNT).
29 Certainly this was the Politburo view, PCE archive microfilm XX, 238, frame 95.
30 ‘^Pactar? Pero ... ^y el pobre soldado de Medellin?’, Vazquez Ocana, Pasiony muerte de la segunda
Republica espanola p. 62. Medellin, in Extremadura, fell to the rebels at the end of July 1938,
,

Chaves Palacio, La guerra civil en Extremadura, p. 250.

400 The Spanish Republic at war

on the ofa supposed reconciliation with Franco. Such an idea had

altar
already been crystallising in the autumn of 1938 when it was articulated
by Besteiro to the PSOE executive. But it posed a far greater danger
in the power vacuum of the centre-south zone in February 1939. For
there was now nothing to contain either the explosive organisational
rivalries between socialists and communists or the desire of some CNT
leaders to avenge their post- 193 7 political marginalisation. These sen-
timents animated David Antona in Ciudad Real, while in the Madrid
region Eduardo Val made an offer of unilateral support to Casado. The
CNT’s support for Casado can be traced back to its organisational ri-
valry with the PCE on the Madrid Defence Council in 1936-7. 31 More
generally, these sectarian tendencies in 1939, as in autumn 1936, were
symptomatic of the collapse of all political centres. But now such ten-
dencies were also propitiated by the international environment, with the
virtual disintegration of the French Popular Front alliance and the ex-
clusion from it of the PCF. Nor, on the threshold of defeat, did many see
any valid reason to hold off any longer from a brutal setding of political
scores. Everywhere in the centre-south zone the opportunism charac-
teristic of the old, ingrained clientelist politics exacerbated tensions and

held up a mirror that reflected the image of defeat.

The widespread feeling that the war was almost over fuelled these
sentiments in the centre-south zone. Morale there was — it is true — less
uniformly poor than it had been in Catalonia after Munich and the Ebro
retreat. The centre-south zone troops had not experienced the crushing
defeat in their own flesh and blood. And some munitions and light arms
could still be turned out in Valencia, Albacete, Alicante and Murcia.
But there were no professional army officers - not even those belonging to the
PCE- who still believed that sustained military resistance was possible. 32
The Ebro defeat, even at one remove, had corroded Negrin’s credibility.
Among the civilian population too, material privations had weighed in-
creasingly heavily since April 1938, when the zone had been cut off from
Catalonia - and thus the French frontier. And in this context of extreme
political isolation, the most effective annihilator of civilian morale was
the bombing raids: thus the population of the port cities of the Levante
coast was now experiencing what Barcelona’s had earlier in 193 8. 33 In
this environment, Falangists and other fifth columnists were growing

in confidence and political daring. Some had established themselves as

31
See chapter 3 above. 32 Togliatti, Escritos sobre la guerra de Espafia, 269-70.
pp.
33 Valencia, Gandia and Alicante. De les bombes a I’exili (exhibition catalogue) (Gandia: CEIC Alfons
el Veil, 2001).
 The collapse of the Republican homefront 401

linksbetween Franco’s military intelligence and key military and political

figures on the Republican side - most notably Julian Besteiro. 34
Negrin sought to stem the rising tide of hostility to the PCE with a
powerful and determined speech in Madrid on 12 February in which he
spelled out the realities of the Republic’s situation: in essence, ‘all of us or
none’ will be saved. 35 But the power of conviction was remained all that
to Negrin by way of a weapon to disarm his opponents. Lacking virtually
any governmental or administrative support, by February 1939 Negrin
was already a prime minister without the means of exercising power.
This governmental void did not, of itself, pose problems in terms of
day-to-day survival. The declaration of a state of war had, as we have
seen, formally transferred power to the military authorities. The real
danger lay by mid February, there was no consensus
in the fact that,
over resistance policy among the Republic’s operational military com-
manders. On 16 February Negrin met with them at the Los Llanos aero-
drome near Albacete. Present at the meeting were Miaja and Matallana,
along with the acting chiefs of the airforce (Camacho) and navy (Buiza),
the head of the Cartagena naval base (Bernal) and the commanders
of the armies of the Centre, Levante, Extremadura and Andalusia —
respectively Casado, Menendez, Escobar and Moriones. At this point
only Casado and Matallana openly declared further resistance to be im-
possible. But most were pessimistic. Only Casado was so far an active
political dissident. 36 (By 5 February he had established direct contact
with the Madrid fifth column. On the 14th he had met with represen-
tatives of Izquierda Republicana to ask them to go to Paris to persuade
Azana to return and sack Negrin.) But Miaja, Matallana and Menendez
all knew of Casado’s activities, and Moriones suspected. What made

this situation particularly precarious was the fact that none of these of-

ficers could see the point of even attempting further military resistance.
Indeed, Admiral Buiza’s threats to withdraw the an end to the
fleet if

war had not been agreed by 4 March must surely haverung alarm bells
for Negrin. 37 After all, a crucial part of his own raison d’etre was an orderly
34 For Besteiro ’s contacts with Madrid’s column and the capitulationist ‘underground’ in the
fifth

capital, see Cervera, Madrid en La ciudad clandestina igg 6-iggg, chapters 7 and 12. These
guerra.
contacts dated back to 1938: for a summary, see Preston, Comrades, pp. 180-3.
35 ‘O todos nos salvamos, o todos nos hundimos en la extermination y el oprobio’, El Socialists
(Madrid), 14 Feb. 1939.
36 Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana, pp. 265-9, 314. Bernal had
refused involvement in Cartagena, ibid., p. 430.
37 The government had SIM intelligence on fifth column activity and the generally poor political
environment/morale in Cartagena: Bahamonde Magro and Cervera Gil, Asi termino la guerra de
Espana, pp. 427-9.
402 The Spanish Republic at war

evacuation of^at-risk groups - something which required access to the

defensive capacity of the fleet 38 .

Negrin opted for one last-ditch attempt to convince the assembled

military personnel of the absoluteneed to hold on until Franco agreed
to takeno reprisals against the defeated. In retrospect, one might be
tempted to argue that Negrin should have taken the initiative after the
Los Llanos meeting and promoted the PGE to all the key army and
defence posts. The
party’s core, ideologically convinced commanders
(men such Modesto and Enrique Lister) and the generation of
as Juan
PCE mandos who had come through militia school, were still committed
to functional resistance - not least in order to organise a guerrilla force
and the party apparatus for clandestinity. But Negrin did not do this -
afterLos Llanos or ever even though he was aware of Casado’s disloy-
:

and of increasing fifth column activity in Madrid. Negrin stood by

alty,

because - even leaving aside all other political objections - he knew

thatsummoning the PCE would have been entirely counter-productive.
By resorting to the one military holding action that might have bought
him some time, Negrin would have provoked the very political collapse
he was trying to stave off. Given the rising tide of anti-war feeling now
manifesting itself as anti-communism, Negrin would have been opening
the floodgates and sweeping away the very thing he was desperate to
protect- the wherewithal to prolong resistance in order to ensure space
and time for an orderly withdrawal and evacuation.
Negrin’s apparent political paralysis in the crucial second half of Febru-
ary was, then, a function of the broader circumstances. He was forced to
take agamble that the loyalty of professional officers like Casado would
outweigh their scepticism over his depiction of what was at stake in a
peace ‘without guarantees’. Negrin also had to deal, simultaneously, with
escalating criticism from Azana in Paris. The President’s complaint that
Negrin was not doing enough to obtain the humanitarian intervention of
Britain and France seems particularly ill made. Britain was demanding
a clear statement of rendition from the Republican government before it
would even deign to raise the issue of guarantees and evacuation with
Franco. But Negrin understood what virtually no one else seemed to -
that once the Republic had liquidated the resistance, it would also have
liquidated any chance of wresting guarantees from Franco.

38 The remnants of the fleet - 3 cruisers, 12 destroyers and a submarine - did not have the capacity
to carry out the evacuation - but they would be vital to protect evacuation boats against the
rebel destroyers patrolling the access to the Republic’s Mediterranean ports.
 The collapse of the Republican home front 403

From his base in the Paris embassy, Azana attempted to intervene

personally with the British via the Republican ambassador in London,
Pablo Azcarate. 39 But the British availed themselves of Negrin’s under-
standable discretion to claim they that had no official interlocutor with
whom to discuss such matters and thus washed their hands of the issue.

Britain was increasingly reluctant to complicate its relations with Franco.

In this it was no doubt influenced by the uncompromising tenor of the re-
cently published Law of Political Responsibilities. Azana was ever more
disheartened. He made no secret of his desire to resign the presidency.
But he held off - hoping against hope that Negrin would achieve a
last-minute diplomatic breakthrough.
Bythe same token, however, Azana refused to take the one action
that Negrin knew was essential if there was to be any last-minute break-
through — namely that, as titular head of the Republic, Azana should
return to the centre-south zone. Every day that Azana remained away
crucially strengthened the French and British arguments for the official
recognition of Franco. But in spite of the frantic efforts of both Alvarez
del Vayo and (an increasingly exasperated) Pascua (ambassador in Paris
since June 1938), Azana would refuse to move until the 26th, when he left
not for Spain but for the south of France - to announce his resignation
°
as Republican president the following day. 4
Throughout the latter part of February Negrin remained sunk in a
deep solitude. This was the consequence of the sheer weight of political
responsibility that by now he was bearing virtually alone. 41 His insistence
on maintaining strategic resistance had since December 1938 put such
a strain on his relations with the republican groups and many in the
PSOE that they had effectively withdrawn active and practical political
support. But nor could the PCE make good the shortfall. For in spite
of Negrin’s speech of 12 February, his relationship with the Communist
Party was also increasingly tense.
For the prime minister the PCE, like the PSOE before it, was a tool,
to be used to the utmost as such without any thought for what that might
39 Spanish sources for this are to be foundArchivo de Negrin, in the archive of the foreign
in the
affairs ministry, Madrid - in own report and misc. correspondence n Jan.
particular Azcarate ’s
to 8 Mar. 1939, (14). Also Moradiellos, ‘Una mision casi imposible: la embajada de
caja RE: 150
Pablo de Azcarate en Londres 1936-1939’, pp. 140. There are also French diplomatic reports
in the Quai d’Orsay which I have not consulted: see Bahamonde Magro and Cervera Gil, Asi
termino la guerra de Espana, p. 442.
40 A critical account of Azana’s behaviour during this period is in Pascua’s unpublished article,
‘Azana en embajada de Paris’, AHN/MP, caja 1 (9).
la
41
On this point see, Vega Diaz, ‘El ultimo dia de Negrin en Espana’, pp. 61-3.
404 The Spanish Republic at war

mean for the liability of the party. But the fall of Catalonia - which had
concentrated Negrin’s own mind on peace conditions - had also capsized
the Spanish communists’ carefully promoted image of ‘strength’, expos-
ing its To make matters worse, while Togliatti was
very real weaknesses.
still in France - after the loss of Catalonia - elements in the party contin-

ued to issue intransigent public statements about iron resistance ‘to the
last’. These Comintern
reflected a current of opinion, identified with the
adviser Stepanov, but present in the Politburo and the wider party, which
saw Negrin’s declaration of a state of war as offering the chance to go
for communist vanguardism and a ‘Numantian resistance’. 42 Exalted
speeches were made by the party’s Madrid leadership at its provincial
conference between 9 and 11 February. 43 In these one can glimpse the
unresolved ideological debate of an earlier period: the united workers’
front (frente unico) versus the inter-class alliance of Popular Front. 44 For
the partisans of ‘Numantia’ tended to be those, such as Pasionaria, who
in1938 had expressed disquiet at what they saw as the neutering of
communism by the requirements of wartime Popular Front.
But it may well be that this exaltation had a strategic purpose too. In
the mould of the PSOE left’s own verbal radicalism of 1934-6, designed
to stay the hand of the military and political right, the PCE’s exaltados
of February 1939 may have been trying to warn off the would-be con-
spirators whose plotting in Madrid was by this point an open secret. But
as with the socialist left, PCE - its strategy was myopic
so too with the
and dangerous because it provoked an enemy which it had no effective
means of countering. For even if it had been prepared to, the PCE was
simply not in position to ‘take power’ in what remained of Republican
Spain after the fall of Catalonia. Far from producing a nascent com-
munist dictatorship, the military take-over of the centre-south zone had
exposed the fragility of the PCE - on both the military and the home
fronts.
The PCE had made Catalonia the focus of its efforts, and the Army of
the Ebro had come to symbolise its resistance strategy. But the Ebro army
had been defeated and Catalonia had fallen. At the same time, a great
many of the PCE’s wartime recruits, in both the military and civilian
spheres, were present in party ranks for a range of broadly opportunistic
reasons - whether economic defence, political protection or individual

42 The fortress of Numantia, near Soria, was the scene of a desperate resistance of Spanish indige-
nous peoples to the Roman conquerors.
43 Togliatti, Escritos sobre laguerra de Espana,
p. 275
44 Elorza and Bizcarrondo, Queridos camaradas,
pp. 428-30.
 The collapse of the Republican home front 405

career advancement. At the very best one can say that the PCE’s bur-
geoning base had for a time reflected the general feeling that the party
represented the Republic’s best hope for winning the war. Precisely be-
cause this was the nature of a great deal of wartime ‘communism’ in
Spain and because it had since become evident that the PCE could not
secure military victory, the party’s reputation was now mortally wounded
in what remained of Republican territory. In the last analysis, the rhetor-
ical ‘Numantia’ of some communist leaders in Madrid has thus to be

read as an admission of weakness. It also further increased the PCE’s

isolation, as the party became a magnet for the anti-war sentiment of
sectors of the political class and of the weary population at large.
Moreover, Negrin read the PCE’s protestations of intransigent resis-
tance as a declaration of autonomous party policy, and thus as a form of
insubordination to himself. On 16 February an exasperated premier de-
clared to the PCE’s cabinet minister, Vicente Uribe, that he would ‘shoot
all the communists’. 45 Fortunately Togliatti’s return from France to the

centre-south zone the same day was able to reinforce less exalted opinion
in the Politburo and thus succeeded in mending relations with the prime
minister. Under Togliatti’s influence, the PCE leadership introduced a
crucial nuance into its manifesto of 26 February. For the first time the
PCE publicly referred to the idea of ‘ending the war’. The function of re-
sistance was now delineated as a means of ensuring Negrin’s three points
for peace from the Figueras parliament - and most crucially a guarantee
of no reprisals. Togliatti had brought the PCE back to a position of fully
supporting Negrin. 46
But this scarcely diminished the gravity of the situation facing Togliatti
and the PCE leadership: how to respond to the evident conspiracy in the
face of Negrin’s own paralysis and the underlying weakness of the party’s
position? By late February even the ideological communists among the
military command had come to realise that serious military resistance
was no longer - given the psychological environment which ob-
feasible
tained. Togliatti repeatedly sought advice from Moscow. In his telegram
of 27 February he identified Casado as the danger. But a reply never
came - or at least was never received. 47 Although holding off Franco’s
entry into Madrid still coincided with Comintern interests, the Politburo
was entirely on its own, as events accelerated. Moreover, the immense

45 Ibid.,
p.431.
46
Indeed, Negrin himself made it clear to Togliatti what the PCE needed to say, ibid., pp. 432-3;
Bolloten, SCW, p. 711.
47 Elorza and Bizcarrondo, Queridos camaradas , p. 434.
 ,

406 The Spanish Republic at war

pressures on die leaders,, the increasing difficulties of communication

inside the zone and the growing fears among rank-and-file members
of becoming a sacrifical offering for Franco meant that party structures
were themselves breaking up.
The PGE in Madrid was largely thrown back on its own resources.
The growing hostility towards the party there awakened a keen sense of
the need to implement measures of self-defence to defend party resources
and indeed lives. These measures were also crucial to ensure that cadres
could be prepared for clandestinity. There was also a growing awareness
that the size of the party - that is, its post- 1 8 July cohort - now posed a
threat to its very survival. For many of these new communists were clearly
not to be trusted in the crucial task of planning protective measures
against the threat of an anti-Negrin rebellion. This is surely part of
the reason why neither the PCE’s central committee nor its Madrid
provincial committee were, as a whole apprised of the party’s self-defence
,

plans. Instead these were worked out by trusted individuals - such as

Isidoro Dieguez and Domingo Giron of the Madrid party leadership,
who liaised with hand-picked communist commanders and commissars
among the Madrid forces. 48
This caution also points up another crucial facet of the PGE’s dilemma
with ‘new’ communists. By this stage there was a feeling that the party
could not afford to rely on the many professional army officers who had
joined the party in early wartime in very different circumstances. This
problem had its remote origins in the conclusions drawn by many of these
officers after the April 1938 division of Republican territory. By February

1939 the PCE had no confidence that professional officers would follow
instructions just because they held a party card. 49 Again, the PGE’s sup-
posed strength - or rather the hybridity that had once been its strength in
different circumstances - was actually now a weakness. It makes no sense,
then, to measure the PCE’s strength by the number of party cardhold-
ers in Madrid’s military command. The hesitations and prevarications
of so many ‘communist’ commanders in the initial phase of the now
imminent Casado rebellion reinforce this crucial point. In Madrid and
the other key cities of the centre-south zone many communist-affiliated
officers in positions of command would choose not to oppose Casado while ,

some would join him. 50 Nor was the PCE - from Politburo to provincial
48 Bahamonde
Magro and Cervera Gil, Asi termino la guerra de Espana, pp. 366-71.
49 Togliatti, Escritos sobre la guerra de Espana
pp. 269-70.
50 Bahamonde Magro and Cervera
Gil, Asi termino la guerra de Espana, pp. 377-8, 383, 386-9, 412,
416-18, 420-1; Taguena, Testimonio de dos guerras, p. 321.
 ,

The collapse of the Republican homefront 407

leaderships - anywhere in possession of decent intelligence about the

mechanics of the Casado coup. This itself raises questions about the na-
ture of ‘communist’ influence in the SIM. If the PCE had been even
half as powerful in the armed forces as is frequently suggested in numer-
ous memoirs, then Casado and his fellow conspirators would not have
made it to the starting line. The fact that the story runs differently should
long ago have alerted us to the need to re-examine the received wisdom
about PCE ‘control’ of the Republican army. 51 All the more so, given that
within the standard narrative of the Casado events, references to PCE
‘hegemony’ have always been accompanied, somewhat problematically,
by descriptions of the party’s ‘paralysis’.

THE CASADO COUP

What Casado ’s long-planned action, however, was the
finally triggered
need to pre-empt Negrin. But what Casado was pre-empting when he
announced the Madrid Defence Council late on the evening of 5 March
was not, as the received wisdom long had it, a wholesale promotion of
communist commanders. This only ever existed as a post hoc justifica-
tion of Casado’s own invention. What Casado had to act to forestall
was Negrin’s bid to remove him and other ‘defeatist’ officers from op-
erational command by kicking them upstairs - out of harm’s way - to
the general staff. 52 This removal was the logical concomitant of their
public declarations at the Los Llanos meeting that no further resistance
was possible. The fact that only Matallana’s reassignment was published
in the official bulletin of the defence ministry for 3 March 53 may have
been an attempt to stay Casado’s hand. Negrin repeatedly insisted on
51
Bolloten, SCW, p. 600. Azana himself complained about this, Obras completas vol. 4, p. 883.
But as I have already suggested, he had no alternative policy with which to replace Negrin’s.
Complaining about the PCE was, ultimately, a means of expressing frustration about the cul-de-
sac in which the Republic found itself.
52
Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana, pp. 364-5 (see also pp. 338-
47). As the authors point out, the infamous blanket of PCE promotions that Casado claimed
had figured in the official bulletin of the defence ministry on 3 and 4 March 1939 never in fact
appeared. Rather it was the bulletin’s 3 March Matallana and Miaja
stipulations as they affected
that explain the timing of the coup (the dissolution of the Republican army group, reducing
Miaja’s post to a figurehead, and Matallana’s appointment to the general staff, where he would
effectively be sandwiched between loyal operational commanders and Negrin himself as defence
minister). As Michael Alpert also indicates, Casado’s claims for the scale of Negrin’s ‘communist
coup’ grew significantly between the first version of his memoirs, The Last Days ofMadrid (London,
r
939), and the second, Asi cayo Madrid (Madrid, 1968) - Alpert, El ejercito republicano, p. 290. Also
Bolloten, SCW, pp. 713-16.
53 Bolloten, SCW, pp. 713-14; Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana,

P- 34 i-
408 The Spanish Republic at war

4March that jllasado and Matallana should travel from Madrid to see
him at his headquarters outside Alicante, in the village of Elda. 54 Casado
declined. But once he had done so, he had no choice but to act. He almost
certainly did so in the deluded belief that he was increasing the Repub-
lic’s chances of achieving a negotiated settlement with Franco. The fifth

column had been assiduous rumour that Franco was

in spreading the
prepared to treat with professional army officers on the Republican side.
Casado and others like him may even at some level have seen this of-
fer, perversely, as a recognition of their status, which they felt had been

threatened by the rise during the civil war of a new kind of officer of
militia origin. 55 Whatever the case, in the oxygen-deficient atmosphere
of those times, Casado was convinced that his own military credentials
plus his preparedness to sacrifice the PCE as the mythical ‘author of the
Republic’s woes’ would be sufficient to appease theenemy and guarantee
some form of national reconciliation thereafter.
But before this, Casado also wanted to ensure an orderly military with-
drawal and the evacuation of the most politically compromised sectors
of Republican personnel. He was to be thwarted, however, by events
at the Cartagena naval base. On the evening of 4 March, as Casado
was in the final stages of preparing the defence council in Madrid,
a confused double rebellion of pro-Casado and pro-Franco supporters
exploded in the south-eastern port city. 56 The government had received
SIM intelligence reports of the likelihood of trouble at Cartagena. But
although Republican forces did eventually manage to regain control, in
the initial panic of whatwas an immensely uncertain situation, Admiral
Buiza ordered the put to sea. Whether because he believed that
fleet to

the pro-Franco forces were gaining control in the port or because he was
looking for a way out of what he saw as an impossible situation, whoever
ended up in control of Cartagena - Buiza ordered the fleet to set sail for

54 This was a country house, Villa Poblet, denominated in war code ‘posicion Yuste’.
55 A. Bahamonde Magro and Cervera Gil allude to this tension, Asi termino la guerra de Espana,
pp. 351, 354. Also suggestive is the fact that in Madrid the most resolutely anti-Casado comman-
der was Ascanio (a Mayor deMilicias), while Ortega and Barcelo, as professional officers who were
also communists, demonstrated much more ambiguous attitudes. To my knowledge, however,
there has been no systematic investigation of these latent ‘class wars’ in the Republicanarmy
between professional officers from the pre-war period and those who had risen to command
positions through the wartime militia schools. It would be interesting to know how important
this factor was in fuelling hostility between the CNT and PCE: see Paz, Durruti en la revolucion

espanola, pp. 726-7. On Ascanio, see Alpert, El ejercito republicano, pp. 356, 361.
56 L. Romero, Desastre en Cartagena (Barcelona: Ariel, 1971), passim] Zugazagoitia, Guerray vicisitudes,

pp. 559-66; Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana pp. 421-38.
,
 The collapse of the Republican homefront 409

Algiers. would be interned by the French at Bizerta and later

There it

delivered to Franco. (France, along with Britain, had recognised Franco’s

government on 27 February.) Without the fleet the Republic had no hope
of implementing any real or effective evacuation.
Meanwhile in Madrid, Casado was assembling the civilian con-
stituents of his defence council. However, it is important to realise that
the real strength of Casado ’s position derived from Negrin’s formal dec-
laration of a state of war in January. After the fall of Catalonia the only
articulated structures in the centre-south zone were, anyway, military
ones. But, additionally, under the terms of the Republic’s public-order
legislation of 1933, of war also formally bestowed on the military
a state
supreme and undivided political authority. This legislation was even more
favourable to the military than its predecessor of 1870, which had at least
required civilian representatives to be included on the ruling council of
war. 57 Casado’s own inclusion of civilians in the Defence Council was
part and parcel of his bid to anchor its legitimacy and appeal to Franco in
the ‘anti-communism’ that united sectors of the military and civilian po-
litical leaderships in the centre-south zone, while also giving the council
a ready message with which to address the zone’s war-weary inhabitants.

