See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236775399

" The Indian Ocean in Eurasian and African World-Systems Before the Sixteenth
Century "

Article  in  Journal of World History · December 2005

DOI: 10.1353/jwh.2006.0014

CITATIONS READS
80 1,227

1 author:

Philippe Beaujard
French National Centre for Scientific Research
27 PUBLICATIONS   225 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Philippe Beaujard on 01 April 2015.

The user has requested enhancement of the downloaded file.

The Indian Ocean in Eurasian and African World-Systems before
the Sixteenth Century*
Beaujard, Philippe.

Journal of World History, Volume 16, Number 4, December 2005,

pp. 411-465 (Article)

Published by University of Hawai'i Press

DOI: 10.1353/jwh.2006.0014

For additional information about this article

http://muse.jhu.edu/journals/jwh/summary/v016/16.4beaujard.html

Access Provided by University of Hawaii @ Manoa at 09/22/11 1:48AM GMT

2JWH_391-465 12/27/05 2:22 PM Page 411

The Indian Ocean in Eurasian and

African World-Systems before the
Sixteenth Century *

philippe beaujard
Centre National de la Recherche Scientifique, Centre d’Études Africaines,
École des Hautes Études en Sciences Sociales, Paris, France

Une nouvelle connaissance de l’organisation est de nature à créer

une nouvelle organisation de la connaissance.
New thinking about organization can lead to a new orga-
nizing of our thinking.
—E. Morin

riting of the Mediterranean, Fernand Braudel has remarked, “It

W isn’t water that links its shores,” but “seafaring peoples.” From a
very early date, the Indian Ocean, too, was traversed by sailors, traders,
religious men, and migrants moving in search of goods, new lands, or
the great unknown. Their movements were shaped by numerous fac-
tors, both geographic and social in origin. Over the centuries, these
exchanges transformed the Indian Ocean into a unified space.1 Trade
—above all, long-distance trade—played a central role in this process.

* I am grateful to Dr. S. Fee for having translated this article from the French. I thank
Drs. C. Coquery-Vidrovitch, E. Alpers, and M. Garden for their comments on this article.
1
On the Mediterranean, cf. F. Braudel, vol. 1, La Méditerranée et le Monde Méditer-
ranéen à l’Époque de Philippe II, 9th ed. (Paris: A. Colin, 1990), p. 253; and Civilisation
Matérielle, Economie et Capitalisme, XVe–XVIIIe Siècle, vol. 3, Le Temps du Monde (Paris: A.
Colin, 1979), p. 12. K. N. Chaudhuri, in Trade and Civilization in the Indian Ocean: An Eco-
nomic History from the Rise of Islam to 1750 (Cambridge: Cambridge University Press, 1985),
explicitly modeled his work on the Indian Ocean after Braudel’s La Méditerranée. On the
unity of the Indian Ocean, cf. M. N. Pearson, The Indian Ocean (London: Routledge, 2003),
p. 5.

Journal of World History, Vol. 16, No. 4

© 2005 by University of Hawai‘i Press

411
2JWH_391-465 12/27/05 2:22 PM Page 412

412 journal of world history, december 2005

A study of the history of trade networks, together with political and

religious factors, allows us to understand the creation of particular soci-
eties, as well as the larger area to which they were linked.
Trade implies not only an exchange of goods, but also an exchange
of knowledge, beliefs, and values. The power of trade to unify, create,
and transform cultures can be appreciated only by viewing it from the
widest possible angle. The Indian Ocean, our case in point, became a
unified space through exchange networks that went far beyond its own
borders, reaching from China to Europe and Africa. Gradually over
time, transcontinental networks—both maritime and terrestrial—
linked the Indian Ocean to the Mediterranean, turning it into a uni-
fied zone wherein events and developments occurred interdependently.
Therefore, the Indian Ocean should not be considered as a discrete
case that underwent processes similar to those of the Mediterranean, as
K. Chaudhuri has suggested, but rather as an area that was integrally
tied to the Mediterranean. The fact that the different regions of the
ancient world united by trade experienced a demonstrable synchro-
nization in their development suggests (but is not yet sufficient proof
for) the systemic nature of their relations.
It is not only the interconnections or the size of the networks but
the regularity, intensity, and speed of the exchanges that resulted in
the different regions being progressively integrated and shaped into a
world-system. This notion was originally conceived by I. Wallerstein
in 1974, in a holistic perspective that “looks for an explanation at the
level of the whole.” This perspective, which E. Morin has rightly
pointed out the limits of, has not clearly defined the concept of “sys-
tem” itself. Before proceeding with the discussion, I should identify
what I mean by the term “system.” I follow the definitions established
by Morin, namely, (1) a system represents a “complex unit and the
complex of relations between the whole and its parts,” (2) a system is
made up of cumulative interactions, (3) which constitute the organiza-
tion of the system.2 The character of this organization is, in essence,

2
E. Morin, Science avec Conscience (Paris: Fayard, 1990), pp. 244–245, 252. A system
is more than the sum of its parts, but is also less than the sum of its parts (ibid., pp. 241–243;
also I. Wallerstein, The Modern World-System, vol. 1, Capitalist Agriculture and the Origins
of the European World-Economy in the Sixteenth Century [San Diego: Academic Press, 1974],
p. 8). An explanation is to be found out not only at the level of the whole but through the
interactions between and within the parts, which constitute the whole. Morin undertook
to transpose the theories of the chemist Prigogine in the anthropological field. Cf. I. Prigo-
gine and I. Stengers, Order Out of Chaos: Man’s New Dialogue with Nature (London: Heine-
mann, 1984), and The End of Certainty: Time, Chaos and the New Laws of Nature (New York:
Free Press, 1997).
2JWH_391-465 12/27/05 2:22 PM Page 413

Beaujard: The Indian Ocean in Eurasian and African World-Systems 413

both complex and dynamic. The system generates both order and dis-
order, unity and diversity. Taking into account these general charac-
teristics of all systems and their implications can aid the interpretation
of the available historical data for the Eurasian and African zone. The
systemic approach provides a new “logic,” and leads to a new under-
standing of world history.
Wallerstein forged the concept of world-system in relation to the
modern era but, in fact, the creation of an Eurasian and African world-
system can be traced much further back in time. He enumerated twelve
characteristics of the “modern world-system,” which include ever-
increasing capital accumulation, a division of labor, growing imbal-
ances of power between cores and peripheries, phases of hegemony—
within a given core—that alternate between a single power exercising
control and several rival powers vying for control, and the existence
of cycles. Frank and Gills argue that these same characteristics have
also been present in world-systems for the past several thousand years.3
According to these authors, for too long scholars have underestimated
the importance of capital accumulation, markets, and individual enter-
prise in ancient societies. Available actual data, however, show the
formation of an Eurasian and African world-system with the Chris-
tian era.
From its origins, the Eurasian and African world-system developed
and was restructured following the rhythm of economic cycles that
lasted several centuries (periods of growth followed by periods of
decline). Understanding these cycles, their nature, and possible ori-
gins provides a key to unlocking the history of the region.
Based on geographic factors and exchange networks, the Asian and
East African maritime zones can be divided into three main areas: the
China Sea, the eastern Indian Ocean, and the western Indian Ocean,
with the latter area being further divided—except during some rare
moments of unity—between the Persian Gulf and the Red Sea. Each
of these subsystems had its own “core” (China, India, western Asia,4
and Egypt) that determined the nature of trade with its peripheries. I
agree with Frank and Gills that transfers of surplus between regions
“necessarily imply” a division of labor, issues of hegemony, and the

3
A. G. Frank and B. K. Gills, eds., The World System: Five Hundred Years or Five Thou-
sand? (London: Routledge, 1993). Before Frank and Gills, cf. K. Ekholm and J. Friedman,
“‘Capital’ Imperialism and Exploitation in Ancient World-Systems,” Review 4, no. 1
(1982): 87–109.
4
As part of my efforts to avoid Eurocentrism, I employ the terms “western Asia” and
“eastern Asia,” in place of the more usual Near East (or Middle East) and Far East.
2JWH_391-465 12/27/05 2:22 PM Page 414

414 journal of world history, december 2005

emergence of core-periphery relations,5 but this is not the only source

of power. One needs to examine the ensemble of mechanisms that lie
behind these transfers, which Frank and Gills, in my view, have not
done sufficiently. The concept of value of transferred surplus has not
yet been properly considered.6 I argue, in addition, that the levels of
integration in the construction of a system must be taken into
account.7 Furthermore, while the surplus the center takes from the
periphery creates a particular type of relation between these two zones,
it affects their respective internal social relations as well.
Cities are sited at the nodes of the networks; they direct production
and exchange according to a hierarchical structure. Within a given
core or periphery, there exists a further hierarchical division between
each metropolis and its dependent zones. Urban points linked by long-
distance trade create a string of conglomerations or relay points—
“archipelagos of towns” to use Braudel’s delightful phrase 8 —whose
interconnections form the spine of the system. Within the constella-
tion of towns, city-states—or, more precisely, certain cultures of city-
states—played a particular role in the evolution of the global system,
especially through their ideological and institutional innovations (see
below).9
Throughout the ages, observers of these metropolises have under-
scored their cosmopolitan Tower of Babel character as places fostering
interrelationships that give birth to innovative creations. Along the
perimeter of the Indian Ocean, trade gave rise to the formation of
“fringe cultures,” a term and concept used by anthropologist P. Ottino
to account for the similarity in many of the syncretic societies that
developed on the Indian Ocean rim.10

5
A. G. Frank and B. K. Gills, “The Five Thousand Year World System in Theory and
Praxis,” in World System History: The Social Science of Long-Term Change, ed. R. A. Dene-
mark, J. Friedman, B. K. Gills, and G. Modelski (London: Routledge, 2000), p. 4.
6
The “surplus” represents the difference between what has been produced and what
has been consumed by the producer. On the concept of value, cf. infra.
7
It seems to me that a certain level of integration is required. For Frank and Gills,
“‘mere’ trade makes a system,” but they speak also of a “regular and significant trade” (Frank
and Gills, “Five Thousand Year World System,” p. 6).
8
Braudel, Civilisation Matérielle, vol. 3, p. 20.
9
Following Hansen, it is possible to define city-states as “self-governed cities which
consider themselves as political units” and are recognized as such by the other political
units of the region. These cities may be independent, or dependent through diverse modes
(within a hierarchical ensemble of city-states, within a federation, as a tributary of a macro-
state, and so on). M. H. Hansen, “Conclusion: The Impact of City-State Cultures on
World History,” in A Comparative Study of Thirty City-State Cultures, ed. M. H. Hansen
(Copenhagen: Det Kongelige Danske Videnskabernes Selskab, 2000), pp. 606, 608–609.
10
M. N. Pearson also underlines the key concept of “littoral society” (Indian Ocean,
pp. 37–41). J. Friedman has pointed out that “the wave of discourses on cultural hybridity
2JWH_391-465 12/27/05 2:22 PM Page 415

Beaujard: The Indian Ocean in Eurasian and African World-Systems 415

Not all coasts enjoy the same opportunities to become a trade

nexus. Zones at the intersection of two subsystems are particularly
favored in this respect, hence the successes of Southeast Asia, southern
India and Ceylon, Aden and Yemen, Hormuz and Oman. The success
of a particular port is tied to the quality of its haven, its geographical
position, and its relation to commercial routes, both maritime and ter-
restrial—the most striking example being that of the famous Silk
Roads that ran through China, central Asia, India, and Persia. The
development of maritime trade also depended in part on the relations
established between the coast and its hinterland. The same held true
for waterways, such as rivers and streams, for the mouths of large rivers
have always served as points in the development of commerce and
trade. Likewise, the availability of resources and labor naturally played
a role in the construction of a system.
Exchanges are not solely shaped by geographic and economic fac-
tors, but also by systems of ideas and by the balance of power. I want
to emphasize two points. First, trade is not the only method for trans-
ferring surplus. Political domination and conflict also play roles (for
example, the imposition of tribute and taxes, looting, and so forth), as
do religious networks and relations of production (the relationship
between the governing elites and the producers). The export and
import of products are closely tied to ideologies, themselves insepara-
bly meshed with political forms. Once an ideology has become ascen-
dant in a given network, it can influence the desirability of certain
products and contribute to the determination of their commercial
value. Thus, studies of world-systems must consider how the economic
can be encompassed by the cultural and the sacred. The organization
of marketplaces and their varying levels of importance are closely
linked to social structures and to the politico-military competition of
states and elites. A study of trade circuits requires a reconstruction of
the political, economic, and religious history of the regions where the
network operated, at both international and local levels. Built around

consist of the analysis of cultural elites and their discourses. [. . .] the ideology of hybridity
is primarily an elitist discourse in a world that is otherwise engaged in the opposite.
Hybridization and balkanization are two simultaneous processes of the global shift in hege-
mony” (“Concretizing the Continuity Argument,” in Denemark et al., World System His-
tory, p. 147). The concepts of “fringe culture,” networks, and hybridity get their full mean-
ing only if they are analyzed at the level of the whole through a systemic approach that goes
beyond a holistic perspective, as the latter “expresses only a partial and simplifying vision
of the whole.” Pascal had already expressed this “new paradigm brought by the idea of sys-
tem: ‘I consider as impossible either to know the parts without knowing the whole, or to
know the whole without knowing each of the parts’” (Morin, Science avec Conscience,
p. 240).
2JWH_391-465 12/27/05 2:22 PM Page 416

416 journal of world history, december 2005

religious sites that are also sites of production and trade, religious net-
works become spaces where, following the paths of pilgrims, wealth
and information circulate. Second, transfers of surplus are not the only
means for cores to achieve dominance. This they also accomplish
through ideological and political power acting via diverse strategies,
such as colonization, alliances, religious conversions, intermarriages,
and so forth.
Understanding the processes in the construction of an area unified
by exchanges and contacts, and capturing both the changes and conti-
nuity in the articulation of the network, can be achieved only through
a study of the long term, through a consideration of the very origins of
the Eurasian and African world(s)-system(s), and through a compara-
tive perspective that encompasses the entire region under considera-
tion, that is to say, whole oceans (Indian Ocean, China Sea, Mediter-
ranean) and continents. The study must be both transdisciplinary and
systemic. In essence, it requires an examination of the relationships
between economic, political, and religious data; technological inno-
vations; climatic changes; demographic trends;11 and the grasp of the
dynamics of interaction and organization between the system as a
whole and its constituent parts.

