’Egypt at the Exhibition’: Reflections on

the Optics of Colonialism

A review of Timothy Mitchell’s Colonizing Egypt (1988)
Charles Hirschkind
The New School for Social Research, New York

Colonizing Egypt (1988) is far more ambitious than either its title or
slender size (179 pp.) suggest: its aim is nothing less than the descrip-
tion of a distinct form of power-one originating in Western capitalist
societies and extending over much of the globe during the period of
European colonial expansion-upon which the ’modern political method’
depends. Nineteenth century Egypt, in this context, appears cast in the
role of a test-case-one illustration of a widespread if not universal
phenomenon-and, as a result, the book exhibits a certain paucity of
historical detail. Indeed, as Mitchell is primarily interested in European
systems of representation and domination, as well as their spread beyond
the national borders of Europe, he elides much of the particularity of
the Egyptian experience in favor of these broader concerns; or rather,
he addresses specifics insomuch as they involve the importation and
imposition of Western forms of power, while neglecting other (poss-
ibly relevant) details of the period. I make this point here simply to
clarify the goals of Mitchell’s inquiry, and not to indicate a flaw in the
analysis.
The originality and importance of Mitchell’s book lies in two related
areas. First, proposes new avenues for the study of both the modern
it
Middle East and colonial history in general, avenues opened up by the
work of such French theorists as Foucault, Derrida and Baudrillard,
but avenues having been only rarely followed by other researchers and
then in relation to rather more specific historical questions.1 Admittedly,
Mitchell does little to extend or reformulate the approaches he borrows:
his chapters on the introduction of Western techniques of distribution and
arrangement into Egypt do little more than restate Foucault’s insights in

Critique of Anthropology
polog © 1991 (SAGE, London, Newbury Park and New
Delhi), Vol. 11(3): 279 -
279-298.
280

Discipline and Punish (1979), with the examples now coming from a non-
European context; his use of Derrida and Baudrillard on Western systems
of representation is similarly conservative and uncritical. Nonetheless, as
the explanatory potential of these approaches has not been explored and
developed to the extent it should be, Mitchell’s work comes as a needed
step, and one worthy of critical attention.
Second, Mitchell attempts to integrate two quite distinct notions of
power, one based on what Foucault has called ’microphysical power’,
the other based on a ’theological effect’ resulting from a specific mode
of representation which divides the world up into surface appearances
and hidden truths, into a realm of the material and a realm of the
conceptual, with this latter ’effecting a certain, metaphysical authority’
(1988: 160). Mitchell identifies the intersection of these two mechanisms-
in his analysis, different ’effects’ of a single ’technique’-as constitutive of
the modern form of ’political power’.
Yet, despite Mitchell’s attempt to demonstrate the connections and
interdependencies of these two notions of power, the linkage, when
examined in terms of its underlying theoretical assumptions, proves less
than stable. As this paper explores, the two notions rest upon quite
different-and incompatible-ideas about the constitution of the subject
and the nature of social reality in general. Moreover, Mitchell’s attempt
reveals the difficulties of applying a ’world-as-text’ perspective to historical
analysis, specifically in regard to the problems of interpretation, and the
relation between meanings and behaviors (a problem of interpretive
approaches in general). Despite these shortcomings, Mitchell’s treatment
of these problems is original and provocative, and his argument merits our
careful examination.

A world divided
The book opens with discussion of the world exhibitions of nineteenth-
a

century Europe, seenin terms of ’a practice that exemplifies the nature

of the modern European state’ (Mitchell, 1988: 2). Mitchell points out
the role of these exhibitions in promoting the global economic and
political transformations ushered in by modern capitalism: the spread of
industrialization and new patterns of consumption, the opening up of all
continents to the unrestricted flow of commodities, the creation of
transportation and communication systems to link the countries of the
world into a single economic system. The exhibition brought together
representatives of industrial and financial interests from a multitude of
nations, fostering increased international cooperation for the expansion
of commerce.
 281

Beyond providing a forum for business, the exhibition accomplished

a pedagogic function as well, serving as a sort of ’theatre of represen-
tation’, where the ideas and values of European capitalist society were
given metaphorical expression through highly developed techniques of
display and simulation. Elaborate presentations of merchandise, highly
detailed panoramas portraying scenes from such ’exotic’ places as Cairo or
Bangkok, immense models of the latest triumphs of technology, displays of
recent scientific discoveries-the exhibition rendered up these creations
for the edification of the wider public in its crucial role in the emerging
world capitalist system:
...
everything [was] collected and arranged to stand for something, to
represent progress and history, human industry and empire; everything
set up, and the whole set-up always evoking somehow some larger truth.
(Mitchell, 1988: 6)
The exhibit of the type Mitchell analyzes presented a remarkably detailed
replica or model of an external world-a street scene, an architectural
style, a manufacturing process, etc.-and arranged it to be viewed by
an observer occupying a specific position in relation to it. The realism
of the model was distinguished from the external reality which it was
made to represent by virtue of having been conceived and constructed
as a spectacle for ’an observing gaze surrounded and set apart by the
exhibition’s careful order’:
If the dazzling displays of the exhibition could evoke some larger historical
and political reality, it was because they were arranged to demand this
isolated gaze. The more the exhibit drew in and encircled the visitor,
the more the gaze was set apart from it, as the mind is set apart from
the material world it observes. (1988: 9)