REBELLION AND RESISTANCE IN MADRID (6-13 MARCH I939)

The Madrid PCE’s opposition to the Casado rebellion was made in the
name of the legitimate Republican government under Juan Negrin. But
it was the party’s isolation and sense of being besieged in the city which
explains why this took the form of an armed response. 58 The plans for
this were sketchy, poorly laid and chiefly geared to self-protection. The
objective was to create a bargaining position from which to negotiate
terms with the Defence Council in such matters as guaranteeing the in-
tegrity of PCE organisations and lives. But the difficulties of resistance
were immediately apparent. First, because of the swift detention by pro-
Casado forces of Domingo Giron, the linch pin of the party’s defence

57 Ballbe, Orden publico y militarismo en la Espafia constitucional, pp. 361-2. By dint of this, however,
Casado’s actions cannot really be seen as direcdy mirroring those of the military rebels in July
1936. In the latter case it was the rebels themselves who declared the full state of war - albeit
assisted by the Republic’s public-order legislation, which contravened its own constitution: see
Ballbe, p. 13.
58 An up-to-date summary of the Madrid fighting (based on important internal PCE documenta-
tion) is in Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espafia, pp. 379-404. See
also Graham, Socialism and War, pp. 231-44, which uses both PCE and PSOE reports.
410 The Spanish Republic at war

plan. Even more crucially, many ‘communist’ military commanders now

refused to take action against Gasado. Three of the four Madrid army
corps were headed by professional officers with party cards: Barcelo (I),

Bueno (II) and Ortega (III). But in spite of desperate exhortations from
the provincial party leadership, all argued that they could respond only
to the orders of their military superiors. 59 Under pressure, Bueno and
Barcelo would eventually participate reluctandy against Casado - but
Barcelo only because of the pragmatic case made to him by his fellow
officer Ascanio, who pointed out the increasing untenability of Barcelo’s
own position.
There was fierce fighting in Madrid on 6 and 8 March from which the
PCE’s forces appeared to emerge with the upper hand. But the illusory
nature of this advantage is immediately apparent on contemplation of
the wider military and political context in the centre-south zone. Inside
the Madrid area the communists had already mobilised all their dis-
posable resources, whereas Casado still had an ace to play: the so far
unused IV army corps commanded by the CNT’s Cipriano Mera. 6 °
The Madrid PCE was also isolated by the virtual absence of communist
resistance elsewhere in the centre-south zone. But what really capsized
the morale of the communists in the capital was the message received
from the Politburo on 9 March that they should end the fighting and
reach a modus vivendi with Casado because the priority now was to pre-
pare party cadres for clandestinity. 61 The message came from Togliatti
and Checa in Albacete - and it was very much their own message. The
62
centre-south zone had been incomunicado since the rising. Togliatti had
remained behind (very much against earlier Comintern advice) to help

59 This emerges clearly from internal PCE documentation: see Bahamonde Magro and Cervera
Gil, Asi termino la guerra de Espafia, pp. 379, 387, 417. See also Togliatti, report of 21 May 1939,
Escritos sobre la guerra de Espafia, p. 227.
60
Mera’s units were still on the outskirts of Madrid. It was thus socialist-commanded units that had

been crucial Casado in the first days of rebellion. Mera himself supported Casado, although
to
the CNT had, somewhat unrealistically, discussed the formation of an all-anarcho-syndicalist
Defence Council.
61
Togliatti, report of 21 May 1939, in Escritos sobre la guerra de Espafia, pp. 290-1; Elorza and
Bizcarrondo, Queridos camaradas, p. 436; Bahamonde Magro and Cervera Gil, Asi termino la guerra
de Espafia, p. 392 (based on internalPCE report by Jacinto Barrios: see p. 367). The message
came by emissary and telegraph were uncertain. One of the emissaries (there
since telephone
were two) was probably a Valencian communist called Fernando Montoliu, Togliatti, p. 297 and
F. F. Montiel, Un coronel llamado Casado (Madrid: Criterio, 1998), pp. 224-5. Montiel was one of

two high-profile transfugas from the PSOE to the PCE at the beginning of 1937 (the other being
Margarita Nelken). During the war Montiel worked for government press relations. In 1939 he
was in the PCE in Madrid — for a time in charge of radio communications, J. Garcia Pradas,
Como termino la guerra de Espafia (Buenos Aires: Editorial Iman, 1940), p. 31.
62
Falcon, Asalto a los cielos , pp. 175-6; Elorza and Bizcarrondo, Queridos camaradas, p. 434.
 1

The collapse of the Republican homefront 41

Checa and Fernando Claudln. 63 The rest of the Politburo left Spain
in the very early hours of 7 March, in the wake of the departure of
Negrin and his government ministers, as Casado’s forces were closing in
on his headquarters in Elda. 64 The knowledge that Madrid’s resistance
was effectively at odds with party policy caused consternation among
provincial leaders like Isidoro Dieguez. Moreover, Negrin’s departure,
which they also now learned of, immediately deprived the resistance of
its larger political justification.
There is no doubt that Negrin’s departure from Spain on 6 March
was the act which collapsed the shell of Republican resistance - its legit-
imacy already undermined by Azana’s resignation of 27 February and
the ensuing vacuum in the presidency 65 But in the end Negrin had been
presented with little choice — other than to go down with the ship. His
Elda headquarters were, geographically, in a Casadista vice. Had he
allowed himself to be taken prisoner, he might well not have survived.
Much has been conjectured about Negrin’s existential despair during his
final hours in Spain after the Casado rebellion. Certainly Negrin was
physically and emotionally exhausted. 66 But what has often been ex-
trapolated from this seems unjustified. Given what we know of Negrin,
there is no reason to suppose that he was psychologically intimidated
by the prospect that he might not survive the war. When he had re-
turned to Spain in February, it had been to negotiate a staged hand-over
of power. He had also assumed that procuring guarantees for the Repub-
lican population would demand that he became the propitiatory victim
of Francoist justice. 67 But for Negrin a sacrifice had to be an intelli-
gent sacrifice - that is one that stood a chance of attaining its purpose.
Self-immolation post-Casado in March 1939 would not have achieved
63 Elorza and Bizcarrondo, Queridos camaradas, p. 434. Irene Falcon attributes it to his close emotional
PCE as well as a hard-headed understanding of the need to plan for
identification with the
clandestinity, Asalto a los debs, pp. 176—8.
64 The balance of military power in the Levante army (discussed below) meant that no military
defence was open to the PCE (as both Modesto and Lister recognised), Togliatti, Escritos sobre b
guerra de Espana, p. 290. The PCE’s HQwas a ramshackle country house - code-named ‘posicion
Dakar’ - in the village of Elda.
65 W. Carrillo, El ultimo episodb de b guerra dvil espanob (Toulouse: n.p., 1945), p. 9. The formal
recognition of Franco by Britain and France on 27 February finally precipitated Azana, who
resigned hours later. He should immediately have been replaced by the vice-president, Diego
Martinez Barrio. But he refused to step in unless Negrin gave him full powers to end the war.
66
Falcon, Asalto a bs debs, pp. 171 —5; Vega Diaz, ‘El ultimo dia de Negrin en Espana’, pp. 61-2.
The author sketches a compelling picture of how Negrin’s immense capacity of will momentarily
succumbed in exhaustion at Elda to the ‘last straw’ of Casado’s game playing. Negrin is reported
to have said that the Colonel was ‘leading [him] like a torero with a bull’ (‘me esta toreando’).
67
Zugazagoitia, Guerra y vidsitudes, p. 541 and cf. p. 383; Negrin in speech, 1 Aug. 1945, Palacio de
Bellas Artes, Mexico, Documents politicos para b historia de b Republica Espanob, p. 26.
412 The Spanish Republic at war

a peace with guarantees; whereas, as Negrin read the political moment

then, the Republic had everything to play for in the European con-
still

flagration that loomed. So more than any personal concerns and, still
less, cowardice, what made Negrin choose survival over self-sacrifice was

his sense that there was still some purpose — a patriotic purpose — to his
political leadership.
That and the PCE did not see matters this way is, of course,
Togliatti
also understandable. Even before the political agendas of the postwar
period complicated communist assessments of Negrin, the PCE was
scarcely likely to be enamoured of a premier who, while relying on
party support on the Republican home front, retreated ever more into
himself as, across the last six months of the war, he had pursued his secret
diplomacy in search of a negotiated peace. 68 Moreover, there is also no
doubt an element of truth in Togliatti’s assessment that the Casado take-
gave Negrin personally an honourable way out of
over, as a fait accompli,
an untenable situation, thus allowing him to move on to the next round of
the confrontation in which the terrain of battle would be all Europe. For
Togliatti, Negrin’s departure was flight pure and simple and, moreover,
one which left the cupola of the PCE high and dry. 69
But back in Madrid even the knowledge that Negrin had left Spain and
what that meant did not deter anti-Casado army officers like Ascanio,
who, in Bueno’s absence, was commanding the second army corps. They
had come too far to turn back - especially as Casado was not offering
terms but demanding capitulation. In spite of plummeting morale, then,
the communist resistance in Madrid continued because many felt that
they had no other choice. Gradually, between 9 and 12 March, Casado
gained the upper hand. This happened partly because of Francoist ac-
tion, clearly designed to assist him, on the westerly Casa de Campo front
and partly because of the entry of new troops (including Mera’s), which
was also facilitated by Franco. 70 The Republican airforce, now under
the control of the Casadista Antonio Camacho, also bombed the PCE’s

68
los cielos, pp. 17 1-2. Negrin nevertheless felt that he had contracted a debt of
Falcon, Asalto a
honour to the PCE which was manifest in the time and effort he spent in the hours after the
Casado coup procuring (as far as precarious communications would allow) the release of detained
communist leaders. See also Togliatti’s comments, Escritos sobre la guerra de Espana, p. 288.
69 Togliatti, Escritos
sobre la guerra de Espana pp. 289-90.
,

70 Franco’s
plan was to Casado to ‘clean up’ in Madrid. The Generalisimo had no interest
allow
in using the internecine war as an opportunity to breach Madrid’s defences. Nor, in any case,
had the internal fighting unmanned the fronts. Both the PCE and Casado had called mainly
on reserve troops. For more on Franco’s strategy, see Bahamonde Magro and Cervera Gil, Asi
termino la guerra de Espana, pp. 397-401.
 The collapse of the Republican homefront 413

strongholds. 71 The communist resisters effectively came to be besieged

inside the capital.
The hostilities concluded on March. But, predictably, Casado re-
13
fused to treat with the PCE. Ostensibly this was because of the assassi-
nation of a number of hostages: three of Casado’s military commanders
and a socialist political commissar. In reprisal Casado ordered the execu-
tions of Colonel Barcelo, whom he erroneously considered the architect
of the PCE’s resistance, and of the communist commissar, Conesa. 72 But
the rising in Madrid had been a war, and both sides had killed and been
killed. 73 Nor was Casado’s decision here fundamentally determined by
an impulse of revenge. Rather his whole policy of negotiating success-
fullywith Franco was predicated on the basis of dismembering the PCE.
This was to be the offering in return for which Franco would, so the
calculation went, concede a peace with guarantees for the defeated. 74
Other civilian members of the Madrid Defence Council concurred in
this tactic - most notably the veteran socialist leader Besteiro, with his
chimerical belief in the resurgence of a ‘responsible’ labour movement
under a patrician dictatorship. 75

THE CASADO EVENTS IN THE REST OF

THE CENTRE-SOUTH ZONE
Elsewhere the PCE’s position in the wake of the Casado take-over ap-
peared relatively less vulnerable. While in some places communists had

71
L. Romero, Elfinal de la guerra (Barcelona: Ariel, 1976), p. 331. Camacho took over from Hidalgo
de Cisneros, who left Spain on 7 March.
72
Of the other three communist commanders in Madrid, Ascanio and Ortega (who had declared
himself neutral and took no part in the Casado fighting) were both shot by Franco after the war.
Bueno, who fought against Casado (in much the same circumstances as Barcelo - see Bahamonde
Magro and Cervera Gil, Asl termino la guerra de Espana, p. 386), was removed from his command
(by Casado) but apparendy survived the war and the aftermath of the war too: Alpert, El ejercito
republicano, pp. 363-4.
73 Graham, Socialism p. 240. There were some 20,000 deaths as a result of the fighting.
and War,
74 For a brutal description of this gambit, see the comments made by the PSOE’s Molina Conejero,
cited in Graham, Socialism and War, p. 237. Doubdess the pressures of war and internal politi-
cal antagonisms also saw a certain internalisation of Francoist propaganda among Republican
sectors.
75 ‘Those of us who have responsibilities, especially in the Union [UGT], we have to stay. I’m sure
that nothing much will happen. We’ll have to see how things turn out, and maybe we’ll be able
to reconstruct a more moderate UGT - along the lines of the British trade unions’, Besteiro ’s
comments on 11 March 1939 to the civil governor of Murcia, Eustaquio Canas, in the latter’s
(unpublished) memoir, ‘Marzo de 1939. El ultimo mes’ (1948), p. 30 - copy in the Archivo de
Ramon Lamoneda (ARLF-172-30) FPI.
414 The Spanish Republic at war

been detained and party premises attacked, nowhere had there been sus-
tained armed confrontations as in Madrid. 76 Moreover, the active or tacit
support for Casado from many military commanders in the centre-south
zone was not predicated so much on anti-communism as on the need for
a mechanism to end the war and to deliver some kind of peace terms that
Franco was manifestly refusing to negotiate with Negrin. Above all, these
commanders were concerned to avoid a civil war within the Republican
army Precisely for this reason, the Casado take-over in Valencia had
been implemented by the police. The army commander there, General
Menendez, held his troops aloof. He was determined to protect their
esprit de corps and morale, which had been built up - irrespective of politi-

cal affiliation - through the combat experience of 1938—9. And precisely

because Menendez ’s objective was a peace with guarantees, he was also
keen keep the Popular Front formally intact in order to bolster Casado
to
in his dealingswith Franco.
In all the major Republican centres - Valencia, Alicante and
Albacete - provincial communist leaders sought an accommodation with
Casado’s forces. Once they knew of Negrin’s departure there was, any-
way, little political point in resisting the new order. But their decision
was motivated chiefly by the desire to guarantee their own safety and
adequate conditions for the future operation of their organisations. Each
provincial organisation acted on its own initiative. No real communica-
tion was possible between the local party organisations in the first days
after the coup. Nor were they able to communicate with Togliatti or
Checa until 9 March, when the Politburo representatives were released
from prison in Alicante. (They had been detained, together with Claudin,
hours after the departure of the rest of the leadership in the early hours of
7 March. ) The line instinctively followed by the provincial communist
77

parties was rapidly endorsed by Togliatti after the fact since it served to
defuse the political tension somewhat — thus buying time to prepare the
PCE for clandestinity. 78
Unfortunately, however, neither the local party negotiations nor Togli-
atti ’s efforts nor the goodwill of some sectors of PSOE and UGT could
gainsay the groundswell of popular anti-communism in the centre-south
zone. The determination of various Republican military commanders
76 The most up-to-date overview of events is in Bahamonde Magro and Cervera Gil, Ast termino la
guerra de Espana, pp. 408-21.
77 Espana p. 291; Falcon, Asalto a los cielos, p. 177.
Togliatti, Escritos sobre la guerra de ,

78 This was Togliatti’s main objective - hence his concern that Negrin should keep trying to reach
an accord with Casado right to the bitter end - i.e. even as the Casadistas were closing in on
Elda on 6 March. Falcon, Asalto a los cielos p. 176.
,
 The collapse of the Republican homefront 415

to keep the peace meant that there was no repetition of the extremes
of political violence seen in Madrid. (For example, military censorship
was used to prevent incendiary articles appearing in the press - and
in particular the anarcho-syndicalist press.) In some places it was even
possible forcommunist party organisations to regain some semblance
of a functioning life. But to speak of the ‘re-entry’ of the PCE, or the
normalisation of political life after Casado, is misleading. 79 The Popu-
lar Front was dead. Its epitaph was written in the brutal organisational
dismemberment of both the UGT and the JSU spearheaded by the an-
gry and vengeful supporters of the ex-prime minister, Francisco Largo
80
Caballero, in the weeks following Casado’s take-over. There were so-
cialist leaders who sought to prevent this unedifying settling of scores
on the threshold of defeat - men such
as the UGT’s Rodriguez Vega
or Antonio Perez, the union’s representative on the Casado Defence
Council, who cast the only vote against the death sentences on Barcelo
and Conesa. 81 But sadly they failed. Moreover, if we deconstruct the
‘anti-communism’ now driving events across the zone, their failure is

scarcely surprising.
It is important to understand that the apparently monolithic phe-
nomenon of anti-communism was a language through which many
different kinds of anger, frustration and despair were being mediated.
In terms of organisational politics, as we have already seen, anti-
communism was members between
part of a clientelist struggle for
the PCE on the one hand and republicans, socialists and the CNT
on the other. This competition was given enormous impetus and ur-
gency by the special conditions of the civil war, which had seen an ac-
celerated mobilisation of the population. Nowhere was the competition
more acute than between the ideologically similar socialist and commu-
nist movements. The outrage of the Caballerista socialists derived in part
from their conviction that the socialist movement was predestined to in-
herit both the Spanish working classes and the state - en route reabsorbing
the Communist Party, whose ‘separateness’ was a constant, painful re-
minder of the wayward and unwarranted split in Spanish socialist ranks
in 1921— 2. But the civil war had rewritten the script. The Caballeristas

79 Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana, pp. 402, 415.
80
This story is told (from PSOE primary sources and the press) in Graham, Socialism and War ;

see especially pp. 239-44. Details of PCE expulsions are in PCE archive microfilm XX, 238,
frames 134-7 (Valencia); 141-3 (Alicante); 149-50 (Ciudad Real/centre-south). For the new
PSOE appointments see F. Ferrandiz Alborz, La bestia contra Espana (Montevideo: n.p. 1951),
pp. 72-3.
81
Graham, Socialism and War, p. 242.
41 6 The Spanish Republic at war

could not acocept that this had happened, however, and consequently

went looking for what were superfluous conspiratorial explanations of

theirown marginalisation.
The choice of the PCE as a scapegoat in 1939 was not, of course, an
arbitrary one. The party had been a pro-active participant in the bruising
war of organisations behind the Republican lines and had shown itself
more than a match for its opponents. There had been many feuds and
it had many enemies. Worker constituencies were also now physically

exhausted after bearing the brunt of over thirty months of war. Some
worker sectors were also hostile to the Popular Frontist role the PCE
had played in bolstering a liberal republican order based on capital
and a respect for private property. Other - urban and rural middling -
constituencies, who were once very happy with the PCE in this respect,
now focused their anxiety and disappointment on the party because
they believed it be an obstacle to peace with Franco. This belief in
to
the possibility of a peace in reconciliation was precisely what Casado
tapped into and in turn reinforced, as had, independently and for its

own reasons, the fifth Madrid and Valencia.

column in
By the second half of 1938 popular anti-communism was thus becom-
ing part of a generalised and deepening war weariness - the product
both of a closed political horizon internationally and cumulative and
worsening material hardship (bombardment as well as hunger and short-
age intensified by the economic effects of Non-Interventionist blockade).
The PCE was targeted especially for blame by different (and sometimes
mutually politically antagonistic) sectors of a disappointed Republican
populace precisely because it had identified itself so closely with the war

effort - and more specifically with victory. The PCE had been projected,
and had projected itself, as the symbol par excellence of inter-class hopes for
Republican success. In 1936 the party had derived kudos from an intense
5
but entirely unrealistic popular belief in the ‘epic power of the Soviet
Union. By 1939 those hopes had been consumed in the unforgiving heat
of a gruelling and messy war. The resulting ‘anti-communism’ had about
it an elemental force (the casting out of a secular god?) which serves to
remind us that a crucial part of the PCE’s original popular appeal lay in
what was effectively a redemptive myth.
What existed through and after Casado, then, was not a rational po-
litical front - still less a rational political front against ‘communist dicta-

torship’ as has sometimes been suggested. Rather it was a visceral and

chaotic amalgam of political and social constituencies brought together
by a desperate war weariness and by the desire for an act of collective
 The collapse of the Republican home front 417

psychological and polidcal unburdening. Someone or something tangi-

ble had to be found to blame for the fact that the war was about to
be lost after so much and superhuman effort. In the end, the
sacrifice
meaning of the Casado episode was that the PCE, like all other forces
in Republican Spain, had ultimately been consumed by the war.
Once Casado was in full political and military control of the centre-
south zone, as he was by the middle of March, he turned his attention to
82
the peace negotiations that he saw as the raison d’etre of his victory. In the
wake of this, there had been a significant level of desertion - especially
from the Madrid front - by ordinary soldiers who felt that peace had
already arrived. Nevertheless, the Republican military fronts and home
front for the time being remained stable. But Casado knew that he had to
move rapidly to gain an assurance of no reprisals from Franco. Otherwise
he would not be able to guarantee public order in the zone. Casado had,
by his own actions, brought an end to effective Republican resistance.
But he still believed he had a certain margin for negotiation with Franco
by dint of their common anti-communism and the service he had done
5
the Generalisimo in ‘cleansing the party from the political life of the
Republic. This, so Casado hoped, would open up the way to his favoured
scenario: the reunification of Spain via a reconciliation of its military
family. For whatever reasons, Casado chose to ignore the awkard detail
that by ending the Republic’s resistance he had also removed the only
reason Franco might ever have had to concede a peace with terms.
Also resolutely ignoring this, Republican political organisations across
the centre-south zone entered into a frenzy of reformation, making ready
for a ‘normality that would never come. 83 The pro-Casado press fed
5

this groundless mass optimism, which clearly originated in a mechanism

of psychological self-preservation. Only the PCE was exempt from the

collective delusion by reason of its particular political situation. This
would be reinforced by the Defence Council’s bringing to trial of six-
teen front-rank communist military commanders and political leaders
on 24 March. The run-up to this did at least have the positive effect
of concentrating the party’s efforts on evacuating its leading cadres. On
the day of the communist trial, Togliatti left To tana (Cartagena) in a
plane bound for Algiers. He was accompanied by a number of party
leaders including Jesus Hernandez, Pedro Checa, Isidoro Dieguez and

82
What follows here is an analytical overview of events. For a detailed account see Bahamonde
Magro and Cervera Gil, Ad termind laguerra de Espana, pp. 439-99.
83 For example, on 24 March the Madrid press published details of the new PSOE executive elected
in the city. This and other examples are in Graham, Socialism and War, pp. 239-43.
 ,

418 The Spanish Republic at war

Jose Antonio TJribes. 84 Over the subsequent four days the PCE also
managed to evacuate other leading cadres (in small boats and the few
merchant vessels that would take refugees) from the main Republican
ports of Cartagena, Valencia, Alicante and Almeria. Precisely because
the PCE had been made brutally aware of the necessity of such an evacu-
ation earlier than any other Republican group, it managed the operation
relatively better. But neither this nor the PCE’s relatively superior prepa-
rations for clandestinity could prevent thousands of communists from
being caught up, along with those of other political affiliations or none,
in the grim last chapter of Republican defeat.
For Franco’s own reasons for facilitating Casado’s take-over were, of
course, quite different from Casado’s interpretation of them. Franco had
previously let it be known
(including via the fifth column) that he would
only negotiate with fellow army officers. He had also issued a text on
5 February detailing certain written ‘concessions’ which, again, reached
Casado via the fifth column. 85 But what neither Casado nor his civilian
supporters seemed to grasp (in spite of the publication of the Law of
Political Responsibilities) was that these concessions were being wielded

as an instrument of war in order to accelerate the end of the conflict

rather than offered as a basis for reconciliation between victors and
86
vanquished. From the start Franco would demand of Casado uncon-
ditional surrender. For this reason, Franco refused to treat with Casado
himself or other officers of the Republican general staff, insisting that
those delegated to attend the joint meeting requested by the Republicans
should be of subordinate rank. As far as Franco was concerned, they were
not attending as interlocutors but simply to receive practical instructions
for handing over the Republic’s troops and territory.

This message was brutally reinforced by Franco’s representatives at

the meeting, which took place on 23 March at the Gamonal air base in
Burgos. They refused to enter into any discussion with their Republican
counterparts 87 about either formal guarantees for the defeated or any

84 Togliatti, Escritos sobre laguerra de Espana, They had to combat Casado’s forces to get into
p. 297.
the aerodrome. Dieguez had escaped Casado by disguising himself as an ordinary soldier on the
sierra front: Romero, Elfinal de laguerra pp. 367-8.
85 These included the provision of safeconducts for those wishing to leave Spain and the statement
that simple political support for the Republic would be excluded from the definition of what was
‘criminal’.
86
Note that these ‘concessions’, which Franco refused to discuss with Casado, would again be
broadcast on the eve of the Republican surrender in order to encourage the compliance of
officers and troops, Bahamonde Magro and Cervera Gil, Asi termino laguerra de Espana, p. 462.
87
The Republican delegates were Lt Colonel Antonio Garijo Hernandez and Major Leopoldo
Ortega Nieto. Garijo was certainly very sympathetic to the opposing camp, if not actively an
agent, as the PCE would subsequently claim.
 The collapse of the Republican homefront 419

modification of the terms of the surrender. Casado had dispatched his

representatives in the hope that it would still be possible to agree a
staggered hand-over of territory in order to allow a slow, staged retreat
of the Republican armies. But it was clear that Franco would accept no
delay.
The outcome of the meeting of 23 March made Casado fully aware
of his powerlessness. He was also painfully cognisant of his total isolation
in this predicament. Britain had looked favourably on his actions thus
88
far. But now that Franco had definitively won, courtesy of those actions,
British policy makers were determined to do nothing that might offend
Spain’s new ruler. The key to Britain’s inaction was a determination
to protect its own position in the Mediterranean by ensuring Francoist
Spain’s neutrality in any future European war. But the political and
social animosity felt by the British establishment towards the Spanish
Republicans also played a part here. For this reason there was absolutely
no possibility that the British government would respond to Casado ’s
desperate plea that it exert meaningful pressure on Franco to offer, at this
late stage, some guarantee of no reprisals against the civilian population.
Nor would it allow the Royal Navy to assist in any evacuation. Britain’s
negative in both regards was made ringingly clear. The essentials of the
position evinced when Britain had recognised Franco on 27 February
basically held throughout March - even though by 5 March the Republic
had lost its own navy. The British authorities would do nothing without
Franco’s express approval. 89 As reasons, they adduced the following. There
was probably no need to engage in a mass evacuation anyway. The
Royal Navy did not want to be involved. 90 There was nowhere for such
a large number of refugees to go. (Britain was disposed to admit only a
reduced number of the economically solvent.) In short, as Lord Halifax
publicly claimed on 9 March, Britain’s stance was eminently justifiable
on the grounds of ‘political realism’. But there was also an unspoken
assumption here: that even if Franco were to carry out mass executions
of Republicans, these would pose no danger to British commercial or
political interests.

88
There is no doubt that the British were by February 1939 already au courant regarding Casado ’s
plans, Bahamonde Magro and Cervera 468-9.
Gil, Asi termino la guerra de Espana, pp.
89 Britain had recognised Franco unconditionally. Neither guarantees for the Republican popula-
tion nor evacuation was raised.
90 In the
early weeks of the war the Royal Navy had, entirely on its own account, taken off affluent
and powerful refugees threatened by the popular violence that had followed in the wake of
Republican state collapse. But in 1939 it chose to argue that it had no remit to ‘intervene’. The
issue was clearly not ‘intervention’ per se, but the fact that the naval high command and officer
corps felt detatched from, when not actively hostile to, the social and political complexion of
those they were now being asked to save.
420 The Spanish Republic at war

The Franco and the ensuing Casado events

British recognition of
also saw Stalin move
away from viewing collective security as a
further
viable strategic defence of Soviet frontiers. In his speech to the XVIII
Congress of the CPSU, inaugurated on io March, he had alluded publicly
for the first time to the possibility of seeking a modus vivendi with Nazi
Germany. 91
Casado was caught, as Negrin had been before him, between the
rock of international isolation and the hard place of Francoist intransi-
gence. On 25 March,
Republican representatives met again with
as the
their Francoist counterparts, Casado wrote a personal letter to Franco
in one final attempt to secure a written guarantee of no reprisals. He
appealed now for the sake of a victorious Francoist public order, arguing
that without such a guarantee there was no certainty that the orders
of the Republican military command would be obeyed - there might
be desperate resistance in some places, with the zone descending into
chaos. Franco’s only response announce the commencement of
was to
his final military offensive simultaneously on all fronts. To all intents
and purposes, the General had broken off contact with the Defence
Council. At this point, Casado, knowing that he had failed, and fearful
above all things of increasing disorder in the centre-south zone, had no
choice but to see through to the bitter end what he had started. Speed
was of the essence. On 26 March the Republican Airforce was hastily
surrendered to Franco. On 28 March Casado issued the order to the
Republican armies to surrender on all fronts. 92 In Madrid some soldiers
took the metro home, others the road east towards the Mediterranean
ports. In some places there was fraternisation between the troops. In the
end, Madrid - capital of Republican resistance and, for a time, of world
anti-fascism - did not have to be captured. Franco’s troops walked in un-
opposed. Inside the city was the extremely well-organised fifth column
it

which oversaw the mechanics of the, largely orderly, hand-over of power.