The Birth of the Eurasian and African World-System

As early as the fourth millennium b.c.e., the rise of the state, espe-
cially interrelated city-states, in Mesopotamia and the expansion of
trade networks with neighboring regions may have resulted in the for-
mation of a world-system, with southern Mesopotamia acting as a
core. This system probably included at least a part of the Persian

11
Unfortunately, we do not have quantitative figures to determine the level of inte-
gration of the various parts of the world-system for the eras that concern us here. Estab-
lishing a direct measure of commercial volume is also, of course, impossible. We must
therefore rely on the indices to be found through archaeological excavations and ancient
writings. They allow us to estimate size and intensity of exchange networks and follow their
expansions and shrinkages (cf. J. Abu-Lughod, Before European Hegemony: The World Sys-
tem A. D. 1250–1350 [New York: Oxford University Press, 1989], p. 368). The number and
size of the principal cities, and their localization, provide precious indications as to the gen-
eral direction of activity (growth or decline) and the internal structure of the world-system
(A. Bosworth, “World Cities and World Economic Cycles,” in Civilizations and World Sys-
tems: Studying World-Historical Change, ed. S. K. Sanderson [Walnut Creek, Calif.: AltaMira
Press, 1995], pp. 206–227). In addition, other methods, such as palynological studies, ice
core analysis, and dendrochronology are useful because they allow for the reconstruction of
historic climatic and environmental conditions, which in turn can be compared to demo-
graphic, economic, and political trends.
2JWH_391-465 12/27/05 2:22 PM Page 417

Beaujard: The Indian Ocean in Eurasian and African World-Systems 417

Gulf.12 In the third millennium, the Indian Ocean was not yet a uni-
fied space, but a world-system took shape between the urbanized soci-
eties of Mesopotamia, Elam, and Indus through maritime roads in the
Persian Gulf and land routes that ran all the way to Turkmenistan
and Bactria (ca. 2600–1800 b.c.e.).13 Frank and Gills support the idea
that a single system united Egypt, Arabia, Mesopotamia, and the rest
of western Asia from the third millennium b.c.e. “The confluence
occurred . . . about 2700–2400 b.c.” 14 The weakness of the ties
between Mesopotamia and Egypt makes this hypothesis hard to accept.
Up to the eighteenth century b.c.e., bronze was rarely used in Egypt.
In fact, Egypt and Mesopotamia during that period do not follow the
same rhythm.15 As would again later be the case, from 1000 to 600
b.c.e., the exchange networks that can be identified point to the artic-
ulation of two different “spheres of interaction” and not so much to a
supposed unique world-system (Figure 1).16

12
G. Algaze, The Uruk World System: The Dynamics of Expansion of Early Mesopotamian
Civilization (Chicago: University of Chicago Press, 1993).
13
The network may have reached all the way to East Africa (C. Chase-Dunn and T. D.
Hall, “Comparing World-Systems to Explain Social Evolution,” in Denemark et al., World
System History, p. 106) if the copal necklace found in a tomb at Tell Asmar (near Bagh-
dad) and dated to 2500–2400 b.c.e. can be proved with certainty to come from Zanzibar
or its environs (C. Meyer, J. M. Todd, and C. W. Beck, “From Zanzibar to Zagros: A Copal
Pendant from Eshnunna,” Journal of Near Eastern Studies [1991]: 296–297). Some scholars,
such as J. Philipps (“Punt and Aksum: Egypt and the Horn of Africa,” Journal of African His-
tory 38 [1997]: 437), have expressed doubts about this origin.
14
B. K. Gills and A. G. Frank, “The Cumulation of Accumulation,” in The World Sys-
tem, p. 82.
15
Chase-Dunn and Hall’s attempt to compare the size of the empires of Mesopotamia
and Egypt shows two very different growth rates for the period 2500–3000 b.c.e. and
opposing trends from 2000 to 1500 b.c.e. (“Comparing World-Systems,” Fig. 4.9, p. 106).
This opposition in terms of political integration nevertheless does not exclude synchro-
nism on an economic level.
16
Frank and Gills (“Rejoinder and Conclusions,” in The World System) trace the begin-
nings of the Eurasian world to the third millennium b.c.e., but they do not offer “proof ” of
definite cycles (with phases of growth and decline) until after 1700 b.c.e. Wilkinson mean-
while dates the origin of his “central civilization” to 1500 b.c.e. (D. Wilkinson, “Central
Civilization,” in Sanderson, Civilizations and World Systems). A comparison of the cycles put
forward by Frank and Gills with Chandler’s data (Four Thousand Years of Urban Growth: An
Historical Census [Lewiston: The Edwin Mellen Press, 1987]) demonstrates well the diffi-
culties faced by scholars for the period before the first century c.e. Analyses of world-systems
for the Bronze Age and the beginning of the Iron Age face very difficult obstacles (Bos-
worth, “World Cities and World Economic Cycles”). I have stressed that, owing to the lim-
its of archaeology and the paucity of texts, we can ascertain only general tendencies and
sometimes chains of dependency. Ongoing debates on the chronologies of the third and
second millenniums b.c.e. clearly show that the phases of growth and retraction for cer-
tain zones put forward by authors such Frank and Gills can often be taken only as mere
hypotheses (cf., for example, “Just in Time: Proceedings of the International Colloquium on
Ancient Near Eastern Chronology [Second Millenium b.c.],” Akkadica 119–120 (2000),
and D. T. Potts, “Tepe Yahya, Tell Abraq and the Chronology of the Bampur Sequence,”
Iranica Antiqua 38 [2003]: 1–11).
2JWH_391-465 12/27/05 2:22 PM Page 418

figure 1. Asian, African, and European world-systems, from 3500 b.c.e.

to 1750 c.e.
2JWH_391-465 12/27/05 2:22 PM Page 419

Beaujard: The Indian Ocean in Eurasian and African World-Systems 419

From 1600/1500 b.c.e., one can discern the emergence of a multi-

centered system (Hittite kingdom, Mitanni, Assyria, Egypt) in which
the western Indian Ocean (Persian Gulf, Red Sea) remained periph-
eral. The eastern Mediterranean, where the Mycenian civilization
blossomed, was then a central maritime space. This system disinte-
grated with the invasions of the thirteenth through twelfth centuries
b.c.e.
In the first half of the first millennium b.c.e., a Mediterranean
sphere grew out of Phoenician and Greek networks that was linked, but
not united, to the sphere of Mesopotamia/Persian Gulf. The develop-
ment of an alphabet in Phoenicia (eleventh century b.c.e.), soon bor-
rowed by the Greeks, represented a major innovation during this
period and in this region, which provided the basis for a transforma-
tion of the relation between the individual and the existing powers.17
The eastern and western parts of the Indian Ocean remained dis-
crete units before the first millennium b.c.e.18 A limited integration
of the different regions of Asia took shape only from the sixth century
b.c.e. 19 At this time, spices such as cinnamon and cassia arrive in
the Persian Gulf and reach the Greeks. Notably, from 550 b.c.e., the
largest cities of western and eastern Asia seem to show similar growth
and transformations.20 The sixth century b.c.e. also witnessed the
emergence of the first empire in western Asia, the Persian empire,

17
The complexity of earlier scripts matched the limited number of their uses and of
their users; simplicity and efficiency of alphabetical scripts transformed not only their social
function but also the relationship of the individual with the different spheres of power.
18
The archaeological discovery of cloves at the site of Terqa on the central Euphrates,
at a stratigraphy dated 1700–1600 b.c.e.—if it can be verified—would, however, point to
contacts between the Austronesian world and the western Indian Ocean from 2000 b.c.e.
G. L. Possehl, “Meluhha,” in The Indian Ocean in Antiquity, ed. J. Reade (London: Kegan
Paul, 1996), p. 190.
19
Evidence of these interconnections includes the introduction of barley, wheat, and
sheep into China in the third millennium b.c.e. and the appearance of horse-drawn char-
iots in China in the second millennium b.c.e. Silk, which has been found in the Sapalli
tombs of Bactria (ca. 2200 b.c.e.) probably came from China (A. A. Askarov, Sapallitepa
[Tachkent: Fan, 1973]). On China in the second millennium b.c.e., cf. L. Liu and X. Chen,
State Formation in Early China (London: Duckworth, 2003).
20
Cf. Figure 10.8 in Chase-Dunn and Hall, Rise and Demise: Comparing World-Systems
(Boulder, Colo.: Westview Press, 1997), p. 219. Frank and Gills (“Five Thousand Year
World System,” p. 12) hold that a simultaneity in phases can be discerned between east-
ern and western Asia from the middle of the first millennium b.c.e. In my opinion, even if
an interconnection did indeed exist between eastern Asia, India, and western Asia from
the sixth to fifth century b.c.e., the relations do not display the regularity and intensity that
characterize a system; for this period it is difficult to demonstrate synchronous evolution in
the various regions, whereas it can be shown for somewhat later periods.
2JWH_391-465 12/27/05 2:22 PM Page 420

420 journal of world history, december 2005

which took possession of the pivotal regions between the Mediter-

ranean and the Indian Ocean and attempted to control all commer-
cial routes. From China to the Mediterranean, social transformations
gave rise to new views on the universe and society, which went hand
in hand with the emergence of individualism. This can be witnessed
in the sudden appearance of the great philosophical and religious doc-
trines claiming universal truths (Jainism, Buddhism, Confucianism,
Mazdeism) and the flourishing of a rationalist humanism in the Greek
world, a scientific current in China (Mohists), and a political ratio-
nalism in India (the writing of the Arthasastra). The “universal” far-
reaching Greek empire built by Alexander and his armies in the
fourth century b.c.e. largely follows the path of the earlier Persian one
in linking the Mediterranean sea and the western Indian Ocean. At
the same time, it founded inland and coastal settlements at nodes of
the preexisting trade networks.
Beginning in the third century b.c.e., the unification of a large part
of India under the Mauryas, together with the fast spread of Buddhism,
the increasing integration of western Asia, and the unification of China
by King qin Shi Huangdi (221 b.c.e.) made possible the opening of
land routes across Central Asia (the Silk Roads) and maritime routes
in the China Sea and Indian Ocean.
However, it is only on entering the first century c.e., with the cre-
ation of links between the various networks and the development of a
regular, high level of commerce, that there emerged a Eurasian and
African world-system 21 in which the different regions evolved in tan-
dem,22 a world-system that would endure without major changes until

21
Africa was only gradually incorporated into the spheres of interaction or the
Eurasian and African system(s): Egypt and its hinterland, the coast of the Red Sea, and the
horn of Africa from the third millennium b.c.e. in a system centered on Egypt and its hin-
terland, North Africa from the second millennium b.c.e. in a Mediterranean space, East
Africa and its hinterland around the first century c.e. (or a little earlier) in the global
world-system that takes shape in that time, sub-Saharan Africa from the seventh century
at the latest (but perhaps from as early as the first millennium b.c.e.) in the Mediterranean,
and Egypto-Nubian spheres.
22
See also I. C. Glover, “The Archaeological Evidence for Early Trade between South
and Southeast Asia,” in Reade, Indian Ocean in Antiquity, p. 368 and W. H. McNeill,
“World History and the Rise and Fall of the West,” Journal of World History 9 (1998): 129.
The inconsistencies in Phase B (250/150–100/50 b.c.e.) put forward by Frank would argue
against the existence of an Eurasian and African world-system for this time period (the
Mediterranean and China were enjoying economic expansion while Egypt, Mesopotamia,
Persia, and perhaps India—in the second century b.c.e.—were experiencing some level of
decline). But maybe an Asian unified space was taking shape at that time (cf. J. Bentley,
Old World Encounters: Cross-Cultural Contacts and Exchanges in Pre-Modern Times [Oxford:
Oxford University Press, 1993], p. 29). If the regularity and intensity of exchanges must be
2JWH_391-465 12/27/05 2:22 PM Page 421

Beaujard: The Indian Ocean in Eurasian and African World-Systems 421

the modern era. The birth of this system accounts for the movements
of Europeans toward the Orient and the voyages of Austronesians
toward the western areas of the Indian Ocean and toward China. It
also explains the development of a pre-Swahili culture in East Africa
and the “Indianization” of Southeast Asia.

Pulsations in the World-System

From its beginnings, the world-system developed at a progressive rate,

following economic cycles that largely corresponded with cycles of
political and religious events (Figure 2).23 This synchronous evolution,
which points out an increasing integration of the different parts of the
world-system, reveals four cycles, each with a phase of expansion
and a phase of contraction from the end of the first millennium b.c.e.
until the sixteenth century. These can be represented as four ascend-
ing waves—on an upward curve—with a corresponding growth in
population, production, the volume and rapidity of trade, and urban
development (especially of the system’s most important cities). 24
These developments are precipitated or accompanied by technologi-
cal improvements and increased capital investment. The cycle lengths
that I have established differ somewhat from those put forward by
Frank and Gills.25 The first cycle ran from the first to the sixth century
(marked by the rise and fall, within the cores, of the Han, Kushan, Sha-
tavahana, Gupta, Parthian, Sassanid, and Roman empires); the sec-
ond, from the sixth to the tenth century c.e. (in the cores, empires
of the Tang, the Pallavas, the Rashtrakutas, the Muslim and Byzantine
empires); the third, from the tenth to the fourteenth century c.e.
(Sung and then Yuan empires, Chola empire, Delhi sultanate, Abba-
sid and then Ilkhanid empires, Egyptian state); the fourth, from the fif-
teenth to the sixteenth century and from then until the Industrial
Revolution in the middle of the eighteenth century (for the fifteenth

taken into account, it is obviously difficult to establish the threshold at which the integra-
tion of several world economies will allow them to constitute a single world-system. Fur-
thermore, the available historical data are richer from the first century c.e., and this might
introduce a bias to the analysis.
23
But see C. Edens on this point (cf. infra).
24
Cf. C. McEvedy and R. Jones, Atlas of World Population History (Harmondsworth:
Penguin Books, 1978), and Bosworth, “World Cities and World Economic Cycles.”
25
B. K. Gills and A. G. Frank, “World System Cycles, Crises, and Hegemonic Shifts,
1700 b.c.e. to 1700 c.e.,” in The World System. But from the beginning of the Christian
era, phases of growth that have been recognized are in accordance to the periods delimited
by Bentley, Old World Encounters, pp. 26–28.
2JWH_391-465 12/27/05 2:22 PM Page 422
 2JWH_391-465
12/27/05
2:22 PM
Page 423

figure 2. Economic cycles in the Eurasian and African world-system (first through eighteenth centuries).
2JWH_391-465 12/27/05 2:22 PM Page 424

424 journal of world history, december 2005

through sixteenth centuries, Ming empire, sultanates of Gujarat, Ben-

gal, Deccan, empire of Vijayanagara, Ottoman, and Egyptian states).26
The sixteenth century does indeed represent an important moment in
world history, with the two sides of the Atlantic being joined to the
primary system and with Europe’s creation of a second world-system
(Europe-Americas-West Africa). But it did not create a rupture in the
Indian Ocean as has commonly been argued.
From the beginnings of the Eurasian and African world-system,
China played a crucial role.27 Every period of growth in the world-sys-
tem followed China’s phases of unification and its economic rises, with
the Chinese momentum spreading like a wave from east to west,
reaching the West with some delay. Already detectable with the Han
dynasty (first century c.e.), China’s influence grew with the ascension
of the Tang (seventh century), increased under the Sung (end of the
ninth century), and continued with the Ming (beginning of the fif-
teenth century). Conversely, each Chinese recession initiated a down-
ward trend in the system and/or its restructuring. The Ming decision
to withdraw from the system in 1433 and the development of capital-
ism in Europe in the fifteenth century laid the groundwork for the
future shift westward for the core of the Eurasian world-system.
Induced by demographic pressure, progress in agriculture went
hand in hand with ascendant phases in each cycle. This in turn led to
a new rise in population 28 and the growth of cities and artisan activ-
ity. Increases in agricultural output were often linked to favorable cli-
matic changes, such as those in China and Central Asia in the sev-
enth and eighth centuries, and in all the northern hemisphere from
the end of the tenth to the beginning of the fourteenth century.