In other words, the exhibit reproduced many of the qualities of the photo-
graph (realism, singularity of perspective, ’truth’), and hence enhanced
the observer’s sense of detachment from the object under observation.
Mitchell, echoing Benjamin in his work on the Paris arcades, goes on
to suggest that the practice of such techniques was in no way limited to
exhibitions alone, but rather informed European architecture and spatial
patterning in general:
It was as though, despite the determined efforts within the exhibition
to construct perfect representations of the real world outside, the real
world beyond the gates turned out to be rather like an extension of the
exhibition. This extended exhibition would continue to present itself as a
series of mere representations, representing a reality outside. (Mitchell,
1988: 10)
282

The department stores and arcades with their goods displayed behind
glass, the use of deliberate facades on the buildings lining the streets,
the theatres, museums, zoos-such a pervasive concern with organizing
and manipulating appearances characterizes what Mitchell defines as ’the
age of the world-as-exhibition ...
refer[ring] not to an exhibition of
the world but to the world conceived and grasped as though it were
an exhibition’ (1988: 13). In this world reality is perceived in terms of
a distinction between material representations and their hidden, more
’real’ referents, between a realm of things and one of meaning or ’truth’
(1988: 149). The basis of this division of the world into two spheres lay
in the discernibility of the representation as representation; the effect
of a conceptual, original ’reality’ results from this discernibility, from
the perception of the material as mere representation. By constructing
the built environment according to the principles of the exhibition, the
planners of nineteenth-century Europe (Mitchell uses Haussmann as a
prime example) reshaped what Europeans would experience as ’real’ or
’true’, transferring it from the material world into a conceptual one, and
hence perceivable only in representational form.
Mitchell identifies this unprecedented capacity to create and fix ‘truth’-
made possible by nineteenth-century advances in science and technol-
ogy-as constitutive of colonial power, the power of the modern political
method. The colonization of Egypt, moreover, is seen as a process which
entailed the complete restructuring of Egyptian society-the military,
schools, cities, towns, government institutions, etc.-in such a way as to
render it legible, productive of a certain (primarily political) truth. Such a
restructuring required that Egypt be reorganized representationally so as
to produce the same division of the world into two domains as had been
accomplished in Europe.
In explaining how such a transformation was realized, Mitchell draws
on Foucault’s analysis of disciplinary power. He suggests that the same

disciplinary mechanisms analyzed by Foucault in regard to France and

Northern Europe-the techniques of confinement, regulation, supervi-
sion and education through which power ’colonized’ individuals, pro-
ducing pliant, industrious subjects-were instrumental in the extension
of European power abroad (Mitchell, 1988: 34-5). By restructuring
Egypt according to the model of the European prison, barracks and
hospital, power would come to operate in a manner that was con-
tinuous and systematic, more akin to training than violent repression.
Fundamental to this form of power was a new method for the organization
of objects in space, what Foucault referred to as the ’architecture of
distribution’:
 283

... a method of dividing up and containing, as in the construction of

barracks or the rebuilding of villages, which operates by conjuring up
a neutral surface or volume called ’space’. In reconstructing the village,
the spacing that forms its rooms, courtyards and buildings is specified
in exact magnitudes, down to the nearest centimeter.... Within these
containers, items can then be isolated, enumerated, and kept.... The
dividing up of such items is also the breaking down of life into a series
of discrete functions-sleeping, eating, cooking, and so on-each with a
specific location. (Mitchell ... 1988: 44)
Mitchell traces the history of the introduction of such methods of organi-
zation and distribution into Egypt. For example, his work provides a
fascinating and careful examination of the plans for model villages, the
meticulous schedules devised for schools and military academies, the laws
passed to regulate agricultural labor and processes of production (1988:
62-82).
It should be noted that this disciplinary power which, following Foucault,
Mitchell finds to be operative in colonial practice, in no way relies on belief,
ideology or the recognition and acceptance of any ’metaphysical authority’
or ’political truth’. Rather, disciplinary power ’is more dependent on bodies
and what they do’ (Foucault,1980:104); its mechanisms of power-systems
of supervision, scientific procedures of investigation and control, etc.-
produce and circulate a knowledge which informs the very practices
by means of which subjects come to be constituted, controlled and
supervised (Foucault, 1980: 102). The production and proliferation of
such knowledge does not depend on a ’theological effect’ engendered
by a system of representation, but on quite mundane and concrete
research methods and practices, particularly in the fields of social science
and medicine. Consequently, ’meaning’-the organizing principle in the
constitution and operation of the ’techniques of the exhibition’-plays
an insignificant role in the Foucauldian notion of power relations, and
in the determination of the course of history which such a notion implies
(Foucault, 1980: 114).
Mitchell seeks to refute the apparent incompatibility of the two types of
power-disciplinary and what I call ‘representational’-and demonstrates
instead their close interdependence as simply ’two aspects of the same
novel strategies of power’ (Mitchell, 1988: 176). These two aspects
find their intersection and integration in the Heideggerian notion of
’enframing’, a concept employed by Mitchell to analyze the cognitive
effects of architectural form. Specifically, Mitchell suggests that ’dis-
ciplinary architecture’, with its emphasis on uniformity, regularity of
distribution, standardization of measurements and strict specification of
rooms in terms of function, creates a type of order in which the placement
284