Indeed, the fact that the fifth column was so well prepared and ubiqui-
tous (including in Casado ’s own staff, in the person of Jose Centano de la
Paz) further problematises the notion of ‘communist military hegemony’
in Republican Spain. 93
But the relative peace of Madrid, as well as being a peace of ex-
haustion, was also in part a function of the maelstrom having moved

91
In fact previous feelers had been put out by Soviet diplomats in Berlin - although so far to no avail.
92 That is, in the south, centre and Levante. For a summary of Franco’s southern advance, see
Bahamonde Magro and Cervera Gil, Asi ter mind la guerra de Espaha , pp. 488-9.
93 Ibid.,
pp. 454, 463.
 1

The collapse of the Republican home front 42

elsewhere. As the news of Franco’s unconditional demands and looming

final offensive became known, the strange becalmed days which in many
places had followed Casado’s take-over gave way to the eruption of mass
panic as the illusion of a peace with guarantees vanished. On 27 March
tens of thousands of desperate Spaniards, in a frantic and disordered
surge of humankind, made for the Republic’s remaining Mediterranean
ports - Valencia, Gandia and Alicante. But they went in search of boats
that would never arrive. 94 Negrin’s resistance policy had for a long time
logically precluded any open preparations for a mass evacuation. And
although the fall of Catalonia changed that,
provoked a simultane-
it also
ous administrative meltdown in government which once again precluded
much effective planning. Nor by this stage could funds readily be accessed
or channelled. 95
was the British diplomatic staff in Valencia and Gandia who, in-
It

tensely alarmed by the flood of refugees and aware that the Royal Navy
would not intervene, sought to channel it to Alicante with vague promises
that there merchant ships awaited. Among those said to be arriving were
vessels chartered on behalf of the French-led International Coordinat-
ing Committee for Information and Assistance to Republican Spain.
(Its French representatives were particularly active in these evacuation

attempts and had chartered boats belonging to the French Communist

Party’s own company, France Navigation.) The British consular officials
were far from happy at this threatened contravention of their own gov-
ernment’s policy of no evacuation without the approval of Burgos. But at
least it had the advantage of removing the refugees from their immediate
vicinity. Moreover, it also allowed the discreet and orderly embarcation

in Gandia of Casado and his immediate circle, for whom the British had
intervened with the Generalisimo. 96 Franco had facilitated the colonel’s

94 Even before the crisis of 27 March, the few merchant vessels in Republican ports that were
prepared to embark refugees were already demanding payment in currencies other than the
Republican peseta (which was by this stage virtually worthless). But foreign exchange was beyond
the reach of many refugees.
95 The battle between the Republic’s two rival exile funding bodies, the Spanish Republican
Evacuation Service (Servicio de Evacuacion de Republicanos Espanoles (SERE) and the
Committee of Assistance to Spanish Republicans (Junta de Auxilio a los Republicanos Espanoles
(JARE)), would be the first of many bitter conflicts that would poison the political life of the
Republican diaspora. Thomas, Spanish Civil War pp. 920-1;
,
P. W. Fagen, Exiles and Citizens. Spanish
Republicans in Mexico (Austin: University of Texas, 1973 ), passim and L. Stein, Beyond Death and Exile.
The Spanish Republicans in France 1939-1955 (Cambridge, Mass.: Harv ard University Press, 1979),
pp. 87-91. On exile more broadly: J. Cuesta & B. Bermejo (eds.), Emigraciony exilio. Espanoles en
Francia 1936-1946 (Madrid: EUDEMA, 1996).
96 This episode iscovered in detail in Bahamonde Magro and Cervera Gil, Asi termino la guerra
de Espaha, pp. 469-70, 472-3, 474, 475-87. As the authors point out, however, this agreement
422 The Spanish Republic at war

departure from Madrid 9 ? and journey to the port, and - in spite of is-

suing a subsequent formal diplomatic protest - was clearly entirely in

agreement that Casado should depart on a British naval vessel. (The fifth
column in Valencia was more or less in control of the city by 29 March, yet
allowed Casado to pass through unmolested.) Indeed, without Franco’s
approval it is clear that the British would not have taken him off from
Gandia. 98 Only those so approved left the port on British vessels. Mean-
while, further down the coast, Alicante had become the territory of last
resort for beleaguered Republicans.
Reduced to its essentials,
the tragedy of Alicante was the result of three
Franco did not want any of the refugees to leave
things. First, the fact that
Spain and was maintaining his naval blockade of the Mediterranean to
ensure that they did not. Second, that the Republic no longer had a
navy of its own. Third, that Britain would not risk Franco’s displeasure
by using the Royal Navy either to perform large-scale evacuations or
to protect the various merchant ships that hovered on the outskirts of
the port, outside Spanish territorial waters, intimidated by the Francoist
warships. 99Nor was the French government prepared to offer any pro-
tection for own merchant vessels in Spanish waters, in spite of the
its

desperate last-minute appeals from the French solidarity delegates who

had chartered them. Had the Republic still had its own navy, it could
have fulfilled this protective function. But in the absence of this, most
of the merchant boats simply turned around and, sometimes in sight of
the refugees, sailed away empty. Even if all these boats had got through,
however, they could not have evacuated everyone who wanted to leave.
For there were somewhere between 12,000 and 15,000 people crammed
onto the quayside at Aicante. Had the boats got through, however, well

did not extend to other leading members of the Defence Council such as General Miaja or the
anarcho-syndicalist leader Cipriano Mera.
97 The Defence Junta (except Miaja, who had already left) made its last broadcast from Madrid -
with the approval of the Francoist authorities - on the evening of 27 March. Casado appealed
for calm and clemency.
98 The British government argued that this was the logical result of its recognition of Franco’s
legitimacy on 27 February. But it was also an eery replication of the military rebels’ own inversion
of legality in July 1936 whereby the Republicans themselves became ‘the insurgents’: see British
cabinet minutes cited in Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana ,

479.
p.
99 The British-owned merchant ship the Stanbrook took off about 2,000 refugees on 28 March.

(Fligher figures (up to 3,200) are sometimes given. But these seem unlikely as the ship was
less than 1,000 tons.) The exact circumstances in which these refugees were accepted remain
unclear. Maybe was bribed; maybe he was sympathetic to the refugees’ plight. But it
the master
is was following company policy. The British government assumed that Franco
unlikely that he
would overlook the activities of the occasional merchant ship.
 The collapse of the Republican home front 423

over half could have been rescued. As it was, it has been estimated that
in the entire month of March the total number of evacuations from the
centre-south zone was between 7,000 and 7,500. But this figure includes

the 3,200 naval officers and ratings who reached Bizerta. 100 So it seems
unlikely that more than around 4,000 refugees managed to get out of
Alicante port at the end.
The column gained control in Alicante before the arrival of the
fifth

first Francoist forces. These troops were Italians commanded by Gen-

eral Gambara. The committee representing the refugees in the port
attempted to secure his agreement that the port area be considered a
neutral zone under international (consular) protection until the evacua-
101
tion could be effected. But Franco, unsurprisingly, refused to accept
this. The rapid arrival by boat of more Spanish troops saw the gradual
evacuation of the port asmany refugees left landwards. By the night of
31 March about 2,000 remained on the quayside. This was the night of
the Spanish Republic’s last stand. Some opted for suicide. For the rest
the concentration camps awaited. 102

THE WAR AFTER THE WAR

From the start Franco had demanded of Casado unconditional surrender.
Indeed, the Generalisimo even choreographed the final stages to rein-
force the symbolism of absolute victory. All contacts with the Defence
Council were broken off on 25 March in order for Franco to initiate his
final offensive. Yet there was no military necessity for this. In the event
there was little Republican resistance. But even in risking further blood-
shed (including on his own side) Franco was effectively making clear his
subsequent political intentions.

100
Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espana, p. 499. I am grateful to
Michael Alpert for providing figures on the Republican Navy.
101
This committee included the socialists Carlos Rubiera and Gomez Ossorio, former and (in
March 1939) current civil governors of Madrid respectively, and both members of the pro-
Casado PSOE executive elected in Madrid during the last days. Both would be shot by Franco.
Graham, Socialism and War, p. 242.
102
There a growing literature on both concentration camps and Francoist prisons. See, for
is

example, F. Moreno Gomez, Cordoba en laposguerra (la represiony la guerrilla, 1939-1950) (Cordoba:
Francisco Baena Editor, 1987); V
Gabarda, Els afusellaments al Pals Valencia (1938-1956) (Valencia:
Edicions Alfons el Magnanim, 1993); M. Nunez Diaz-Balart and A. Rojas Friend, Consejo de guerra.
Los fusilamientos en el Madrid de la posguerra (1939-1945) (Madrid: Compania Literaiia, 1997);
T. Cuevas, Prison of Women: Testimonies of War and Resistance in Spain 1939-1975 (Albany, n.y.:
SUNY, 1998); Richards, A Time of Silence', C. Mir, Vivir es sobrevivir. Justicia, ordeny marginacion en
la Cataluna rural de posguerra (Lleida: Milenio, 2000); R. Vinyes, ‘“Nada os pertenece. .” Las .

presas de Barcelona, 1939-1945’, Historia Social, 39 (2001), 49-66.

424 The Spanish Republic at war

By ending the war through military action rather than parley, Franco
reinforced the point made by his consistent refusal to offer any remotely
defined guarantees to the defeated. 103 The Republicans would not be
collectively recognised in any way as a legitimate interlocutor since this
could be perceived as giving them rights or claims. Unconditional victory,
not reconciliation, was Franco’s aim. The new Spain would be made in
the image of the victors and only of the victors. But nor had Franco been
prepared to let the defeated go. He had blocked all evacuation attempts -
except in the specific case of Casado. If Franco would not recognise the
defeated or let them leave, then clearly he can only have had in mind
blanket punishment - or ‘penitence’, as it was termed in the regime’s
104
own discourse. What this meant, however, is the beginning of another
story, and one that locates 1940s Spain in a broader European context
of ongoing civil war.
These European civil wars, as this book began by outlining, had
erupted in the aftermath of the Great War of 1914-18 in the heat of
unresolved tensions and anxieties stemming from the vast processes of
industrialisationand urbanisation occurring in uneven fashion across
the European continent. Exacerbated by the inter-war crises of econ-
omy, polity and socio-cultural identity, these civil wars would run on
into the overarching conflict of 1939-45. In Spain too, there would be
ongoing irregular warfare throughout the 1940s as Republican guerril-
lasopposed the regime in a conscious effort to merge with the broader
struggle being fought across Europe against the Nazi new order. 105
The wars would exact a terrible price. For
‘resolution’ of all these civil
the reconstruction of societyand polity and the remaking of European
‘nations’ would happen, in Spain and elsewhere, through large-scale exe-
cutions and the mass imprisonment of compatriots. In short, the recasting
of power across Europe occurred through the creation of categories of
the anti-nation and of non-persons deprived of civil rights. Precisely
how (and when) ‘the other’ was constituted would depend on the
103
For all Negrin’s aspirations, however, nor would a formal set of written guarantees have assured
the rights of the defeated. In a very similar situation after the Greek Civil War, the conservative
victors agreed an amnesty. But they then manipulated its conditions in order to criminalise
virtually the entire political opposition. For the Greek case - which closely parallels the Spanish
experience - see M. Mazower (ed.), After the War was Over: Reconstructing the Family, Nation and State
in Greece 1943-1960 (Princeton, n j Princeton University Press, 2000).
. .:

104
That this punishment was intended to be blanket rather than exemplary can also be inferred
from Franco’s lack of interest in Casado’s attempts to seek the extradition of Negrin to Spain:
Bahamonde Magro and Cervera Gil, Asi termino la guerra de Espaha, p. 456.
105
There is a massively expanding literature on the guerrilla. See most recendy F. Moreno Gomez,
La resistencia armada contra Franco. Tragedia del maquisy la guerrilla (Barcelona: Critica, 2001).
 The collapse of the Republican homefront 425

specific historical andBut whether Jew, Slav, enemy of

cultural location.
the people, communist, Republican, urban worker, or the all-embracing
Francoist epithet, ‘red’, the function and effects would be the same. So
when Franco issued his famous dispatch of 1 April 1939, declaring the
war to be over, 106 in fact only one phase - that of conventional military
conflict - had ended. The Spanish Civil War itself still had years, indeed
107
decades, to run.

106
‘On this day, with the capture and disarming of the Red Army, the National troops have achieved
their final military objectives. The war is over. Burgos, April 1939 - Year of Victory.’ (‘En
i

el dia de hoy, cautivo y desarmado el Ejercito rojo, han alcanzado las tropas nacionales sus

ultimos objetivos militares. La guerra ha terminado. Burgos 1 de abril de 1939 - Ano de la

Victoria.’) Franco’s final war dispatch, published in A B C (Seville), 2 April 1939 and reproduced
in Diaz-Plaja, La guerra de Espaha en sus documents, p. 509.
107
It is obviously highly problematic to extrapolate the nature of any post-victory Republican
state, given that this moves us into the realm of counterfactual speculation. We do know that
the Republic represented a set of liberal and constitutional political values - indeed an ethic -
which opposed the principles and values of Francoist Spain. However imperfect Republican
liberalism (as I have pointed out, Spanish democracy was in its infancy when it was forced
to fight a total war), this ethic was real, not a ‘fagade’. Nevertheless, we cannot extrapolate
an entire Republican practice from such ideals and values. Conservative commentators have
long tended to suggest - either implicidy or explicidy - that the big question is how severe the
Republic’s ‘inverse’ repression of Spain’s affluent classes and elite groups would have been. But
the supposition that such repression would occur does not square with the judicial history of the
wartime Republic. The Tribunal de Responsabilidades Civiles was also relatively inactive, and
the Negrin government refused to dismiss state employees purely on grounds of their political
opinions or past political affiliations (referred to in chapter 6 above). Moreover, as this account
has implied, the violence of the wartime Republican state (in the form of imprisonment and work

brigades, but also work/ factory discipline) tended to be far more in evidence as an instrument
in the reconstruction of political, economic, social and cultural hierarchies. It was directed at
controlling the Republican masses - not elite groups or ‘conservative Spaniards’. In this respect
at least there were some common concerns with the Franco regime (as indicated, for example, in
chapter 5 above, p. 283, n. 120) - even though Republican
constitutionality always supposed a
crucial qualitative difference. There
an interesting (anonymous) attempt to wrestle with these
is

questions in the Azana material lodged in the Archivo de Barcelona. It is entitled ‘Utopico
plan de gobierno’ and is contained in an envelope marked ‘particular y reservado para S.E.
Divulgaciones de un loco’ (n.d.) in Archivo Particular de Manuel Azana (Apartado 7) caja 133,
carpeta 16.
I think that the key question with regard to a post-war Republican state is rather different

from the one usually posed. Although it is hard to answer, what we have to ask is how Republican
post-war reconstruction policies would in practice have affected Spain’s urban and rural popular
classes and, in particular, working-class sectors. How much of the integral liberal project (or, to
put it more basically, how much welfare spending) could have been afforded by the Republic?
For without it (or with little of it) the highest material cost of reconstruction would have been

borne, as was under Franco, by the urban and rural working classes. This, in turn, raises the
it

question of how popular belief in, or commitment to, a Republican social contract could have
survived the harsh economic consequences of such a peace. In other words, while the Republic’s
survival as a democracy would have depended on its finding a means of mediating relatively
harsh economic policies, the Franco regime never faced this problem. In overtly excluding
huge numbers of urban and rural workers (as the ‘defeated’, the ‘reds’ or ‘anti-Spain’) from its
definition of the national community, it acquired an ideological justification for their economic
exploitation in the name of ‘national rebirth’.
 Glossary

Only terms used repeatedly in the text are included here. Others are
explained when they arise, either in the main text or in the rele-
vant footnote. An asterisk indicates that the term also features in the
glossary.

Africanistas - the officers who commanded Spain’s colonial troops in

Morocco
ANV - Accion Nacionalista Vasca, a non-confessional and politically
liberal minority that split off from the PNV* at the end of 1930
Association of Anti-fascist Women (AMA) - one of the mass Popular
Frontist organisations promoted by the Spanish Communist Party
(PCE*)

Bloc Obrer Camperol (BOC) (Workers and Peasants Bloc) (see also
i

POUM*) — a leninist party based in Catalonia and with a strong

Catalan nationalist component

CADCI (Centre Autonomista de Dependents del Corner^ i de la

Industria) - Barcelona shop and office workers’ union

Carabineros — customs and frontier police
Carlists — originally a rival monarchist faction in the early nineteenth
century, Carlism was a brand of Catholic traditionalism attracting
mainly rural Spaniards who wanted to abolish the dominant develop-
ments of the modern age which they saw as enshrined in the Second
Republic. Based in Navarre, the Carlists opposed the Republic from
the start and supported the military rising in July 1936, contributing
militia fighters known as the requetes*
CEDA (Confederacion Espanola de Derechas Autonomas/Spanish
Confederation of Right Wing Groups) — a nationwide mass Catholic
party formed in 1933

426
 Glossary 427

cenetista - member of the CNT*

checa - a private prison. The different political groups (but especially
PCE* and CNT*) set these up in the coup-induced aftermath of
Republican state collapse in July 1936. As state power was recon-
structed, so these were eradicated - although not completely until
after the moments of high political tension in May 1937 when some
private interrogation centres reappeared, usually under the auspices
of the PCE*
CNT (Confederacion Nacional del Trabajo) - anarcho-syndicalist
labour union founded in 1910
Comintern - the Communist (or Third) International

Ertzaina - Basque police

esparto lista - adjective referring to a vehemently centralist Spanish na-
tionalism exalting all things Castilian
Esquerra Republicana de Catalunya (ERC) - Catalan Republican Left
(a coalition which included Estat Catala* and the Rabassaires*)
Estat Catala - radical Catalan nationalist party

FAI (Federation Anarquista Iberica) - anarchist federation founded in

1927 with the ostensible aim of defending core anarchist principles in

the CNT*
faista- member of the FAI*
Falange - Spanish Fascist Party
fifth column - the name given engaging in pro-rebel activ-
to those
ity (i.e. spying and various forms of sabotage) within the Republican
zone
Fifth Regiment (Quinto Regimiento) — the military unit of the Spanish
Communist Party (PCE*), which served as a training unit and model
of militarised discipline for the Republican militia
FNTT - socialist landworkers’ federation and a constituent union of the
UGT*
FOUS - POUM* trade union that joined the UGT* in 1936

Generalitat — Catalan regional government established by the autonomy

statute of 1932
GEPCI (Gremis i Entitats de Petits Comerciants i Industrials) - UGT-
affiliated Federation of Small Traders and Manufacturers, founded by
the PSUC* at the start of the war.
Gudari(s) — Basque soldier(s)
428 Glossary

incontrolables —* literally the ‘uncontrollable ones’ - a term coined to

describe the perpetrators of killings and violent acts in the period of
Republican state paralysis following the military coup. Implicit in the
use of the term is the notion that ‘revolutionary’ violence was often
a cover for all manner of less exalted motivations, including common
crimes
Izquierda Comunista - the small, dissident Left Communist party led by
Andreu Nin which amalgamated with the BOC* in September 1935
to form the POUM*
Izquierda Republicana - Left Republican party (led by Manuel Azana)

Juventudes Libertarias ( JJ LL) the anarcho-syndicalist youth organisa-

tion
Juventudes Socialistas Unificadas ( JSU) - the united socialist-communist
youth federation initiated in April 1936 and by November 1936 (with
the siege of Madrid) entirely in the political orbit of the Spanish Com-
munist Party (PCE*)

ley de fugas - extra-judicial murder carried out by the Spanish police

against those in custody and particularly prevalent in Barcelona during
1920—2 when General Martinez Anido was Civil Governor. Prisoners
would be and then shot while ‘attempting to escape’
‘released’
libertarian movement (libertarians) - alternative term used to describe
the anarcho-syndicalist movement (CNT*, FAI* andJJ LL*)

mando unico - single political-military command which the wartime

Republican coalition wanted to achieve in emulation of the rebels the
better to prosecute the war, but also as a reflection of certain political
preferences which sought the reconstruction of the liberal state
MAOC (Milicias Antifascistas Obreras y Campesinas) - the pre-war
militia of the Spanish Communist Party (PCE*). It constituted the
nucleus of the Fifth Regiment*
miliciana — militiawoman
miliciano - militiaman

NKVD - Soviet intelligence service

oligarchy - state ruled (directly or indirectly) by a small, exclusive group,
or, by derivation, this group elite was constituted
itself. In Spain this
by a senior partner — the large landowners - and by a junior one - the
industrial bourgeoisie
 Glossary 429

ORGA - Galician nationalist and republican party

OVRA - political police /intelligence service of Fascist Italy
paseo - the extra-judicial killings carried out in the aftermath of the July
1936 coup (literally, the ‘stroll’ taken to death)
patria chica (literally, ‘the little country’) - village or neighbourhood, the
lived unit of experience to which loyalty is primarily owed/felt
PCE - the official Spanish Communist Party, affiliated to the Communist
International (Comintern*)
PNV - Basque Nationalist Party
POUM - dissident Communist Party formed in September 1935 by the
merger of BOC* with the leninist groupuscule, Izquierda Comunista*.
The POUM’s base was overwhelmingly Catalan and constituted by
BOC cadres
poumista — member of the POUM*
PSOE - Spanish Socialist Party, founded in 1879
PSUC - United Socialist Party of Catalonia, formed in July 1936
from the merger of four small parties of which the most important
was the social democratic and Catalanist USC (Unio Socialista de
Catalunya)

Rabassaires - Catalan tenant farmers’ and smallholders’ association in-

corporated into the Esquerra* and the special power base of Esquerra
leader, Luis Companys
regulares — indigenous troops from Spanish Morocco, commanded by
Spanish officers (the Africanistas*)
Republican(s) - denotes all those who supported the Republic during the
civil war of 1936—9
republican(s) - denotes members of specifically republican parties and
groups
requetes - Carlist* militia

saca - the removal from gaol and extra-judicial murder of prisoners

SIM (Servicio de Investigacion Militar) — Republican military police/
intelligence service established in August 1937
STV - Basque Nationalist trade union (recruiting from ethnic Basque
urban workers)

Tercio de Extranjeros (tercios) — Spanish Foreign Legion

transfuga - someone who abandons one political organisation for another
(more emotively, a political renegade)
43° Glossary

treintismo/treintistas - moderate syndicalist sectors (from Catalonia and

the Valencia region) which split from the CNT in 193 1-2 because
of opposition to what was seen as the damage being done to the
organisation by the violent direct action of the FAI*
Tribunales Populares - the Popular Courts established by the Republic
in August 1936 in order to curtail extra-judicial violence and as a
stepping stone to the reconstruction of liberal juridical norms

ugetista — member of the UGT*

UGT (Union General de Trabaj adores) - socialist-led trade union
founded in 1888, traditionally strongest in Madrid and in the Asturian
mining and Basque industrial zones
Union Republicana (Republican Union) - a centrist republican party
formed in 1934 when the more liberal wing split from Lerroux’s Rad-
ical Party over its collusion with the CEDA*

Workers’ Alliance (Alianza Obrera) - a failed initiative (of the BOC*) in

1934 to create a nationwide alliance of trade unions and left political

parties
 permission)

1850

kind

since
his

with

development

reproduced

urban

and

Rider

Barcelona:

Nick

3
by
Map

drawn

(Plan
 1937

March

territory

Spanish

of

division

The

4
Map
1938

July

territory

Spanish

of

division

The

5
Map
 Bibliography

PRIMARY SOURCES
ARCHIVES
Archivo Historico Nacional (Madrid) (archives of Marcelino Pascua and Luis
Araquistain) (AHN/MP) (AHN/LA)
Archivo Historico Nacional (Seccion Guerra Civil) (Salamanca) (AHN-SGC)
Archivo del Ministerio de Asuntos Exteriores (Archivo de Barcelona) (AB)
Archivo del PCE
Fundacion Pablo Iglesias (FPI) (Madrid) (Archivo Historico de Moscu,
(PSOE/UGT)) (AHM)
Public Records Office
Quaker Archive (Friends Service Council (Spain), Friends House, London)
(FSC/R/Sp)

MICROFILMED ARCHIVAL MATERIAL

Southworth Collection (University of California)
Blodgett Collection

NEWSPAPERS
Adelante (Valencia)
Claridad (Madrid)
La Correspondencia de Valencia
Frente Rojo (Valencia/Barcelona)
Mundo Obrero (Madrid)
El Socialista (Madrid and Barcelona)
Solidaridad Obrera (Barcelona)
Spain and the World (London)
Treball (Barcelona)
La Vanguardia (Barcelona)

434
 Bibliography 435
In facsimile

Milicia Popular (Diario del Quinto Regimiento) (fascsimile edition, Barcelona: Edi-
torial Hacer, 1977) (comprises 15 1 issues from 26 July 1936 to 24 January
1937 when the Quinto Regimiento was incorporated in the new Republi-
can Army under construction)

SECONDARY SOURCES
UNPUBLISHED DOCTORAL THESES
Ealham, C., ‘Policing the Recession: Unemployment, Social Protest and Law-
and-Order in Barcelona 1930-1936’ (University of London, 1995)
Gomez, M. T., ‘El largo viaje/The Long Journey: The Cultural Politics of
the Communist Party of Spain 1920—1939’ (McGill University, Montreal,
’999)
Montero, E., ‘The Forging of the Second Spanish Republic: New Liberalism,
the Republican Movement and the Quest for Modernisation 1898-1931’
(University of London, 1989)
Schauff, F., ‘Sowjetunion, Kommunistische Internationale und Spanischer
Biirgerkrieg 1936—1939’ (University of Cologne, 2000)

CONTEMPORARY PRINTED SOURCES/CONTEMPORARY MEMOIRS

Abad de Santillan, D., Por que perdimos la guerra, Buenos Aires: Editorial Iman,
I 94°
Acier, M. (ed.), From Spanish Trenches: Recent Lettersfrom Spain , London: The Cresset
Press, 1937
Aguirre, J. A., Veinte ahos de gestion del Gobierno Vasco (igg6igg6), Paris, 1956
Alonso, B., La flota republicanay la guerra civil de Espaha. Memorias de su Comisario
general, Mexico: n.p., 1944
Alvarez del Vayo, J., Freedom’s Battle, London: Heinemann, 1940
Bahamonde, A., Un aho con Queipo : memorias de un nacionalista Barcelona: Edi-
,

ciones Espanolas, n.d. [1938]

Bertran Giiell, F., Preparacion y desarrollo del alzamiento , Valladolid: Libreria
Santaren, 1939
Buckley, H., Life and Death of the Spanish Republic London: ,
Hamish Hamilton,
1940
Bullejos, J., Europa entre dos guerras igi8-ig4y, Mexico: n.p., 1945
Canel, J. (pseud.), Octubre rojo en Asturias , Madrid: Agencia General de Libreria
y Artes Graficas, 1935
Carrillo, W., El ultimo episodio de la guerra civil espahola , Toulouse: Secretaria de
PublicacionesJSE, 1945
‘En torno al tragico fin de nuestra guerra’, Timon (Buenos Aires), Nov. 1939
,

Casado, S. The Last Days ofMadrid, London: Peter Davies, 1939

Cowles, V, Lookingfor Trouble, London: Hamish Hamilton, 1941
436 Bibliography

Cox, G., DefencwfMadrid, London: Victor Gollancz, 1937

Cuevas (Colonel), Recuerdos de la Guerra de Espaha, Montauban: n.p., 1940
Delapree, L., Mort en Espagne Paris: Eds Pierre Tisne, 1937
,

Documentos politicos para la historia de la republica espahola , Mexico City: Coleccion

Malaga, 1945
Dominguez, E., Los vencedores deNegrin, Mexico: Nuestro Pueblo, 1940
Epistolario Prieto y Negrin. Puntos de vista sobre el desarrolloy consecuencias de la guerra

civil Imprimerie Nouvelle, 1939

espahola , Paris:
Fidalgo, P., A in Franco's Prisons London: United Editorial, 1939
Young Mother ,

Fischer, L., Men and Politics London: Jonathan Cape, 1941

,

Foxa, A. de, Madrid de corte a cheka, San Sebastian: Libreria Internacional, 1938
Franco in Barcelona London: United Editorial, 1939 (the (anonymous) author was
,

Muriel McDiarmid, a Quaker relief worker)

Gallndez, Jesus de, Los vascos en el Madrid sitiado, Buenos Aires: Ed. Ekin, 1945
Garcia Pradas, J., Como termino la guerra de Espaha Buenos Aires: Editorial Iman,
,

1940
Garrachon Cuesta, A., De Africa a Cadiz y de Cadiz cl la Espaha Imperial, Cadiz:
n.p.,1938
Gerahty, C., The Road to Madrid, London: Hutchinson, 1937
Gorkin, J., Canibales politicos, Mexico, 1941
Hanighen, F. C., Nothing but Danger, New York: NTC, 1939
Jellinek, F., The Civil War in
London: Victor Gollancz, 1938
Spain,
Jordan, R, There Is No Return, London: Cresset Press, 1938
Knickerbocker, H. R., The Siege of the Alcazar, London: Hutchinson, 1937
Knoblaugh, H. E., Correspondent in Spain, London/New York: Sheed & Ward,
1937
London: Victor Gollancz, 1937
Koestler, A., Spanish Testament,
Largo Caballero, Francisco, La UGTy la guerra. Discurso en el cine Pardihas de

Madrid, Valencia, n.p., 1937

Lizarra, A. de, Los vascos y la Republica espahola, Buenos Aires: Ekin, 1944
Lojendio, L. M. de, Operaciones militares de la guerra de Espaha 1936-1939,
Barcelona: Montaner y Simon, 1940
Lopez Fernandez, A., Defensa de Madrid, Mexico: Editorial A. P. Marquez, 1945
Lunn, A., Spanish Rehearsal, London: Hutchinson, 1937
Martin Blazquez, J., I Helped to Build an Army, London: Seeker and Warburg,
1939
Matthews, H. L., Two Wars and More to Come, New York, Carrick & Evans, 1938
Maurin, J., Hacia la segunda revolucion, Barcelona: n.p., 1935
Mola, E., Obras completas, Valladolid: Libreria Santaren, 1940
Mora, C. de la, In Place of Splendor. The Autobiography of a Spanish Woman,
New York: Harcourt, Brace & Co., 1939
Moron, G., Politica de ayery politica de mahana, Mexico: n.p., 1942
Nunez Morgado, A., Los sucesos de Espaha vistos por un diplomatic, Buenos Aires,
i94i
Paul, E., The Life and Death of a Spanish Town, New York: Random House, 1937
 ,

Bibliography 437

Prieto, I., Como y por que sail del Ministerio de Defensa Nacional. Intrigas de los rusos en

Espana, Paris: n.p., 1939

Discursos en America, Mexico: Ediciones de la Federation dejuventudes Socia-
de Espana, n.d. [1944]
listas

Ramon Laca, J. de, Bajo la ferula de Queipo: como fue gobernada Andaluda, Seville:
Imprenta del Diario Fe, 1939
Ruiz Vilaplana, A., Doy fe. .un aho de actuacion en la Espana nacionalista, Paris:
.