26
The nature of the recession that occurred in the middle of the seventeenth century
is still being debated (cf. infra).
27
Various authors have underscored China’s pre-eminence in the world-system, cf.
W. H. McNeill, The Pursuit of Power: Technology, Armed Forces and Society since A. D. 1000
(Chicago: Chicago University Press, 1982), chap. 2, and McNeill, “World History and the
Rise and Fall of the West,” pp. 219–220; A. G. Frank, ReORIENT: Global Economy in the
Asian Age (Berkeley: University of California Press, 1998), and especially J. Needham, Sci-
ence and Civilisation in China, 20 vols. (Cambridge: Cambridge University Press, 1959–
1998). Some scholars have placed an emphasis on the role of central Asia and its nomadic
herders in the system’s dynamics. I would argue that central Asia— even when it attained
political ascendancy in the thirteenth century—was only dancing to the rhythm set by
other players.
28
I agree with Frank (ReORIENT)—as against G. Arrighi (“The World According
to Andre Gunder Frank,” Review 22, no. 3 [1999]: 336)—that demographic growth and
economic gains went hand in hand. Braudel (Civilisation Matérielle, vol. 1, Les Structures du
Quotidien, p. 17). argued a similar point: “if men become more numerous, there is an
increase in production and trade.” Besides the role of the demographic pressure, political
power can force farmers to increase their labor input and to transform the use of land, but
this is more likely to happen in periods of global growth of the system.
 2JWH_391-465
12/27/05
2:22 PM
Page 425

map 1. The Eurasian and African world-system from the first to the third century.
 2JWH_391-465
12/27/05
2:22 PM
Page 426

map 2. The Eurasian and African world-system from the seventh to the ninth century.
 2JWH_391-465
12/27/05
2:22 PM
Page 427

map 3. The Eurasian and African world-system from the eleventh to the early thirteenth century.
 2JWH_391-465
12/27/05
2:22 PM
Page 428

map 4. The Eurasian and African world-system in the thirteenth and fourteenth centuries.
 2JWH_391-465
12/27/05
2:22 PM
Page 429

map 5. The Eurasian and African world-system in the fifteenth century.

2JWH_391-465 12/27/05 2:22 PM Page 430

430 journal of world history, december 2005

Each period of growth was accompanied by technological innova-

tions in the domains of production, transportation, and trade, which
allowed populations to overcome ecological constraints and human
limitations by increasing productivity and transforming production.29
In a beneficial cycle, these innovations led to a rise in production,
trade, and social complexity; they encouraged the development of hier-
archies and inequalities that in turn modified the nature of constraints.
By innovation, I mean the greater use of inventions, some of them
known from earlier times. We can find many examples. Iron ploughs
using different kinds of molds were perfected at the period of the ear-
lier Hans. Paper was invented at the same period (second century
b.c.e.). The Romans made use of cement at the end of the first cen-
tury b.c.e. or just after. Xylography, invented in China in the eighth
or ninth century, developed only in the Sung period. The compass
appeared on Chinese ships in the eleventh century, even if the prop-
erties of lodestone were known since the fourth century b.c.e. In the
eleventh century, the Chinese built junks of a larger size.
Ideological and institutional innovations also emerged in phases
of expansion in the system. And thus, new philosophical doctrines
or religions—“tools of ideological power” in the words of M. Mann
(1986)—appeared during the initial phases of growth within a large
region or within the entire world-system: Christianity is born in the
first century c.e. and Islam at the beginning of the seventh century;
neo-Confucianism appears in the eleventh century as part of the
intellectual movement of the Sung period. The role played by the

29
It is processes of growth, however, that tend to give rise to innovations. The use of
inventions takes place only when it becomes profitable to invest in technological develop-
ment. If it is true that a certain demographic density is necessary to allow for the develop-
ment of more efficient transport, I do not agree with the position of Chase-Dunn and Hall
(“Comparing World-Systems,” p. 98) that demographic pressure is generally at the root of
technological progress and is the reason for political expansion. In the case of agriculture,
the spread of plants occurred during moments of increased trade. In the seventh and eighth
centuries, the Arabs transported as far away as Spain fifteen types of vegetal species from
the Indian peninsula. In the tenth century, the Sung dynasty promoted the introduction of
rice varieties from Champa, which made possible two harvests per year. However, while
innovations in manufacturing, the realm of science, and “techniques of power” occur most
often during moments of economic growth, or at the beginning of growth, it appears that
innovations in agricultural domains also take place during phases of decline (indeed,
because they provide solutions). For the relations between demographic pressure and tech-
nological change, see E. Boserup, The Conditions of Agricultural Growth: The Economics of
Agrarian Change under Population Pressure, new ed. (London: Earthscan Publications, 1993).
In certain cases, unfavorable climatic conditions can lead to innovations in agriculture,
usually through intensification. This is no doubt what happened during the phase of cool-
ing of the Younger Dryas (ca. 11,000–9800 b.c.e.), which played a role in the development
of agriculture and the domestication of plants in western and eastern Asia.
2JWH_391-465 12/27/05 2:22 PM Page 431

Beaujard: The Indian Ocean in Eurasian and African World-Systems 431

major religions in the processes of integration has often been under-

lined.30 Religions and philosophical doctrines have been a base for
many large political organizations, which in turn have contributed to
their expansion (cf. infra). They have also promoted the rise of exten-
sive trade networks. The Indian merchant guilds and the city-states of
East Africa and Southeast Asia— sites of self-governance 31 and inno-
vation—flowered in periods of economic upturn.32
Increases in trade led to competition between states. Although this
could stimulate commerce and encourage innovation, it could also lead
to wars of expansion.33 Each ascendant phase was accompanied by the
rise of empires 34 whose investments in public works (for example, the
building of roads and canals by the Han, Tang, and Sung in China in
the first, sixth, and tenth centuries; the Muslim empire of the seventh
century; and so forth) often contributed, at least in the early stages, to
economic growth until the state’s level of complexity reached a point
of diminishing returns on investments.35 India offers a good example of

30
Cf. H. P. Ray, The Archaeology of Seafaring in Ancient South Asia (Cambridge: Cam-
bridge University Press, 2003), p. 11; on Buddhism and Hinduism, Chaudhuri, Trade and
Civilization in the Indian Ocean; and on Islam, A. Wink, “‘Al Hind’: India and Indonesia in
the Islamic World-Economy, c. 700–1800 c.e.,” in Comparative History of India and Indone-
sia, vol. 3, ed. P. J. Marshall et al. (Leiden: E. J. Brill, 1988), pp. 33–72.
31
Self-government is one of the main characteristics of city-states. Not all of the city-
states have developed democratic institutions but “even in monarchies, the percentage of
the population involved in the direction of the government is much higher than in other
types of states” (M. Hansen, “The Concepts of City-State and City-State Culture,” and
“Conclusion: The Impact of City-State Cultures on World History,” in A Comparative
Study of Thirty City-State Cultures, p. 18 and p. 607).
32
These trends are even clearer in the economically growing European continent from
the twelfth century. The discovery of Greek and Arabic philosophy and sciences (eleventh
through twelfth centuries) is a prelude to the development of corporations and autonomous
universities (self-governed) and to the initiation by R. Bacon and other scholars of the
experimental method in sciences (thirteenth century). Republican institutions of Italian
city-states allow the individual—within certain limits—freedom of thinking and freedom
of enterprise, which will develop in the fifteenth century during the Renaissance.
33
The development of the state (internal and external) can result from different fac-
tors: growth in population and in production and commerce, increases in social complex-
ity, and innovations. Competition between city-states and states probably explains part of
the technical progress at the time of the Warring States in China (steel production in the
fifth century b.c.e., the invention of the crossbow in the fourth century b.c.e.). It led to
the ascendancy of the Qin state in the third century b.c.e.
34
Imperialist expansion might also represent “an alternative to the domestic redistri-
bution of riches,” a sort of release valve for social tensions.
35
“Investments in socio-political complexity as a problem-solving response often
reaches a point of declining marginal returns. . . . Once a complex society enters the stage
of declining marginal returns, collapse becomes a mathematical likelihood” (A. Tainter,
The Collapse of Complex Societies [Cambridge: Cambridge University Press, 1988], pp.
194–195).
2JWH_391-465 12/27/05 2:22 PM Page 432

432 journal of world history, december 2005

these competitions. States periodically have difficulty trying to control

both the oriental and occidental maritime shorelines of the peninsula
(the Gupta empire in the fourth and fifth centuries, the Delhi sultanate
in the thirteenth and fourteenth centuries, the Bahmani sultanate and
the Vijayanagara empire in the fourteenth and fifteenth centuries).
However, India is most often structured as a multicentered core, as seen
on the maps supra (for example, between the seventh and twelfth cen-
turies, with the Gurjara Pratiharas in the north, the Pala dynasty in
Bengal, the Rashtrakutas and then the Chalukyas in Deccan, the
Pallavas on the eastern coast, the Cholas in the southeast and the
south; in the fourteenth and fifteenth centuries, with the sultanates of
Gujarat, Bengal, and the Deccan states [Bahmani sultanate and Vija-
yanagara empire]). Control of Egypt, Palestine, and Syria, situated at
the articulation of different maritime and land spaces, has also repre-
sented a major issue for the great powers in periods of growth of the
system (cf. infra).
Between periods of prosperity (for the elites, at least) there came
periods of decline and disorder in the system whose causes were usually
multiple: internal contradictions within states and societies, state poli-
cies that discouraged production and commerce (for example, interfer-
ence from state structures, state-sponsored wars that interrupted invest-
ments in the productive sector, excessive taxation), political struggles
for the control of wealth and state apparatus, and so on. Such cycles
are probably inherent to the structure of the system itself (Figure
3). 36 K. Ekholm and J. Friedman adduce internal economic factors to
explain the ending of a period of growth: overaccumulation at the core
leads to an increase in prices and a decrease in the levels of profit and
investment, which result in decentralization and disintegration.37 In a
more general fashion, the demographic increase that accompanies

36
The model I describe differs in several ways from that of Chase-Dunn and Hall
(“Comparing World-Systems,” p. 98). Here, I can offer only its outlines. It is indeed likely
that there exists “a plurality of logics,” but all the same I hold, as against Wilkinson, that
there is a “systemic logic” (D. Wilkinson, “ Civilizations, World Systems and Hegemonies,”
in Denemark et al., World System History, p. 78). One example of possible other interactions
is that linking the environment to technological innovations (for example, in eighteenth-
century England, deforestation led to the increasing rarity and value of wood, which led
to the use of charcoal in blast furnaces). Frank and Gills (“Rejoinder and Conclusions,”
p. 305) have called for a new reading of world history that would take into account how
“world system development both altered and was in turn altered by the natural environ-
ment,” but they have not really pursued this line of inquiry.
37
K. Ekholm and J. Friedman (“‘Capital’ Imperialism and Exploitation in Ancient
World Systems,” in Frank and Gills, The World System, p. 73 n. 19) and I. Wallerstein (His-
torical Capitalism [London: Verso, 1983], pp. 59–60) have noted the high military price of
maintaining hegemony.
 2JWH_391-465
12/27/05
2:22 PM
Page 433

figure 3. Cyclical logic of the world-system before the Industrial Revolution.

2JWH_391-465 12/27/05 2:22 PM Page 434

434 journal of world history, december 2005

periods of economic growth creates, in the end, social tensions and

environmental problems, which in turn lead to a decrease in produc-
tion and an increase in the cost of production. In addition to ideolog-
ical causes, environmental perturbations probably explain the migra-
tions out of central Asia during various eras. Examples of this include
the movements of peoples from the end of the third to the sixth cen-
tury: the Xiongnu toward China; the Huns toward India, Persia, and
Europe.38 The bringing into contact—direct or indirect—of distant
peoples favors the spread of disease.39 Moreover, Chew has emphasized
the damaging processes of deforestation and soil erosion that are the
results of the development of the cores at each phase of growth, and
their negative impact on this growth.40 But first of all, environmental
disruptions and the cycles themselves might be linked to climatic
changes, probably connected to cycles in solar activity. Recent research
seems to show the existence of long solar cycles during the Holocene
(200, 400/500, 950, and 1500 years).41
In times of recession, the world-system does not disappear but
undergoes a phase in which its networks are restructured, as are its
implicated states and their societies (at economic, political, and ideo-
logical levels). Then comes the next upswing brought on by various
factors (cf. supra). In sum, a group of forces working together produces
a pulsation in the system, leading to an upward movement in the curve.
Recessions can be either brief (about seventy years in the four-
teenth century) or drawn out (third through sixth centuries, eighth
through tenth centuries). They do not affect, however, all parts of
the system to the same degree, either because of the partial integration

38
These migrations appear to be tied to increased aridity in central Asia, the south of
Russia, and northern China during this period (R. Brown, History and Climate Change: A
Eurocentric Perspective [London: Routledge, 2001], p. 73).
39
W. H. McNeill, Plagues and Peoples (New York: Anchor Books Editions, 1998).
40
S. C. Chew, World Ecological Degradation: Accumulation, Urbanization, and Deforesta-
tion 3000 B.C.– A. D. 2000 (Walnut Creek, Calif.: AltaMira, 2001). Nevertheless, the rise
in the density of population often induced technological innovations in agriculture that
brought about improvement in soil fertility, thereby initiating a new cycle of economic
growth (Boserup, Conditions of Agricultural Growth, pp. 21–22).
41
G. Bond et al., “Persistent Solar Influence on North Atlantic Climate during the
Holocene,” Science 294, no. 5549 (December 2001): 2130–2136; Feng Sheng Hu et al.,
“Cyclic Variation and Solar Forcing of Holocene Climate in the Alaskan Subarctic,” Sci-
ence 301 (September 2003): 1890–1893. At Arolik Lake in Alaska, “increases in tempera-
ture and moisture apparently corresponded to intervals of elevated solar output and reduced
advections of ice-bearing (cooler) surface waters eastward and southward in the North
Atlantic. Conversely, decreases in temperature and moisture corresponded to intervals of
reduced solar output and increased advections of North Atlantic ice-bearing waters.”
2JWH_391-465 12/27/05 2:22 PM Page 435

Beaujard: The Indian Ocean in Eurasian and African World-Systems 435

of certain subsystems or, more often, because of particular local con-

ditions. Developments in India and Southeast Asia, most notably,
appear to occur sometimes out-of-sync with the rest of the world-sys-
tem.42 For example, the Gupta empire (fourth through fifth centuries
c.e.) emerged at a time of recession. Indeed, the intense contacts
between India and Southeast Asia during this time led to the Indian-
ization of the latter region. Although the recession in western Asia
from the ninth to the tenth century produced a sharp decrease in its
urban populations—Baghdad being a notable example—Indian cities
experienced modest growth during the same time period, and trade
between India and Southeast Asia remained strong. In the fourteenth
century, the downward spiral greatly affected the two far ends of the
world-system (perturbations in China; outbreaks of the plague in
China, Egypt, and western Asia; economic problems in the Ilkhanid
empire and its ensuing demise), while trade between India and South-
east Asia continued with little change, and Mojopahit (Java) enjoyed
great prosperity. In the fifteenth century—after 1433 to be precise—
when China, along with its huge market, practically cut itself off from
foreign trade,43 commerce was thereby forcibly disrupted in the China
Sea, but in Southeast Asia commerce continued to flourish, and
Malacca was described by Portuguese at the start of the sixteenth cen-
tury as “the biggest warehouse in the world.” This relative stability in
Southeast Asia might be attributed to a stability in climatic condi-
tions, to the strategic importance of the Straits of Malacca, and/or to
the high demand for spices from Moluccas, the main product from the
region. This asynchrony also appears to be tied up to the pace of China,