of both material structures (walls, floors, doors, etc.) and their contents
appears to be governed by a framework, a ’system of magnitudes’ defining
space itself (1988: 45). As Mitchell describes,
The act of distributing and fixing in place, repeated again and again in
a sequence of exact and equal intervals, creates the impression that the
intervals themselves are what exist, rather than the practices of distri-
bution.... This structural effect of something pre-existent, non-particular
and non-material [i.e. a plan] is what is experienced as order, or, the same
thing (since it seems to exist apart from the material realization), as the
’conceptual’. (Mitchell, 1988: 79)
With the advent of the techniques of distribution and arrangement, the
effect of the exhibition extended its reach far beyond the exhibition walls.
The world of visible constructions now appeared as the material realization
of an invisible plan or framework, or, once again, as a representation
of a hidden reality, a hidden truth and order. In a world organized
to produce such a cognitive maneuver, as Mitchell argues throughout
the book, political power seems to reside outside the realm of material
forces, and hence to lay claim to a certain suprahuman authority, a certain
naturalness; the efficacy of such a power stems precisely from this effect
(1988: 159-60).
Much of Colonizing Egypt is devoted to examining how this perception
of a dichotomous world was effected in such different domains as writing,
the body and government, in addition to architecture as discussed above.
For example, drawing on Derrida’s investigation into the operation of the
sign and the production of meaning, Mitchell suggests that pre-colonial
Arabic was ’closer than European languages to the play of difference that
produces meaning’ (1988: 149). He argues that, unlike Western languages
in which written words are understood as ’transparent’ representations
of singular, unequivocal meanings, Arabic writing is read or interpreted
with much more sensitivity to the semantic reverberations produced by
syntagmatic and phonic associations among words. In the West, poetry is
read in this manner, but ’ordinary language’ is thought to overcome such
ambiguity, with words operating in a manner akin to signals. Mitchell
also cites the absence of the vowel in most Arabic writing as another
feature of the language mediating against the fixing of meanings; the
dependency of a word’s meaning on contextual factors is seen to foster
an understanding of language as an interpretive process, where truth
is unstable, shifting, contingent, negotiated and so on (Mitchell, 1988:
142-50).
The restructuring of Egypt, according to Mitchell, required a linguistic
transformation whereby Arabic would be rendered capable of re-presenting
 285

authorial intention unambiguously. Mitchell, rather sketchily, attributes

the possibility of this transformation, primarily, to the introduction of
printing, and, to a lesser degree, to new types of writing such as a
state-sponsored pedagogic literature and a new style of journalism (1988:
153). This new literature employed a more ’telegraphic’ style of writing
than previous forms, with language operating more and more as a precise
system of signals (1988: 53). The overall effect was to make an author’s
meaning in a text appear as something unproblematic, the result of an
essentially mechanical process. Moreover, Mitchell relates this question
of authorial intention to that of authority in general:
The unproblematic presence of an author in writing would now correspond,
in all the other realms of ordering that characterize the world-as-exhibition,
to the production of an essentially unproblematic and mechanical presence
of authority in political life. This political authority, produced in the modem
state the way a modem text produces the unambiguous effect of an author,
would appear continuous and mechanically present. (Mitchell, 1988: 154)

Incidentally, as a purely empirical point, the lack of vocalization in

Arabic script, identified by Mitchell as a major factor accounting for
the language’s ingenerate indeterminacy, was at no point ’corrected for’
during the process of linguistic restructuring: vowels continue today to
be absent from nearly all forms of writing; even language pertaining to
the most technical or mechanical procedures functions effectively without
any supplemental vocalization.

Language
While this review of Mitchell’s argument is only cursory and skeletal,
it allows us nonetheless to identify some of the problems in such an
analysis of the form and operation of political power in the colonial (and
post-colonial) age. To begin, it is worthwhile to note that although Mitchell
brings a far more sophisticated theoretical framework to his analysis of the
functioning of language than did the earlier philologers and orientalists
and despite the fact that his general sympathies tend toward the opposite
pole than did theirs, as evident in his quite positive valuation of Arabic,
Mitchell nonetheless reproduces the assertion of many of his nineteenth-
century predecessors concerning the fundamental otherness of the Arabic
language. Specifically, relative to Western languages, Arabic is found once
again to be essentially indeterminate, imprecise, ambiguous, etc.2 While
Mitchell has reversed the polarity-it is the Europeans who are now
entrapped by illusion, while the Arabs live closer to the truth-the
286

orientalist dictum still holds true: ’East is East, and West is West’ and
so on. In addition, this radical separation in language produces a corollary
effect on historical vision. Namely, Egyptian history is telescoped into two
essential phases, pre-colonial and colonial, without any acknowledgement
of the long history of shifts, disjunctures and revolutions subsumed within
the former of these. Viewed in light of an argument about the essential
nature of language, pre-colonial Egypt is made once again to stand outside
history.
Mitchell is drawn to this conclusion, I suggest, by addressing language
as an ideal category, detachable from the specific social contexts in
which utterances are generated. Ironically, Mitchell levels a somewhat
similar critique against the Saussurean distinction between parole and
langue, wherein the latter is conceived of as a structure having an
ideal existence independent of any actual speech act (Mitchell 1988:
140-1). Furthermore, Saussure’s understanding of the linguistic sign,
as combining both a material and a conceptual element, is shown to

repeat yet again the ’theological effect of a distinct realm of meaning’