Editions Imprimerie Cooperative Etoile S.A., 1938

Sanchez del Arco, M., El sur de Espana en la reconquista de Madrid, Seville: Editorial
Sevillana, 1937
Sanz, R., Buenaventura Durruti, Toulouse: Ediciones El Frente, 1945
Schlayer, F., Diplomat in roten Madrid, Berlin, 1938
Sheean, V, Not Peace But a Sword, New York: Doubleday, Doran & Co., 1939
Solano Palacio, F., La tragedia del norte, Barcelona, 1938
Somoza Silva, L., El General Miaja. Biografia de un heroe, Mexico: Editorial Tyris,

*944
Steer, G. L., The Tree of Gernika, London: Hodder and Stoughton, 1938
Torrent Garcia, M., {Que me dice Usted de los presos?, Alcala de Henares, Talleres
Penitenciarios, 1942
Uribarri, M., La quinta columna espahola, vol. 1, Havana: n.p., 1943
El SIM de la Republica, Havana: n.p., 1943
Vazquez Ocana, F., Pasiony muerte de la segunda Republica espahola, n.p.: Editorial
Norte, 1940?
Whitaker, J. T., We Cannot Escape History, New York: Macmillan, 1943
‘Prelude to World War: A Witness from Spain’, Foreign Affairs, 21 (1) (October
1942)
Woolsey, G., Death’s Other Kingdom, London: Longmans, Green & Co., 1939
Worsley, T. C., Behind the Battle, London: Robert Hale, 1939

BOOKS AND ARTICLES (INCLUDING PROTAGONISTS’ MEMOIRS

written/published later)
AA. W, La iglesia catolicay la guerra civil espahola. Cincuenta ahos despues, Madrid:

Fundacion Friedrich Ebert/Instituto Fe y Secularidad, 1990

Abella, R., La vida cotidiana durante la guerra civil. La Espana republicana, Barcelona:
Planeta, 1975
Finales de enero igjg. Barcelona cambia depiel, Barcelona: Planeta, 1992
Abramson, P. and A., Mosaico roto, Madrid: Compania Literaria, 1994

Aguilar, P., Memoriay olvido de la guerra civil espahola Alianza, Madrid, 1996
Aguirre, J. A., De Guernica a Nueva York pasando por Berlin, Buenos Aires: Editorial
Vasca Ekin, 1944
El informe del Presidente Aguirre al gobierno de la Republica (sobre los hechos que de-
terminaron el derrumbamiento del frente del Norte), Bilbao: Editorial La Gran
Enciclopedia Vasca, 1978
Diario de Aguirre, Navarre: Txalaparta, 1998
 y

438 Bibliography

Alba, V, Historyi de la segunda republica espahola, Mexico: Libro Mex, i960

El marxismo en Espaha 1313-1333. Historia del B.O.C. y del P.O.U.M., 2 vols.,

Mexico: Costa- Amic, 1973
Dos revolucionarios: Joaquin Maurln, AndreuNin, Madrid: Seminarios y Ediciones,
1975
La revolucion espahola en la practica. Documentos del POUM, Madrid: Ediciones
Jucar, 1977
Alba, V
and Ardevol, M. (eds), El proceso del P.O.U.M. Documentos judiciales
Barcelona: Ediciones Lerna, 1989
policiales,

Alba, V, Durgan A., Gabriel, P. et al., Andreu Nin i el socialisme , Barcelona, 1998
Alba, V and Schwartz, S., Spanish Marxism Versus Soviet Communism: A History of
the POUM New Brunswick/ Oxford:
,
Transaction Books, 1988
Alcaraz, R., La Unio Socialista de Catalunya 1323-1336, Barcelona: La Magrana,

1987
Alcofar Nassaes, J. L., Los asesores sovieticos en la guerra civil espahola , Barcelona:
Dopesa, 1971
Alexander, M. and Graham, H. (eds), The French and Spanish Popular Fronts.
S.

Comparative Perspectives Cambridge: Cambridge University Press, 1989

,

Alexander, R., The Anarchists in the Spanish Civil War 2 vols., London: Janus, 1999
,

Alfaya, J. L., Como un no de fuego. Madrid 1336, Barcelona: Ediciones Interna-

cionales Universitarias, 1998
Alia Miranda, La guerra civil en la retaguardia, Ciudad Real (1336—1333), Ciudad
F.,

Real: Diputacion de Ciudad Real, 1994

Almendros, J., Situaciones espaholas: 1336/ 1333. Vivencias del quefue Secretario Militar
del PSUC, Barcelona: Dopesa, 1976
Alpert, M., La reforma militar de Azaha (1331 -1333), Madrid: Siglo XXI, 1982
La guerra civil espahol en el mar Madrid: Siglo XXI, 1987
,

El ejercito republicano en la guerra civil, 2nd edn, Madrid: Siglo XXI, 1989
A New International History of the Spanish Civil War, Basingstoke: Macmillan,
1994
‘La diplomacia inglesa y el fin de la guerra civil espanola’, Revista de Politica
Internacional (Madrid) (March-April 1975)
‘Juan Negrin e Inglaterra’, Boletin Institucion Libre deEnsehanza 24-5 (December ,

!
99 6)
Altaffaylla Kultur Taldea, Navarra 1336. De la esperanza al terror, 2 vols., Tafalla:
author-editor, 1992
Alted, A., Egido, A. and Mancebo, M. F. (eds), Manuel Azaha: pensamientoy accion,
Madrid: Alianza, 1996
Alvarez, S., Memorias II. La guerra civil de 1336/1333, A Coruna: Edicios do Castro,
1986
Osorio-Tafall. Su personalidad, su aportacion a la historia, A Coruna: Edicios do
Castro, 1992
Juan Negrin. Personalidad historica, 2 vols., Madrid: Ediciones de la Torre, 1994
Historia politica y militar de las Brigadas Internacionales, Madrid: Compania
Literaria, 1996
 , , ,

Bibliography 439
Alvarez del Vayo, J.,The Last Optimist London: Putnam & Co., 1950
Alvarez Junco, J., El emperador del Paralelo. Lerrouxy la demagogia populista Madrid: ,

Alianza, 1990
La ideologla politico del anarquismo espanol (i 868-1g 10), 2nd edn, Madrid: Siglo
XXI, 1991
Ambou, J., Los comunistas en la resistencia nacional republicana. Laguerra en Asturias ,
el

Pais Vasco y Santander, Madrid: Editorial Hispamerca, 1978

Andrade, J., Apuntes para la historia del PCE, Barcelona: Fontamara, 1979
La revolucion espanola dia a dia, Barcelona: Nueva Era, 1979
Notas sobre la guerra civil. Actuacion del POUM, Madrid: Ediciones Libertarias,
1986
Anso, M., To fui ministro deNegrin, Barcelona: Planeta, 1976
Araquistain, L., Sobre laguerra y civilMadrid: Espasa-Calpe, 1983
en la emigracion ,
Ardiaca, P, La fundacio del PSU de Catalunya Barcelona, 1986 ,

Arenillas, J. M., The Basque Country. The Nationalist Question and the Socialist Revo-
lution, Barcelona, 1937; reprinted by Leeds ILP, 1970

Arias Velasco, J., La hacienda de la Generalitat igji-igg8, Barcelona: Ariel, 1977

Arnaiz, A., Retrato hablado de Luisa Julian, Madrid: Compania Literaria, 1996
Arnal, J., Tofui secretario de Durruti. Memorias de un cura aragones en lasfilas anarquistas,

Zaragoza: Mira Editores, 1995

Arnau, R., Marxisme catala i qiiestio nacional catalana, igjo-igg6. Textos de dirigents
i militants (vol. 1); Textos de movements i partits politics (vol. 2), Paris: Edicions
Catalanes de Paris, 1974
Arostegui, J. (ed.), ‘Violencia y politica en Espana’, Ayer (Madrid), 13 (1994)
Arostegui, J., Calleja, E. G. and Souto, S., ‘La violencia politica en la Espana
XX’, Cuadernos de Historia Contemporanea, 22 (2000)
del siglo
Arostegui,J. and Martinez, J. A., La Junta deDefensa de Madrid, Madrid: Comu-
nidad de Madrid, 1984
Aviles, Farre, J., Pasion y farsa. Franceses y britanicos ante la guerra civil espanola,
Madrid: Eudema, 1994
Lafe que vino deMoscu. La revolucion bolcheviquey los espaholes igiy-igji Madrid:
Biblioteca Nueva/UNED, 1999
‘Francia y la guerra civil espanola: los llmites de una politica’, in Espacio, tiempo

y forma, Serie V, 5 (Madrid: UNED, 1992)

Aya, R., Rethinking Revolutions and Collective Violence. Studies on Concept, Theory and
Method, Amsterdam: Het Spinhuis, 1990
Azana, M., Los espanoles en guerra, 3rd edn, Barcelona: Critica, 1982
Causas de laguerra de Espana, Barcelona: Critica, 1986
Apuntes de memoriay cartas ed. Enrique Rivas, Valencia: Pre-Textos, 1990
Obras completas, vols. 3 and 4 (Madrid: Ediciones Giner, 1990)
Discursos Parlamentarios, ed. J. Paniagua Fuentes, Madrid, 1992
Azcarate, M., Derrotas y esperanzas. La Republica, la guerra civil y la resistencia,

Barcelona: Tusquets Editores, 1994

Azcarate, P. de, Mi embajada en Londres durante la guerra civil espanola, Barcelona:
Ariel, 1976
 ,

440 Bibliography

Azzuz Hakim,* M. I., La actitud de los moros ante el alzamiento. Marruecos 1336,

Malaga: Editorial Algazara, 1997

Bahamonde Magro, A. and Cervera Gil, J. Asi termino laguerra de Espaha Madrid:
, ,

Marcial Pons, 1999

Balcells, A., Crisis economica y agitacion social en Cataluna de 1330 a 1336, Barcelona:
Ariel, 1971
Trabajo industrial y organizacion obrera en la Cataluna contemporanea 1300-1336,
Barcelona: Laia, 1974
‘El socialismo en Cataluna hasta la guerra civil’, in S. Julia (ed.), El Socialismo
en las nacionalidades y regiones , Anales de Historia vol. 3, Madrid: Fundacion
,

Pablo Iglesias, 1988

Balfour, S., The End of the Spanish Empire 1838-1323, Oxford: Clarendon Press,

1997
Balfour, S., and Preston, P (eds), Spain and the Great Powers London: Routledge,
,

1999
Ballbe, M., Orden publico y militarismo en la Espaha constitucional (1812-1383),
Madrid: Alianza, 1983
Ballester, R., La batalla del Ebro, Barcelona: Bruguera, 1974
Baquero Gil, G., Laboratorio de retaguardia (Diario de laguerra en Madrid 1936-1333 ),
Madrid, 1997
Bar, A., La CNT en los ahos rojos. Del sindicalismo revolucionario al anarcosindicalismo
(1910-1926 Madrid: Akal, 1981
'),

Barber, J. and Harrison, M., The Soviet Home Front ig4 1 -1943 Harlow: Longman, ,

1991
Barcelona any zero (1939) (historia grafica de Vocupacio de la ciutat), Barcelona: Proa,

1999
Bardavio, J., Sabado Santo Rojo Madrid: Ediciones Uve, 1980
,

Barral, C., Ahos de penitencia 1975; Barcelona: Tusquets, 1990

,

Barranquero Texeira, E., Malaga entre la guerra y la posguerra, Malaga: Editorial

Arguval, 1994
Barrio Alonso, A., Anarquismoy anarcosindicalismo en Asturias (1890-1936), Madrid:
Siglo XXI, 1988
Barrull Pelegri, J., Violbicia popular ijusticia revolucionaria. El Tribunal Popular de
Lleida (1936-1937), Lleida:Pages editors, 1995
Bateman, I)., Joaquin Maurin 1833-1373. Life and Death of a Spanish Revolutionary
Leeds: Leeds ILP, 1974
Bayo, A., Mi desembarco en Mallorca 1944; Majorca: Miquel Font, 1987
,

Bedmar Gonzalez, A., Lucena: de la segunda republica a laguerra civil Cordoba: ,

n.p.,

r
9 98
Republica, guerray represion. Lucena 1331-1333 Lucena: Ayuntamiento de ,

Lucena, 2000
Ben-Ami, S., The Origins ofthe Second Republic in Spain Oxford: Oxford University
,

Press, 1978
Benavides, M. D., Guerra y revolucion en Cataluna ,
Mexico: Ediciones Roca, 1978
Benet, J., La mort del President Companys, Barcelona: Edicions 62, 1998
 , , , ,

Bibliography 44i

Bengoechea, S., Organitzacio patronal i conflictivitat social a Catalunya. Tradicioy corpo-

rativisme entrefinals del segle i la dictadura de Primo de Rivera Barcelona: l’Abadia
de Montserrat, 1994
Vuitanta-quatre dies de lock-out a Barcelona, igig-20. Els precedents de la dictadura de
Primo de Rivera Barcelona: Curial: 1998
Beramendi, J. G. and Maiz, R. (eds), Los nacionalismos en la Espana de la IIRepublica
Madrid: Siglo XXI, 1991
Bernad, E. and Forcadell, C. (eds), Historia de la Union General de Trabajadores en
Aragon.Un siglo de cultura sindicaly socialista, Zaragoza: Institucion ‘Fernando
el CatolicoYDiputacion de Zaragoza, 2000

Bernecker, W. L., Colectividades y revolucion social. El anarquismo en la guerra civil

espahola, 1936-1939 Barcelona:
,
Critica, 1982
Berneri, C., Guerra de clases en Espana 1936-1937, Barcelona: Tusquets Editores,
J
977
Biglino,P., El socialismo espaholy la cuestion agraria 1830—1336 Madrid: Ministerio ,

de Trabajo y Seguridad Social, 1986

Bizcarrondo, M., ‘Democracia y revolucion en la estrategia socialista de la
Segunda Republica’, Estudios de Historia Social, 16-17 (
j 98i)
Blanco Escola, C., Franco y Rojo. Dos generales para dosEspahas Barcelona: Editorial ,

Labor, 1993
Blanco Rodriguez, J. A., El quinto regimiento en la politica militar del P.C. E. en la guerra
civil Madrid: UNED, 1993
,

Blinkhorn, M., Carlism and Crisis in Spain 1331-1333 Cambridge: Cambridge ,

University Press, 1975

(ed.), Spain in Conflict 1331-1933 London: Sage, 1986
,

Bolloten, B., The Spanish Civil War. Revolution and Counterrevolution ,

Hemel Hemp-
stead: Harvester /Wheatsheaf, 1991
(De les) bombes a I’exili (exhibition catalogue), Gandia: CEIC Alfons el Veil, 2001
Bonamusa, F., El Bloc Obreri Camperol (1930-1932), Barcelona: Curial, 1974
Andreu Mny el movimiento comunista en Espana, Barcelona: Anagrama, 1977
Bookchin, M., The Spanish Anarchists. The Heroic Tears, New York: Free Life
Editions, 1977
Borja de Riquer, El ultimo Cambo 1936-1947 Barcelona: Grijalbo, 1997
Borkenau, F., The Spanish Cockpit, London: Faber & Faber, 1937; London: Pluto
Press, 1986
Borras, J., Aragon en la revolucion espahola, Barcelona: Cesar Viguera, 1983
Bosch Sanchez, A., Ugetistasy libertarios. Guerra y revolucion en el Pais
civil Valenciano,
I 93^~I939 •>
Valencia: Institucio Alfons el Magnanim, 1983
y dinero. El nacionalcatolicismo en Espana (1881-1973), Madrid:
Botti, A., Cielo
Alianza Universitaria, 1992
Bowers, C., My Mission to Spain, London: Victor Gollancz, 1954
Boyd, C., Praetorian Politics in Liberal Spain, Chapel Hill: University of North
Carolina Press, 1979
Brademas, J., Anarcosindicalismo y revolucion en Espana (1930-1937), Barcelona:
Ariel, 1974
 , , ,

442 Bibliography

Braojos Garrick), A., Alvarez Rey, L. and Espinosa Maestre, E, Sevilla 36: suble-

vacion fascistay represion Seville: Munoz Moya y Montraveta, 1990

Brenan, G., The Spanish Labyrinth 1943; Cambridge: Cambridge University
Press, 1990
Personal Record 1320-1372, London, 1974; New York: Alfred A. Knopf, 1975
Bricall, J. M., Politico economica de la Generalitat (1336-1333), 2nd edn, Barcelona:
Edicions 62, 1978
Broue, P., Staline et la revolution. Le cas espagnol (1336-1333), n.p.: Editions Fayard,
1993
Broue, R, Fraser, R. and Vilar, P., Metodologia de la guerray la revolucion espaholas

Barcelona: Fontamara, 1982

Broue, P. and Temime, E., The Revolution and the Civil War in Spain ,
London:
Faber & Faber, 1972
Buchanan, T., The Spanish Civil War and the British Labour Movement, Cambridge:
Cambridge University Press, 1991
Britain and the Spanish Civil War Cambridge: Cambridge University Press, 1997
,

Bueno Madurga, J. I., faragoza 1317-1336. De la movilizacion popular y obrera

a la reaccion conservadora, Zaragoza: Institution ‘Fernando el Catolico’/
Diputacion de Zaragoza, 2000
Bueso, A., Recuerdos de un cenetista 2 vols., Barcelona: Ariel, 1976, 1978
,

Bullejos, J., La Comintern en Espaha, Mexico: n.p., 1972

Cabanellas, G., Laguerra de los mil dias 2 vols., Buenos Aires: Grijalbo, 1973
,

Cabezas, J. A., Asturias: catorce meses de guerra civil Madrid: G. del Toro, 1975 ,

Cabrera, M., Lapatronal ante la II Republica. Organizacionesy estrategia. 1331-1336,

Madrid: Siglo XXI, 1983
Calder, A., The Myth of the Blitz, London: Pimlico, 1991
Callahan, W. J., Church ,
Politics and Society in Spain 1750-1874, Cambridge,
Mass.: Harvard University Press, 1984
Calleja, J. J., Tagiie: un corazon al rojo, Barcelona: Editorial Juventud, 1963
Caminal, M.,Joan Comorera. Catalanisme i socialisme, 1313-1336, vol. 1, Barcelona:
Empuries, 1984
Joan Comorera, Antologla ed. Miquel Caminal, Barcelona: La Magrana, 1987
,

Canellas, C. et al., Historia de lapreso Model de Barcelona Lleida: Pages, 2000 ,

Carabantes, A. and Cimorra, E., Un mito llamado Pasionaria, Barcelona: Planeta,

1982
Carcel Orti, V, La persecucion religiosa en Espafia durante la Segunda Republica
~ Madrid, 1990
(I93 I I939)>
Cardona, G., El poder militar en la Espafia contemporanea hasta laguerra civil, Madrid:
Siglo XXI, 1983
Carr, E.H., The Twilight of the Comintern, London: Macmillan, 1982
The Comintern and the Spanish Civil War, London: Macmillan, 1984
Carr, R., The Civil War in Spain 1336-33, London: Weidenfeld & Nicolson, 1986
Carrasquer, F., Las colectividades de Aragon, Barcelona: Laia, 1986
Carrillo, S., Memorias, Madrid: Planeta, 1993; 6th edn, 1994
La Segunda Republica. Recuerdos y reflexiones, Barcelona: Plaza & Janes, 1999
 , , ,

Bibliography 443

Carroll, P. N., The Odyssey of the Abraham Lincoln Brigade. Americans in the Spanish
Civil War, Stanford, Calif.: Stanford University Press, 1994
Casado, S Asi cay 6 Madrid, 1968; repr. Madrid: Ediciones 99, 1977
Casanova, J., Caspe 1936-1938. Conflictos politicos y transformaciones sociales durante
la guerra civil, Zaragoza: Institucion Fernando el Catolico, 1984

Anarquismo y revolucion en la sociedad rural aragonesa 1936—1938, Madrid: Siglo

XXI, 1985
De la calle al frente El . anarcosindicalismo en Espaha (1931-1939), Barcelona:
Critica, 1997
‘Anarchism and Revolution in the Spanish Civil War: The Case of Aragon’,
European History Quarterly, 17 (1987)
Anarchism, Revolution and Civil War in Spain: The Challenge of Social
History’, International Review of Social History, 37 (1992)
‘Guerra civil ^lucha de clases?: el dificil ejercicio de reconstruir el pasado’,
Historia Social, 20 (1994)
‘Civil Wars, Revolutions and Counterrevolutions in Finland, Spain and
Greece (1918—1949): A Comparative Analysis’, InternationalJournal ofPolitics,
Culture and Society, 13 (3) (2000)
(ed.), El sueho igualitario. Campesinado y colectivizaciones en la Espaha republicana

i936~i939, Zaragoza: Institucion Fernando el Catolico, 1988

Casanova, J. et al., El pasado oculto. Fascismoy violencia en Aragon 1936-1939, 2nd
edn, Zaragoza: Mira Editores, 1999
Casas de la Vega, R., El terror: Madrid 1936, Madrid: Editorial Fenix, 1994
Seis generales de la guerra civil. Vidas paralelas y desconocidas, Madrid: Editorial

Fenix, 1998
Castell, E., Falco, L., Hernandez, X., Junqueras, O., Luque, J. C. and
Santacana, J., La batalla de I’Ebre. Historia, paisatge, patrimoni Barcelona:
Portic, 1999
Castells, A., Las brigadas internacionales de la guerra de Espaha, Barcelona: Ariel, 1974
Casteras, R., LasJSUC: ante laguerray la revolucion (1936-1939), Barcelona: Nova
Terra, 1977
Castilla del Pino, C., Preterito impefecto. Autobiografia Barcelona: Tusquets
Editores, 1997
Castro Delgado, E., Mife se perdio en Moscu, Barcelona: Luis de Caralt, 1964
Hombres made in Moscu Barcelona: Luis de Caralt, 1965
Catalunya i la guerra civil (1936-1939), Barcelona: L’Abadia de Montserat, 1988

(various authors)
and the Spanish Civil War, Berkeley, Calif: University
Cattell, D. T., Soviet Diplomacy
of California Press, 1957
Communism and the Spanish Civil War, New York: Russell and Russell, 1965
Caudet, F., Las cenizas del Fenix. La cultura espahola en los ahos 30, Madrid:
Ediciones de la Torre, 1993
Cazorla Sanchez, A., Las politicos de la victoria. La consolidacion del Nuevo Estado
franquista (1938-1933), Madrid: Marcial Pons, 2000
Cebrian, C., Estimat PSUC, Barcelona, 1997
444 Bibliography

Cenarro, A., Bl fin de la esperanza. Fascismo y guerra civil en la provincia de Teruel

936~i939
i >
Teruel: Diputacion Provincial de Teruel, 1996
Cruzados y camisas azules. Los origenes del franquismo en Aragon 1936-igq^^
Zaragoza.: Prensas Universitarias de Zaragoza, 1997
Cervera, J., Madrid en guerra. La ciudad clandestina 1936-1939, Madrid: Alianza,
1998
Chamorro, V Historia de Extremadura (6 vols., Madrid: Victor Chamorro, n.d.,

[
I98 5l)
Chaves Palacios, J., La represion en la provincia de Caceres durante la guerra civil

(1936—1939), Caceres: Universidad de Extremadura, 1995

La guerra civil en Extremadura , 2nd edn, Caceres, 1997
Violencia polltica y conflictividad social en Extremadura. Caceres en 1936, n.p.:
Diputacion de Badajoz/Diputacion de Caceres, 2000
Chiapuso, M., El gobierno vasco y los anarquistas. Bilbao en guerra, San Sebastian:
Editorial Txertoa, 1978
Christie, S., We, the anarchists! A Study of the Iberian Anarchist Federation (FAI)
ig2y-iggy, Hastings: The Meltzer Press/Jura Media, 2000
Cierva, R. de la, igjg. Agoniay victoria Barcelona: Planeta, 1989
,