42
Gills and Frank have remarked on these phenomena of desynchronism that, they
argue, allow for the restructuration of the world-system. “It is precisely because some regions
and states get out of phase that the transformation and development of and in the system
can take place” (“World System Cycles,” p. 149). They postulate that the rise of certain
powers at a moment of global crisis stems from the weakening of others, but I do not find
this to be sufficient explanation. Why, precisely, is one state better positioned for ascent
than another? Chase-Dunn and Hall (Rise and Demise, pp. 149, 221, 224; “ Comparing
World-Systems,” p. 107) have likewise noted that South Asia and Southeast Asia have
“dynamics . . . less tightly (coupled) to the emerging Afroeurasian world system” than other
regions such as China or western Asia. Growth and decline of states in Southeast Asia
occur often in opposition to those of Chinese and Indian empires. Wilkinson shows that
subregions of the Eurasian continent are often out of sync during phases of slowdown in the
world-system (“Civilizations Are World Systems!” in Sanderson, Civilizations and World
Systems). A final example of desynchronism: the expansion of Byzantium from the late
ninth into the tenth century.
43
For reasons still debated: threat from the Mongols, fear of a massive outflow of cop-
per cash through trade, reactions against the excessive costs of Zheng He’s expeditions,
environmental problems (deforestation due to excessive shipbuilding).
2JWH_391-465 12/27/05 2:22 PM Page 436

436 journal of world history, december 2005

which often moved ahead of the rest of the system: enjoying links to
two systems (India and China), Southeast Asia could more easily
compensate for recessions and take quickest advantage when China
entered an upward trend. For the same reasons, southern India and
Ceylon likewise appear to have escaped the ups and downs of the sys-
tem. Moreover, unlike China or West Asia, these regions were usually
not the targets of foreign invasions. The Indian peninsula was contin-
uously able to take advantage of the multiplicity of commercial routes
to which it was connected.
The rise and fall of cities—whose interconnections form the sys-
tem’s “basic foundation”—serve as useful indicators of economic activ-
ity, the level of interconnectedness of regions, and transformations in
the system(s). T. Chandler’s 1987 calculations of the population of
the twenty-five largest cities are useful in this sense, but they need to
be approached with at least two caveats. First, his estimations for the
oldest periods are questionable, especially for India and China.44 Sec-
ond, his method tends to underestimate the important role of city-
states, whose economic and cultural influences are usually greater than
their demographic standing. The prominent cities belong mainly to the
great powers that control the cores of the system: Luoyang, Chang’an
(Han empire), Rome, Alexandria (Roman empire), and Ctesiphon
(Parthian empire) in the first centuries of the Christian era; Chang’an,
Luoyang (Tang empire), Baghdad (Abbasid empire), Byzantium (Byz-
antine empire), and Cordoba (Umayyad caliphate) from the seventh
till the tenth century; Kaifeng, Hangzhou, Tanzhou, Jizhou (Sung
empire), Merv (Seljukid empire), and Kalyani (Chalukya empire) in
the eleventh and twelfth centuries; Pekin, Hangzhou, Quangzhou
(Yuan empire), Delhi (sultanate of Delhi), and Cairo (Mamluk empire)
in the thirteenth and fourteenth centuries; and Pekin, Hangzhou
(Ming empire), Vijayanagara (Vijayanagara empire), Cairo (Mamluk
empire), and Istanbul (Ottoman empire) in the fifteenth and begin-
ning of the sixteenth century. But in that last period, we find also
important towns that are city-states belonging to semiperipheries:
Malacca and Venice.

44
The work of T. Chandler should today be revived and improved upon. The evolu-
tion in the number and size of cities with more than forty thousand inhabitants shows an
abrupt increase in 1400, which can be attributed in large part to the fact that the data used
by Chandler are more consistent than for preceding periods. It is nevertheless remarkable
that the data collected by Chandler show a decrease in size for the twenty-five biggest towns
in the world between 100 and 361 c.e. During this period, the Han, Kushan (ca. 200), and
Parthian (226) empires disappear.
2JWH_391-465 12/27/05 2:22 PM Page 437

Beaujard: The Indian Ocean in Eurasian and African World-Systems 437

Cores and Peripheries, Elites and Dominated Classes

From its very beginnings, the world-system has been characterized by

hierarchy, its cores and peripheries experiencing interactions that pro-
duce inequalities between regions and within the interconnected soci-
eties. These power relations are structured—most notably—by the
transfers of surplus that occur all along exchange networks.45 A crucial
factor for the success of a core—both in relation to its periphery and
its competition with other cores—is its efficient extraction of surplus.
This efficiency is based on commercial and financial mechanisms, pro-
duction processes, ideological power, and sometimes military might
(Figure 4).46
Four permanent cores can be identified: China, India, western Asia,
and Egypt, their success based on both geographic and historical fac-
tors.47 Agricultural productivity in Mesopotamia, the Nile valley, the
Ganges, the Yellow River, and the Yangzi was obviously a vital attribute
in the creation of these cores, as were the ensuing high rates of urban-
ization and extension of trading networks. Egypt and western Asia
enjoyed a vital advantage: they lie between two or three maritime areas
(the Persian Gulf and the Red Sea, the Mediterranean, and the Black
Sea). With the exception of Persian and Arab ships, which plied the
entire route from China to the Persian Gulf from the seventh to the
ninth century, trade was divided into three zones or subsystems: the
western Indian Ocean, the eastern Indian Ocean, and the China Sea.
These large zones were linked by intermediary areas. Various
regions of Arabia, Persia, India, and Southeast Asia served as cross-
roads and hinges between the sections. They were the end points of
trade routes that changed little over the ages, owing in part to the pat-
tern of trade winds, but also, in the case of some regions, to their posi-

45
A. G. Frank, “Bronze Age World System Cycles,” Current Anthropology 34, no. 4
(1993): 387. Chase-Dunn and Hall (“Comparing World-Systems,” p. 91) contend that the
core/periphery distinction does not necessarily imply hierarchy, but the mechanisms of the
world-system itself would seem to contradict this possibility.
46
Gills and Frank (“Cumulation of Accumulation” and “World System Cycles”) have
not sufficiently explained the mechanisms that assured the transfers of surplus in ancient
times.
47
In effect, India was generally a core with multiple centers (cf. supra). Rome and its
empire formed another center or “core” for the first cycle of the system. Western Europe
acquired the status of a center again only at the end of the fifteenth century in the ancient
Eurasian and African world-system and in a new system that had as peripheries North and
West Africa and the Americas. As cores, Egypt and western Asia were sometimes separate,
sometimes united.
2JWH_391-465 12/27/05 2:22 PM Page 438

figure 4. Diagram of the mechanisms of dependence core/periphery or

semiperiphery.
2JWH_391-465 12/27/05 2:22 PM Page 439

Beaujard: The Indian Ocean in Eurasian and African World-Systems 439

tion in regard to land or river routes. Trade winds, moreover, caused

merchants to spend prolonged periods in the ports of these areas, which
favored the exchange of techniques and religions.
In the modern world-system, cores dominate peripheries and bring
about their “underdevelopment.” 48 Were such economic trends also
the norm before the sixteenth century, in other words, does the nature
of core-periphery relations change after the Industrial Revolution or
after the sixteenth century? The evolution of the world-system before
the sixteenth century shows in fact that cores did exploit their periph-
eries, but also that the two evolved together, pointing to a phenome-
non of “co-evolution” or “spread effects,” made possible by the advan-
tages in geography or human resources that some regions were able to
turn to their profit. Different authors argue that it is difficult to verify,
with any certainty, the existence of the same clear division of labor
that Wallerstein claims as a characteristic of the modern world-sys-
tem. However, the pattern of a core producing manufactured goods and
extracting raw materials from a periphery is well in evidence before the
sixteenth century.49
Besides geographic factors, the long-term advantages enjoyed by
some regions (cores and, to a lesser extent, semiperipheries) in regard
to production and commerce depended on their output of manufac-
tured goods. Gains in productivity were made possible, on the one
hand, by technological innovations and, on the other hand, by an effi-
cient harnessing of labor, whether it be a slave or salaried work force,
controlled by state or private capitalism, or a system of tax or tribute
imposed by an elite. Once the system took shape, textile products con-
stituted efficient commercial weapons for the cores: silks from China
and then Persia, cotton fabrics from India, linen and woolen fabrics
from western Asia and Egypt. But there were also ceramics (porcelain
from China from the seventh or eighth century, ceramics from the
Persian gulf), glass (Roman world, western Asia, Egypt), metal objects
(Roman world, western Asia, India, China), beads (India), and so
forth.
Cities attracted cheap labor from the countryside; they also gave
rise to an elite class of artisans. Cores enjoyed a flexibility in their pro-

48
On the “development of underdevelopment” and “unequal exchange,” see A. G.
Frank, Le Développement du Sous-développement (Paris: Maspéro, 1970); A. Emmanuel,
L’échange Inégal (Paris: Maspéro, 1973); and S. Amin, L’échange Inégal et la Loi de la Valeur:
La Fin d’un Débat (Paris: Anthropos, 1981).
49
In this regard, I do not share the reservations of Chase-Dunn and Hall, “Compar-
ing World-Systems,” p. 91. Ancient Mesopotamia, which these authors cite, in fact pro-
vides a good example of this process, as it was exporting textiles and metal goods.
2JWH_391-465 12/27/05 2:22 PM Page 440

440 journal of world history, december 2005

duction and a fluidity in their organizational systems, much more so

than their peripheries, where the chances of increasing supply could
be limited or nonexistent (the growing difficulty of procuring gold and
ivory in certain regions of East Africa is a good example).50 The stream-
lining of production and commercial practices as well as the power of
ideology and military might could all tip the balance in favor of cores
over peripheries. In the creation of dominance between centers and
peripheral regions, the control of routes where exchanges occur appears
to be of fundamental importance. Merchants of peripheries use these
routes little or not at all, unlike merchants of the cores, who are able to
use them to take advantage of the different exchange values of prod-
ucts in different markets (Figure 4).
The notion of exchange value can be considered from various
angles. It can be evaluated by the amount of labor time used in obtain-
ing raw materials and slaves or in producing a commodity. On the other
hand, it can be measured by what has been obtained in the exchange
process. The exchange value in the periphery remains low because of
(1) inefficiency (for example, in capturing slaves and in gold mining),
but more so (2) the exchange rate dictated by the core through differ-
ent mechanisms. Cores were in a position to impose their products at
their own prices or exchange rates, thereby structuring the interna-
tional order of labor. The useless nature of the raw exported goods in
the periphery increases the loss in exchange value. These raw materi-
als were most often luxury goods (for example, ivory, gold, aromatic
resins from East Africa and Madagascar), but “luxury” only from the
standpoint of the importing core. In exchange, cores offered cloth,
beads, and pottery that they mass produced. The contrast between
the desirability of imported manufactured products and raw local
goods along with the gap between the societies trading with each
other appear to be fundamental elements in the formation of unequal
exchange,51 on the East African coast for example. This difference in
desirability tends to grow over time because imported manufactured
products contribute in part to the social status of the elites within

50
Moreover, because cores control commercial routes and markets, gains in their indus-
tries’ productivity do not bring about a decrease in profit for their manufactured goods; con-
versely, the profits from gains in productivity that occur in peripheries are recouped by
cores.
51
Versus M. N. Pearson, Port Cities and Intruders: The Swahili Coast, India, and Portugal
in the Early Modern Era (Baltimore: Johns Hopkins University Press, 1998), p. 116, and
P. Sinclair and T. Hakansson, “The Swahili City-State Culture,” in Hansen, A Compara-
tive Study of Thirty City-State Cultures, p. 475.
2JWH_391-465 12/27/05 2:22 PM Page 441

Beaujard: The Indian Ocean in Eurasian and African World-Systems 441

peripheries. The imbalance of power in an exchange was proportional

to the gaps—in terms of technology, military might, social complexity,
and so forth—between the two concerned parties. Besides the export
of raw materials, peripheries also served as a labor pool, supplying the
ever-voracious cities with artisans, soldiers, servants, or food produc-
ers. Indeed, slaves from East Africa were traded throughout the Indian
Ocean, all the way to China, from the time of the Tang dynasty. So too,
the Muslim world in western Asia and Egypt was importing slaves from
East Africa and Sudan, as well as from central Asia and the “Slavic
lands.” This “international division of labor” becomes sharper with
each succeeding cycle in the system.52
The movement of surpluses through networks—which occurs in
various ways 53 —gives rise to political and economic hierarchy between
regions, states, and classes and to the creation of “chains of depen-
dency.” This pattern holds true for zones lying outside of, but interact-
ing with, cores, for whom economic success is also tied to the exploita-
tion of their own “fringes” (rural areas, underprivileged social classes).
The concept of semiperipheries—which combine organizational and
institutional characteristics of both cores and peripheries—developed
by I. Wallerstein, C. Chase-Dunn, and T. D. Hall, captures well the
nature of the intermediate zones that develop between cores and the
lands they exploit. This notion helps to shed light on the success of
trading centers on the East African coast and its Swahili civilization.
It also accounts for the “Indianization” of Southeast Asia, which lay
between the cores of China and India, a process marked by the creation
of the Funan state in the first century, the thalassocracy of Sriwijaya

52
From the sixteenth century, the African inland was a hinge between the two exist-
ing world-systems. At the center of the slave trade networks, Africa provided the West and
the East with slaves. See the map of the slave trades published by C. Coquery-Vidrovitch,
L’Afrique et les Africains au XIXè Siècle (Paris: Colin, 1999), p. 190. Moreover, earlier evi-
dence for links established through the African inland are cowries, which represented a
well-established currency in West Africa from the twelfth until the nineteenth century
(C. Coquery-Vidrovitch, personal communication). These shells—which came ultimately
from the Maldivian islands—were found even earlier at Awdaghust (western Sahara) in
levels belonging to the ninth through tenth centuries (J. Devisse, “Commerce et Routes du
Trafic en Afrique Occidentale,” in Histoire Generale de l’Afrique [UNESCO, NEA, 1990]
p. 450).
53
Unlike S. Amin (“History Conceived as an Eternal Cycle,” Review 22, no. 3, [1999]:
308), I believe that core-periphery relations in periods before the sixteenth century are
already “defined in terms of economic exploitation” (although not uniquely so). Waller-
stein and Amin have unnecessarily assumed that politics and ideology dictated the econ-
omy before the end of the fifteenth century in terms of a “tributary system.” Indeed, the
notion of “a tributary age” (ibid., p. 316) before 1500 runs squarely against the historic
facts.
2JWH_391-465 12/27/05 2:22 PM Page 442

442 journal of world history, december 2005

in the seventh century, the expansion of Mojopahit in the thirteenth

and fourteenth centuries, and the development of the Pasai (from the
thirteenth century) and Malacca city-states (fifteenth century). The
current debate over whether these semiperipheries came about through
“colonization” or indigenous initiative is misplaced. In fact, “agents”
from cores (or centers of semiperipheries) were present and active in
the merchant circles of the peripheries—for example, the Persians,
Egyptians, Arabs, and Indians residing on the coast of East Africa. The
alliances they created with the local elite, or the elite that eventually
formed in the area—through commercial alliances, but also through
intermarriage, the formation of “blood brotherhood” pacts and the reli-
gious conversion of elites—gave rise to centers that played an inter-
mediary role between the main core and the periphery (for example,
the African hinterland, and the Comoros Islands and Madagascar for
East Africa). These alliances resulted in the ideological, economic,
and political configuration of the periphery or semiperiphery, which
increased the dependency and flow of goods toward the core, a depen-
dency that profited the elites at the expense of the “fringes” of their
own sociopolitical entities. Indeed, it is the addition of a new periph-
ery to the world-system that brings about the creation of new commer-
cial centers that serve as “pivots” between the distant land and the
main core. “For the various craft activities,” the Swahili, for example,
“were able to exploit the technological advantage gained through their
wide-ranging contacts, and possibly by the actual settlement of over-
seas craftsmen within the towns.” The Swahili restricted the supply of
many items to which they had access through their Indian Ocean part-
nerships, and they have “actively kept Islam as a coastal monopoly.” As
Horton and Middleton underline it, “The trading relationship between
the coast and the interior has always been asymmetrical—one of
exploitation of Africa by the outside world. Typically, the interior has
exchanged its natural and raw materials for manufactured commodi-
ties.” 54 Scholars such as M. Pearson, who claim that on the East Afri-
can coast exchange was not unequal but beneficial for all trading part-
ners willfully ignore the mechanisms of exploitation, the size of the