(1988: 144). Yet, after having noted the problems stemming from this
sort of essentialization of language, Mitchell, following certain paths
opened up by Derrida’s work, goes on himself to repeat the error.
With the exception of certain instances to be discussed below, words
whose meanings are ’ambiguous’, ’proliferating’, ’problematic’ and open
to multiple interpretations, are precisely words removed from context,
isolated from the concrete social configurations which serve to implement
and delimit their potential meanings.
Volosinov’s (1973) work bears directly upon this point. Anticipating
the investigations of Derrida and others, Volosinov defined ’meaning’
as ’those aspects of the utterance that are reproducible and self-identical
in all instances of repetition’ (1973: 100). These aspects, however, have
no ’autonomous existence in an artificially isolated form’; that is, one
cannot say that particular meanings belong to particular words; rather
they come into being only in the interaction between speaker and listener,
an interaction determined by both verbal and extraverbal elements. In

fact, meaning, inherently undetermined, requires the imposition of a

value judgement, what Volosinov terms ’evaluative accent’ (1973: 105).
Judgements of this type are easily recognized when conveyed through
intonation or facial expression. More important here, however, is the
extent to which they enter into the production of referential meaning
itself. As Volosinov states:
Referential meaning is molded by evaluation; it is evaluation, after all, which
determines that a particular referential meaning may enter the purview of
 287

speakers-both the immediate purview and the broader social purview of

the particular social group. (1973: 105)
Most importantly, the locus of this determination does not lie within
language, rather:
The generative process of signification in language is always associated with
the generation of the evaluative purview of a particular social group, and the
generation of an evaluative purview-in the sense of the totality of all those
things that have meaning and importance for a particular group-is entirely
determined by expansion of the economic base. (Volosinov, 1973: 106)
While Volosinov’s terminology reflects the Marxian tradition within which
he is writing, the point to be emphasized is that, in the dialectical process
of signification, social practice preponderates over linguistic potential.
Mitchell’s upending of this hierarchy in favor of the linguistic appears
clearly in his analysis of the evolution of the Arabic word ’hizb’. Prior
to the late nineteenth century, hizb was used to refer to any one of a
number of different social groupings, be they either political, ethnic or
national in nature. Sometime after this point, however, the word came
to denote ’political party’ alone. Mitchell cites this shift to illustrate how
the colonization of the Arabic language involved the transformation of
certain words-and of the Arabic language in general-from ’a state of
confusion to one of clarity-a movement, in other words, from hesitation
to certainty, from ambiguity to unambiguity, from instability to stability’
(Mitchell, 1988: 138). The change, according to Mitchell, resulted from a
new ’approach to language’ in which words were to be conceived of as

‘clear, stable and univocal’ (1988: 138).

In relying too heavily on the traditional methods of historical linguistics,
Mitchell winds up overlooking the social practices in which the term hizb
was embedded. For an examination of these factors suggests that the nar-

rowing of referential content of the term hizb reflected (and presupposed

commitment to), not a new approach to language, but the increasing
salience of the modern political party in Egyptian society. Should there
be any doubts about the weakness of such a language-privileging approach
to history, I cite an analogous example, drawn this time from a European
context: before the advent of nuclear physics in the late nineteenth and
early twentieth centuries, the term ’atom’ could refer to any number of
the indivisible particles postulated by philosophers, whereas now the term
refers almost exclusively to that unit of analysis central to the experimental
practices of chemists and physicists, and which, incidentally, is considered
quite divisible. Clearly, these terminological histories illustrate simple
cases of a language drawing on its resources of idiom in adapting to an

evolving social reality.

288

As mentioned above, Mitchell’s discussion of political authority, to

its detriment, is likewise grounded in the notion of the indeterminacy
of language. For example, he states that, ’[p]olitical argument always
worked through this power of language, seeking out the contradic-
tions in words or their ability to evoke alternative meanings’ (1988:
137). While this sort of observation is true enough, it fails to convey
the nature of the relationship between interpretation and authority.
Mitchell seems to understand interpretation (the interpretive status of
assertions) to imply the possibility of a plurality of perspectives, and
consequently, of open political contestation, argument and negotiation.
In other words, political claims in pre-colonial Egypt were recognized
by all contending parties as mere interpretations; sometime thereafter,
however, with the arrival of a Western (structuralist) conception of
language, such claims acquired the status of ’truth’, their meanings
now fixed and beyond debate. This analysis obscures the crucial issue
at stake here; namely, that interpretation is precisely the act by which
meanings are fixed, by which truth is produced. Far from opening up
the possibility of argument, interpretation signifies the exclusion of
argument.
This misconception underlies Mitchell’s suggestion that Ibn Khaldun’s
Muqaddima was ’a unique attempt to overcome the essential weakness
of writing’ by identifying the limits of historical possibility and hence of
interpretive freedom (Mitchell, 1988: 152). For, as Mitchell himself notes,
’Ibn Khaldun wrote in a period of political crisis in the Arab world’;
now ’political crisis’, by definition I would suggest, refers precisely to
a moment in which no social faction is able to impose an interpretation