Cifuentes Chueca, J. and Maluenda Pons, P, El asalto a la Republica. Los origenes

del franquismo en Zaragoza (ig36-iggg), Zaragoza: Institution ‘Fernando el
Catolico’, 1995
Cimorra, Espaha en las trincheras Madrid: Editorial Nuestro Pueblo, 1938
E., ,

Claudin, The Communist Movement. From Comintern to Cominform Harmonds-

F., ,

worth: Penguin, 1975

Santiago Carrillo: cronica de un secretario general, Barcelona: Planeta, 1983
Cobb, C., Ij)s milicianos de la cultura, Bilbao: Universidad del Pais Vasco, 1994
‘The Educational and Cultural Policy of the Popular Front Government
in Spain 1936—9’, in M. S. Alexander and H. Graham (eds), The French
and Spanish Popular Fronts. Comparative Perspectives, Cambridge: Cambridge
University Press, 1989
Cobo Romero, F., La guerra civil
y la represion franquista en la provincia de Jaen
-
(i J5 °), Jaen: Diputacion dejaen, 1993
93 fj IC
Conflicto rural y violencia polltica, Jaen: Universidad de Jaen/Universidad de
Granada, 1998
‘El vo to campesino contra la II Republica. La derechizacion de los pequenos
propietarios y arrendatarios agricolas jiennenses 1931-1936’, Historia
Social, 37 (2000)

Coll, J. and Pane, J., Josep Rovira. Una vida al servei de Catalunya i del socialisme,

Barcelona: Ariel, 1978

Collier, G. A., Socialists of Rural Andalusia, Stanford: Stanford University Press,
I98 7
Collins, L. and Lapierre D., Or Pll Dress You in Mourning, London: Weidenfeld
& Nicolson, 1968
Comin Colomer, E., Historia del Partido Comunista de Espaha, 3 vols., Madrid:
Editora Nacional, 1967
 , ,

Bibliography 445
Contreras, M., El PSOE en la II Republica. Organizacion e ideologla Madrid: CIS,
1981
Corbin, J. R., The Anarchist Passion. Class Conflict in Southern Spain 1810-1965,
Newcasde upon Tyne: Avebury, 1993
Corbin, J., An Anthropological Perspective on Violence’, International Journal

ofEnvironmental Studies 10 (1977) ,

Cordon, A., Trayectoria Paris: Coleccion Ebro, 1971

,

J. W. (ed.), A City in War:

Cortada, American Views on Barcelona and the Spanish Civil
War 1936—1939, Wilmington: Scholarly Resources, 1985
Costa Vidal, E, Villena durante la guerra civil 1936-1939, Alicante, 1998
Costello, J. and Tsarev, O., Deadly Illusions ,
London: Century, 1993
Crome, ‘Walter (1897-1947):
L., A Soldier in Spain’, History Workshop Journal,

9 (Spring 1980)
Cruells, M., Mayo sangriento. Barcelona 1937, Barcelona: Editorial Juventud, 1970
Lexpedicio a Mallorca. Any 1936 Barcelona: Editorial Juventud, 1971
La societat catalana durant la guerra civil, Barcelona: Edhasa, 1978
Cruz, R., El Partido Comunista de Espaha en la II Republica, Madrid: Alianza,
1987
Madrid: Biblioteca Nueva, 1999
Pasionaria,
Cruz, R., ‘jLuzbel vuelve 2d mundo! Las imagenes de la Rusia y la accion
colectiva en Espana’, in R. Cruz and M. Perez Ledesma (eds.), Culturay
movilizacion en la Espaha contemporanea, Madrid: Alianza, 1997
Cuadernos de la guerra civil, Los sucesos de Mayo (1937), Madrid: Fundacion

Salvador Segui, 1987

Las relaciones CNT-UGT, Madrid: Fundacion Salvador Segui, 1989
Las transformaciones colectivistas en la industriay los servicios de Barcelona (193 6—1939),
Madrid: Fundacion Salvador Segui, 1992
Cuadrat, X., Socialismoy anarquismo en Cataluha 1890—1911, Madrid: Revista del
Trabajo, 1976
Cuesta,J. and Bermejo B. (eds), Emigraciony exilio. Espaholes en Francia 1936-1946,
Madrid: Eudema, 1996
Cueva Merino, J. de la, ‘El anticlericalismo en la Segunda Republica y la
guerra civil’, in E. la Parra and M. Suarez (eds), El anticlericalismo espahol
contemporaneo, Madrid: Biblioteca Nueva, 1998
Cuevas, T., Prison of Women: Testimonies of War and Resistance in Spain 1939-1975
Albany, n.y.: SUNY, 1998
Dallin, A. and Firsov, F. I. (eds), Dimitrov and Stalin 1934-1943. Letters from the

Soviet Archives, New Haven: Yale University Press, 2000

Deak, I., Gross, J. andjudt, T. (eds.) The Politics of Retribution in Europe. World
War II and its Aftermath (Princeton: Princeton University Press, 2000)
Diaz, J., Tres ahos de lucha, Toulouse, 1947; 3 vols., Barcelona: Laia, 1978
Diaz Nosty, B. La comma asturiana. Revolucion de octubre de 1934 Madrid: Zero, 1974
, ,

Diaz-Plaja, F., La guerra de Espaha en sus documentos, Barcelona: Ediciones GP, 1969
Los grandes procesos de la guerra civil espahola, Barcelona: Plaza & Janes, 1997
DolgofT, S., The Anarchist Collectives , New York: Free Life Editions, 1974
44 6 Bibliography

Dona, J., Desd&la nochey la niebla. Mujeres en la carceles franquistas, Madrid, 1978;
2nd edn, 1993
Durgan, A. C., B.O.C. igjo—igj6. El Bloque Obrero y Campesino, Barcelona:
Laertes, 1996
Durgan, A., ‘Trotsky, the POUM and the Spanish Revolution’, Journal of Trotsky
Studies , 2 (1994)
Durruti i8g6—igj6, Madrid: Fundacion Anselmo Lorenzo, 1996
Ealham, C., Anarchism and Illegality in Barcelona, 193 1-7’, Contemporary
European History 4 (2) (1995) ,

‘La lluita pel carrer, els vendedors ambulants durant la II Republica’, L’Aveng,
230 (November 1998)
‘
“Revolutionary Gymnastics” and the Unemployed. The Limits of the
Spanish Anarchist Utopia 1931-37’, in K. Flett and D. Renton (eds), The
Twentieth Century: A Century of Wars and Revolutions ?, London: Rivers Oram,
2000
Egido Leon, A. (ed.), Azahay los otros, Madrid: Biblioteca Nueva, 2001
Ehrenburg, I., Memoirs ig2i-ig4i, New York: Grosset & Dunlap, 1966
Ellwood, S., ‘Spanish Newsreels 1943-1975* The Image of the Franco Regime’,
Historical Journal of Film, Radio and Television 7 (3) (1987)
Elorza, A., La utopia anarquista bajo la Segunda Republica Madrid: Ayuso, 1973 ,

‘La estrategia del POUM

en la guerra civil’, in La II Republic. Una esperanza
frustrada (Actas del congreso Valencia Capital de la Republica), Valencia:
Edicions Alfons el Magnanim, 1987
‘Stalinisme et internationalisme en Espagne 1931 -1939’, in S. Wolikow and
M. Cordillot (eds), Proletaires de tous les pays - unissez-vous? Les difficiles chemins
de Vinternationalisme 1848—igg6, Dijon, 1993
Elorza, A. and Bizcarrondo, M., Queridos camaradas. La Internacional Comunistay
Espaha igig-igjg Barcelona: Planeta, 1999
,

Escobal, P R, Death Row. Spain igg6, New York/ Indianapolis: Bobbs-Merrill

Co., 1968
Escofet, F., Al servei de Catalunya i la Republica Paris, 1973
,

Espin, E., Azaha en el poder. El partido deAccion Republicana Madrid: CIS, 1980 ,

Espinosa Maestre, F., La guerra civil en Huelva Huelva: Diputacion Provincial de

,

Huelva, 1996
Lajusticia de Queipo. (Violencia selectivay terror fascista en la II Division en igg6)
Sevilla, Huelva, Cadiz, Cordoba, Malaga y Badajoz, Seville: Centro Andaluz
del Libro, 2000
Estruch, J., Historia delP.C.E. (ig20-iggg), vol. 1, Barcelona: El Viejo Topo, 1978
Historia oculta del PCE, Madrid: Temas de Hoy, 2000
Fagen, P. W., Exiles and Citizens. Spanish Republicans in Mexico Austin: University,

of Texas Press, 1973

Falcon, I., Asalto a los cielos. Mi vida junto con Pasionaria, Madrid: Temas de Hoy,
1996
Febo, G. di, Resistenciay movimiento de mujeres igj6-igy6 Barcelona: Icaria 1979 ,

Febo, G. di and Natoli, C. (eds), Spagna anni Trenta. Societa, cultura, istituzioni,
Milan: FrancoAngeli, 1993
 , ,

Bibliography 447

Fernandez, C., Paracuellos deljarama: Carrillo culpable?, Argos Vergara, 1983

Fernandez Santander, C., Alzamimto y guerra civil en Galicia (1936-1939), 2 vols.,
A Coruna: Edicios do Castro, 2000
Ferrandiz Alborz, F., La bestia contra Espaha. Reportaje de los ultimos dlas de la guerra
espaholay los primeros Montevideo: n.p., 1951
de la bestia triunfante ,
Figueras, J. M., El Consell de guerra a Lluis Companys Barcelona: Proa, 1997 ,

Foltz, C., Jr, The Masquerade in Spain Boston: Houghton Mifflin, 1948
,

Fontana, J. M., Los catalanes en la guerra de Espafia Barcelona: Ediciones Acervo,

,

1951
Forner Munoz, S., Industrializacion y movimiento obrero. Alicante 1323-1336,
Valencia: Institucio Alfons el Magnanim, 1982
Fraser, R., In Hiding The Life ofManuel Cortes London: Allen Lane, 1972 ,

Blood of Spain. The Experience of Civil War 1336—1333, 1979; Harmondsworth:

Penguin, 1981
Fyrth, J., The Signal was Spain.
The Aid Spain Movement in Britain 1336-33

London: Lawrence and Wishart, 1986

Gabarda, V, Els afusellaments al Pais Valencia, (1338-1336), Valencia: Edicions
Alfons el Magnanim, 1993
Gabriel, P, ‘Poblacion obrera catalana’, Estudios de Historia Social 32—3 (1985) ,

‘Sindicalismo en Cataluna 1888-1938’, Historia Social, 8 (1990)

Garcia, C. Las carceles de Soledad Real. Una vida Madrid, Ediciones Alfaguara, 1982
,

Garcia, M., Miguel Garda's Story ed. A. Meltzer, Orkney: Cienfuegos Press, 1982
,

Garcia de Consuegra Munoz, G., Lopez Lopez, A. and Lopez Lopez, F., La
represion en Pozoblanco Cordoba: Francisco Baena, Editor, 1989
,

Garcia Delgado, J. L. (ed.), El primer franquismo: Espaha durante la segunda guerra

mundial, Madrid: Siglo XX, 1989
Garcia Duran, J., La guerra civil espahola: Fuentes, Barcelona: Critica, 1985
Garcia Garcia, C., Aproximacion al estudio de la represion franquista en
Asturias: “paseos” y ejecuciones en Oviedo (1936-1952)’, El Basilisco, 2 (6)
(* 99 °)
Garcia Oliver, J., El eco de los pasos Paris: Ruedo Iberico, 1978
Garcia Pradas, J., jTeniamos que perder!, Madrid: G. del Toro, 1974
Garitaonandia, C. and Granja, J. L. de la (eds), La guerra civil en el Pais Vasco, go
ahos despues Bilbao: Universidad del Pais Vasco, 1987
,

Garrido Gonzalez, L .,Colectividades agrarias en Andalucia: Jam (1331-1333),

Madrid: Siglo XXI, 1979
Gazur, E. P, Secret Assignment. The FBPs KGB General, London: St Ermin’s Press,
2001
Gerhard, C., Comissari de la Gmeralitat a Montserrat (1336-1333), Barcelona,
L’Abadia de Montserrat, 1982
Gibaja Velazquez, J. C., Indalecio Prieto y el socialismo espahol, Madrid: Fundacion
Pablo Iglesias, 1995
Gibson, I., The Assassination of Federico Garcia Lorca London: W. H. Allen, 1979 ,

Paracuellos:como fue, Barcelona: Argos Vergara, 1983

Queipo de Llano: Sevilla, verano del 1336, Barcelona: Grijalbo, 1986
Lorca. A Biography London:,
Faber & Faber, 1989
 , y
,

448 Bibliography

Gil Andres, C* Echarse a la calle. Amotinados, huelguistas y revolucionarios. La Rioja

1830-1336, Zaragoza: Prensas Universitarias de Zaragoza, 2000
Gil Bracero, R., Revolucionarios sin revolucion. Marxistasy anarcosindicalistas en guerra.
Granada-Baza 1336-1333 Granada, 1998 ,

Ginard i Feron, D., Heriberto Quinones y el movimiento comunista en Espana (1331-

ig42), Palma/Madrid: Edicions Documenta Balear/Compama Literaria,
2000
Goldman, E., Vision on Fire New York: Commonground Press, 1983
,

Gomez Casas, J., Los anarquistas en el Gobierno 1336—33, Barcelona: Bruguera 1977
Gomez Ortiz, J. M., Los gobiernos republicanos. Espana 1336-1333, Barcelona:
Bruguera, 1977
Gonzalez, V
(‘El Campesino’), Comunista en Espanay antiestalinista en la U.R.S.S.

Mexico: Editorial Guarania, n.d.

Gonzalez Calleja E. and Rey, F., El Mausery el sufragio. Orden publico, subversion
violencia politica en la crisis de la Restauracion, igij-31 Madrid: CSIC, 1999
Gonzalez Echegaray, R., La marina mercantey el trafico marltimo en la guerra civil,
Madrid: San Martin, 1977
Gonzalez Egido, L., Agonizar en Salamanca. Unamuno, julio-diciembre 133 6, Madrid:
Alianza 1986
Gonzalez Martinez, C., Guerra civil en Murcia. Un analisis sobre el poder y los
comportamientos colectivos Murcia: Universidad de Murcia, 1999
,

Gonzalez Portilla, M. and Garmendia, J. M., La guerra civil en el Pais Vasco.

Politicay economia, Madrid: Siglo XXI, 1988
Goodway, D. (ed.), For Anarchism. History, Theory, Practice London: Routledge, ,

1989
Gorkin, J., Espana. Primer ensayo de democracia popular, Buenos Aires: Asociacion
argentina por la libertad de la cultura, 1961
ElprocesodeMoscu en Barcelona, Barcelona: Ayma, 1974
Graham, H., Socialism and War. The Spanish Socialist Party in Power and Crisis,

Cambridge: Cambridge University Press, 1991

I93 ^ I939 ,
‘The Socialist Youth in the JSU’, in M. Blinkhorn (ed.), Spain in Conflict
1331-1336, London: Sage, 1986
‘Community, Nation and State in Republican Spain 1931-1938’, in C.
Mar-Molinero, and A. Smith (eds), Nationalism and the Nation in the Iberian
Peninsula, Oxford: Berg, 1996
‘War, Modernity and Reform: The Premiership of Juan Negrin’, in P.
Preston and A. MacKenzie (eds), The Republic Besieged. Civil War in Spain
i936-i939, Edinburgh: Edinburgh University Press, 1996

‘Spain 1936. Resistance and Revolution: The Flaws in the Front’, in T. Kirk
and A. McElligott (eds), Opposing Fascism. Community, Authority and Resistance
in Europe, Cambridge: Cambridge University Press, 1999

Graham, H. and Labanyi, J. (eds), Spanish Cultural Studies. An Introduction,

Oxford: Oxford University Press, 1995
Graham, H. and Preston, P. (eds), The Popular Front in Europe, Basingstoke:
Macmillan, 1987
 ,

Bibliography
449
Grand, A. de, In Stalin’s Shadow. Angelo Tasca and the Crisis of the Left in Italy and
France igio-1343, Northern Illinois University Press, 1986
Granja, J. L. de la, JVacionalismoy II Republica en el Pals Vasco ,
Madrid: Centro de
Investigaciones Sociologicas/Siglo XXI, 1986
Republica y guerra civil en Euskadi. Del Pacto de San Sebastian al de Santoha Onati: ,

HAEE/IVAP, 1990
Grossi Mier, M., La insurreccion de Asturias, 1935; Madrid: Edicionesjucar, 1978
Guarner, V, Cataluha en la guerra de Espaha Madrid: G. del Toro, 1975
,

L’aixecament militar i la guerra civil a Catalunya Barcelona, 1980

,

Guerra y revolucion en Espaha 1936-33, 4 vols., Moscow: Editorial Progreso,

1966-77
Guillamon, A., The Friends ofDurruti: r937 -1939, Edinburgh: Press, 1996 AK
Gullon, R., ‘Justice et guerre civile: souvenir d’un procureur’, in C. Serrano
Madrid 1936-1939, Paris: Editions Autrement, 1991
(ed.),

Guzman, La muerte de la esperanza Madrid: G. del Toro, 1973

E. de, ,

Aho de la Victoria (Memorias de la guerra de Espaha), Madrid: G. de Toro, 1974

Haro Tecglen, E., Arde Madrid, Madrid: Temas de Hoy, 2000
Haslam, J., The Soviet Union and the Struggle for Collective Security in Europe
1933 ~I939 London: Macmillan, 1984
’>

Hermet, G., The Communists in Spain, Lexington, Mass.: Lexington Books, 1974
Hernandez, J., En el pals de la gran mentira, Madrid: G. del Toro, 1974
To fui un ministro de Stalin Madrid: G. del Toro, 1974
Herrmann, G., ‘The Hermetic Goddess: Dolores Ibarruri as Text’, Letras
Peninsulares (Spring 1998)
Heywood, P, Marxism and the Failure of Organised Socialism in Spain 1873-1336,
Cambridge: Cambridge University Press, 1990
Hidalgo de Cisneros, I., Cambio deRumbo, Vitoria-Gasteiz: Ikusager, 2001
Hills, G., The Battlefor Madrid, London: Vantage Books, 1976
Hofmann, B., Joan i Tous, R and Tietz, M. (eds), El anarquismo espaholy sus tra-
diciones culturales, Frankfurt-on-Main/Madrid: Vervuert-Iberoamericana,
J
995
Hopkins, J. K., Into the Heart of the Fire. The British in the Spanish Civil War,
Stanford, Calif: Stanford University Press, 1998
Howson, G., Aircraft of the Spanish Civil War, London: Putnam Aeronautical
Books, 1990
Armsfor Spain, London: John Murray, 1998
Huertas Claverla, J. M., Obrers a Catalunya, Barcelona: Avene, 1982
Ibarruri, D., En la lucha. Palabras y hechos 1336-1333, Moscow: Editorial
Progreso, 1968
El unico camino,Madrid: Editorial Castalia, 1992
Iglesias, La fasefinal de la guerra civil, Barcelona: Planeta, 1977
I.,

J., A Tentorn del centenari

Iglesias, I., Cabo, F. de and Andrade, dAndreu JVin
1832-1332, Barcelona: Fundacio Andreu Nin, 1993
Imagenes en guerra. Memoria estampada en la Espaha de los ahos 30, Coleccion de

J. Diazi Prosper y J. Roca Boix, Valencia, 1998

 , ,

45° Bibliography

International Solidarity with the Spanish Republic 1336-1333, Moscow: Progress

Publishers, 1975
Irujo, M. de, Un vasco en el ministerio dejusticia , Memorias 3 vols., Buenos Aires:
Editorial Vasca Ekin, 1976-9
Iturralde, J. de, El catolicismoy la cruzada de Franco Bayonne, 1955
,

Jackson, G., The Spanish Republic and the Civil War 1331 -1333, Princeton, n.j.:

Princeton University Press, 1965

Historian’s Quest, New York: Alfred A. Knopf, 1969
Jimenez de Aberasturi, L. M., ‘Guerra y movilizacion popular en el Pais
Vasco’, inj. Arostegui (ed.), Historiay memoria de la guerra civil. Encuentro en
Castilla y Leon Valladolid, 1988
,

Julia, S., La PSOE (1335-1336'), Madrid: Siglo XXI, 1977

izquierda del
Ongenes del Frente Popular en Espaha 1334-1336, Madrid: Siglo XXI, 1979
Madrid, 1331 -1334. De la fiesta popular a la lucha de clases, Madrid: Siglo XXI,
1984
Historia del socialismo espanol 3 of M. Tunon de Lara (ed.),
1331—33 (vol.

Historia del socialismo espanol, Barcelona: Conjunto Editorial, 1989)

Manuel Azaha. Una biograjia polltica Madrid: Alianza, 1990 ,

‘Manuel Azana: la razon, la palabra y el poder’, in V A. Serrano and J. M.

San Luciano (eds), Azaha Madrid: Edascal, 1980
,

‘Sindicatos y poder politico en Espana’, Sistema 97 (July 1990) ,

‘La doble derrota de Juan Negrin’, El Pais 26 February 1992 ,

‘Presidente por ultima vez: Azana en la crisis de mayo de 1937’, in A. Alted,

A. Egido and M. F. Mancebo (eds), Manuel Azaha: pensamiento y accion,
Madrid: Alianza, 1996
(ed.) Socialismo y guerra civil Anales de Historia vol. 2, Madrid: Editorial Pablo
, ,

Iglesias, 1987
(ed.) ‘Politica en la Segunda Republica’, Ayer 20, 1995
Julia, S. et al., Victimas de la guerra civil, Madrid: Temas de Hoy, 1999
Justicia en guerra (Jornadas sobre la administracion de justicia durante la guerra civil
espahola: institucionesyfuentes documentales) (Salamanca, 26-8 November 1987),
Madrid: Ministerio de Cultura, 1990
Kalyvas, S. N. The Logic of Violence in Civil War, Madrid: Fundacionjuan March,
2000
Kaplan, T., Red City, Blue Period. Social Movements in Picasso’s Barcelona (1888-1337),
Berkeley, Calif.: University of California Press, 1992
Kelsey, G., Anarchosyndicalism, Libertarian Communism and the State. The CNT in
Zaragoza and Aragon 1330-1337 ,
Amsterdam: Institute for Social History,
1991
‘Anarchism in Aragon during the Second Republic: The Emergence of a
Mass Movement’, in M. Blinkhorn (ed.), Spain in Conflict, London: Sage,
1986
Kemp, P, Mine Were of Trouble, London: Cassell, 1957
Kern, R. W., Red Tears, Black Tears. A Political History of Spanish Anarchism
1311-1337 Philadelphia: ISHI, 1978
 ,, ,

Bibliography 451

Kisch, R., They Shall Not Pass. The Spanish People at War 1936-9, London:
Wayland Publishers, 1974
Klehr, H. and Haynes, J. E., The Secret World of American Communism New ,

Haven: Yale University Press, 1995

Koltsov, M., Diario de la guerra espahola Madrid: Akal, 1978
,

Labanyi, J., ‘Women, Asian Hordes and the Threat to the Self in Gimenez
Caballero’s Genio de Espaha!, Bulletin of Hispanic Studies 73 (1996) ,

Laboa, J. M., Iglesia e intolerancia: la

guerra civil, Madrid: Atenas, 1987
Ladron de Guevara, M. P, La esperanza republicana. Reforma agraria y conflicto

campesino en la provincia de Ciudad Real, (1931-1939), n.p.: Diputacion de

Ciudad Real, 1993
Lama, J. M., Una biografia frente al olvido: Jose Gonzalez Barrero, Alcalde de Zafra m
la II Republica, Badajoz: private edition, 2000
Lamoneda, R., Ramon. Lamoneda. Posiciones politicos; documentos; correspondencia
Mexico: Roca, 1976
Lannon, F., Privilege, Persecution and Prophecy. The Catholic Church in Spain 1823-1923,
Oxford: Clarendon Press, 1987
Lannon, E and Preston, P (eds), Elites and Power in Twentieth-Century Spain. Essays
in Honour of Sir Raymond Carr, Oxford: Clarendon Press, 1990

Largo Caballero, F., Mis recuerdos, 2nd edn, Mexico DF: Ediciones Unidas, 1976
Escritos de la Republica with introduction and notes by S. Julia, Madrid:
Fundacion Pablo Iglesias, 1985
Laruelo Roa, M., Asturias, octubre del 32 /el Cervera a la vista!, 2nd edn, Gijon:
Imprenta la Industria, 1998
Leval, G., Collectives in the Spanish Revolution, London: Freedom Press, 1975
Lincoln, B., ‘Revolutionary Exhumation in Spain, July 1936’, Comparative
Studies in Society and History, 27 (2) (1985)
lister, E., Nuestra guerra, Paris: Coleccion Ebro, 1966
Memorias de un luchador vol. 1, Madrid: G. del Toro, 1977
Little, D., Malevolent Neutrality. The United States, Great Britain and the Origins of the
Spanish Civil War, Ithaca /London: Cornell University Press, 1985
Litvak, L., Musa Libertaria. Arte, literatura y vida cultural del anarquismo espahol
(1880-1913), Barcelona: Antoni Bosch, 1981
Llarch, J., La Muerte de Durruti, Barcelona: Ediciones Aura 1973
Cipriano Mera. Un anarquista en la guerra civil espahola, Barcelona: Producciones

Editoriales, 1977
es veneer!, Barcelona: Planeta, 1985
Negrin /Resistir
Lloyd D., and Thomas, P., Culture and the State, London, Routledge: 1998
Lopez, J., Una mision sin importancia. Memorias de un sindicalista, Madrid: Editorial
Nacional, 1972
Lopez Fernandez, A., General Miaja, Defensor de Madrid, Madrid: del Toro, 1975
Lorenzo, C. M., Los anarquistas espaholes y el poder 1868-1939, Paris: Ruedo
Iberico, 1972
Low, M. and Brea, J., Red Spanish Notebook, 1937; San Francisco: City Lights
Books, 1979
 ,

452 Bibliography

Lozano, J., La fegunda Republica. Imageries, cronologia y documentos, Barcelona:

Ediciones Acervo, 1973
Lynam, S., The Spirit and the Clay Boston, Mass.: ,
Little, Brown & Co., n.d.
Macarro Vera, M., La utopia revolucionaria. Sevilla en la Segunda Republica Seville:
J. ,

Monte de Piedad y Caja de Ahorros de Sevilla, 1985

Socialismo, Republica y revolucion en Andalucia (igji -igj6), Seville: Universidad
de Sevilla, 2000
‘Causas de la radicalizacion socialista en la II Republica’, Revista de Historia
Contemporanea (Seville), 1 (Dec. 1982)
‘Economia y polltica en el Frente Popular’, Revista de Historia Contemporanea
7 (^96)
Madariaga, R. M. de, ‘The Intervention of Moroccan Troops in the Spanish
Civil War: A Reconsideration’, European History Quarterly 22 (January ,