54
M. Horton and J. Middleton, The Swahili: The Social Landscape of a Mercantile Soci-
ety (Oxford: Blackwell Publishers, 2000), p. 102. Cf. also E. A. Alpers, Ivory and Slaves:
Changing Patterns of International Trade in East Central Africa to the Later Nineteenth Century
(Berkeley: University of California Press, 1975); and A. Sheriff, “Trade and Underdevel-
opment: The Role of International Trade in the Economic History of the East African
Coast before the Sixteenth Century,” in Hadith 5: Economic and Social History of East
Africa, ed. B. A. Ogot (Nairobi: East African Literature Bureau, 1976), pp. 1–23.
2JWH_391-465 12/27/05 2:22 PM Page 443

Beaujard: The Indian Ocean in Eurasian and African World-Systems 443

exchanges—in volume, space, and time55 — social stratifications pro-

duced by long-distance trade,56 and the existence of a sustained and
prolonged slave trade.57
While semiperipheries adopt many technological and social aspects
of the core, they can also create novel ones; among many examples,
one can cite the creation of an alphabet in Ugarit and Palestine toward
the thirteenth century b.c.e., the development of navigation in South-
east Asia around the first century c.e., and the banking systems and
“Republican” forms of organization in Italian cities after the thirteenth
century c.e.
In the modern world-system, the proof for the underdevelopment
of peripheries is readily visible—and quantifiable—in the “decline in
the rates of exchange,” but such figures are not available for the Indian
Ocean before the sixteenth century. Some peripheries were no doubt
impoverished by their weaker position in a network structured by
power relations, but maybe in a less permanent and irremediable man-
ner than in the modern era. It is probable that the peripheries most

55
Ivory was not only a by-product of elephant hunting, and searching for gold was not
just an occasional activity to exchange gold for cloth. See The Book of the Wonders of India,
which speaks of Africans working like ants in galleries, and Abu al-Fida’, who quotes Ibn
Sa‘id (beginning thirteenth); G. S. P. Freeman-Grenville, The East African Coast: Select
Documents from the First to the Earlier Nineteenth Century (Oxford: Clarendon Press, 1975),
pp. 15, 24. Cf. also Portuguese accounts about the Manica country and the chiefdoms of
Amçoçe and Mazofe (Mazoe) in 1512 and 1573, W. G. L. Randles, L’Empire du Monomo-
tapa du XVe au XIXe Siècle (Paris: Mouton, 1975), pp. 81, 122. Pearson himself acknowl-
edges an export of ten tons of gold a year from the coast of Sofala before the end of the fif-
teenth century (Indian Ocean, p. 84).
56
Cf. the effects of long-distance trade on the construction of hierarchical societies as
in the Limpopo area and Great Zimbabwe between the twelfth and fifteenth centuries.
According to M. N. Pearson, the Swahili and foreign merchants were able to make their
“huge profits” without any recourse to exploitation. Instead, “Overall, the advantage lay
with Africa. . . . Africans could work as much or as little as they wanted” to extract gold
and acquire imported products that “were discretionary rather than necessities in their
agricultural and hunting lives, except for cloth on the plateau.” Thus, for Pearson, cloth is
bought only for protection against the cold. It is not tied to social status, about which Pear-
son has little to say anyway. Pearson, Port Cities and Intruders, pp. 117, 124. Moreover, it is
certainly erroneous to consider that in the fifteenth or sixteenth century, “the traders from
the [East African] coast had to try and create a market de novo” (ibid., p. 113). Exchanges
have been developing between coast and interior for more than a thousand years. If it was
so difficult for the traders to sell their products, how did they manage to make such “huge
profits”?
57
It is as if the razzias for slaves carried out by coastal groups did not inflict any harm
or social damage on the peripheral societies. These razzias probably are just part of the
“benign” effects of the trade mentioned by Pearson (Port Cities and Intruders, p. 119). The
effects of the slave trade were all the more devastating because women probably formed the
larger number of slaves. On the East African coast, women play an essential role in agri-
cultural production.
2JWH_391-465 12/27/05 2:22 PM Page 444

444 journal of world history, december 2005

subject to domination were those within the core itself, both the geo-
graphic edges and weaker social groups. With the advent of European
capitalism, the chasm between the core and the periphery widened
deeply, owing in part to the rapid rate of technological innovation,
which was much more marked than in the system(s) of the ancient
world. But innovation was only one of the tools of power.
The instruments of economic exploitation and/or cooperation were
moreover embedded in ideological influence and domination. The
spread of Islam and Christianity appear to have been tied to economic
and political relations. “The Missionaries teach the natives that they
are naked and the merchants sell them cloth.” 58 Religious men often
become traders. Thus scholars who argue that societies have “choice”
in their trade relations overlook the glaring issues of power relations
and the fact that the desirability of products is largely the result of pro-
cesses of domination.
Contacts between cores, semiperipheries, and peripheries can be
violent or pacific. In East Africa, the establishment of relations with
the outside world stimulated trade far inland and gave rise to the cre-
ation of ruling chiefdoms and city-states on the coast, whose relations
with the interior tended to be formed through alliances rather than
military force,59 even if the continued existence of local slavery and
the exportation of slaves resulted in raids and wars.60 This ability to
form alliances—combined with their geographic advantage—no
doubt explains the stability of the Swahili cities despite their seem-
ingly fragile position.
As a different strategy for cores or semiperipheries, the creation of
an empire is a policy to control directly—through force—the totality
of routes and accumulation centers in a system (or a part of a system)
and therein to increase the level of surplus extracted. Various political
constructions have aspired to become “universal” empires: the Achae-
menid empire, the Greek empire under Alexander the Great, and the
Mongol empire under Gengis Khan and his successors.
When points of trade are separated by long distances, the extrac-
tion of resources tends to be accomplished through commerce and

58
A. Sherratt, “Envisioning Global Change: A Long-Term Perspective,” in Denemark
et al., World System History, p. 121.
59
For the case of Southeast Asia, see P.-Y. Manguin, “The Amorphous Nature of
Coastal Polities in Insular Southeast Asia: Restricted Centres, Extended Peripheries,”
Moussons 5 (2004).
60
Cf. for example Ibn Battuta on Kilwa: “They are devoted to holy wars because their
country is near pagan Zenj” (Ibn Battuta, Voyages, vol. 2, De La Mecque aux Steppes Russes
[Paris: F. Maspéro, 1982], p. 90).
2JWH_391-465 12/27/05 2:22 PM Page 445

Beaujard: The Indian Ocean in Eurasian and African World-Systems 445

taxes, and only rarely through raids or a military presence. Dominance

was achieved, in large part, through economic and ideological power.
Yet occasionally we find examples of a core undertaking military incur-
sions in a semiperiphery: the raids of the Chola empire against Sriwi-
jaya in the eleventh century, the expedition of the Yuan of China
against Java and Champa in the thirteenth century, and the cam-
paigns of Egypt into Yemen in the twelfth century and again in the fif-
teenth century. Before the arrival of the Portuguese, there occurred
only one large-scale economic and politico-military intrusion across
the length of the Indian Ocean, that of the imperial Ming flotilla at
the start of the fifteenth century. Conversely, however, a (semi)periph-
ery frequently conquered a core or several cores, as was the case for the
Arab semiperiphery in the creation of the Muslim empire, for a Cen-
tral Asian semiperiphery in the emergence of the Kushan empire in
India (first century), and for the Mongol invasions of Mesopotamia and
China (thirteenth century) and later of Mesopotamia and India (from
the end of the fourteenth through the early fifteenth century).
Power relations evolve within the spatial and temporal context of
the pulsations within the system. Cores and peripheries underwent
alternating phases of integration and disintegration: after holding a
dominant position, a power went into decline, followed by the emer-
gence of several regional rival powers. This tended to happen during
times of downturn. The history of India, particularly northern India,
provides a good example of these alternating moments of centraliza-
tion and decentralization. In a similar way, between the Han, Sui then
Tang, Sung, and Ming empires, China went through periods of politi-
cal fragmentation and often social chaos. The balance of power oscil-
lated, too, between regions, the result of economic competition and
changing politico-military strength. In the west, politico-economic
preeminence went back and forth between the Persian Gulf (third
through ninth centuries) and the Red Sea (second century b.c.e.
through second century c.e., tenth through fourteenth centuries c.e.).
Periodically, a dominant power was able to simultaneously control both
of these maritime trade “corridors” to the Indian Ocean (for example,
the Assyrian empire in the seventh century b.c.e., the Persian Achae-
menids, Greece in the time of Alexander the Great, and the Muslim
caliphate), or at least it made the attempt (for example, the Roman
empire in the first century b.c.e. and the second century c.e., the Mon-
gols in the thirteenth century). In Southeast Asia, the route to China
and the spice route first passed over the northern end of the Malay
Peninsula (until the sixth century), then moved to go around the
southern end of the Malay Peninsula and the southeast coast of Suma-
2JWH_391-465 12/27/05 2:22 PM Page 446

446 journal of world history, december 2005

tra (seventh through thirteenth centuries), next shifting to Java (thir-

teenth through fourteenth centuries) before finally returning to the
Malay Peninsula at Malacca in the fifteenth century.
Hierarchies are not immutable in space or time and the history of
the world-system is not “an eternal cycle” (S. Amin, 1999). Spurred by
technological innovations, along with processes of accumulation and
demographic increase, the general trend, from one cycle to another,
has been for ever-increasing hierarchy and specialization among the
zones of the system—and within zones—and for ever-increasing polit-
ical and economic integration.61 In general, the given data point to
incrementally increasing expansion.

Continuity and Change: The Expansion of

the World-System

The succeeding cycles in the system are at once similar and different.
Any changes that do occur do not affect the structure or the nature of
the system. Fundamental transformations are noticeable at only two
moments: (1) from 3500 to 2400 b.c.e. in western Asia and Egypt and
in the second millennium in China, when the state was born along
with private means of accumulation, which broke with the former
mode of accumulation based on kinship relations, and (2) toward the
middle of the eighteenth century, during the Industrial Revolution—
which also represents an ideological revolution—when the capitalist
mode of accumulation became, for the first time, preeminent in a
world-system that would henceforth link America, Africa, and Eura-
sia. The sixteenth century, however, also represents a decisive moment,
when a new ocean, the Atlantic, was joined to the Indian and Med-
iterranean Oceans—linked since ancient times—a moment when
Europe emerges as core of a new space.
Each new period of growth creates a new spatial division of labor,
although it builds on the preceding one. With the exception of the
oscillations between rival regions mentioned above, commercial routes
—at least the principal ones—vary little over time. Secondary net-
works are more flexible, widening and becoming more dense during
phases of growth in the world-system and contracting during periods
of regression, as these secondary routes arise or disappear according to

61
Cf. C. Chase-Dunn and T. D. Hall, “The Historical Evolution of World-Systems,”
Sociological Inquiry 64, no. 3 (1994): 271.
2JWH_391-465 12/27/05 2:22 PM Page 447

Beaujard: The Indian Ocean in Eurasian and African World-Systems 447

the resources on offer (raw materials, manufactured goods) and to

those in demand, as well as in relation to geographic and political fac-
tors. Trade winds, for instance, are of great importance to ports located
between two oceans. Pivotal regions (for example, Southeast Asia,
Oman, Yemen, and Ceylon) retain a primordial role in the networks
throughout history. The same holds true for ports situated at the
mouths of rivers or harbors that serve as the final destination for major
land routes (for example, the mouths of the Tigris and the Euphrates,
the Indus and the Ganges, the Yangzi, and the Gujarat, which abuts the
roads leading to and from the Ganges valley) and for certain regions
that lie close to essential raw materials (the gold of Zimbabwe, the
spices of the Moluccas, and so forth).62
In the far hinterland, particular zones can exert a marked influ-
ence, creating a link or, conversely, a barrier. For example, central Asia,
through which passed the Silk Roads, played a crucial role in the com-
merce between China, India (by way of the Hindu Kush), and west-
ern Asia. The confederations and empires of the peoples of the steppes
developed hand in hand with those of China. All the great empires,
whether of China, northern India, Persia, or the Abbasid caliphate—
with the exception of the Sung empire—sought to control central
Asia, a strategy that was profitable, however, only during periods of
expansion in the system. Mention should be made too of the “steppe
routes” lying farther to the north and the importance of the routes
between China and Burma and the trans-Indian passage between the
Ganges and Gujarat or across the Deccan.
Land routes tend to complement sea routes rather than compete
with them. They tend to develop in conjunction with sea passages,
even if a state may concentrate its efforts on a certain route, with the
purpose of getting around another state’s control over a key region. In
the first century, the Romans were able to bypass Arab merchant car-
avans by the building up of shipping in the Red Sea. Byzantium relied
on a route across the steppes of Khazaria in order to avoid Persian com-
petitors in the sixth century. But the flourishing of the Silk Roads and
the route across the Mongol steppes in the thirteenth century did not
weaken trade in the Indian Ocean.63

62
Some ports, however, that served as warehouses and crossroads had practically no
hinterland and owed their success entirely to their strategic geographic location. This was
the case for Hormuz and Malacca.
63
I therefore disagree with Bosworth who sees a “tension” between land and sea routes
and argues for the existence of cycles that alternated between silk routes by land and spice
routes by sea (Bosworth, “The Evolution of the World-City System, 3000 bc to ad 2000,”
in Denemark et al., World System History, p. 282).
2JWH_391-465 12/27/05 2:22 PM Page 448

448 journal of world history, december 2005

Not only was there continuity in commercial routes, there was also
continuity in the types of products being exported. India, for example,
was exporting cotton cloth as early as the first century c.e., along
with beads and precious stones. The general trend has been for certain
basic types of goods and products to influence commerce and the move-
ments of peoples over the course of several centuries; such products
include metals (both precious and nonprecious), clothing items, lux-
ury goods for body adornment, perfumes, medicines (for religion and
health), aromatics (for health and cooking), means of production
(slaves, tools, and so forth), means of transport (for example, ships),
instruments of war (weapons, horses, elephants), food, rare goods, and
sacred objects.
This continuity was embedded in a general trend of expansion and
intensification that itself was linked to technological innovation and
social developments brought about, in part, through economic con-
straints and sometimes—especially in the case of agriculture—demo-
graphic and ecological factors.64 This rhythm could intensify in certain
times and in certain places. The free trade promoted by the Sung in
China from the tenth to the thirteenth century gave rise to innova-
tion and growth. The competition between states in Europe from the
sixteenth century had the same effects. As I have underscored, progress
in agriculture played a crucial role, as it is the cornerstone for urban-
ization and the development of handicraft or semi-industrial produc-
tion that fuels trade in both internal and external markets.
Progress in ship building, especially in the twelfth and thirteenth
centuries, reduced transport times and influenced the types of products
that were shipped. It allowed for the transport of ever-larger loads of
heavy and inexpensive materials (for example, raw materials, agricul-
tural products, and so forth). This point is important for our discussion
of world-systems. Wallerstein rejects “Frank’s so-called world-system”
before the sixteenth century on the grounds that it traded only in lux-
ury goods, not staple items, and thus, in consequence, it could not
have been organized by the “axial division of labor” that characterizes
the modern world. This reasoning, however, does not hold up under