(a political claim) with sufficient authority so as to gain its widespread

recognition and acceptance. In other words, it was not the weakness of
writing which brought turmoil to North 3Africa in the fourteenth century,
but the weakness of political authority.3
Moreover, a broad reading of the history of Koranic interpretation
(ta’wil), particularly with respect to the technique of ijtihad (personal
judgement) further underscores the point being made here. Notably,
the consolidation of orthodoxy in the ninth century with the concomi-
tant ’closing of the doors’ of ijtihad (the exclusion of non-orthodox
interpretations) was rendered possible by the rise of a social faction
capable of commanding widespread authority, namely, the ulama (the
class of religious specialists). Alternatively, during times of rapid social
change or political contestation and reorganization, as in the nineteenth
century, ijtihad was often invoked anew. I do not mean to imply that one
can map the political history of Muslim societies onto the succession of
 289

arguments concerning the application or abrogation of ijtihad with the

hope of showing any sort of exact correspondence. Clearly, in some

periods ijtihad did not figure in political strategy at all. My point is
simply that opening and extending the range of acceptable interpre-
tations (political claims) has generally not been a practice advocated
by groups who have already consolidated and secured their power and
authority.
This brings us back, however, to the questions of what we mean when
we speak of interpretation, and how we understand the operation of

language. Postmodernist theories emphasize the ambiguity and inde-

terminacy of texts. As I have noted, this possibility emerges when
traditional texts are removed from the context of tradition-the more
or less coherent set of beliefs, practices and institutions which act to
stabilize meaning, to impose an interpretation. Thus, theorists working
in the postmodernist vein tend to speak of Arabic, for example, rather
than Arabic-as-used-in-such-and-such-a-period. The current prevalence
of such essentialist approaches reflects the peculiar characteristics of
what MacIntyre calls the ’international languages of modernity’, those
idioms used by speakers across a wide range of cultural traditions, such
as late twentieth-century English, German and Japanese (MacIntyre,

1988). The seemingly universal availability of such languages is due to

the fact that,
...
they are tied very loosely to any particular set of contestable beliefs,
but are rich in modes of characterization and explanation which enable
texts embodying alien schemes of systematic belief to be reported on-not
in the light of some other rival scheme of belief, by reference to which they
would necessarily be exhibited as true or false, reasonable or unreasonable,
but rather in detachment from all substantive criteria and standards of truth
and rationality. (1988: 384)
Insofar as the use of such a language does not imply a commitment to
any specific body of beliefs, texts employing it can easily appear unstable,
subject to a near infinite number of interpretations. To draw such a
conclusion, however, presupposes that one understands ’interpretation’
to mean the deciphering of external signs, the acceding to a state of
’correct consciousness’ in relation to an object. Mitchell’s analysis of
how Egyptians came to read colonial authority, to accept and believe
in colonial truths, suggests just such an understanding. The adequacy or
inadequacy of language, however, or any other mode of representation,
does not depend on the accuracy of perception, but rather, is always
relative to a particular purpose (MacIntyre, 1988: 357). A successful
reading in this regard can better be described as the ability to reproduce,
290

both in actions and words, the projects of power, and thus is entirely
independent from the question of ’belief’ understood as the subject’s inner
conviction (see Asad, 1988). Consequently, we can conditionally agree
with Mitchell when he ascribes to Western languages an unprecedented
power to fix meanings, if we add the caveat that ’meaning’ here refers
to the concrete effects produced in the interaction between subject and
object, speaker and listener, and not to the accompanying internal states
of mind.
Interestingly, in some cases the modern political method may depend
far more on the production of ambiguity than of truth itself. For example,
Lomnitz and Lomnitz (1990) in their study of the Mexican presidential
elections of 1988, suggest that the hegemonic strategy of the ruling
party (PRI) may in part rely on a social practice they term ’popular
hermeneutics’. They observed that events during the electoral campaign
are subject to multiple interpretations by the press and general
populace,
and are considered by all to be highly ambiguous until the election is over.
This period of uncertainty and speculation during which no positions are
final nor statements irrevocable, allows for the negotiation and resolution
of whatever differences impede the formation of consensus around the
PRI’s choice of candidate. In other words, the indeterminacy of all
political signs leading up to the election guarantees that the party will
be able to make the necessary adjustments so as to retain its political
dominance.
Indeed, the operation of political power upon such fragmented, cultur-
ally heterogeneous populations as are found in many modern nation states
may depend precisely on its ability successfully to reconcile a radical diver-
gence of opinions (’interpretations’) with a convergence of behavior. A
cogent account of such processes will require a conceptualization of power
which is more flexible than that provided by either the panoptical model
or (theoretical objections aside) Mitchell’s model of representational

power.

Colonial power: corporeal and visual

As the fundamental problematic in Mitchell’s work concerns the inter-
relationship of power and ’truth’, it will be helpful to return again to
Foucault’s own work on the question. In PowerlKnowledge, Foucault
offers the following definition of truth: ’the ensemble of rules according
to which the true and the false are separated and specific effects of power
are attached to the true’ (Foucault, 1980: 132). These rules take the form
of scientific discourse and, as such, are products of specific institutional
 291