'
992 )
Maddox, R., ‘Revolutionary Anti-Clericalism and Hegemonic Processes in an
Andalusian Town. August 1936’, American Ethnologist 22 (1) (1995) ,

Maisky, I., Spanish Notebooks, London: Hutchinson & Co., 1966

Majuelo, E., Luchas de clases en Navarra (igji —igg6), n.p.: Gobierno de Navarra,
1989
Malefakis, E., Agrarian Reform and Peasant Revolution in Spain ,
New Haven/London:
Yale University Press, 1970
Malefakis, E. (ed.), Laguerra de Espaha igg6-igjg, Madrid: Taurus, 1996
Manent i Segimon, A. and Raventos i Giralt, J., LEsglesia clandestina a Catalunya
durant laguerra civil (igg6-iggg). Els intents de restablir el culte public, Barcelona:
L’Abadia Montserrat, 1984
Mangini, S., Memories of Resistance. Women’s Voices from the Spanish Civil War, New
Haven/London: Yale University Press, 1995
Marichal, J., La vocacion de Manuel Azaha, Madrid: Cuadernos para el Dialogo,
1968
El intelectualy la politica, Madrid: CSIC, 1990
Marquez Rodriguez, J. M. and Gallardo Romero, J. J., Ortiz . General sin dios ni
amo, Barcelona: Hacer, 1999
Marti Gilabert, F., Politica religiosa de la Segunda Republica espanola, Pamplona:
Eunsa, 1998
Marti Gomez, J., La Espaha del estraperlo (igg6-igg2), Barcelona: Planeta, 1995
Martin, B., The Agony of Modernisation. Labor and Industrialisation in Spain, Ithaca,
n.y.: ILR Press, Cornell University, 1990
Martin, B. The Hypocrisy of Justice
F., in the Belle Epoque, Louisiana State
University Press, 1984
Martin Ramos, J. L., Els origenes del Partit Socialista Unificat de Catalunya
(ig30-igg6), Barcelona: Curial, 1977
Martinez Amutio, J., Chantaje a un pueblo, Madrid: G. del Toro 1974
Martinez Bande, J. M., Los cien ultimos dias de la Republica, Barcelona: Luis
Caralt, 1973
Martinez Barrio, D., Memorias Barcelona: Planeta, 1983
,
 , , ,

Bibliography 453

Martinez Leal, J., Republica y guerra civil en Cartagena (1931 -1939), Murcia:
Universidad de Murcia, 1993
Martinez Paricio, J. I., Los papeles del general Rojo, Madrid: Espasa-Calpe, 1989
Martinez Saura, S., Memorias del secretario deAzaha Barcelona: Planeta, 1999
,

Martorell, M., Jesus Monzon. El lider comunista olvidado por la historia, Pamplona:
Pamiela, 2000
Massot i Muntaner, J., El desembarcament de Bayo a Mallorca. Agost-Setembre de

1996, Barcelona: L’Abadia de Montserrat, 1987

La persecucio religiosa de 1936 a Catalunya. Testimoniatges a cura de J. Massot i
Muntaner Barcelona: L’Abadia de Montserrat, 1987
,

El Consol Alan Hillgarth i les Illes Balears (1936-1939), Barcelona: L’Abadia de

Montserrat, 1995
Mateo Merino, P, Por vuestra libertad y la nuestra. Andanzas y rejlexiones de un
combatiente republicano , Madrid: Editorial Disenso, 1986
Matthews, H. L., Half of Spain Died New York: Charles Scribner’s Sons, 1973
Maura, M., Asi cay 6 Alfonso XIII, Barcelona: Ariel, 1995
Maurin, J., La revolucion espanola, Barcelona: Anagrama, 1977
Maurin, J., Como se salvo Joaquin Maurin. Recuerdosy testimonios, Madrid: Jucar, 1980
Mazowe r, M., Inside Hitler’s Greece. The Experience of Occupation 1941-44, New
Haven/London: Yale University Press, 1993
Mazower, M. (ed.) After the War was Over. Reconstructing, the Family, Nation and State
in Greece 1943-1960 Princeton, n.j.: Princeton University Press, 2000

Meaker, G. H., The Revolutionary Left in Spain 19 14-1923, Stanford: Stanford

University Press, 1974
Meer, F. de, El Partido Nacionalista Vasco ante la guerra de Espana (1936-1937),
Baranain-Pamplona: Eunsa, 1992
Melossi, D., The State of Social Control. A Sociological Study of the Concepts of State and
Social Control in the Making of Democracy Cambridge: Polity Press, 1990
Mendez, R., Caminos inversos: vivencias de cienciay guerra, Mexico: Fondo Cultura
Economica, 1987
Mera, C., Guerra, exilioy carcel de un anarcosindicalista, Paris: Ruedo Iberico, 1976
Mercader, L. and Sanchez, G., Ramon Mercader. Cincuenta anos despues, Madrid:
Espasa Calpe, 1990
Miliband, R., The State in Capitalist Society. The Analysis of the Western System,
London: Quartet, 1973
Mintz, F., La Autogestion en la Espana revolucionaria, Madrid: Ediciones de la
Piqueta, 1977
Mintz, F. and Pecina, M., Los amigos de Durruti, los trotsquistas y los sucesos de mayo,
Madrid: Campo Abierto, 1978
Mintz, J. R., The Anarchists of Casas Viejas, Chicago: University of Chicago
Press, 1982
Mir, C., Vivires sobrevivir. Justicia, ordeny marginacion en la Cataluna rural de posguerra,
Lleida: Milenio, 2000
Miralles, R., El socialismo vasco durante la II Republica, Bilbao: Universidad del
Pais Vasco, 1988
 , , , , , , y

454 Bibliography

‘La gran huelga minera de 1890. Los orlgenes del movimiento obrero en el

Pais Vasco’, Historia Contemporanea 3 (1990)

‘Juan Negrin: al frente de la politica exterior de la Republica (1937-1939)’,
Historia Contemporanea 15 (1996)
‘Paz humanitaria y mediacion inter nacional: Azana en la guerra’, in
A. Alted, A. Egido and M. F. Mancebo (eds), Manuel Azana: pensamiento
accion Madrid: Alianza, 1996
Miravitlles, J., Episodis de la guerra civil espanyola, Barcelona: Editorial Portic, 1972
Modesto, J., Soy del Quinto Regimiento, Barcelona: Laia, 1978
Molins i Fabrega, N., UHP. La insurreccion proletaria de Asturias, Madrid: Ediciones
Jucar, 1977
Monjo, A., ‘Militantes y afiliados cenetistas en los anos treinta’, in M. Vilanova
(ed.), Elpoder en Barcelona: Antoni Bosch, 1986
la sociedad,

Monjo, A. and Vega, C., ‘La clase obrera durante la guerra civil: una historia
silenciada’, Historia y Fuente Oral 3 (1990)
Monleon, J., El mono azul. Teatro de urgencia de la guerra civil, Madrid: Ayuso, 1979
Monreal, A., El pensamiento politico de Joaquin Maurin, Barcelona: Ediciones
Peninsula, 1984
Montero Moreno, A., Historia de la persecucion religiosa en Espaha 1936—1939, 1961;
Madrid: BAC, 1999
Montiel, F. F., Un coronel llamado Casado. Mentirasy misterios de la guerra de Stalin en

Espana Madrid: Criterio, 1998

,

Montoliu, P, Madrid en la guerra civil Madrid: Silex, 1998

,

Montseny, F., Mis primeros cuarenta anos, Barcelona: Plaza y Janes, 1987
Moradiellos, E., Neutralidad benevola. El go bierno britanico y la insurreccion militar

1936 Oviedo: Pentalfa, 1990

espanola de
La perpdia de Albion. El gobierno britanico y la guerra civil espanola, Madrid: Siglo
XXI, 1996
‘The Origins of British Non-Intervention in the Spanish Civil War:
Anglo-Spanish Relations in early 1936’, European History Quarterly 21 (1991) ,

‘British Political Strategy in the Face of the Military rising of 1936 in Spain’,
Contemporary European History, 1 (2) (1992)
‘Appeasement and Non-Intervention: British Policy during the Spanish Civil
War’, in P Catterall and C. J. Morris (eds), Britain and the Threat to Stability
in Europe 1918-43 Leicester: Leicester University Press, 1993
‘Una mision casi imposible: la embajada de Pablo de Azcarate en Londres
durante la Guerra Civil (1936-1939)’, Historia Contemporanea, 15 (1996)
Moreno de Alboran y de Reyna, F. and S., La guerra silenciosay silenciada. Historia

de la campaha naval durante la guerra de 1936—39 4 vols., Madrid: Graficas

Lormo, 1998
Moreno Gomez, F., La guerra civil en Cordoba (1936-1939), Madrid: Alpuerto, 1985
La resistencia armada contra Franco. Tragedia del maquisy la guerrilla, Barcelona:
Critica, 2001
Morrow, F., Revolution and Counter-Revolution in Spain, 1938; London: New Park
Publications, 1976
 , , ,

Bibliography 455

Mosher, J. R. The Birth ofMass Politics in Spain: Lerrouxismo

,
in Barcelona igoi -1303
New York: Garland, 1991
Munis, G.,Jalones de derrota, promesa de victoria. Criticay teoria de la revolucion espahola

(1930-1939), Madrid: Zero, 1977

Nadal, A., Guerra civil en Malaga Malaga, Editorial Arguval, 1984
Nadal, J., El Jracaso de la revolucion industrial en Espaha, 1814-1313, Barcelona:
Ariel, >975
Nash, M., Defying Male Civilisation: Women in the Spanish Civil War, Denver, Colo.:
Arden Press, 1995
Nenni, P, La guerra de Espaha, in Italian, 1958; 4th edn, Mexico: Ediciones Era,
1975
Neves, M
., La Matanza deBadajoz Badajoz: Editorial Regional de Extremadura,
,

1986
Nin, A., Los problemas de la revolucion espahola (^93 1-1337), n.p.: Ruedo Iberico,
1971
Por la unificacion marxista, Madrid: Castellote, 1978
Socialisme i nacionalisme, Barcelona, 1985
Nunez Diaz-Balart, M., La prensa de guerra en la zona republicana durante la guerra
civil espahola (1936-1939), 3 vols., Madrid: Ediciones de la Torre, 1992
Nunez Diaz-Balart, M. and Rojas Friend, A., Consejo de guerra. Los Jusilamientos
1939-1943), Madrid: Compania Literaria, 1997
en el Madrid de laposguerra (

Nunez Florencio, R.. Militarismo y antimilitarismo en Espaha 1888-1006, Madrid:

CSIC, 1990
Octubre 1334 (various authors), Madrid: Siglo XXI, 1985
Olaya, F., El oro deNegrln, Madrid: Ediciones Madre Tierra, 1990
Ollivier, M. and Landau, K., Espagne. Les Fossoyeurs de la revolution sociale, Paris:

Spartacus, 1975
Orlov, A., questionnaire response (1968), Forum fur osteuropaische Ideen und
feitgeschichte, 4 (2000)
Ors Montenegro, M., ‘La represion de guerra y postguerra en Alicante’, in
Violencia politica i ruptura social a Espanya (1336-1343), Lleida, 1994
La represion de guerra y posguerra en Alicante (1336-1343), Alicante: Instituto de
Cultura Juan Gil-Albert, 1995
Ortiz Heras, M., Violencia politica en la II Republica (Albacete 1336-1333), Madrid:
Siglo XXI, 1996
Ortiz Villalba, J., Sevilla 1336. Del golpe militar a la guerra civil, Seville: Diputacion
Provincial de Sevilla, 1997
Ossorio y Gallardo, A., Vida y sacrificio de Companys, Buenos Aires, 1943;
Barcelona: Nova Terra, 1976
Oyon, J. L. (ed.), Vida obrera en la Barcelona de entreguerras Barcelona: n.p., 1998
Paez-Camino Arias, F., Juan Negrin en nuestra historia’, Ifpna Abierta, 23 (1980)
Pages, P, AndreuNin: Su evolucion politica 1311-133 7, Bilbao: Zero, 1975
El movimiento trotskista en Espaha (1330-1333), Barcelona: Ediciones Peninsula,
1977
Historia del partido comunista de Espaha, Barcelona: Hacer, 1978
 , , ,

45 6 Bibliography

Pages i Blanch, P, La preso Model de Barcelona. Historia d’un centre penitenciari en

temps de guerra (1936-1939), Barcelona: L’Abadia de Montserat, 1996

La guerra civil espanyola a Catalunya (1936-1939), 2nd edn, Barcelona, 1997

Pamies, T., Cuando eramos capitanes. Memorias de aquella guerra, Barcelona: Dopesa,
r
974
Paniagua, X., La sociedad libertaria. Agrarismo e industrializacion en el anarquismo
espahol, Barcelona: Critica, 1982
Parga, C., Antes que sea tarde , Madrid: Compania Literaria, 1996
Pastor Petit, D., Espionaje. Espafia 1936-1939 Barcelona: Bruguera, 1977
,

Payne, S., The Spanish Revolution ,

London: Weidenfeld & Nicolson, 1970
Basque Nationalism, Reno, University of Nevada Press, 1975
Spain’s First Democracy. The Second Republic 1931-1936, Madison: University of
Wisconsin Press, 1993
Fascism in Spain 1923-1977, Madison: University of Wisconsin Press, 1999
Paz, A., Durruti. The People Armed, New York: Free Life Editions, 1977
Durruti en la revolucion espahola, Madrid: Fundacion Anselmo Lorenzo, 1996
Peirats, J., La CNT
en la revolucion espahola, 3 vols., Paris: Ruedo Iberico, 1971
Peiro, J., Escrits 1917 -1939, Barcelona, Edicions 62, 1975
Trayectoria de la CNT (Sindicalismo y anarquismo), Madrid: Ediciones Jucar,
1979
Perez-Baro, A., 30 meses de colectivismo en Cataluha Barcelona: Ariel, 1974
Perez Verdu, Cuando
Valencia fue capital de Espaha, Valencia: Generalitat
F.,

Valenciana, 1993
Perez Yruela, M., La conflictividad campesina en la provincia de Cordoba 1931-1936,
Madrid: Ministry of Agriculture, 1979
Pernau, J., Diario de la caida de Cataluha, Barcelona: Ediciones B, 1989
Pestana, A., Lo que aprendi en la vida, Madrid, 1972

Trayectoria sindicalista ed. A. Elorza, Madrid: Tebas, 1974

Pike, D. W., Vae Victis! Los republicanos espaholes refugiados en Francia 1939-1944,
Paris: Ruedo Iberico, 1969
Les Franqais et la guerre dEspagne 1936-1939, Paris: Presses Universitaires de
France, 1975
Ponamariova, L. V, La formacion del Partit Socialists Unijicat de Catalunya,
Barcelona: Icaria, 1977
Pons, S., Stalin e la guerra inevitable 1936-1941, Turin, 1995
Pons i Porta, J. and Sole i Sabate, J. M., Anarquia y Republica a la Cerdanya
( 3
(1936-1939). El Cojo de Malaga i els fets de Deliver, Barcelona: l’Abadia de
Montserrat, 1991
Porte, P. La, La atraccion del iman. El desastre de Annual y sus repercusiones en la politico
europea (1921 -1923), Madrid: Biblioteca Nueva, 2001
Prados de La Escosura, L., De imperio a nacion: crecimiento y atraso economico en
Espaha 1780-1930, Madrid: Alianza, 1988
Preston, R, The Politics of Revenge. Fascism and the Military in Twentieth-Century
Spain, London: Unwin Hyman, 1990
Franco. A Biography London: HarperCollins, 1993
 , , , , ,

Bibliography 457

The Coming of the Spanish Civil War. Reform, Reaction and Revolution in the Second
Republic 2nd edn, London: Routledge, 1994
Comrades London: HarperCollins, 1999
Doves of War. Four Women of Spain, London: HarperCollins, 2002
‘The Great Civil War 1914-1945’, in T. C. W. Blanning (ed.), The Oxford
History ofModern Europe Oxford: Oxford University Press, 2000
,

(ed.), Revolution and War in Spain 1991 -1939, London: Methuen, 1984

Preston, P and Mackenzie, A. (eds), The Republic Besieged: Civil War in Spain
I93^~I939 Edinburgh: Edinburgh University Press, 1996
?

Prieto, H. M., Posibilismo libertario Val-de-Marne, France: n.p., 1966

,

Prieto, I., Mexico: Ediciones Oasis, 1969

Palabras al Viento 1942; ,

Toy Moscu, 2nd edn, ed. de Mauricio Carlavilla, Madrid: Nos, i960
Convulsiones de Espana. Pequehos detalles de grandes sucesos, 3 vols., Mexico:
Ediciones Oasis 1967-9
Discursosfundamentales , Madrid: Ediciones Turner, 1975
Textos escogidos ed. R. Miralles, n.p.: Clasicos Asturianos del Pensamiento
Politico, 1999
Puelles Benitez, M. de, ‘El sistema educativo republicano: un proyecto
frustrado’, Historia Contemporanea , 6 (1991)
Pugliese, S. G., Carlo Rosselli: Socialist Heretic and Antfascist Exile ,
Cambridge,
Mass.: Harvard University Press, 1999
Quilis-Tauriz, F., Revoluciony guerra civil. Las colectividades obreras en la provincia de

1936-1999 Alicante: Institute de Cultura ‘Juan Gil- Albert’, 1992

Alicante ,

Quirosa-Cheyrouze y Munoz, R., Politicay guerra civil en Almeria Almeria: Cajal, ,

1986
Almeria 1936-9 7. Sublevacion military alteraciones en la retaguardia republicana

Almeria: Universidad de Almeria, 1997

Radcliff, P., From Mobilisation to Civil War. The Politics of Polarisation in the Spanish
1900-1992, Cambridge: Cambridge University Press, 1996
City of Gijon
Radosh, R., Habek, M. R. and Sevostianov, G. (eds), Spain Betrayed. The Soviet
Union in the Spanish Civil War, New Haven /London: Yale University Press,
2001
Ragsdale, H., ‘Soviet Military Preparations and Policy in the Munich Crisis:
New Evidence’, Jahrbiicherfur Geschichte Osteuropas,
47 (1999)
Raguer, H., La espaday la cruz . La iglesia 1996-1999, Barcelona: Bruguera, 1977
Salvador Rial, Vicari del Cardenal de la Pau, Barcelona: l’Abadia de Montserrat,
l
993
La polvoray el incienso. La iglesia
y la guerra civil espanola (1996-99), Barcelona:
Ediciones Peninsula, 2001
Rama, C., Fascismoy anarquismo en la Espana contemporanea, Barcelona: Bruguera,
!
979
(ed.), Camillo Berneri: Guerra de clases en Espana 1996-1992 Barcelona: Tusquets,
1977
Ramos, V, La guerra civil (1996-1999) en la provincia de Alicante, 3 vols., Alicante:
Biblioteca Alicantina, 1974
 , , ,

45 8 Bibliography

Ranzato, G., Luiha de closes y lucha politico en la guerra civil espahola Barcelona:
Anagrama, 1977
Rees, ‘The Highpoint of Comintern Influence? The Communist Party
T.,

and theCivil War in Spain’, in T. Rees and A. Thorpe (eds), International

Communism and the Communist International 1919-43 Manchester: Manchester ,

University Press, 1998

Regler, G., The Owl ofMinerva, London: Rupert Hart-Davis, 1959
Reig Tapia, A., Ideologia e historia. Sobre la represion Jranquista y la guerra civil

Madrid: Akal, 1986

Violenciay terror. Estudios sobre la guerra civil espahola Madrid: Akal, 1990
,

Memoria de la guerra civil Los mitos de la tribu ,

Madrid: Alianza, 1999
Rey, F., Propietarios y patronos. La politicode las organizaciones economicas en la Espaha
de la Restauracion, 1914-23 Madrid: Ministerio de Trabajo y Seguridad
,

Social, 1992
‘Capitalismo Catalan y golpe de Primo’, Hispania, 168 (1988)
Richards, M., A Time of Silence. Civil War and the Culture of Repression in Franco’s
Spain, 1936-1949 Cambridge: Cambridge University Press, 1998
,

‘Civil War, Violence and the Construction of Francoism’, in P. Preston and

A. MacKenzie (eds), The Republic Besieged. Civil War in Spain 1936-1939,
Edinburgh: University of Edinburgh Press, 1996
Richards, V, Lessons of the Spanish Revolution, 2nd edn, London: Freedom Press,

972
T

Richardson, R. D., Comintern Army. The International Brigades and the Spanish Civil
War Lexington: The University Press of Kentucky, 1982
Ripa, Y., ‘La tonte purificatrice des republicaines pendant la guerre civile
espagnole’, Identites femenines et violences politiques (1936-1946). Les cahiers de
ITnstitut d’Histoire du temps present, 31 (October 1995)
Rittersporn, G. T., Stalinist Simplifications and Soviet Complications. Social Tensions
and Political Conflict in the USSR 1933-1953, Chur, 1991
Rivas Cherif, C. de, Retrato de un desconocido, Barcelona: Grijalbo, 1979
Roberts, G., ‘Soviet Foreign Policy and the Spanish Civil War 1936-1939’,
in C. Leitz and D. J. Dunthorn (eds), Spain in an International Context
i936~i959, Oxford/New York: Berghahn Books, 1999

Robinson, R. H., The Origins ofFranco’s Spain. The Right, The Republic and Revolution
1931-1936, Newton Abbott: David & Charles, 1970
Rodriguez, E., Acotaciones para la historia del POUM, Barcelona: Fundacion
Andreu Nin, 1989
Rodriguez Miaja, F., Testimoniosy remembranzas. Mis recuerdos de los ultimos meses de
la guerra de Espaha (1936-1939), Mexico:
1997 n.p.,
Rojas, C., La guerra en Catalunya, Barcelona: Plaza & Janes, 1979
Rojo, V, Alerta los pueblos!, 1939; Barcelona: Ariel, 1974
Espaha heroica. Diez bocetos de la guerra espahola, 1942; Barcelona: Ariel, 1975
Asifue la defensa de Madrid, 1967; Madrid: Comunidad de Madrid, 1987
(Rojo) Homenaje al General Vicente Rojo en el primer centenario de su nacimiento, Sagunto:
Fundacion Municipal de Cultura, 1998
 , ,

Bibliography
459
Romero, F., Spain 1314-1318. Between War and Revolution , London: Roudedge,
1999
Twentieth- Century Spain. Politics and Society i8g8-igg8 Basingstoke: Macmillan,
1999
‘Spain and the First World War: The Structural Crisis of the Liberal
Monarchy’, European History Quarterly ,
25 (1995)
Romero, L., Tres dias dejulio, Barcelona: Ariel, 1967
Desastre en Cartagena , Barcelona: Ariel, 1971
Elfinal de la guerra, Barcelona: Ariel, 1976
c
Romero Maura, J., La Rosa del Fuego\ Republicans y anarquistas: la politica de

los obreros barceloneses entre el desastre colonial y la Semana Tragica i88g-igog ,

Barcelona: Grijalbo, 1989

Romeu Alfaro, F., El silencio roto: mujeres contra elfranquismo , n.p.: n.p., 1994
Rosado, A., Tierra y libertad. Memorias de un campesino anarcosindicalista andaluz
Barcelona: Critica, 1979
Rosal, Amaro del, Historia de la UGT de Espaha 1301-1333 2 ,
vols., Barcelona:
Grijalbo, 1977
Ruiz, D., Insurreccion defensiva y revolucion obrera. El octubre espahol de 1334,
Barcelona: Labor, 1988
Ruiz-Castillo Basala, J., Funcionario republican de reforma agrariay otros testimonios
Madrid: Biblioteca Nueva, 1983
Sabin, J. M., Prisiony muerte en la Espafia de la postguerra, Madrid: 1996
Saborit, A., Julian Besteiro Buenos Aires: Losada, 1967
,

Sacaluga, J. A., La resistencia socialists en Asturias 133^-1362, Madrid, 1986

Safon, A. and Simeon Riera, J. D., Valencia 1336-y. Una ciudad en guerra
Valencia: n.p., 1986
Sahlins, P, Boundaries: The Making of France and Spain in the Pyrenees, Berkeley:
University of California Press, 1989
Saiz Valdivielso, A., Indalecio Prieto. Cronica de un corazon, Barcelona: Planeta, 1984
Salas, N., Sevilla fue la clave: republica, alzamiento, guerra civil (1331 -1333), 2 vols.,
Seville: Castillejo, 1992
Salas Larrazabal, R., Historia del ejercito popular de la republica , 4 vols., Madrid:
Editora Nacional, 1973
Sanchez, J.M., The Spanish Civil War as a Religious Tragedy, Notre Dame, Ind.:
University of Notre Dame Press, 1987
Sanchez Albornoz, N. (ed.), The economic modernisation of Spain 1830-1330,
New York: New York University Press, 1987
Sanchez Recio, G .,Justiciay guerra en Espaha: Los Tribunates Populares (1936-1333),
Alicante: Instituto de Cultura ‘Juan Gil-Albert’, 1991
La Republica contra los rebeldesy los desafectos. La represion economica durante la guerra
civil, Alicante: Universidad de Alicante, 1991

Los quefuimos a Madrid, Toulouse: n.p., 1969

Savater,F., Para la anarquia. Y otros enfrentamientos, Barcelona: Orbis, 1984

Saz Campos, I., Mussolini contra la II Republica: hostilidad, conspiraciones, intervencion

(ig3i-ig36), Valencia: Edicions Alfons el Magnanim, 1986
460 Bibliography

Schauff, F., ‘Company Choir of Terror: The Military Council of the 1930s -
The Red Army between the XVII and XVIII Party Congresses’, Journal
(2) (June 1999)
of Slavic Military Studies 12 ,

Sedwick, The Tragedy of Manuel Azana and the Fate of the Spanish Republic n.p.:
F., ,

Ohio State University Press, 1963

Seidman, M., Workers against Work. Labor in Paris and Barcelona during the Popular
Fronts Berkeley: University of California Press, 1991
,

Semprun-Maura, C., Revolucion y contrarrevolucion en Cataluna (1936-1937),

Barcelona: Tusquets, 1978
Sender Barayon, R., A Death in £amora, Albuquerque: University of New
Mexico Press, 1989
Serrano, C., Final Espaha i8gg-i8g8, Madrid: Siglo XXI, 1984
del Imperio.