64
Building on Julian Steward’s notion of “cultural ecology” (Theory of Culture Change:
The Methodology of Multilinear Evolution [Urbana: University of Illinois Press, 1955]), J. Dia-
mond uses data from archeology, history, genetics, and molecular biology to argue that envi-
ronmental conditions gave rise to the development of agriculture and conferred a decisive
advantage on Eurasian societies (Guns, Germs, and Steel: The Fates of Human Societies [New
York: W. W. Norton, 1997]).
2JWH_391-465 12/27/05 2:22 PM Page 449

Beaujard: The Indian Ocean in Eurasian and African World-Systems 449

scrutiny.65 Agricultural products—and raw materials in general—were

part of trade networks from the very beginnings of the system, evi-
dence of which can be found in the Periplus of the Erythrean Sea 66 and
recent archaeological excavations of the Egyptian port of Berenice.67
Well before this time, in the third millennium b.c.e., evidence shows
tar, oils, and grain moving from Mesopotamia to Dilmun and Oman,
wood and dairy products from India to Mesopotamia and Dilmun, and
copper and hard stone from Oman to Mesopotamia. Moreover, some
authors, as Gills, think that the systemic nature of trade does not
derive primarily from the types of products that are exchanged (“pre-
ciosities”/“necessities”), but from the fact that there exists a transfer of
surpluses between the zones of the system.68 In addition, one could
question Wallerstein’s notion that the trade in luxury goods has no
important systemic impact.69 Indeed, some authors have argued the
opposite, that the circulation of luxury goods has structural implica-
tions because of their close connection to ruling elites, who wish to
control them, any change in the supply of these goods having reper-
cussions for political hierarchies.70
Significant gains in production and commerce are made possible
through the use of money and the buildup of banking services. Thus
we find that financial institutions arise from the very start of Islam.
They built on the experiences of the Sassanid Persians, the Byzantine
empire, and from even further back, Persia and Mesopotamia of the
first millenium b.c.e. In the western Muslim world and in India, guilds
and merchant associations played the role of bankers, prime examples
being the Ayyavole of Deccan and the Manigramam of Tamil Nadu

65
I. Wallerstein, “World System Versus World-Systems: A Critique,” in World System:
Five Hundred Years, pp. 293–294; and Chase-Dunn and Hall (“Comparing World-Sys-
tems,” p. 92) also contend there was “no movement in raw materials” in ancient times.
66
Cf. L. Casson, Periplus Maris Erythreai (Princeton, N.J.: Princeton University Press,
1989).
67
Cf. R. T. J. Cappers, “Archaeobotanical Evidence of Roman Trade with India,” in
Ray, Archaeology of Seafaring, pp. 51–69, W. Z. Wendrich et al., “Berenike Cross-Roads:
The Integration of Information,” Journal of the Economic and Social History of the Orient 46,
no. 1 (2003): 46–87.
68
B. K. Gills, “ Capital and Power in the Processes of World History,” in Sanderson,
Civilizations and World Systems, p. 146.
69
Wallerstein, Modern World-System, vol. 1, pp. 41–42.
70
J. Schneider, “Was There a Pre-Capitalist World-System?” Peasant Studies 6, no. 1
(1977): 20–29; P. Peregrine, “Prehistoric Chiefdoms on the American Mid-Continent: A
World System Based on Prestige Goods,” in Core/Periphery Relations in Precapitalist Worlds,
ed. C. Chase-Dunn and T. D. Hall (Boulder, Colo.: Westview Press, 1991), pp. 193–211;
Chase-Dunn and Hall, Rise and Demise, pp. 13–14.
2JWH_391-465 12/27/05 2:22 PM Page 450

450 journal of world history, december 2005

from the eighth to the thirteenth century, and the Karimi Egyptians
from the twelfth to the fourteenth century (even if the latter was not
strictly a guild). Increasing monetarization occurs over the centuries,
although at different, sometimes opposing, rhythms from region to
region.71 The creation of large political entities (for example, the Abba-
sid empire, the Delhi Sultanate, the various Chinese empires) natu-
rally facilitated the spread of monetary systems over wide zones.
Precious metals—in the form of merchandise and/or money—and
their flow played an important role in the structuring of the world-sys-
tem and its evolution. Precious or highly sought-after objects, which
sometimes served as money (a standard of value and/or means of
exchange), were spread along commercial routes. Cowries were used
from Bengal to Assam, in Burma and Yunnan, as well as in Thailand
and Annam. Silk, of course, was ubiquitous in caravan and maritime
networks and served in different regions as a means of exchange and
also as a unit of value (for example, Nanzhao in the ninth century, or
Cambodia from the eighth to the thirteenth century).
Cores that lacked precious or semiprecious metals sometimes over-
came the shortfall and created a tool of economic expansion through
the development of credit, as was the case for the Muslim world from
the eighth to the tenth century, in Ayyubid Egypt, and in China from
the ninth century.
From ancient times, market prices played a role in the movements
of goods. In the first century c.e., the Periplus records that Egypt
exported copper to Barygaza (Gujarat), and that this same port also
exported copper (either of Egyptian origin or coming from production
centers in the Northwest of India) to Apologos (near Basra). Marco
Polo described a favorable gold-to-silver ratio that encouraged mer-
chants to transport silver from Yunnan to Burma.
Commercial routes, however, were not the only networks that con-
tributed to the birth and evolution of the world-system; other types of
networks intersected with those of merchants. Religion, for example,
played a key role in the history of the Indian Ocean and the China
Sea. The development of Buddhism, Christianity, and then Islam,
which saw the worshipper’s relation to the divinity move toward an
increasingly individualized encounter, occurred within the context of

71
For Southeast Asia, see, for example, R. S. Wicks, Money, Markets, and Trade in Early
Southeast Asia: The Development of Indigenous Monetary Systems to AD 1400 (Ithaca, N.Y.:
Cornell University Press, 1992).
2JWH_391-465 12/27/05 2:22 PM Page 451

Beaujard: The Indian Ocean in Eurasian and African World-Systems 451

increasing private enterprise.72 Buddhism and Islam spread along mer-

chant routes, contributing to their growth. H. P. Ray has argued for a
tie between Buddhism and entrepreneurial spirit, and D. Lombard has
made a similar case for Islam.73 Clearly tied to commerce, the expan-
sion of Islam went hand in hand with the growth of the world-system
and contributed to the integration of its different parts. Indeed,
D. Lombard has called the Indian Ocean before the fifteenth century
the “Islamic Sea.” In Islam, religion and trade formed a continuum.
Pilgrimages to shrines or to tombs of saints gave rise to merchant gath-
erings and even to fairs, notably in India.74
Other religious diasporas created fairly powerful networks over the
centuries. Jewish communities, based on the coast of Kerala from the
time of the Romans, were present in China from the eighth century
and probably earlier. Their networks of the eleventh through twelfth
centuries, which radiated from Egypt, are well known, thanks to the
Geniza documents of Cairo.75 I have already made reference to the
role of merchant organizations such as the Tamil guilds (active until
the fourteenth century). On a more modest scale, Nestorian Christians
were also active throughout the Indian Ocean and along the routes of
central Asia. Along with trade networks, large political organizations
have also contributed to the expansion of the major religions: the

72
L. Dumont (Essays on Individualism: Modern Ideology in Anthropological Perspective
[Chicago and London: University of Chicago Press, 1986], p. 27) writes, “There is no
doubt about the fundamental conception of man that flowed from the teaching of Christ:
as Troeltsch said, man is an ‘individual-in-relation-to-God.’” Cf. also M. Augé, Génie du
Paganisme (Paris: Gallimard, 1982), p. 139, about the figure of an individual human destiny
shaped by Christianity. “Salvation, Transcendence and the Mystery,” Augé continues (ibid.,
p. 14), “are essentially absent . . .” from paganism. The importance of the individual was
established in schools of thought from Hellenistic times (e.g., Cynics, Epicurians, Stoics).
Dumont holds that Christianity’s emphasis on the individual reached its peak with Calvin
during the Reformation: “the individualist value now rules without contradiction or restric-
tion” (Dumont, Essays on Individualism, p. 57).
73
H. P. Ray, The Winds of Change: Buddhism and the Maritime Links of Early South Asia
(New Delhi: Manohar, 1994); D. Lombard, “Y a-t-il Une Continuité des Réseaux March-
ands Asiatiques?” in Marchands et Hommes d’Affaires Asiatiques dans l’Océan Indien et la Mer
de Chine: 13e–20e Siècles, ed. D. Lombard and J. Aubin (Paris: EHESS, 1988), pp. 11–18.
74
As M. N. Pearson underlines it (“The Indian Ocean and the Red Sea,” in The His-
tory of Islam in Africa, ed. N. Levtzion and R. L. Pouwels [Athens: Ohio University Press;
Oxford: James Currey; Claremont: David Philip, 2000], p. 44), “the hajj is a remarkably
efficient method of integrating the worldwide [Muslim] community.”
75
See S. D. Goitein, “From the Mediterranean to India: Documents on the Trade to
India, South Arabia, and East Africa from the Eleventh and Twelfth Centuries,” Speculum:
A Journal of Mediaeval Studies 29 (1954): 181–197, for the networks linking Egypt and
India.
2JWH_391-465 12/27/05 2:22 PM Page 452

452 journal of world history, december 2005

Roman, then the Byzantine and Carolingian empires to the develop-

ment of Christianity, the Umayyad and Abbasid empires, then the
sultanate of Delhi to the spread of Islam.
Cities served as the point of articulation between commercial
routes and religious networks. Increasing urbanization along with gen-
eral demographic growth and improvements in communication were
integrally tied to the creation and expansion of the world-system.
Although originating in the cores of the system, urbanization also
spread to the peripheries. The engines of production and trade, cities
were everywhere marked by social divisions, although these were more
pronounced in cores than in peripheries. The development of networks
resulted, moreover, in a hierarchical relationship between metropo-
lises and their satellite cities. “Clusters” of cities were thus formed, the
intensity of their interactions being shaped by pulsations in the system.
Over the ages, metropolises rose, fell, and replaced one another, such
changes accompanying a restructuring of networks and hierarchies,
which could be political and economic in nature, sometimes also social
or ideological. Commercial centers moved between competing zones.
In the Persian Gulf, they moved from Siraf (seventh to ninth century)
to Qays and Hormuz (from the eleventh century, and especially from
the fifteenth century in the case of Hormuz); from the mouth of the
Indus (Daybul from the eighth to the tenth century) in the Sind to
Gujarat (Somnath in the eighth to the tenth century and Cambay
from the thirteenth to the fifteenth century) in the northwest of India;
in East Africa, from Rhapta (at the mouth of the Rufiji River, from the
first century) to the archipelago of Lamu, Pemba, and Zanzibar (ninth
through tenth centuries especially) then to Kilwa (after the eleventh
century, and especially from the fourteenth through fifteenth cen-
turies), then to Mogadishu and Mombasa (in the fourteenth and fif-
teenth centuries). D. Lombard has shown that in Java in the thirteenth
century (with the momentum building in subsequent centuries) the
major centers moved from the “hydraulic” cities of the interior to the
political and economic ports on the coasts.
A distinguishing feature of cities throughout the world-system, and
from its very beginnings, is their cosmopolitan nature, which only
increased as the network developed. This went hand in hand with reli-
gious tolerance: tolerance at the level of political authorities and also,
at certain moments, through the intermingling of religious networks
(for example, by Muslim and Jewish merchants in Egypt and Yemen in
the eleventh century), and although it may not have been the general
rule, it is nevertheless a remarkable feature. In this regard, J. Aubin
has justly remarked on “the indifference of the economic world to reli-
2JWH_391-465 12/27/05 2:22 PM Page 453

Beaujard: The Indian Ocean in Eurasian and African World-Systems 453

gious matters.” 76 As counterpart to the cosmopolitan cities was the

transnational character of networks (cf. for example, the Gawan fam-
ily of the fifteenth century).77 S. D. Goitein’s description of Egypt in
the twelfth century is applicable to other cities in other times: the per-
tinent division “was less in terms of religion and nationality than in
terms of the warrior leaders and the merchant entrepreneurs.” 78

States and Private Capital

Data show that from the first millennium b.c.e., private capital has
played an important role, alongside state capital. Characteristics con-
sidered particular to the modern capitalist system—increasing produc-
tion of goods, enterprises seeking maximal profit, salaried labor, tech-
nological progress—are in fact present before the sixteenth century
in numerous states. In some periods, the state is the major engine of
accumulation; in others, private entrepreneurs. Recent research has
questioned the presumed importance of the state in the development
of commerce in ancient Mesopotamia, Maurya India (fourth through
third centuries b.c.e.), and the Roman empire in its relations with
south Asia.79 In any event, it is clear in other cases such as the period
today known as “medieval,” in the Abbasid caliphate (eighth and
ninth centuries), southern India (eighth to eleventh centuries), Ayyu-
bid Egypt (twelfth to thirteenth centuries), the sultanates of Bengal
(fourteenth to fifteenth centuries) and Gujarat (fifteenth century),
that capitalism came to dominate the spheres of production and com-
merce, a phenomenon that has been qualified incorrectly as “mer-
chant” capitalism. As would later be the case in Europe, these entre-
preneurs were at the same time producers, merchants, and financiers.
Indeed, a general trend can be observed for “a complex mixture or
articulations of modes (of accumulation) at all times in the develop-

76
J. Aubin, “Y a-t-il Interruption du Commerce par Mer Entre le Golfe Persique et
l’Inde du XIe au XIV e Siècles?” in Lombard and Aubin, Marchands et Hommes, p. 197.
77
As but one example, Mahmud Gawan was a learned man and horse merchant whose
family originated in Gilan. He served as a minister in the Bahmani sultanate of India while
his brother, Ahmad, was based in Egypt. One of Ahmad’s sons did business with India while
two others, Koranic scholars, lived in Mecca. On the importance of diasporas and networks
that bypassed “ethnicity” and nations, see also P. Curtin, Cross-Cultural Trade in World His-
tory (Cambridge: Cambridge University Press, 1984).
78
Goitein, “From the Mediterranean to India,” p. 197.
79
For example, cf. Ray, Archaeology of Seafaring, pp. 188–189, 192.
2JWH_391-465 12/27/05 2:22 PM Page 454

454 journal of world history, december 2005

ment . . . of the world-system,” 80 with alternating phases of develop-

ment and decline in the private sector.81 It is true that capitalism in
ancient times rarely assumed a preeminent position, which contrasts
with Europe after the sixteenth century, when capitalism permeated
the structures of the state, thereby laying the groundwork for the birth
of the capitalist world-system.82 The Chinese Sung empire provides an
example of a state and its society being transformed by burgeoning
capitalism, with the creation of an urban bourgeoisie and internal
markets. This evolutionary path in China would, however, be halted
by the foreign threat that weighed on the country and internal con-
tradictions in Chinese society. Under the Ming, commerce would be
more strictly controlled by the state, especially after 1433. State poli-
cies vis-à-vis private long-distance commerce varied greatly, ranging
from encouragement (under the Abbasids, the Egyptian Ayyubids, and
the Sung) to restrictive measures (direct or indirect) to outright con-
trol. Control could take various forms. Sriwijaya in the eleventh cen-
tury established a monopoly in the trade of sandalwood (and probably
in other products). The Yemeni Rasulids, in the 1420s, discouraged
merchant activity through imposing high taxes and making various
exactions. In Egypt, from 1429, the Burji Mamluks created a monop-
oly on the trade in spices, fixed market prices, and overburdened mer-
chants through taxation and the strict restricting of their business,
thereby killing “the goose that laid the golden egg” in what was an
already difficult international scene.
Capitalism developed not only within states but also within the
context of the transnational networks that I have described.
Another setting for the development of capitalism, one that was