structures, scientific procedures, techniques of intervention and control,

what Foucault has termed ’apparatuses of knowledge’ (1980: 102). While
these apparatuses may be accompanied by ideological constructs-for
example, an ’educational ideology’-their existence is rooted in specific
material practices, ones which impact upon human beings in precise and
directed ways. In other words, the knowledge, or ’truth’ circulated by
such apparatuses is intimately tied to the exercise of power on the body,
primarily through the very disciplinary mechanisms which provide its
conditions of existence: power produces truth and truth extends the
effects of power in a continuous feedback loop; what Foucault terms a
’regime of truth’ (Foucault, 1980: 133).
The ’truth’ produced by the power of representation, in contrast to the
above, operates on the level of appearances, and, hence, is essentially an
ideological phenomenon: ’order is, in effect, a framework that seems to
precede and exist apart from the actual individuals or objects ordered’
(Mitchell 1988: 176); ’The world appears to the observer as a relationship
between picture and reality’ (1988: 60); ’Seeming to exist apart from a
world &dquo;outside,&dquo; the apparatus of politics and schooling must touch the
exterior world’ (1988: 157). Verbs of visual perception-to seem, to
appear, etc.-dominate Mitchell’s work, suggesting a world manufactured
by optical illusions, simulations, trompe l’oeil. ’Truth’ in such a world is
a function of perception, and thus something on the order of ’belief’,
or ’ideology’. Whereas in Foucault’s analysis, ’truth’ is inseparable from

power in both its material and discursive manifestations, no such claim

can be made for the term as developed by Mitchell: there is no ’regime
of belief’ connecting it perforce to specific material practices. Hence the
relationship between ’truth’ and behavior in such a formulation remains
problematic.
The world structured according to the logic of representation required
and produced a subject who perceived reality in terms of the material/con-
ceptual division. Mitchell argues that through his continuous subjection to
the strategies of schooling and bodily discipline, this subject was trained
to interpret correctly the code of his semiolically constituted universe.
The effectiveness of the newly imported disciplinary mechanisms to exert
their unprecedented hold upon the body depended, or so Mitchell asserts,
on the subject having the ability to ’read’ this code, that is, to ’read’

power. Moreover, the diminution or ’disappearance’ of power, of its

visible restraints and coercive mechanics, is made possible precisely by
its textual re-presentation on the external surfaces of the world, a sort of
writing-on-the-wall about power and authority: ’The appearance of order
means the disappearance of power’ (Mitchell , 1988: 79).
292

But has it really disappeared? Even Mitchell agrees that it has not:
while disciplinary power operates without recourse to the more dramatic
visual displays which accompanied earlier methods of control, it none-
theless remains intensely present, pushing with a gentle hand whenever
needed, supervising, checking, correcting (Mitchell, 1988: 34-48). Yet,
in Mitchell’s view, disciplinary power, insufficient in itself, for some
unexplained reason, requires the interpretive act by the subject, the per-
ception recognition of a (metaphysical) authority. This conclusion marks
a departure from the notion of disciplinary power outlined by Foucault,

particularly in regard to the functioning of power in the constitution of

the subject. A subject on whom power operates through the incessant
reading of textual inscriptions is sharply distinct from one whose mind
and body themselves are nothing other than products of the workings of
power. To illustrate this divergence I return to Foucault.
In Discipline and Punish (1979), Foucault speaks of the exercise of
power on the body as giving rise to a ’soul’, understood not in the sense
of a ’theological effect’, as Mitchell would argue, but as a new form of
subjectivity, an actual, historical product of disciplinary methods:
...
[the soul] is the element in which are articulated the effects of a
certain type of power and the reference of a certain type of knowledge,
the machinery by which the power relations give rise to a possible corpus
of knowledge, and knowledge extends and reinforces the effects of this
power. On this reality-reference, various concepts have been constructed
and domains of analysis carved out: psyche, subjectivity, personality,
consciousness, etc. (Foucault, 1979: 29-30)
Disciplines such as education and psychology have made of this soul
their object of investigation, building upon it their extensive institutions,
the ’apparatuses of knowledge’. In turn, these apparatuses have made
the ’soul’ bearer of a number of truths about itself, such as the truth
of its sexuality recognized by Freud (Foucault, 1980a), or the truth of
its aggressiveness uncovered by animal behaviorists (Lorenz , 1963) and
sociobiologists (Wilson, 1978). The point I wish to emphasize here is that
the discursive practices which imbue the ’soul’ with its specific nature,
its content, do not socialize an already formed subject, nor inscribe
themselves on a tabula rasa we might call mind, consciousness, etc.
Rather, it is the operation of these practices alone (in all of their
materiality) which brings the soul, and hence the subject, to life: ’[A]
&dquo;soul&dquo; inhabits him and brings him to existence, which is itself a factor in
the mastery that power exercises over the body’ (Foucault, 1979: 30).
Mitchell clearly draws on Foucault’s work in his analysis of the coloni-
zation of ’the Egyptian character’. His examination of nineteenth-century
 293

journalistic and scholarly productions reveals the keen interest held

by both the British and local elites in the refashioning of Egyptian
subjectivities. This interest was pursued by means of the disciplinary
apparatuses discussed above:
The descriptive process of ’ethnology’ and the disciplinary practice of the
school worked together in this way to create the new subject of colonial
politics, the individual character or mentality.... Modern, educative
politics is an ethnological process, predicated upon the formation and
maintenance of this mind or character.... The true nature of this
character, moreover, was to be aproducer. (Mitchell, 1988: 105)
The education of this new Egyptian character became the subject
of intense concern and debate among scholars and administrators, as
evidenced in numerous publications circulated in the decades just prior to
the turn of the century. The works of Western orientalists, now translated
into Arabic for the first time, further fueled this debate; the disparaging
commentaries on Egyptian ’indolence’ and ’greed’ found in the works of
such notable scholars as Edward Lane and Georg Bernard Depping, as
well as the social Darwinist theories of Gustave Le Bon, gave shape to
the discussions on educational reforms. According to Mitchell, Le Bon
in particular played a salient role in this regard: ’Le Bon was probably
the strongest individual European influence in turn-of-the-century Cairo
on the political thought of Egypt’s emergent bourgeoisie’ (1988: 123).
Mitchell attempts to demonstrate how these reformist debates became
linked with the new educational practices by which the Egyptian character,
or soul came to be constituted.
However, having sketched out the productive role of power in con-
stituting the Egyptian colonial subject, Mitchell reasserts the primacy of
representation as the means by which power was able to achieve its hold
on subjects:
[Disciplinary powers] were to operate in terms of a distinction between
the physical bodythat could be counted, policed, supervised and made
industrious, and an inner mental space within which the corresponding
habits of obedience and industry were to be instilled. But more importantly,
this new divided personhood was to correspond to a divided world.
77!C world
worM wasw~M something
~o/~~~~g to~o be co~rMcfc~ and
~ constructed c~d ordered
o~y~d according
acco~~g to fo
The
an equivalent distinction between physical things and their non-material
structure. (Mitchell,126-7, emphasis added)
1988:

The division of the world into two realms, as Mitchell repeatedly stresses,
was a ’metaphysical effect’ produced by the techniques of representation.
Moreover, the mind also was such an effect: ’In the same way as it divided
the world, this division separated the human into two distinct parts, a body
294

and a mind’ (1988: 176). In other words, the truths which constitute the
mind-its industriousness, for example-exist only as representations,
surface images in the ’theatre of certainty’ set up before the individual
(Mitchell, 1988: 178). It is as observer in this theatre that one gains
knowledge of one’s soul, that one recognizes the truth about oneself.
Mitchell links this practice of reading truth, of recognizing the self, with
Descartes’ concept of the mind:
The Cartesian mind was conceived ... as an interior space in which
representations of external reality are inspected by an internal eye-in other
words, again, like an exhibition set up before an observer. (1988: 177)
In positing that the operation of power in the constitution of the
mind is predicated on perception, Mitchell, by implication, places a
phenomenological subject at the core of his analysis. A subject who
by constitution is industrious, obedient, etc., does not need to perceive
these elements in herself in order to act correspondingly. Her actions
reflect these qualities because they are her very nature. Mitchell’s analysis,
on the other hand, implies an a priori subject who, through the act
of perception, acquires his specific historical formation. The reality
of this subject therefore is purely a phenomenological one, rooted in
appearances and simulations; it is the product of what in Marxian terms
would be called an ideological construct.
Despite numerous points of overlap, it is in regard to this notion of the
subject that we find the greatest divergence between Mitchell’s project
and Foucault’s genealogical approach. As Foucault relates,
...
genealogy ... [is] a form of history which can account for the
constitution of knowledges, discourses, domains of objects, etc., without
having to make reference to a subject which is either transcendental in
relation to the field of events or runs in its empty sameness throughout
the course of history. (Foucault, 1980: 117)
Mitchell’s analysis suffers precisely because of its dependency on such a
trans-historical subject. What is the nature of this perceiving subject? On
what grounds can we claim to have isolated and described its characteristics?
The power of the genealogical approach stems from its ability to examine
history from outside the notion of the subject, hence circumventing the
insuperable problems incurred in posing these questions.
Moreover, Mitchell’s insistence on the necessity of an ideological com-
ponent in the functioning of modern power opens up his analysis to a
whole other set of problems. In arguing that under the colonial ’regime of
representation’ things always only ’seem’ or ’appear’ to be real, Mitchell
implies that there is some deeper level of reality which is hidden behind
 295

this somewhat illusory semiotic one. That is, if the notion of an ’external
reality’was simply an effect produced by a particular mechanism of power,
as Mitchell contends, then what was the other reality which this effect was

concealing? While Mitchell’s answer to this question might well be ’power’,

he makes no attempt to address the problem directly.
Mitchell’s description of the Middle East in the period before European
colonial expansion, however, does reveal his general orientation to this
question. For Mitchell, the pre-colonial Middle East, as a region which
’had not yet been organised representationally’, was free from the ’strange
effects’ which characterized the West (Mitchell, 1988: 29); reality here
was not ’theological’ or ’metaphysical’. Mitchell draws on Bourdieu’s

description of a Kabyle house4 to clarify the distinction. For the Kabyle,

as Mitchell explains:

...
space itself [within the house] is polarised, according to the oppositions
Bourdieu describes, and the polar oppositions invest every activity in the
house.... [S]uch polar forces occur themselves not as a structure of
oppositions but as an unstable play of differences. The male, the light
or the dry is each nothing more than the process of excluding or deferring
the female, the dark or the wet.... Difference, as Derrida would tell us,
is not a pattern of distinctions or intervals between things, but an always
unstable deferring and differing within. (Mitchell, 1988: 50)

Notably, this description bears a definite resemblance to that which

Mitchell uses in addressing the Arabic language. In other words, ’reality’
in the Middle East is entirely exhausted in its semiotic aspects: for Mitchell,
language is clearly the model for understanding the world.