LEnjeu espagnol: PCF et guerre dEspagne (Paris: Messidor, 1987)

Le Tour du Peuple. Crise nationale, mouvements populaires et populisme en Espagne
(i8go—igio), Madrid: Casa de Velazquez, 1987
(ed.), Madrid igg6-iggg. Un peuple en resistance ou Vepopee ambigue, Paris: Editions

Autrement, 1991
Serrano, V
A. and San Luciano, J. M. (eds), Azana Alcala de Henares: ,

Fundacion Colegio del Rey, 1991

Servicio Historico Militar (Col. Martinez Bande, J. M.)
La marcha sobre Madrid, Madrid: Editorial San Martin, 1968
La lucha en torno a Madrid Madrid: Editorial San Martin, 1968
,

La campana de Andaluda, Madrid: Editorial San Martin, 1969

Laguerra en el norte Madrid: San Martin, 1969
,

La ofensiva sobre Segoviay la batalla deBrunete Madrid: Editorial San Martin, 1972
,

Nueve meses de laguerra en el Norte, Madrid: Editorial San Martin, 1980

Elfinal de laguerra Madrid: Editorial San Martin, 1985
,

La batalla del Ebro Madrid: Editorial San Martin, 1988

,

La Invasion de Aragon y el desembarco en Mallorca Madrid: Editorial San Martin,

,

1989
Shubert, A., The Road to Revolution
in Spain. The Coal Miners ofAsturias i86o-igj4 ,

Urbana/Chicago: University of Illinois Press, 1987

A Social History ofModern Spain London: Unwin-Hyman, 1990
,

Simeon Riera, J. D., Entre la rebelio y la tradicio (Lliria durante La Republicay la

Guerra Civil. ig3i-rggg), Valencia: Diputacio de Valencia, 1993
Simpson, A. W. B., In the Highest Degree Odious. Detention without Trial in Wartime
Britain Oxford: Oxford University Press, 1994
,

Smith, A., Anarchism, the General Strike and the Barcelona Labour Movement
1899-1914’, European History Quarterly 27 (1) (1997) ,

Smyth, D., ‘The Politics of Asylum. Juan Negrln in 1940’, in R. Langhorne

(ed.), Diplomacy and Intelligence during the Second World War Cambridge:
,

Cambridge University Press, 1985

Smyth, T. M., La CNT alPais Valencia igjS-iggy, Valencia: Eliseu Climent, 1977
Sola i Gussinyer, P, Educacio i Movement Libertari a Catalunya (igoi—iggg),
Barcelona: Edicions 62, 1980
 ,

Bibliography 461

Sole Sabate, J. M. and Villarroya, J., Hocupacio militar de Catalunya Barcelona:

,

L’ Aveng, 1987
Sole i Sabate, J. M. and Villaroya i Font, J., La repressio a la reraguarda de Catalunya
(1936-1939)) 2 vols., Barcelona: l’Abadia de Montserrat, 1989—90
Souchy, A., With the Peasants ofAragon, Orkney: Cienfuegos Press, 1982
Souchy, A. et al., The May Days. Barcelona 1937, London: Freedom Press, 1987
Southworth, H. R., Guernica! Guernica! A Study of Journalism, Diplomacy, Propaganda
and History, Berkeley, Calif.: California University Press, 1977
El mito de la cruzada de Franco, 2nd edn, Barcelona: Plaza & Janes, 1986
Spain 1936—1939. Social Revolution and Counter Revolution, London: Freedom Press,
1990
(The) ‘Spanish Ci vil War. The View from the Left’, Revolutionary History 4 (1-2) ,

(winter 1991 -2)

Starinov, I. G., Over the Abyss. My Life in Soviet Special Operations New York: Ivy
Books, 1995
Stein, L., Beyond Death andExile. The Spanish Republicans in France 1939-1955 •>

Cambridge, Mass.: Harvard University Press, 1979

Stone, D., ‘Ontology or Bureaucracy? Hannah Arendt’s Early Interpretations
of the Holocaust’, European Judaism, 32 (2) (1999)
Suarez, A., El proceso contra el POUM, n.p.: Ruedo Iberico, 1974
Sueiro, D., La flota es roja, Barcelona: Argos Vergara, 1983
Suero Roca, M. T., Militares republicans de la guerra de Espafia, Barcelona:
Peninsula, 1981
Suero Sanchez, L., Memorias de un campesino andaluz en la revolucion espahola,
Madrid: Queimada Ediciones, 1982
Taguena Lacorte, M., Testimonio de dos guerras, Mexico: Ediciones Oasis, 1974
Tavera, ‘La premsa anarco-sindicalista (1868-1931)’, Recerques, 8 (1977)
S.,

Tavera, and Ucelay da Cal, E., ‘Grupos de afinidad, disciplina belica y

S.

periodismo libertario, 1936-1938’, Historia Contemporanea, 9 (1993)

Tavera, S. and Vega, E., ‘La afiliacio sindical a la CRT de Catalunya: entre
l’euforia revolucionaria i l’ensulsiada confederal 1931—36’, in (various
authors) Revolucio i Socialisme, vol. 2, Barcelona: Universitat de Barcelona,
1990
Thomas, H., The Spanish Civil War, Harmondsworth: Penguin, 1977
Togliatti, P, Opere r935 -1944 , Rome: Editorial Riuniti, 1979
Escritos sobre la guerra de Espaha (Barcelona: Crltica, 1980)
Tortela, G., El desarrollo de la Espaha contemporanea. Historia economica de los siglos

XIXy XX, Madrid: Alianza Universidad, 1994

Toryho, J., No eramos tan malos, Madrid: G. del Toro, 1975
Townson, N., The Crisis of Democracy in Spain. Centrist Politics under the Second
Republic 1931-1936, Brighton/Pordand: Sussex Academic Press, 2000
(ed.), El republicanismo en Espaha (1830-1977), Madrid: Alianza, 1994

Tunon de Lara, M. (ed.), Gernika: 50 ahos despues (1937-1987). Nacionalismo,

republica, guerra civil, San Sebastian: Universidad del Pals Vasco, 1987

(ed.), Los origenes culturales de la II Republica, Madrid: Siglo XXI, 1993

 , ,

462 Bibliography

Tunon de Lara* M. et al., La guerra civil espahola. 50 anos despues Barcelona:

Labor, 1985
Tunon de Lara, M., Miralles, R. and Diaz Chico, B. N., Juan Negrin Lopez. El
hombre necesario Las Palmas: Gobierno de Canarias, 1996
,

Ucelay da Cal, E., La Catalunya populista. Imatge, cultura i politica en I’etapa

republicana (1931 -1939), Barcelona: La Magrana, 1982

‘Socialistas y comunistas en Cataluna durante la guerra civil: un ensayo de
interpretation’, in S. Julia (ed.), Socialismo y guerra civil. Anales de Historia,
vol. 2, Madrid: Editorial Pablo Iglesias, 1987
‘Cataluna durante la guerra’, in E. Malefakis (ed.), La guerra de Espana
i936~i939, Madrid: Taurus, 1996
Ugarte J., La nueva Covadonga
Telleria, insurgente. Origenes sociales y culturales de

la sublevacion de 1936 en Navarra y el Pais Vasco ,

Madrid: Biblioteca Nueva,
99 8
!

Ullman, J. Connelly, The Tragic Week: A Study ofAnticlericalism in Spain 1875-1912,

Cambridge, Mass.: Harvard University Press, 1968
Vargas, B., Rodolfo Llopis (1895-1983) Una biograjia politica, Barcelona: Planeta,
1999
Vazquez M. and Valero, J., La guerra civil en Madrid, Madrid: Tebas, 1978
Vazquez Montalban, M., Barcelonas, London: Verso, 1992
Pasionariay los siete enanitos Barcelona: Planeta, 1995
Vega, C., Monjo, A. and Vilanova, M., ‘Socialization y hechos de mayo’,
Historiay Fuente Oral, 3 (1990)
Vega, E., El trentisme a Catalunya. Divergencies ideoldgiques en la CNT (1930—1933),
Barcelona: Curial, 1980
Anarquistasy sindicalistas durante la Segunda republica. La CNTy los Sindicatos de

Oposicion en el Pais Valenciano, Valencia: Institucio Alfons el Magnanim,

1987
Vega Diaz, F., ‘El ultimo dia de Negrin en Espana’, Claves de Razon Practica, 22
(May 1992)
Ventin Pereira, J. A., La guerra de la radio (1936-1939), Barcelona: Editorial
Mitre, 1986
Vidal, C., Las Brigadas Internacionales, Madrid: Espasa Calpe, 1998
Vidarte, J. S., Todosfuimos culpables, 2 vols., Barcelona: Grijalbo 1978
Vignaux, P., Manuel de Irujo , Ministre de la Republique dans la guerre d’Espagne
1936-1939, Paris: Beauchesne, 1986
Vila Izquierdo, J., Extremadura: la guerra civil, Badajoz: Universitas Editorial, 1983
Vilanova, M., Las mayorias invisibles. Explotacion fabril, revolucion y represion,

Barcelona: Icaria, 1996

‘Anarchism, Political Participation and Illiteracy in Barcelona between 1934
and 1936’, American Historical Review, 97 (1) (1992)
Villarroya i Font, J., Els bombardeigs de Barcelona durant la guerra civil (1936-1939),
1981; 2nd edn, Barcelona: L’Abadia de Montserrat, 1999
Vinas, A., El oro espahol en la guerra civil, Madrid: Instituto de Estudios Fiscales,

1976
 ,

Bibliography 463

El oro de Moscu. Alfay omega de un mito franquista, Barcelona: Grijalbo, 1979

Guerra, dinero,dictadura. Ayuda fascista y autarquia en la Espaha de Franco

Barcelona: Critica, 1984

Franco, Hitler y el estallido de la guerra civil. Antecedentes y consecuencias, Madrid:
Alianza Editorial, 2001
Vinas, A., ‘Gold, the Soviet Union and the Spanish Civil War’, European Studies
Review, 9 (1) (Jan. 1979)
‘The financing of the Spanish Civil War’, in P. Preston (ed.), Revolution and
War in Spain 1931-1939, London: Methuen, 1984
‘Las relaciones hispano-ffancesas, el gobierno Daladier y la crisis de Munich’,
Espaholesyfranceses en la primera mitad del siglo XX, Madrid: CEH-CSIC, 1986
Vinas, R., Laformacion de las Juventudes Socialistas Unificadas (1934-1936), Madrid:
Siglo XXI, 1978
Vincent, M., Catholicism in the Second Spanish Republic. Religion and Politics in

Salamanca 1930-1936, Oxford: Clarendon Press, 1996

‘The Martyrs and the Saints: Masculinity and the Construction of the
Francoist Crusade’, History Workshop Journal, 47 (1999)
Vinyes, R., La Catalunya Internacional. El frontpopulisme en Vexemple catala,

Barcelona: Curial, 1983

.
‘“Nada os pertenece ” Las presas de Barcelona, 1939—1945’, Historia
. .

Social, 39 (2001)

Zafon Bavo, J., El Consejo Revolucionario de Aragon, Barcelona: Planeta, 1979

Zaragoza, C., Ejercito popular y militares de la Republica (1936-1939), Barcelona:
Planeta, 1983
Zugazagoitia, J., Guerra y vicisitudes de los espaholes, Buenos Aires, 1940; 3rd edn,
Barcelona: Critica, 1977

NOVELS
Barea, A., La forja de un rebelde, 3 vols., Madrid: Turner, 1984
Bates, R., Lean Men, Harmondsvvorth: Penguin, 1934
Lewis, N., The Day of the Fox, London, 1957 and reprints
Malraux, A., Days of Hope, 1938; Harmondsvvorth: Penguin, 1982
Olaizola, J. L., La guerra del General Escobar, Barcelona: Planeta, 1983
Vazquez Montalban, M., El pianista, Barcelona: Seix Barral, 1985
Wolff, M., Another Hill, Urbana/Chicago: University of Illinois Press, 1994
 Index

Aalto, Bill, 343 n.93 construction of Republican (Popular) Army,

Abad de Santillan, Diego, 227 n.47, 228 138-44, 147-9, 176, 177, 178, 179, 180,
Accion Catolica, 30-1 208
Aguirre, Jose Antonio, 32, 131 n.2, 243, 244, ideology of officer corps, 1, 12-13, 120-1
245, 246, 248, 249, 308-11, 349 professional officers in Republican Army,
Aguirre, Jose Maria, 134 144-6, 149-50, 323-4
Aiguader, Artemi, 233, 238, 261, 266, 268, reorganisation after territorial division
268-9 n.66 ( 1938 ), 358

Aiguader, Jaume, 287 n.136, 380 n.239 Republican officers and Casado coup, 400,
Alba, duque de, 384 401, 408 n.55, 410, 414
Alcala-Zamora, Niceto, 19, 20 {see also International Brigades)
Alianza Obrera (AA OO) (Workers’ Alliance), Ascanio, Guillermo, 408 n.55, 4 10 4 12 413 , ,

54-8,74,76,98,430 n.72
Alicante, 91, 100, 106, 264, 414, 421-3 Ascaso, Domingo, 277, 281
Almeria, 106, 264 Ascaso, Francisco, 39, 232, 262 n.41, 276-8
Alvarez del Vayo, Julio, 133, 166, 212, 362, Ascaso, Joaquin, 232, 278 n.99
368, 398, 403 Asensio, General Jose, 143, 146, 165, 168, 207,
Andalusia, 6, 7, 94, no, 120 208-9, 323 n -27, 385
Andrade, Juan, 275, 285, 289-90, 291 Asociacion de Mujeres Antifascistas (AMA),
n T 57
-
59, 180, 182, 183,426
anti-clericalism, 5-6, 10, 28-30, 77 Assault Guards, 37, 38 n.2, 79 n.4, 92, 94,
after 18 July 1936, 84, 85-6 94 nn.63-4, 191, 191 n.232, 203, 218, 234,
Antona, David, 400 261, 285
Antonov-Ovseenko, Vladimir, 237, 239 Asturias, 2, 8, 47, 71, 127, 183, 246, 247, 314,
Aragon, 91, 92, 93, 231, 272, 273-4, 296, 323* 323
352, 353,356, 358, 359, 364 and military rebellion, 94
anarchist collectivisation in, 7, 101 n.97, 102, October 1934 rising in, 57-8, 62-4, 66, 109,
103, 139, H0 , 149, 224-6, 230-2, 314-15 132
Council of Aragon in, 231, 232-3, 297, and workers’ committees in, 97-100, 100
312-15, 335 n.87
‘great retreats’ in (1938) 375, 386 Austria, 80, 357,
358
political competition231-2, 313, 314
in, Azana, Manuel, 25, 27, 33, 72, 134, 135, 160,
Araquistain, Luis, 44, 72-3, 133-4, 156-7, 302, 164, 166, 172, 173, 207, 280, 313, 322, 336,
3°3, 305, 307 338, 35°> 360, 362, 378, 402-3 plate 10 ,

Army, Spanish attitudes to the state, 25

Army of Africa, 104, 105, 108, no, in, 113, becomes Republican president, 73
1 15—17 , 120, 122, 177 and discontent with Largo/May 1937
collapse in Republican territory, 79, 83, cabinet crisis, 298, 299-300, 301, 302
137-8 and gold reserves, 154
commissars in Republican (Popular) Army, and Largo’s nomination as prime minister,
146-7, 33!-3, 355 i3b 132, 132 n.5

464
 1

Index 465

leaves Spain, 397 and Non-Intervention, 124-5, I2 6, 316-17

and military rebellion, 80, 81-3 pro-rebel sympathies, 105 n.115, 116,
and negotiated peace, 213, 318, 359, 363, 316 n.2
3 6 9. 37°-2> 39b 403 recognises Franco, 409
and Nin’s disappearance, 286 n.134 and republican refugees, 419, 421-2
and Popular Front, 63, 64 Brunete, battle of, 319, 323
relations with Negrin, 305-7, 370-2 Bueno Nunez de Prado, Emilio, 410, 412,
resigns as Republican president, 41 4 J 3 n -7 2
Aznar, Rita, in n.136 Bugeda, Jeronimo, 154, 155 n.91, 159 n.103,
307
Badajoz, 92, 1 1 1 —13, 116, 118, 128 Buiza, Admiral Miguel, 401, 408
Baldwin, Stanley, 316 Burillo, Ricardo, 287 n.135
Balearic Islands, 95 n.67
Baraibar, Carlos de, 44 Caceres, 95, 111

and nationalist rebellion in Morocco, caciques, 5, 5 n.8, 9, 19, 20

300 n.192 CADCI (Barcelona shop and office workers’
Barbieri, Francesco, 294 union), 224, 263 n.45, 426
Barcelo, Luis, 408 n.55, 410, 413, 415 Cadiz, 104
Barcelona, 2, 8, 9-12, 13, 35, 37-8, 39, 54, 68, Caja de Reparaciones, 338
70, 91, 104, 135, 163, 171, 201, 210, Camacho, Antonio, 401, 412
215-18, 231, 234, 252-3, 254, 258-61, Carabineros (Customs police), 202-3, 202
262-6, 314, 351-2, 359, 364 n.267, 234, 261, 307, 380, 426
collectivisation in, 215, 223-4, 243, 263-5, Carcel Modelo (Model Prison, Madrid), 83,
284 128, 160, 189, 190, 192, 192 n.235
to rebels, 395-7
falls Carlists, 31-2, 83 n.2 1, 95, 426
government moves to, 336-8 Requetes, 93, 117, 429
May days in (1937), 266-83 Carrillo, Santiago, 76 n.154, 169, 170, 172,
and military rebellion, 79, 92-4 181-2, 194 plate 6,

subsistence crisis in, 253, 254-6, 257, 258-9, and Paracuellos killings, 190-1, 193, 193
260 n.238
and workers’ committees, 96—7, 101-2 Cartagena, 91, 106, 152, 408-9
Basque Country, 2, 8, 31, 32, 33, 127, 149 n.66, Casado, Colonel Segismundo, 143, 392, 401,
183, 190 n.226, 201, 214, 240-53, 308-12 402, 405, 406, 421-2
wartime divisions in, 95, 241, 243 attempts at peace negotiations, 417-20
Belchite, batde of, 323, 356 coup, 407-10, 412-13
Berneri, Camillo, 294-6 leaves Spain, 42 1 -2
Besteiro, Julian, 362 n.167, 371-2, 381, 391-2, tries communists, 417
400, 401, 413 Casares Quiroga, Santiago, 80-1, 82, 83
Bilbao, 8, 10 n.15, 31, 63, 171, 190 n.226, 241, Castellon, 91, 100, 346, 371
244, 249, 252, 253, 308-9 Catalonia, 2, 10, 12, 17, 92, 100, 104, 129, 139,
falls to Franco (June 1937), 311-12 140, 149, 167, 182-3, i8 3 n.197, i 8 7> i9 6 >
and military rebellion, 95, 241-2 201, 214, 215-24, 226, 233-4, 238, 240,
Blanco, Segundo, 363 254-84, 296, 319, 319 n.12, 335, 337,
Bloc Obrer i Camperol (BOC), 17, 48-9, 34°-b 34 6 349-5°> 35 b 3 6 4> 3 6 5>
?

49 n 6 9> 55 n -95> 5 6 n -9 8 426

-
> 379-80
attempts at merger of Catalan left, 66-7, Autonomy Statue in, 33, 34, 282, 335
221-2, 226, 236 conflict over POUM in, 198-9
and Workers’ Alliance, 54-7, 66 falls to rebels, 395-8
(see also Joaquin Maurin; POUM) influence of trientistas in, 48-9
Blum, Leon, 125, 316, 358, 367 Popular Front in, 65-70, 337
Bonnet, Georges, 367-8 war industries in, 283, 335-6
Britain, 129, 136, 164, 165, 212, 319, 320, Workers’ Alliance 54-7 in,

321-2, 343, 357-9, 364, 367, 373, 383, workers’ committees in, 101-2
397,39^ 402-3 Catholic Church, 4-6, 10, 27-31, 124, 340-1,
and Casado coup, 419 386
 1

4 66 Index

Cazorla,Jose, 169, 194, 195-6, 200 ‘invertebrateness’, 47, 49, 50, 51, 219-20,
Centano de la Paz, Jose, 420 278
Chamberlain, Neville, 316, 321, 357, 368 and Largo government, 135-6, 164-5, 298
Checa, Pedro, 410-11, 414, 417 and May cabinet crisis, 302-5
checas, 284, 290, 377 and May days (1937), 266-84
Ciutat, Captain Francisco, 246, 247 membership crisis in, 48
Civil Guard, 6, 9, 38, 38 n.32, 40, 40 n.44, and military rebellion, 79, 92-5
79 n.4, 84, 92, 94, 94 nn. 63-4, 95, 109 and negotiated peace, 372
n.129, 138 n.25, 161, 218, 234 and PCE, 135, 194-200, 354, 374-5
Claudin, Fernando, 181, 41 1, 414 plate 6 ,
and Popular Front, 7
clientelism (as political culture), 147, 158, 17 1, and popular violence/ incontrolables, 87-9,
182, 197, 263 n.46, 293, 294 n.168, 328-9, 293
332, 339? 34 1-2 , 376, 381, 400, 415 and POUM, 225-7, 236, 237, 238
CLUEA (Valencian fruit-exporting and registration of collectives, 230-3
cooperative), 234 n.73, 312 n.234, 335 rivalry with UGT, 42-3, 45-6, 48,
8, 14,
Codovilla, Vittorio, 61, 134 n. 13, 203-5, 210, 5^4, 55, 9b 162, 227, 262 n.44, 292
211,328, 330 similarity with UGT, 16
collectivisation, 96, 97, 100-4, 225, 230-3, sindicatos unicos (industrial unions), 15, 50,
- 278
3 2 5 7 353-4 5

Comintern, 60, 68, 70, 76, 135, 169, 185, 192, support for Casado, 400
197-8, 203-4, 214, 222, 239 n.93, 240, trientistas in, 48-9, 51, 54, 70, 100, 167
285, 289, 326, 327-8, 330, 334, 367, wartime centralisation of, 275-6, 278
386-7,394, 404-5,427 and workers’ committees, 97-104
and Popular Front, 59, 61, 68-9 Zaragoza congress, 51, 136
and POUM, 198-9, 235, 284-6, 290, Cordoba, 108, no, 168
291-2, 296, 345 Cordon, Antonio, 143, 143 n.41, 145, 146
Comorera, Joan, 66, 67-8, 69-70, 221-3, 225, Cortada, Roldan, 262, 266, 292 n.160
2 33, 237, 238, 240, 256 Cuatro Vientos, 93, 126
Companys, Luis, 92, 217-18, 220-1, 227, 229, Czechoslovakia, 358-9, 383
237-9,263,335
and conflict with Negrin, 335-8, 349, 380 Daladier, Edouard, 367-8
and May days (1937), 269, 270-1, 279-80, Diaz, Jose, 60-1, 74 n.150, 365 n.179, 366 n.180
282, 283 Diaz Tendero, Captain Eleuterio, 138
(see also Catalonia, war industries in; Dieguez, Isidoro, 139 n.29, 184, 406, 411, 417,
Esquerra; Generalitat) 418 n.84
Conesa, Jose, 413, 415 ‘Disaster of 1898’, 2, 9
Confederacion Espanola de Derechas Durruti, Buenaventura, 39, 177-9, 277-8 platej ,

Autonomas (CEDA), 30, 31, 32, 43, 58,

65, 426 Ebro, battle of the, 373-4, 382, 383-4, 387,
Confederacion Nacional Catolico-Agraria 39 b 392 395 ,

(CNCA), 31,31 n.13 Eden, Anthony, 316 n.12, 357

Confederacion Nacional del Trabajo (CNT), Escobar, General Antonio, 281, 281 n.m, 401
6-7, 8, 9, 27, 38, 42, 66, 70, 76, 83, 87, Escofet, Frederic, 92,94 n.94
90, 128, 13 1, 134, 140, 163, 166, 167, 184, Esquerra Republicana de Catalunya (ERC),
200, 208, 236, 350, 352-4, 360, 362, 33-4, 48, 56, 58, 68, 70, 131, 218, 218 n.io,
427 221, 223, 227, 227 n.45, 229, 233, 234,
in Aragon, 224-6, 312-14 236, 237, 238-9, 256, 263, 266 n.55, 268
in Barcelona/ Catalonia, 11-12, 68, 216-21, n.65, 302 n.200, 335, 338, 349, 380,
218 n.12, 223-4, 227-30, 238, 240, 427
255-6, 257, 259-60, 261-4, 292, 294-6, and May days (1937), 269, 270-1, 279, 282
297,396 and Popular Front, 65-6
inBasque Country, 248, 252, 308-9 (see also Catalonia; Companys, Luis;

and defence of Madrid, 170, 177-9 Generalitat)

internal differences in, 14-16, 47-51, 226, Euskadi(see Basque Country)

228, 275-6 Extremadura, 6, 7, 95, in

 1

Index 467

Falange, 93, 94, 106, 108, no, 112, 117, 123, 427 Gero, Erno, 69, 222, 330
column, 287, 342-3, 400-1
as fifth Getafe, 93, 126, 132
Fanjul, General Joaquin, 93 Gibraltar, 106
Federacion Anarquista Iberica (FAI), 15, 16, Gigi-Bibi, 295

39, 49-5°, 7b 97 n -74, 164-5, 217-19, 217 Gijon, 14, 57, 98, 99 n.86, 100, 312, 323
n.8, 218 n.12, 227, 259, 276 n.91, 278, 427 Giral,Jose, 82-3, 126, 13 1, 137-8, 362, 363,
and May days (1937), 272, 275-6 399, plate 10
and negotiated peace, 372 and goldreserves, 150-1, 154-5
and popular violence/ incontrolables, 87-9, Girauta Linares, Vicente, 19
293 Giron, Domingo, 406, 409
Federacion de Juventudes Socialistas (fJS), Girona, 67, 215 n.2, 226, 398
- Goded, General Manuel, 92
75 6 83, i8 5 186
. >

Federacion Nacional de Trabajadores de la Goff, Irving, 343 n.93

Tierra (FNTT), 42, 45, 58, 71, 102, 103, Gomez, Paulino, 363, 380, 399 n.28
103 n.103,305 n.210, 427 Gonzalez Pena, Ramon, 363, 372, 399 n.28
Federacion Universitaria Espanola (FUE), Gorkin, Julian, 292 n.159, 346 n.no, 347 n.112,
17 385
Fifth Regiment (Quinto Regimiento), 84, 140, Gracia, Anastasio de, 133, 302
147-9, 170, 176, 177, 180, 427 Granada, 106-8, no
First World War, 1, 3, 6, 9, 11, 12 Gremis i de Petits Comerciants i
Entitats
France, 4, 23, 127, 129, 136, 152, 164, 165, 176, (GEPCI), 258, 263, 427
Industrials
212, 319, 322, 343, 358, 359, 360, 364, Guadalajara, 94 n.64, 101 n.97, 190 n.226
-
3 6 5 . 3 6 7 8 3 8 3 397 5 4 ° 3 421-2
> , ,
battle of, 212-13
and Non-Intervention, 125, 126, 318, 320, Guernica, 308
356-7 Guipuzcoa, 241-2, 251
petitioned for arms, 83, 105 n.115, 124-5 Gustavo, Soledad, 165
recognises Franco, 409 guerrilla warfare, 150 n.70, 343 n.93
Franco, General Francisco, 58, 104, 105, 109,
no, 113, 120, 124, 150 n.70, 164, 177, 179, Hernandez, Jesus, 166, 167 n.140, 360, 363,
2i3> 3° 8 322, 356, 357, 362, 364-5,
, 393 4 i 7
,