80
Frank and Gills, “The 5000-Year World System: An Interdisciplinary Introduc-
tion,” in The World System, p. 46. Here I am in agreement with Frank, at least for the peri-
ods preceding the eighteenth century. Frank holds that it is illusory to try to distinguish a
qualitative difference in phases of the world-system based on modes of production. There
was never a progressive succession of modes of production; instead, the various types of
modes could be operating simultaneously, in different combinations and relations. Indeed,
Frank goes so far as to say “This received conceptualization has continued to divert our
attention away from the much more significant world systemic structures and processes”
(ReORIENT, p. 331).
81
Chase-Dunn and Hall (Rise and Demise, pp. 212–213, and “Comparing World-
Systems,” p. 101) make reference to the “oscillation between state-based and private cap-
italist accumulation.”
82
Chase-Dunn and Hall, Rise and Demise, p. 47. Gills, however, argues, “the entire
Eurasian world-economy/system of the thirteenth century was perhaps already ‘capitalist’
(and perhaps even that of) the tenth century” (“Capital and Power,” p. 139). It does, in
fact, appear that capitalism played an increasingly important role from one cycle to the
next.
2JWH_391-465 12/27/05 2:22 PM Page 455

Beaujard: The Indian Ocean in Eurasian and African World-Systems 455

historically primordial, was the city-states and small merchant sul-

tanates that flourished from the thirteenth century in India, Southeast
Asia,83 and on the Swahili coast. Particular cases of note include
Honavar and Calicut in India in the fourteenth and fifteenth cen-
turies; in Southeast Asia, Samudra-Pasai from the end of the thir-
teenth to the fifteenth century; cities on the north coast of Java in the
fourteenth through fifteenth centuries—dependencies of Mojopahit
that became increasingly independent from the fifteenth century;
Malacca in the fifteenth century; and Aceh after 1511. For the Swahili
coast, the most important cities were Kilwa from the twelfth to the fif-
teenth century and Mogadishu in the fourteenth and fifteenth cen-
turies. As seats of power but also the results of social contracts between
political elites and merchant-producers, these cities blossomed in the
interstitial regions of empire, in the semiperipheries of the system.
Occasionally, however, a more or less autonomous city-state devel-
oped within a central state of the system (for example, Hormuz in the
fifteenth century). In other cases, they have been able to build real
empires, as Sriwijaya from the seventh century 84 and, much earlier in
history, Phoenician ports and Greek cities in the first millennium b.c.e.
The latter represent a good example of these city-states where a mer-
chant capitalism flourished, based on investment in production aimed
at export.
The notion of city-states as interstitial formations (but we have
seen that they have not only developed in that framework) has been
used by G. Arrighi, following F. Braudel. It is to them, and to transna-
tional commercial networks, that can be traced the essential origins of
the modern capitalist system. The obvious examples are the Italian
city-states—Pisa, Venice, Genoa, Florence—which, from the twelfth
century, took advantage of the absence of a strong Italian “country-
state” 85 to form a semiperiphery. Other cities likewise emerged in
northern Europe (Bruges, Lubeck, and the ports of the “Hanseatic
League”).
Specialized in specific forms of production, in long-distance com-
merce, and intermediary activities, city-states—those “cités à l’état

83
The city-states of the Malaysian world in fact appear much earlier (P.-Y. Manguin,
“City-States and City-State Cultures in Pre-15th Century Southeast Asia,” in Hansen, A
Comparative Study of Thirty City-State Cultures, pp. 409–416), as well as the city-states of
the Swahili East African coast (Sinclair and Hakansson, “Swahili City-State Culture”).
84
Manguin, “ City-States and City-State Cultures.”
85
Hansen, “ Concepts of City-State and City-State Culture,” p. 16. Hansen proposes
to use this term “country-state” or “macro-state” instead of “territorial state.”
2JWH_391-465 12/27/05 2:22 PM Page 456

456 journal of world history, december 2005

pur” 86 —were extremely efficient, owing in part to the low cost of the
state sector and, even more so, to their dynamic momentum, which
attracted capital and skilled labor. But they were also evidently a vul-
nerable and tempting prey for territorial states. Thus, they were always
careful to build up forms of defense, create alliance networks (for exam-
ple, the relations between Swahili cities and groups in the interior),
and take the best advantage of their geographic location (for example,
the cities of the Straits of Malacca). Even guilds and private merchants
made recourse of men at arms; examples include the great Indian
guilds of the fifth to fourteenth centuries, the ships of the west coast
of India in the fourteenth century, and Chinese junks during the same
time period. Concerning the princes who engaged in commerce, two
interpretations are possible: they might represent merchants who pre-
figured capitalist entrepreneurs, or they may have acted as predators.87
Ibn Battuta provides numerous examples of flotillas owned by gov-
ernment officials. His descriptions reveal the close ties that existed
between commerce, political power, and also piracy, as illustrated in
the case of the sultan of Honavar, Jamal al-Din (fourteenth century).88
Likewise, within the core of a system, certain periods of weak polit-
ical integration seem to have encouraged the development of produc-
tion and commerce: competition between states or city-states stimu-
lated the economy, and private entrepreneurs enjoyed greater liberty
than those operating in stronger states. C. Edens has clearly demon-
strated this phenomenon for Early Dynastic III (in the middle of the
third millennium b.c.e.) and for the Isin-Larsa period in western Asia
(beginning of the second millennium b.c.e.).89 This point is essential
for understanding the evolution of capitalism in Europe.

86
Braudel, Civilisation Matérielle, vol. 3, p. 88.
87
Cf. M. Morineau, “Eastern and Western Merchants from the Sixteenth to the Eigh-
teenth Centuries,” in Merchants, Companies and Trade: Europe and Asia in the Early Modern
Era, ed. S. Chaudhury and M. Morineau (Cambridge: Cambridge University Press, Maison
des Sciences de l’Homme, 1999), pp. 116–144.
88
S. Digby, “The Maritime Trade of India,” in The Cambridge Economic History of
India, vol. 1, c. 1200– c. 1750, ed. T. Raychaudhuri and I. Habib (Cambridge: Cambridge
University Press, 1982), p. 155.
89
C. Edens, “Comments” on the article of A. G. Frank, Current Anthropology 34, no.
4 (1993): 408. The end of the Spring-and-Autumn period and the Warring States period
(fifth through fourth centuries b.c.e.) might represent another example of growth in a time
of intense competition. I do not, however, agree with Edens’s rejection of Frank’s causal tie
between volume of trade and levels of urbanization. Regarding the first point, Friedman
goes further than Edens when he writes “centralized empires were often a symptom of slow-
down or even decline. . . . Expansion is most often and systematically linked to political
decentralization” (J. Friedman, “Comments” to the article of A. G. Frank, Current Anthro-
pology, 34, no. 4 [1993]: 409). He disputes the idea that the decline of an empire serves as
evidence for a B phase in the cycle (slowdown). Available data, however, would seem to
support Frank.
2JWH_391-465 12/27/05 2:22 PM Page 457

Beaujard: The Indian Ocean in Eurasian and African World-Systems 457

It would be erroneous, however, to assume there was an automatic

opposition between private modes of accumulation and state modes of
accumulation. There is evidence for competition between the state
and private enterprise, but also instances of cooperation, for example
in the Sung Chinese dynasty or the Abbasid empire. In large agricul-
tural empires, princes relied on merchant-bankers to convert wealth in
foodstuffs into forms of revenue that could be used by the state.90 Mer-
chants, who furnished merchandise as well as services to elites, did not
simply coexist with bureaucratic structures, but rather the two worlds
penetrated one another. States often called on merchants to serve in
their administrations. During the Tang dynasty, traders were given the
task of collecting the tax in salt. Likewise, in the Abbasid caliphate,
some tax collectors were merchants. In Egypt and Iraq of the tenth and
eleventh centuries, representatives of Jewish merchant families were
established as both bankers and tax collectors. Also in the tenth cen-
tury, the king of East Java used merchants—Javanese, but also Sinhal-
ese, southern Indian, or Burmese—to collect taxes. In China during
the time of the Yuan, Muslim merchants held a virtual monopoly on
the gathering of taxes.
The state created favorable conditions for economic development
through investments in public works such as the digging of canals and
the improvement of roads in China under the Sui, Tang, and Sung
dynasties, and sometimes through a policy of economic stimulation.91
So too, in providing security the state encouraged commerce, as trade
could only hope to flourish in a peaceful and relatively predictable set-
ting. The state might also play a role in the redistribution of wealth,
and so indirectly contribute to production and social equilibrium. Con-
versely, political instability or violent competition between states neg-
atively affected commerce, as did an excessive expansion of the state
bureaucracy and the corruption that often went with it. Or, the state
itself could function as producer or merchant, as can be seen for the
Chinese Sung or the Abbasid caliphate.
Another site for capital accumulation was religious institutions,
which enjoyed relative autonomy; it could also bring them into con-

90
K. N. Chaudhuri, Asia before Europe: Economy and Civilization in the Indian Ocean
from the Rise of Islam to 1750 (Cambridge: Cambridge University Press, 1990), pp. 256, 387.
Gills, “ Capital and Power,” p. 139.
91
Ancient Egypt offers an example of growth being directly instigated by actions of the
state, through its technological innovations (the invention of writing, irrigation, etc.) and
especially through its efforts to stimulate demand (D. Warburton, “Before the IMF: The
Economic Implications of Unintentional Structural Adjustment in Ancient Egypt,” Jour-
nal of the Economic and Social History of the Orient 43, no. 2 [2000]).
2JWH_391-465 12/27/05 2:22 PM Page 458

458 journal of world history, december 2005

flict with political authorities.92 Religious networks might intersect

with merchant networks and thereby provide the state with a means
of political integration, as for example the Buddhist and Brahman
centers in South and Southeast Asia.
The volatility of merchant-elites and their desire for independence
—some guilds achieved true autonomy—could pose challenges for
local authorities, who had to choose between cajoling and controlling
them.93 The operating rationale of states and private enterprise can be
very different. Several authors have underscored this ambiguous nature
of the relationships between political powers and merchants and their
networks. D. Lombard has noted that “we must conclude that mer-
chants in Asia were kept out, or kept themselves out, of politics,” 94 but
there are, in fact, many examples which show the opposite. In Fatimid
Egypt in the eleventh century, the Tustari brothers, who were bankers
and traders in luxury goods in the Indian Ocean, rose to the position
of vizier. In China, under the Sung, merchants played important roles
in the commercial centers and sometimes obtained official positions.
Likewise, in India, in Deccan, and on the Malabar coast (among the
Islamic Mapilla communities), trade and politics intermingled, leading
M. Gaborieau to justly remark on “the lack of a dividing line between
the merchant and the soldier-administrator.” 95 In Yemen under the
Rasulids, merchants—often foreigners—could hold official positions.
But rather than enter directly into political structures, merchants usu-
ally preferred to aim at influencing leaders.96 The evidence appears to
be on the side of Aubin and Lombard and against Y. Lacoste when he

92
Burma provides a good example of the contradictions that can arise between the
development of religious entities and state interests. Periodically, in phases of “purifica-
tion,” the state seized the wealth that had been amassed by religious orders. From 843 to
845, the Chinese emperor likewise confiscated the belongings of Buddhist monasteries.
93
Ibn Battuta notes for example the warm welcome given to traders by the lords of
Kulam and Calcutta, but also the practices of extortion and piracy inflicted by local lords
of the west coast of India (Voyages, vol. III, Inde, Extrême-Orient, Espagne et Soudan, pp.
206–207, 213–214).
94
Lombard, “Y a-t-il une Continuité,” p. 117. Also Aubin, “Y a-t-il Interruption du
Commerce,” pp. 88–89. Aubin (ibid.), however, makes reference to “condottieri mer-
chants,” “who were both viziers and military leaders.”
95
M. Gaborieau, “L’Islamisation de l’Inde et de l’Asie Orientale,” in Etats, Sociétés et
Cultures du Monde Musulman Médiéval, ed. J. C. Garcin, M. Balivet, T. Bianquis, H. Bresc,
J. Calmard, M. Gaborieau, P. Guichard, and J.-L. Triaud (Paris: PUF, 1995), p. 456. Indeed,
the merchant Ibn al-Kawlami became governor of Cambaye through the auspices of the
sultanate of Delhi (Ibn Battuta, Voyages, vol. 3, pp. 102–103).
96
S. Digby (“Maritime Trade of India,” p. 52) gives the example of the Tibi family,
which played the role of kingmakers in the Pandya kingdom of South India at the end of
the thirteenth century.
2JWH_391-465 12/27/05 2:22 PM Page 459

Beaujard: The Indian Ocean in Eurasian and African World-Systems 459

writes that “in the Indies, in China, in the Arab world . . . merchants
[were] permanently integrated into the aristocratic minority and so had
no reason to wish to change the society.” 97
Capitalist entrepreneurs oscillated between two strategies: remain-
ing outside politics to try to reduce the role of the state (but capitalist
networks could hardly do without the state because they require a sta-
ble world to develop their operations and/or a military force to defend
their access to vital resources), or investing in the state. On the other
hand, the elites of the state had to choose between taking control of
the economy or encouraging success in the private sector and then
taxing its activities.