Final comment
I have chosen to discuss in some detail the conceptual problems found
in Mitchell’s work, not with the intention of discrediting or devaluing the
endeavor, but, on the contrary, because I believe the type of investigation
he has undertaken to be an important step toward a different account of
colonial history. Mitchell’s documentation of nineteenth-century colonial
discourses, and especially of the reformist debates and policies within
Egypt, serves as a persuasive argument for a new articulation of the
history of European colonialism in terms of the genealogy of the modern
subject. As for the various problems in the work which I have addressed
here, they can in great measure be attributed to an undervaluing of
historical particularity. Assumptions about the trans-historical operation
of language dominate over and obscure the specific encounter of elements
and strategies which led Egyptians to speak and write differently to how
296

they had before. Likewise for Mitchell’s use of ’representation’, a concept

deeply embedded in certain recent philosophical discourses, but one
whose relevance in other contexts-such as nineteenth-century Egypt-
should not be assumed. Granted, Mitchell does attempt to show how the
conditions of possibility for the operation of representational power were
exported to Egypt during the process of colonization. His investigation
falls into epistemological error, however, to the extent that it is guided
(and thus, blinded) by an understanding of the concept ’representation’
rooted solidly in quite contemporary discussions: the chasm separating
this modern philosophical notion from the world of practical experience
in colonial Egypt cannot be convincingly bridged in this way.
This brings up another point concerning the importance of contextual-
ization, one which Bourdieu himself has noted in his recent work on the
’scholastic point of view’. As he affirms:
I discovered that the scholastic vision destroys its object every time it
is applied to practices that are the product of the practical view and
which, consequently, are very difficult to think of, or are even practically
unthinkable for science.... (Bourdieu, 1990: 382)
Mitchell’s comparison of a certain ’scholastic vision’ (a postmodernist
theory of representation), with the practical knowledge embodied in such
social forms as Egyptian or Algerian self-built housing constitutes just
such an epistemological mistake. The productions of a reflexive knowl-
edge will almost inevitably be far more schematic, integrated and regular
than those resulting from practical forms of thought. Consequently,
any comparison of architectural practices, for example, must take into
consideration who does the planning and construction (class, ethnicity,
etc.), and where their knowledge originates. An investigation into these
sorts of empirical questions would have led Mitchell to confront some
of the conceptual difficulties I have discussed above. While his attempt
to combine a language-based approach with one (Foucault’s) based on
a quite different set of premises about the world ultimately fails in
the particular regards I have discussed here, it remains nonetheless a
useful document for assessing the potential value of certain theoretical
approaches, and in pointing out directions for further research.

NOTES
1 For recent works in this vein see American Ethnologist 16 (4), particularly the
contributions of Ann Stoler, John Comaroff and Randall Packard The pioneering
studies in this area remain those of Talal Asad (1973) and Edward Said (1978)
2 This conclusion was even adopted by a number of the nineteenth-century Egyptian
 297

reformers, as the following observations of Ahmed Amin, a reform-minded writer of

the penod, makes clear ’If one uses the English language, each assertion leads
to the next, concise and to the point It is rare that there is any play on words,
digression from the subject, or repetition of arguments However, if one uses
the Arabic language, the discussion drags on, becomes more prolix, and ideas
often are linked, not with sister-ideas, but with more distant cousin-ideas’ (Amin,
1965, transl. Hirschkind)
3 Until the twentieth century, the Maghrib was rarely placed under the control of
a single centralized regime Numerous writers, Ibn Khaldun among them, have
commented on the form of political instability endemic in the region, one often
analyzed—in a less than satisfactory manner—in terms of a constant struggle
between an imperial state centered in the cities and various tribal groups residing
in the mountains (see Gellner, 1981)
4 The Kabyle studied by Bourdieu live not in Egypt, but in Algeria In choosing to rely
on Bourdieu’s work, Mitchell implicitly accepts the highly problematic assumption of
cultural homogeneity across North Afnca Once again, the specifics of the Egyptian
experience are made subordinate to the requirements of his conceptual argument.

REFERENCES

Amin, Ahmed,
1965 ’Logique et langue’, in A Abdel-Malek (ed.) Anthologie de la littérature Arabe
contemporaine
Asad, Talal,
1973 Anthropology and the Colonial Encounter Atlantic Highlands, NJ: Humanities
Press
Asad, Talal,
1987 ’Are There Histories of Peoples Without Europe? A Review Article’, Society for
the Comparative Study of Society and History
Bourdieu, Pierre,
1990 ’The Scholastic Point of View’, Cultural Anthropology 5(4). 380-91.
Foucault, Michel,
1979 Discipline and Punish The Birth of the Prison New York Vintage Books
Foucault, Michel,
1980 Power/Knowledge Selected Interviews and Other Writings, 1972-1977 New
York Pantheon Books
Foucault, Michel,
1980a The History of Sexuality Volume 1 An Introduction New York Vintage Books
Gellner, E ,
1981 Muslim Society Cambndge Cambndge University Press
Lomnitz, Claudio and Lomnitz, Lanssa,
1990 ’EI fondo de la forma actos publicos de la campana presidencial del Partido
Revolucionario Institutional Mexico, 1988’, Nueva Anthropologia, April
Lorenz, Konrad,
1963 On Aggression New York Harcourt, Brace, and World.
MacIntyre, Alasdair,
1988 Whose Justice? Which Rationality? Notre Dame, IN University of Notre Dame
Press
298

Mitchell, Timothy,
1988 Colonizing Egypt Cairo: The American University in Cairo Press.
Said, Edward,
1978 Orientalism New York: Vintage Books.
Volosinov, V.N ,
1973 (1929) Marxism and the Philosophy of Language. Cambridge, MA Harvard
University Press.
Wilson, E O.,
1978 On Human Nature Cambndge, MA: Harvard University Press.