- - Hidalgo de Cisneros, Ignacio, 364, 391

3 6 7 8 370 373 3 8 2 3 8 4 395 8 412,
, , , , , ,

412 n.70, 418-425 Hitler, Adolf, 105, 212 n.313, 322, 383
Huelva, 109 n.129
Gaikis, Leon, 152 n.76, 213 Huesca, 224
Galarza, Angel, 133, 189, 191, 295
Gamir Ulibarri, Manuel, 310 Ibarruri, Dolores, 61 n.109, 148 n.63, 166
Garces, Santiago, 376 n.226, 398 n.137, 184 n.201, 330, 360 n.157, 365
Garda Oliver, Juan, 164-5, *66 n.133, 184, n.179, 366 nn. 183-4, 367 n.185, 404
262 n.41, 350—1 International Brigades, 175-6, 177, 185, 197,
and May days (1937), 270, 271-2, 273, 276, 201, 356, 376 n.222, 383, 384
277-8 Iron Column (Columna de Hierro), 88, 178
Generalitat, 34, 56, 58, 70, 92, 167, 217-19, n.182, 276
220, 227, 229, 230, 232, 233, 234, 237, Irujo, Manuel de, 131 n.2, 242, 244, 245 n.no,
239, 256 n.15, 257, 258, 260-1, 264, 267, 285, 290, 298 n.189, 339, 345, 348-9,
278, 335 338 345 349-5 °, 352 n.131,
, , , 363, 380 n.239
379 427 ,
Inin, 127, 244, 247
and May days (1937), 268-9, 270, 271, 274, Italy, 12, 80, 176, 316, 351, 357-8, 364, 366,
279, 280-2, 284 423
Catalonia, war industries in;
(see also aid to rebels, 105-6, in, 115 n.144, 124, 124
Gompanys, Luis; Esquerra) I ° n
n.179,5 7 °, 175 2I 3 3°8, 3 10
-
, , ,

Germany, 35, 77, 80, 123-4, 125, I7 6 3 64 , , 317-18, 319, 320, 322, 335, 365, 373
366, 383 Izquierda Communista (Left Communists), 55,
aid to rebels, 105-6, in, 115 n.144, 124, 150 66, 67, 198 n.258, 235, 428
n 7 °, 175 213, 3 IQ ,
-
, 322, 382 Izquierda Republicana (Left Republicans), 82,
Condor Legion, 179, 308, 311 170, 302 n.200, 313, 401, 428
 3 1

4 68 Index

Jaen, 92, 94 n.64, 101 n.97, 103, 106 Llano de Encomienda, General Francisco,
Jimenez de Asua, Luis, 39 n.41, 358, 370, 381 247,310
Juventudes Socialistas Unificadas (JSU), 18, Lleida, 67, 215 n.2, 226, 364
76, 83-4, 148, 169, 174, 176, 181-2, 183, Lliga Regionalista, 33 n.17, 34 n.20, 249 n.127
184, 214, 303, 366, 382, 415, 428 Lopez, Anita, in n.136
Lopez, Juan, 164, 166
Kollontai, Alexandra, 165
Koltsov, Mijail, 141 nn. 34-5, 169 n.150, 192, Madrid, 2, 8, 13, 47, 53, 63, 67, 75, 76, 79, 89,
210 n.299 >
plate 104, 105, no, 116, 126-8, 137-8, 140, 141,
Krausism, 25 H2 ,
J 59, 198-200, 254, 264, 265, 342
and Casado coup, 409-13
Lamoneda, Ramon, 186, 305, 307 nn. 217-18, entry of Franco’s armies, 420
333, 361 n.161, 381, 391, 392 n.8 and military rebellion, 79-84, 92-4
Largo Caballero, Francisco, 27, 40-1, 53, 62, siege of, 150-1, 164-5, 167-82, 187-8,

75, 163, 167, 168, 186, 189, 200, 201-2, 189-96, 201, 215
207-8, 232, 243, 244, 245-6, 309, 310, workers’ committees in, 96-7

132 , 354-5 pkte ,

I. Malaga, 90, 92, 94 n.64, 2 54, s64
becomes prime minister, 129, 13 1-4 falls to the rebels, 205-9

breaks with republicans, 43 Marchenko, Sergei Grigoryevich (Soviet

brings CNT into government, 135—6, 164, charge d’affaires), 152 n.76, 342 n.90, 375
228 n.222
and Casado coup, 415 Martin, Antonio, 262, 268
and gold reserves, 154 Martin Blazquez, Jose, 143, 146
and May cabinet crisis, 299-303 Martinez Anido, General Severiano, 1

and May days (1937), 270, 271, 280 Martinez Barrio, Diego, 82, 95, 300, 305
and militarisation, 138-43, 146-7 Martinez Cabrera, General Toribio, 208, 209
moves government to Valencia, 165-6 Martinez Monje, General Fernando, 208, 209
opposition to Prieto becoming prime Marty, Andre, 135 n.15, 177 n.179
minister, 73-4 Matallana, General Manuel, 395, 398, 399,
political worldview, 136-7, 299, 303—5 401, 407-8, 407 n.52
and Popular Front, 64, 72-3, 212 Mateo Merino, Pedro, 398 n.24
relations with CNT, 8, 13-14, 16, 27, 39, Maura, Miguel, 19, 339
45-8, 51-4, 140, 302-5, 372 Maurin Joaquin, 55 n.95, 66-7, 236
relations with Miaja, 199-200, 199 n.261 Mendez Aspe, Francisco, 154, 338 n.77, 398,
relations with PCE, 203—5, 209-12, 214, 399
297-300, 327-8, 329, 333, 381 Menendez, General Leopoldo, 401, 414
and socialist left (caballeristas), 43-7, 62-3, Mera, Cipriano, 178 n.179, 410, 412
64, 132-4, 304-5, 355 Merida, in
and socialist youth (IJS), 75-6 Mexico, 154
targeted by employers’ federation, 42 Miaja, General Jose, 113, 209, 301, 395, 398,
trial after October 1934,
59 399, 401,407 n.52 plate 10 ,

wartime anti-communism, 14 1, 210-12, in charge of defence of Madrid, 167-73, l

l8
303-4,328,329,355,415-16 relations with Largo, 199-200, 199 n.261
latifundios, 3, 5, 6, 102, 109, 120 Mije, Antonio, 61 n.109, 140-1, 169 n.151,
Law for the Defence of the Republic, 37, 38 169-70, 184, 193 n.238, 205
Law of Political Responsibilities (February Milicias Antifascistas Obreras y Campesinas
397,403,418
!939)> (MAOC), 76, 84, 147-8, 428
League of Nations, 322, 368 Spanish
Military, (see Army)
Left Communists (see Izquierda Communista) Minorca, 398
Left Republicans (see Izquierda Republicana) Model Prison (see Carcel Modelo (Madrid);
Lerroux, Alejandro, 10-11, 27 Preso Model (Barcelona))
Ley de Fugas, 11, 276 n.94 Modesto, Juan, 170, 402
Ley de Vagosy Maleantes (anti-vagrancy Mola, General Emilio, 82, 93, 105, no, 111,
legislation), 39, 71 117, 124, 126, 192, 308
Lister, Enrique, 402 Montiel, Francisco, 410 n.61
 Index 469
Montoliu, Fernando, 410 n.61 NKVD, see Soviet Union
Montseny, Federica, 165, 184 Non-Intervention, 125, 125 n.182, 129, 135,
Moriones, Domingo de (Marques de 150, 150 n.70, 151, 153, 155, 157-8, 185,
Oroquieta), 401 201, 204, 257, 305, 316-22, 347, 357, 368
Morocco, 4, 79, 104, 105, no, 300, 300 n.192 effects on Republic summarised, xi-xii,

Mozos de Escuadra, 94, 94 nn. 63-4 157-8, 282, 310, 318, 323, 352, 388-9
Munich agreement (September 1938), 383-4,
387-8, 392, 393, 396 Orlov, Alexander, 150 n.70, 192 n.235, 287
Mussolini, Benito, 12, 105, 1
13 n. 142, 213 n.136, 288-9, 32i n.20, 344 n.99, 346
n -3 I 3, 3 22 357
>
n.no, 347 n.m, 376 n.224
Ortega, Antonio, 287, 290, 408 n.55, 410, 413
National Republican Guard, 203, 234, 261 n.72
Negrin, Juan, 73, 74, 90, 126, 132, 133, 152, Oviedo, 97,99, 129, 183
159, 173, 186, 202-3, 234, 254, 302, 309, and military rebellion, 94, 94 n.64
324, 33°, 332, 34M, 354-5, 357, 359 .
OVRA (Italian intelligence service), 295-6,
376, 378-8o, 382, 387, 395-6, plate 10 342
becomes prime minister, 305-7
building the state, 155-9 (
see a ^° Negrin, Pact of San Sebastian, 20 n.47, 33
political worldview; Negrin, wartime Paracuellos del Jarama, 190-1, 191 n.227, 192,
judicial policy)
J93
and Casado coup, 407-8, 411-12 Partido Comunista de Espana (PCE), 9, 46, 90,
and Catalonia, 254, 335-8, 379-80 127 133, 134, l6 6, 198-99, 200, 207-8,
,

and Council of Aragon, 312 237, 247, 307, 3 X 3, 352-3, 360, 363,
diplomacy, 316, 318-19, 321, 322, 357-9, 380-1, 387, 392-5, 399-400, 402, 429
3 6 8-9. 373. 378 n.222, 383-4, 397 appeal of, 75, 176, 182-4, 185-7, 213-14,
and freedom of worship, 339-41 327,328-9,416
and gold reserves, 154-6 army officers and, 145, 147
and ‘invisibility’, 159 n.ioo and assassination of Nin, 289-90, 292
and Nin assassination, 290-1, 344-5, 346 and Asturian rising (October 1934), 59
and PCE, 324-5, 360, 365-7, 393, 399-401, in Basque Country, 247
403-4, 405, 412 and Casado coup, 406-7, 409-14, 415-18
and political worldview, 132—3, 159, 161, 163, and Catalonia, 65-7, 68-9, 221, 221 n.19,
338-42, 350 (see also Negrin, building the 336, 337
state; Negrin, wartime judicial policy) and CNT, 135, 194-200, 332, 353, 374-5,
and POUM trial, 384-6 400, 410, 415
and PSOE, 155, 307, 364, 380-2, 390-3, and collectivisation, 230, 231-2, 313, 325-7,
394, 403 352
relations with Araquistain, 156-7 and Comintern, 59, 60, 61, 197, 204, 205,
relations with Azana, 306, 369-72 285, 297, 327, 328, 333, 334, 387, 404,
relations with Prieto, 132, 155, 306, 307, 4°5, 4 IQ
360-2, 364 n.172, 371-2, 381 and defence of Madrid, 169-77, 184
reshuffles cabinet (April 1938), 362-4 and FJS, 75-6, 185
reshuffles cabinet (August 1938), 379-80 and Jrente unico 59, 404
,

resistance policy, 318-19, 360, 361-2, limited growth of before 1936, 46, 59-61
370-2, 382, 390-1, 397, 399, 402, 403 and militarisation, 140-1, 145-6, 147-9, 239
and Soviet credits 368-9, 391 and military rebellion, 84, 92
temporarily departs Spain, 397-8 and political commissars, 146-7
and Thirteen Points (May 1938), 366 and Popular Front, 59, 60-2, 64, 71, 74-6,
wartime judicial policy, 339, 345,
347-9, 203, 324, 325-8, 355, 388
378-9, 384-6 relations with Largo Caballero, 13 1, 132,
Nelken, Margarita, 186, 205, 410 n.61 203-5, 209-12, 214, 297-303, 327-8,
Nin, Andreu, 55, 66, 67, 198 n.258, 235, 329, 333
236-7,385 relations with Negrin, 324-5, 334-5, 336,
assassination, 287-90, 344-6 360, 365-7, 393, 399-401, 401, 402,
and May days (1937), 274-5 403-4, 405, 412
470 Index

Partido Comunista de Espana (PGE)' (cont.) the Motorizada (socialist militia), 83, 128
relations with POUM, 198, 289, 291-2, 299, and Negrin, 155, 307, 364, 380-2, 390-3,
384 394) 403
relations with Prieto, 203, 301, 307, 328, and October 1934 rising, 58-9
329> 33 1 -3 and PCE, 46, 59-61, 184-7, 201-5, 214,
relations with PSOE, 46, 59-61, 184-7, 307, 328, 327-30, 376, 380, 387, 414-15
201-5, 214, 307, 328, 327-30, 376, 380, (see also Lamoneda, Ramon; Largo
387,414-15 Caballero, Francisco; Negrin, Juan;
relations with PSUC, 68, 337 Prieto, Indalecio)
and resistance policy, 355, 366-7, 387, 392, Partido Socialista Unificado de Cataluna
393) 394) 4°°) 403-7 )4 l8 (PSUC), 67-70, 182-3, 210, 218 n.12,
and SIM, 375-6, 387 221-3, 239-40, 256-7, 259, 262-3, 265,
and UGT, 61, 62, 63, 75, 303, 327-8, 329, 268 n.65, 294-5, 325, 335, 337, 380, 396,
354) 3^3) 4I4-I5 429
and wartime mobilisation, 180, 183-4, and CNT, 223, 225, 229, 233-4, 256-8,
324-5, 327 279, 280-1
and Workers’ Alliance, 55 Comintern accepts separate membership of,

(see also Fifth Regiment) 68, 222 n.24

Partido Nationalista Vasco (PNV), 31 n.14, expanding membership, 221 n.19, 223, 239
n -59, l 3 *> l 3 l n 2 2 4i-53,
3 I_3, 34, 9 1 -
,
n.92, 262-3, 265
302 n.200, 309, 31 1, 429 and May days (1937), 269, 270-1, 279, 281,
Partido Obrero de Unification Marxista 282
(POUM), 49, 67, 68, 92, 178 n.180, 215 and POUM, 236-40, 291-2, 292 n.159
n.2, 218 n.12, 221-2, 223, 225-7, 234, Partit Catala Proletari (PCP), 69, 221

239 n.92, 259, 314, 429 Pascua, Marcelino, 152, 152 nn. 77 and 79,
conflict with other groups in Catalonia, 211-12, 211 n.308, 354 n.141, 369, 385
235740 n.258, 403
exclusion from Madrid Defence Council, Pasionaria, La (see Ibarruri, Dolores)
I9M Peiro,Juan, 47, 47 n.62, 49 n.72, 88 n.45, 164,
and May days (1937), 273, 274-5, 2 79, 164 n.127, 166, 166 n.133
296-7 Perez, Antonio, 415
repression of, 284-92, 299, 344-7, 349~5° Pestana, Angel, 47, 47 n.62, 49, 170, 198-9
trial of (October 1938), 350, 374-5, 384-6 Pinuela, Fernando, 393-4
(see also Bloc Obrer i Camperol; Maurin, Popular Front, 46, 49, 53, 56, 58 n.101, 59,
Joaquin; Nin, Andreu) 64-8, 70, 71, 72, 73, 77, 84, 85, 88, 90, 91,
Partido Republicano Radical (Radical 98-9, 100, 108, 128, 134, 135, 136, 140,
(Republican) Party), 10, 27, 54 150 n.70, 157, 160, 170, 182, 183, 212-13,
Partido Socialista Obrero Espanol (PSOE), 8, 214, 230, 232, 259, 302, 313, 325, 326,
11, 16, 20 n.47, 23, 31, 42-3, 45, 136, 137, 327, 328, 330, 333-4, 354, 366, 388-9,
155, i6 5, 2 °3) 208, 33 6 354-5, 3 62
, , 393) 4°°) 4 : 5) 425 n i0 7
391-4, 429 in Asturias, 98-9
anti-Catalanism/centralism of, 54-5, 68 in Catalonia, 33, 56-7, 65, 66, 69-70, 259
and Basque Country, 242, 243 and Comintern, 59, 61, 328, 330
‘bolshevisation’ of, 44 and mass political mobilisation, 71, 77
and Catalonia, 11 n.21, 65, 69-70, 221, 222, and PCE, 60-2, 64, 71, 74, 76, 182, 183-4,
271 201-2, 203, 214, 325-6, 328, 330, 333,
and defence of Madrid, 170, 184-5 334) 3 88 404) 4 l6
)

fear of mass political mobilisation, 26—7, 60, and POUM, 66-7, 235, 235 n.79
77, 185-6 and PSOE/UGT, 62-5, 71, 72-3, 201-2,
internal divisions in, 43-4, 62—3, 73-4, 203, 214, 333, 366
132-3,212,298-9, 303-4 reconstituted (January 1937), 201-2, 203,
and Largo Caballero becoming prime 214
minister, 131, 132, 133 structural function of, 61, 333-4
and May cabinet crisis, 298-9, 301-3 wartime erosion of, 329, 334, 354, 388-9,
and military rebellion, 79-80, 83-4, 90-2 393, 400, 415
 1

Index 47i

Portella VaUadares, Manuel, 339 Rodriguez Vega, Jose, 415

Portugal, hi, 113, 125 n.182 Rojo, General Vicente, 143, 172, 174, 209
Pozas, General Sebastian, 167, 208 n.293, 297 n 2 97, 309. 3 2 4> 364, 3 8 7, 395> 396,
-

Preso Model (Model Prison, Barcelona), 351 397-8 39 8 n -26

3

n.128 and defence of Madrid, 169, 17

Prieto, Horacio, 136, 165-6, 165 n.128, 275, Rosenberg, Marcel, 141, 151, 152 n.76, 154,
275 n.86, 372 198, 205, 211, 212, 213
Prieto, Indalecio, 32, 36 n.24, 58, 76, 80 n.9, Rubiera, Carlos, 185, 185 n.206, 423 n.101
9°> Hh i55 *57* i65> l l% i 8 3-4 l8 6,
3 ?
Rybalkin, Lt-Col. Yuri, 152-3
202, 243, 244, 280, 290, 307, 309, 310,
311, 324, 328, 335 n.64, 336, 348, 352, Salazar, Oliveira, 113

355, 372, 381 San Sebastian, 93, 241-2, 245, 249

anti-communism, 333, 344 Sanchez Guerra, Jose, 19
bid to become prime minister, 73—4, 75 Sanjurjo, General Jose, 82 n.16, 161, 349
‘defeatism’ of, 359-60 Santander, 127, 246, 247, 309, 312, 323
and gold reserves, 154 Segura, Cardinal Pedro, 28
and Largo government, 131, 132, 133 Serrano Poncela, Segundo, 19 1, 193
and May cabinet crisis, 73-4, 300, 301-3, Servicio de Investigacion Militar (SIM), 344,
306 3493 355-6, 375-8, 386, 387, 407, 408,
and Popular Front, 64—5, 71 n.139 429
relations with Negrin, 132, 155, 306, 307, Sese, Antoni, 280-1
360-2, 364 n.172, 371-2, 381 Seville, 2, 52, 104, 105, no, in, 116, 129
relations with PCE, 203, 301, 307, 328, 329, as anarchist stronghold, 49, 272
331-3 and military rebellion 94, 108-10
sacked as war minister, 361, 364 as PCE stronghold, 60
and SIM, 344, 376 Sindicatos Libres, n, 52
Primo de Rivera, General Miguel, 12, 13, 14, Sindicatos de Oposicion {see CNT, treintistas in)

!
5, !7, J9, 304 Socorro Blanco, 340
Solano, Wilebaldo, 274 n.8o, 275 n.85, 286
Queipo de Llano, General Gonzalo, 108, 122 n.134
n.169 Soviet Union, 18, 105 n.115, 134, 175, 197, 344,
Quinto Regimiento {see Fifth Regiment) 354, 375, 386-7
as agent of modernisation, 18, 69, 181
Rabassaires, 34, 55, 218 nn.io and 12, 227, communication channels to Republic,
227 n.45, 229, 238, 279, 429 197-8
Radical (Republican) Party {see Partido foreign policy of (1933-9), 152, 320-1, 334,
Republicano Radical) 383, 394, 420
Republican Union {see Union Republicana) and gold reserves, 153-4
republicans images of in Republican Spain, 18, 164, 174,
and Church, 28-30 176, 181, 197,354-5,416
and economic policy, 35, 36, 37, 41, 77 NKVD (Soviet intelligence service), 192,
and education, 36 284, 288-9, 345, 376 n.222, 428
fear of mass political mobilisation, 40, 60, and Non-Intervention, 126, 204, 320-2
7 b 77 personnel in Spain, 126, 174, 285
‘idealism’ of, 25, 39-40 and POUM, 198, 235, 237, 239, 284,
and public order, 37, 38-9, 72 285-6, 345, 346
and state, 24-5 supplies aid to the Republic, 134 n.13,
Requetes, see Carlists 150 n.70, 151-3, 158, 164, 174-6, 192, 192
requisition, 83, 87, 88, 100, 101, 225, 231, n.235, 201, 239, 307, 317, 321-2, 356-7,
255-6, 260, 314, 327, 352-3 358-9, 369= 383, 39b 393-45 398, 398 n.25
Reverter, Andreu, 265 Stalin, Joseph, 126, 132, 134 n.13, Hb *52, x
75>
Rios,Fernando de los, 126 n.184, 381 192 n.235, !97, 204-5, 210, 211-12, 235,
Riquelme, General Jose, 113 297-8, 307, 321-^2, 328, 33°-b 3345 3555
Rivas Cherif, Cipriano, 370 3 6 35 3695 3835 393-45 420
Rodriguez Salas, Eusebio, 268, 292 n.160 Stashevsky, Arthur, 154, 307
 1

472 Index

Stepanov, Stepan Minev Ivanov, 286, 330, 366 and Popular Front, 64
n.181, 404 rivalry with CNT, 8, 14, 42-3, 45-6, 48,
Sverchevsky, Col. (‘General Walter’), 175 n.171, 5 I_4, 55, 9 1 ,37, 162, 227, 262-4, 292
!

386 n.265 and similarity with CNT, 16

and socialist left, 43-7, 62-3, 133, 304, 305
Talavera de la Reina, 115, 124, 127 structural limitations, 16, 50-1 n.8o
Tangier, 106 and workers’ committees, 97-104
Taradellas,Josep, 229, 238, 271 n.73 Union Militar Espanola (UME), 168, 172
Tarragona, 67, 215 n.2, 226, 279, 395 Union Militar Republicana y Antifascista
Teruel, 88, 95, 224, 356, 358 (UMRA), 138
Tetuan, 104, 105 Union Republicana, 82, 170, 302 n.200, 430
Togliatti, Palmiro, 325, 330,
333-4, 334 n.6o, Urales, Federico, 165
366 n.184, 393-4, 404, 405, 417 Uribarri (Barutell), Manuel, 376 n.224
and Casado coup, 410-11, 412, 414 Uribe, Vicente, 325, 363, 398, 405
Toledo, 101 n.97, 14 1, 150, 164 Uribes, Jose Antonio, 418
Tribunal de Responsabilidades Civiles, 338,
425 n.107 Val, Eduardo, 400
Tribunales Especiales de Guardia, 161 n.no, Valencia, 2, 63, 67, 88, 91-2, 139, 171, 177,

347 378-9
, 182, 184, 187, 201, 202, 264, 265, 313
Tribunales de Espionaje y Alta Traicion, 161 n.240, 335, 346, 365, 373, 414, 421
n.no, 345, 347, 378-9 anarchist collectivisation in, 7, 223
Tribunales Populares, 128, 161-2, 339 chosen as new capital, 166-7
and CNT, 48, 49
Unio Democratica, 340, 341 and military rebellion, 94-5, 102
Unio Socialista de Catalunya (USC), 17, 56 and workers’ committees, 100—
n.98 Varela, General Jose Enrique, 104, no
and formation of PSUC, 67-70, 221-2 Vatican, 245, 340-1
and Popular Front, 65-7 Vazquez, Mariano R., 354, 362
and Workers’ Alliance, 54-7 Vidal i Barraquer, Cardinal Francesc, 340
Union General de Trabajadores (UGT), 7-9, Vidiella, Rafael, 262, 292 n.160
23, 38, 39, 49, ! 33, !3 6 J 37, i 6 3, 207, 218
,
Villalba, Col. Jose, 207, 209, 323 n.27
n.12,224, 353-5,430 Vinaroz, 364, 365
and Casado coup, 413 n.75, 414-15 Vizcaya, 2, 8, 10 n.15, 59-60, 91 n.59, 95, 129,
in Catalonia, 11 n.21, 55, 56 n.98, 70, 223, 214, 241, 242-3, 246, 248, 249, 251,
226, 262, 263, 264, 266, 292 252-3, 308, 310-11
collaboration with Primo dictatorship, 13-14
and defence of Madrid, 170 ‘Walter’, Gen., see Sverchevsky, Col.
fear of mass political mobilisation, 26-7, 60, Workers’ Alliance ( see Alianza Obrera)
185-6
and Largo’s resignation as prime minister, Yagiie, Colonel Juan de, no, in -12, 124, 127
3°2, 303, 3°4-5 Yagiie, Pablo, 170
and May days (1937), 270, 279, 280-1
and military rebellion, 79-80, 83-4, 90-2, Zaragoza, 2, 129, 136
93 as CNT stronghold, 49, 224
and negotiated peace, 372 and military rebellion, 94, 94 n.64
and October rising (1934), 58-9 Zugazagoitia, Julian, 185, 288 n.140, 290, 339,
and PCE, 60-2, 63, 75, 303, 327-9, 354, 346, 363-4, 364 n.172, 379, 380, 398
414-15 n.2 6, plate 6
 BOSTON PUBLIC LIBRARY

3 9999 05181 986 8

m
Mo lon< *
gf fliO'
Boston Pl viic Library'*
mmof ttifenmtertiii benefits the Library.
 by some distance the best book I have read in any language on the Spanish
Republic duringthe Civil War. The detailed - and thoroughly analytical - narrative
ofthe politics ofthe Republic outshines anythingthat has been written before ...

Avery major work by a mature historian writingatthe height of her powers.’

PROFESSOR PAUL PRESTON
London School of Economics and Political Science

This is a newand comprehensive analysis oftheforces ofthe Spanish left-

interpreted broadly-duringthecivil warof-1936-9, and the fi rst of its kind for

more than thirty years.

The book argues two crucial propositions. First, that the wartime responses (anc
limitations) ofthe Spanish left- republicans, socialists, communists and anarcho
syndicalists -can be understood only in relation to their pre-war experiences, work
views, organisational structures and the widerSpanish context of acute uneven
development which had moulded theirorganisations over previous decades.
Second, thattheoverarchinginfluencethatshaped theevolution of the Republic
between i936and 1939 was the war itself: the book explores the complex, cumulat-
ive effects of a civil war fought under the brutally destabilising conditions ofan

international arms embargo.

Helen graham is Readerin Spanish History,

Royal Holloway, University of London. Herbook
Socialism and War: the Spanish Socialist Party in

Poiver and Crisis, i gj6-t gjg was published by

Cambridge University Press in 1991. She has
otherwise published widely on the political,
social and cultural history ofSpain in the 1930s
and 1940s.

Republican poster Cambridge

( 1938 ) reinforcing the importance of home UNIVERSITY PRESS
front production in sustainingthe wareffort. www.cambridge.org
Private collection (Jordi Carulla). ISBN 0-521 -45932-X

780521 459327