The Advent of the Modern Capitalist World-System

The history of the Indian Ocean to the sixteenth century shows its dif-
ferent regions becoming progressively integrated into the Eurasian and
African world-system, as can be seen in the economic cycles synchro-
nized with political, social, and ideological evolutions; the develop-
ment of urbanization; the general growth of commerce and production;
and the simultaneous development of hierarchical relations between
cores and peripheries within the international division of labor. This
early history also sheds light on the period that would follow—the
emergence of the capitalist world-system—and perhaps also provides
some hints as to the possible futures of the system.
Whether or not the beginnings of the sixteenth century represent
a true break in world history is still debated. The arrival of the Portu-
guese in the Indian Ocean occurred during an ascendant phase of a
cycle, a phase that would continue throughout the sixteenth century.
Their appearance most likely does not represent a rupture, but rather
a temporary disturbance. Far too few in number, the Portuguese did not
have the means to implement the maritime policy that they tried to
assert. Moreover, the real break had already occurred earlier in China
in 1433, when the Ming prohibited trade with its southern seas. All
the same, coming just after the discovery of America, the arrival of the
Portuguese in the Indian Ocean signifies the creation of a new order

97
This situation contrasts with that of Europe, which accounts for the fact, according
to Y. Lacoste (La Géographie du Sous-développement, 5th ed. [Paris: PUF, 1982], p. 270),
that it was in Europe that capitalism emerged as the dominant system and paved the way
for the Industrial Revolution.
2JWH_391-465 12/27/05 2:22 PM Page 460

460 journal of world history, december 2005

in the world-system, one in which the center of gravity moved toward

the West and that indicates a structural transformation.
The history of the Indian Ocean completely undermines the idea
that capitalism is a European invention. Nevertheless— during the
seventeenth century?—capitalism would become characteristic of the
world-system that began to grow up around Europe toward the end of
the fifteenth century. Beyond the Italy of the thirteenth century, cap-
italism has its origins in the Afro-Eurasian world-system in which
Europe was simply a periphery. Braudel, it is well known, did not share
Wallerstein’s fascination with the sixteenth century.98 Frank and Gills
meanwhile are totally opposed to the idea that a qualitative change
took place in the world-system of the sixteenth century.
The interstitial growth of capitalism, which Braudel has demon-
strated for the European area, is in fact found in other regions of the
world-system at various points in time. However, it was only in Europe
that capitalism could fully impose its rationale on “country-states” and
thereby become the dominant mode of production in the world-sys-
tem. Regarding this phenomenon, G. Arrighi has adduced three sup-
plementary causes.99 First, the competition of relatively weak central-
ized states for mobile capital resulted in the latter setting the terms.
The system of relations existing between city-states and the very phi-
losophy of these cities became a model for emerging nation-states.
Second, the military competition between European states encour-
aged advances in technology and organization that led to the expan-
sionist orientation of the modern world-system. The bypassing by
Europe of the three “central corridors” of the world-system (central
Asia, Persian Gulf, Red Sea) led to the establishment of trans-Atlantic
connections. Third, the plundering of the Americas and the African
slave trade—an imperialist policy that differs little from that of the
Roman empire—constituted the foundations of Europe’s capitalist
takeoff. European states built what can be considered a new world-sys-
tem 100 —including the Americas and parts of Africa—one that would

98
Braudel, Civilisation Matérielle, vol. 3, p. 44. Marx also expressed that the “biogra-
phy of capital begins in the sixteenth century.”
99
G. Arrighi, “Capitalism and the Modern World-System: Rethinking the Nondebates
of the 1970s,” Review 21, no. 1 (1998): 127–128, and The Long Twentieth Century: Money,
Power and the Origins of Our Times (London: Verso, 1994), chaps. 1–2.
100
Frank and Gills, however, do not support this interpretation. According to these
authors (“Rejoinder and Conclusions,” p. 304), the Americas were a periphery to a core
situated in Europe, but the latter was not the core of a discrete world-system. The gold and
silver from the Americas only “(permitted) the Europeans to participate more actively” in
exchanges in the system (Frank and Gills, “Five Thousand Year World System,” p. 7). Yet,
2JWH_391-465 12/27/05 2:22 PM Page 461

Beaujard: The Indian Ocean in Eurasian and African World-Systems 461

eventually subsume the old system, after the Industrial Revolution of

the eighteenth century. Henceforth, the world-system became coter-
minous with the whole world, with Europe and then the United States
acting as the dominant centers.
Moreover, the Industrial Revolution was preceded and accompa-
nied by an agricultural revolution marked by technological innovations
(for example, abandonment of the fallowing of land and the introduc-
tion of new crops), as well as by juridical changes, the enclosure move-
ment, and changes to the Corn Laws in England. The majority of the
yeomen disappeared, reduced to becoming salaried employees in agri-
culture or forced to flee to the towns.101
Scholars have long wondered why China or another Asian coun-
try did not follow a similar path. The advantages Europeans enjoyed
from their spoils of silver 102 and gold from the Americas cannot alone
account for their success. Paradoxically, it is probably the large size
and strength of Asian states that prevented their evolving as Europe
had done, thereby discouraging both initiative and capitalist accumu-
lation. If the progression of capitalism in Asia did not end in an indus-
trial revolution, it was not from a lack of technological evolution, but
because entrepreneurs did not create a great enough force of change.
Asia never experienced an industrial revolution, but it also never
underwent the ideological revolution of Britain, the United States, and
France in the eighteenth century.103 The Declaration of the Rights of
Man and Citizens adopted by the Constitutional Assembly of 1789
“marks in a way the apotheosis of the individual.” 104 Along with the

elsewhere, Frank (ReORIENT, pp. 277–279, 316) does describe a world-system centered
in Europe. This debate ties in with the question of whether European states of the six-
teenth century were qualitatively different from other states and if the logic of the system
changed with the rise of modern Europe. It could be argued that the Netherlands in the
seventeenth century, before England, represent the first example of a capitalist state. But,
asks Braudel (Civilisation Matérielle, vol. 3, p. 161), “Can the United Provinces be called a
‘state’?” The East India Companies, in the seventeenth and eighteenth centuries, offer a
good example of association between state and capitalism.
101
M. Mazoyer and L. Roudart, Histoire des Agricultures du Monde du Néolithique à la
Crise Contemporaine (Paris: Seuil, 1998), p. 313.
102
The discovery of silver mines at Potosí (Bolivia) in 1545 and at Zacatecas (Mex-
ico) in 1548 would, in the decades to come, greatly augment the silver supply in the trade
circuits of the ancient world. Access to this silver, along with gold, enabled Europe to com-
pensate for its commercial trade deficit with Asian states. The influx of precious metals also
became a factor for growth in production and commerce in the Eurasian and African zones.
103
The notion that mass education can provide a structural base for technological
innovation was developed only in the twentieth century.
104
L. Dumont, Essays on Individualism, p. 92. In addition, see the “Bill of Rights
adopted by certain (American) States and particularly that of Virginia of 1776” (ibid.,
p. 93).
2JWH_391-465 12/27/05 2:22 PM Page 462

462 journal of world history, december 2005

agricultural expansion, the Industrial Revolution was also preceded

and accompanied in Europe by a scientific revolution that saw the birth
of an experimental method in science. Yet this method played practi-
cally no role in the technological innovations of the Industrial Revo-
lution of the eighteenth century.105 Europe alone cannot be credited
for its developments. “The rise of modernism is the product of an eco-
nomic and technological synergy that was generated over several mil-
lenniums in different parts of Eurasia.” 106 There is no necessary and
direct tie between the growth of modern science and capitalism in
Europe, but the two events are both tied to the granting of liberty to
individuals by city-states (Italy, Netherlands) and then European
states: the liberty to conduct business, to make profits, and also to think
and experiment, a liberty free from interference by political and reli-
gious powers.107 In this connection, some scholars, such as Y. Lacoste,
have argued that the feudal system which developed in Europe—and
Japan—by removing merchants and bankers from power, resulted in
the “individualization of the bourgeoisie” as a revolutionary class, and
also conferred on it “its essential role in economic and social evolu-
tion.” 108 The bourgeoisie’s takeover of power opened the way for the
Industrial Revolution, coming at a time when all the conditions for
major technological innovations were ripe in Europe. In the eigh-
teenth century, Asia, on the other hand, found itself in a phase of

105
J. E. McClellan III and H. Dorn, Science and Technology in World History: An Intro-
duction (Baltimore: Johns Hopkins University Press, 1999), p. 292.
106
D. Christian, “Silk Roads or Steppe Roads? The Silk Roads in World History,”
Journal of World History 11 (2000): 25.
107
It is noticeable that the emergence of rational thinking in Greece took place in the
context of city-states that developed democratic institutions. The Mohist Chinese wave of
logician thinking appeared at the same period (fifth through fourth centuries b.c.e.), but it
will not have the same posterity, because of very different political and social contexts
(within a strong central state marked by Confucianism). As in the Greek cities of antiq-
uity, some freedom of thinking along with liberty to conduct business existed in Sung
China, with different kinds of limitations (the importance of external threats nevertheless
constituted an inhibiting factor for growth in both cases). Freedom is restricted in Ming
China by state power, based on a neo-Confucianist philosophy; in the Muslim world, by
religious power itself.
108
Lacoste, La Géographie du Sous-développement, pp. 269–270. According to Lacoste
(and to Marx), it is the unique character of the European feudal system that explains the
creation of a revolutionary bourgeoisie, the triumph of its ideology, and the advent of cap-
italism as the dominant mode of production. Open to question, however, is Lacoste’s con-
tention that “in a country without feudal structures, merchants did not constitute a bour-
geoisie,” but were part of the ruling aristocracy (ibid., pp. 260, 267, 273). Likewise refutable
is Lacoste’s assertion—made in order to fit with Marx’s “Asian mode of production”—that
there was no private ownership of land in Asia. For the Tang, Sung, and Ming, cf. J. Ger-
net, Le Monde Chinois (Paris: Armand Colin, 1999), pp. 230, 275–278, 360.
2JWH_391-465 12/27/05 2:22 PM Page 463

Beaujard: The Indian Ocean in Eurasian and African World-Systems 463

regression, demographic increase having generated economic, social,

and environmental pressures; the overabundance of labor and the
scarcity of capital put a brake on investment and technological inno-
vation. The situation in Europe was just the opposite, with a scarce
and costly work force and much capital available for investment.109
The thesis from Lacoste that I evoked may be rather simplistic. It
ignores or underestimates the changes in Europe in the twelfth and
thirteenth centuries (with the discovery of the Greek and Arab phi-
losophers, the legalist and institutional changes, the emergence of cor-
porations, the birth of universities) and then in the Renaissance. As
Hansen underlines it, “the roots of the political culture of the con-
temporary world lie in the cultures of city-states,” in particular those,
Greek and Latin, that developed republican institutions. The funda-

109
Frank, ReORIENT, pp. 301–306. Before broaching the topics of population
increase and the resulting imbalances, Frank puts forward a seemingly paradoxical argu-
ment. The assets of Asian cores were also the source of their weakness: the growing supply
of silver resulted in “increased purchasing power, income, and demand on domestic and
export markets in the world economy. That presumably increasingly skewed the distribution
of income, which could have led to constraints on effective demand and growing political
tensions” (ReORIENT, p. 267). But why would the influx of money be extremely benefi-
cial to Asian economies in the sixteenth and seventeenth centuries and then suddenly
become nefarious in the eighteenth century? The argument makes sense only if it assumes a
reverse in the cycle, induced by growing demographic pressure and a decline in the levels
of profits. Frank emphasizes the unequal distribution of income in Asia, which I mentioned
above. Other authors see a phase of decline in the seventeenth century, but the character
of the recession of the mid-seventeenth century is still under debate. Was it a global event
affecting the entire world-system or a more localized recession? Was it a slowdown at the
end of a long phase or the second phase of a Kondratieff cycle? Frank (ReORIENT, p. 231)
today rejects the idea of a generalized economic crisis in the seventeenth century. How-
ever, there was a decrease in the population of the ancient world between 1600 and 1650
(McEvedy and Jones, Atlas of World Population History; C. Clark, Population Growth and
Land Use [London: McMillan, 1977]) and a global decrease in temperature in what is called
the Little Ice Age (1640–1705). Moreover, “in the period 1640–1650, the white metal from
America no longer arrived in Spain in large quantities” (F. Braudel, “Monnaies et Civili-
sations de l’Or du Soudan à l’Argent d’Amérique: Un Drame Méditerranéen,” Annales-
économie, Sociétés, Civilisations 1 [1946]: 20). According to Frank, Europe has (only) taken
advantage of the weakness of the Asian cores in the eighteenth century, thus managing to
acquire a dominant position, which leads I. Wallerstein to say ironically that “Frank proves
the European miracle” (“Frank Proves the European Miracle,” Review 22, no. 3 [1999]:
365–371). Considering the existence of only one world-system in the eighteenth century,
which is Frank’s view, we are led to ask ourselves why this B phase (phase of decline) that
he describes (a B phase valid “at least for Asia,” says Frank, ReORIENT, p. 259), brings
about a general demise in Asia and a spectacular upturn in Europe. It seems to me neces-
sary to think of the competition between two world-systems, the old Eurasian and African
one (maybe truly in a B phase before the Industrial Revolution) and a new one centered
on Europe (and then North America)—a situation that Frank himself seems to describe
(ReORIENT, p. 283): “While the Europeans were gathering strength from the Americas
and Africa, as well as from Asia itself, Asian economies and polities were also becoming
weakened during part of the eighteenth century—so much so that the paths finally crossed.”
2JWH_391-465 12/27/05 2:22 PM Page 464

464 journal of world history, december 2005

mental ideas and the institutional innovations of these city-states are

in fact rediscovered by Europe from the thirteenth century. Rousseau,
one of the thinkers who influenced the French Revolution, took Rome
in its early republican period as a model, as Machiavelli did earlier. In
1787, says Hansen, some American federalists refer to “the Greek
Amphictyonic council and hellenistic federations.” In almost all the
city-states, even those built on monarchic systems, “decisions are taken
in assemblies, by a vote of the majority, and after a debate among the
participants.” 110
From its inception in the eighteenth century, the capitalist world-
system continued its path of ascent (the first half of the twentieth cen-
tury representing only a relatively short-lived phase of regression).
The pace of economic growth in the system and the integration of its
different parts accelerated even further after World War II. However,
demographic and ecological pressures, which exist now at both local
and global levels, may in the near future mark the upper limits of
growth in this system,111 for this particular phase of it, or even for the
system as a whole, unless new technological (and ideological as well)
advances or adaptations in the system can be found. Already in 1970,
a team directed by D. H. Meadows of the Massachusetts Institute of
Technology prepared an alarming report for the Rome Club that
revealed that the limits for growth had previously been reached, as
seen in the exhaustion of certain resources (notably fossil fuels), of land
and fresh water, and in the pollution created by growth. At the current
(1970) rate of growth, wrote the authors of the report, the world’s eco-
system is bound to collapse in a brutal manner. They recommended the
passage from a state of growth to one of equilibrium, knowing that the
longer we wait to break with growth that we cannot control and that
is destructive for the environment, the less chance we have of ever
attaining equilibrium. Since 1970, little progress has been made in the

110
Hansen “Conclusion,” pp. 612–613. The United Provinces after 1579 provide a
more recent example of a federation (ibid.). “Republicanism and federalism represent
major aspects of the modern state which have their roots in the cultures of city-states.
Before the end of the eighteenth century, they were practically only to be found in the city-
states” (ibid., p. 616). When they eventually managed to emerge (in India, Southeast Asia,
but not in China), Asian city-states were not able to impose their ideology to the regional
states.
111
Chase-Dunn and Hall (“ Comparing World-Systems,” p. 108) predict a collapse
circa the “year 2020.” Some experts think that oil supply could stay at reasonable prices in
the world until 2030, but “these forecasts are not the most probable” (P. Testard-Vaillant,
“Pétrole: Les Raisons de la Tourmente. Pourquoi les Prix Flambent-ils?” Le Journal du
CNRS 178 [2004]: 19).
 2JWH_391-465 12/27/05 2:22 PM Page 465

Beaujard: The Indian Ocean in Eurasian and African World-Systems 465

direction advocated by Meadows and his team, and although the con-
clusions struck many at the time as being overly pessimistic, today their
accuracy is becoming more apparent. “The period 2000–2050 will
thus be chaotic.” 112 The pressing question for us now then is “whether
the fall will take place in totalitarianism and barbarity or whether it
will . . . be channeled through humanism and democracy,” 113 guided by
a new economic model. Here, echoing Wallerstein, I believe we must
decide what kind of society we want to see develop in the new world-
system that will evolve.114

112
I. Wallerstein, “The World We Are Entering, 2000–2050 (32 Propositions),” in
The World We Are Entering, 2000 –2050, ed. I. Wallerstein and A. Clesse (Amsterdam:
Dutch University Press, 2002), p. 22. “The modern world-system (which is a capitalist
world-economy) is in a structural crisis and it will not be able to go on functioning under
the historic way it took. The system enters a chaotic period of transition towards something
different” (ibid., p. 17). “The crisis of the system disfavours the survival of the capitalist
world-economy as a historic system.” “During this period (of chaos), there will be a politi-
cal struggle about the nature of the system which will follow” (“Introduction,” in World We
Are Entering, p. 7).
113
La Décroissance, 2004, p. 2. Cf. N. Georgescu-Roegen, La Décroissance: Entropie,
Écologie, Économie (Paris: Sang de la Terre, 1995). S. Latouche, Survivre au Développement
(Paris: Mille et Une Nuits, 2004).
114
I. Wallerstein, Utopistics, or Historical Choices of the Twenty-first Century (New York:
New Press, 1998), and “Introduction,” in World We Are Entering, p. 8.

View publication stats