Society for Comparative Studies in Society and History

After Orientalism: Culture, Criticism, and Politics in the Third World Author(s): Rosalind O'Hanlon and David Washbrook Reviewed work(s): Source: Comparative Studies in Society and History, Vol. 34, No. 1 (Jan., 1992), pp. 141-167 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/178988 . Accessed: 14/02/2012 07:52
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Cambridge University Press and Society for Comparative Studies in Society and History are collaborating with JSTOR to digitize, preserve and extend access to Comparative Studies in Society and History.

http://www.jstor.org

After Orientalism:Culture, Criticism, and Politics in the Third World

ROSALIND O'HANLON Clare College, Cambridge DAVID WASHBROOK

University of Warwick Over the last decade, studies of "third world" histories and cultures have come to draw to a very considerableextent upon the theoreticalperspectives and postmodernism.With the publication in provided by poststructuralism 1978 of Edward Said's work, Orientalism, these perspectives-now fused and extended into a distinctive amalgam of cultural critique, Foucauldian approaches to power, engaged "politics of difference," and postmodernist emphaseson the decenteredand the heterogeneous-began to be appropriated historiesand cultures.Certainly in a majorway for the study of non-European in our own field of Indian colonial history, Said's characteristicblending of these themes has now become virtually a paradigmfor a new generationof historiansand anthropologists.These directionshave been most recently and sharply endorsed in Gyan Prakash's discussion, "Writing Post-Orientalist Histories of the Third World:Perspectivesfrom IndianHistoriography."' of We sharePrakash'sconcernwith the emancipation previouslysubmerged colonial historiesand identities. However, we are deeply concernedat the way in which his "postfoundational" history would set aboutthese tasks. Prakash sees this history, and the postmodernistand poststructuralist perspectivesthat of underlieit, as ourbest futurehope for a genuinelycriticalunderstanding the Indianpast. We question this, given the mannerin which these perspectives have come to be interpretedand absorbedinto the mainstreamof historical and anthropologicalscholarship,particularlyin the United States. We argue thatpostfoundational historyoffers us ways of "knowing"the Indianpast that are quite inadequateto its supposed political concerns. In emancipatingourselves from what Prakashcalls foundationalism,we need also to ask rather
We would like to thankAjay Skaria,CrispinBates, Saurabh Dube, David Ludden,FredReid, and Burt Stein for their reading and comments on this paper. I Gyan Prakash, "WritingPost-OrientalistHistories of the Third World:Perspectives from Indian Historiography," ComparativeStudies in Society and History, 32:2 (April 1990), 383408. 0010-4175/92/1609-0300 $5.00 ? 1992 Society for ComparativeStudy of Society and History

I4I

142

ROSALIND

O'HANLON,

DAVID WASHBROOK

more carefully what exactly we are emancipatingourselves into. We argue that these approachesprescribe remedies which actually create new and in many cases much more serious difficultiesof their own, in partbecause they have, of course, as much to do with argumentsaboutthe politics of representation in Westernintellectualand academiccircles, as they do with imposing on thatmannerof representation the thirdworld'shistory.We discuss what we in the difficultiesof these approaches the contextof Indianand other see to be historicalwriting and suggest that they have arisen in partfrom non-Western the widely sharedbut mistakenassumptionthat EdwardSaid's work provides a clear paradigm for a history that transcends older problems of representation.
POST-FOUNDATIONAL HISTORY: DILEMMAS AND PROBLEMS

TakingEdwardSaid's definition of Orientalismas his startingpoint, Prakash moves througha rangeof approachesfor the studyof Indiansociety, showing how each has inheritedand reproducedsome of Orientalism'skey assumpfor tions and techniquesof representation.Indiannationalisthistoriography, example, has been unable to transcendOrientalism'spreoccupationwith essences and its teleologies of modernity.Its historiansunderstoodknowledge of as a "more or less adequaterepresentation the real," and India itself as India itself appeared having an existence independentof its representations.2 for them as an undivided subject struggling to transcend colonial backwardnessand to realise itself as a modem nationalstate. Likewise, the area studies programmesthat dominated South Asian history and anthropology from the 1950s searchedfor an authenticIndianhistoryand culture, fixing on caste as Indian society's essence and scrutinizingits structuresin terms of their potential as vehicles for political and economic modernization. foundationalhistories." By this he Prakashthen turnsto "post-nationalist and what he calls "social historiansoriented toward world means Marxist history,"such as C. A. Bayly, who have been concernedwith Indianpolitical economy, particularlyin its relationshipto world-historicaltransitions. Although Prakash carefully points out their gains, he finds them ultimately because theirhistories are "foundational." They use categories unsatisfactory which are at some level fixed and essential, as if history were "ultimately foundedin and representable throughsome identity-individual, class, structure-which resists furtherdecompositioninto heterogeneity."Such categories cannotbut have an "objectivistbias" built into them.3 Theiremphasison the theme of capitalist transitionleads, moreover, to a teleological account
2 Ibid., 390. 3 Ibid., 397.

THE THIRD WORLD AFTER ORIENTALISM

143

that sees India principallyas an instance of abortedcapitalist modernityand cannot this exploreand exposethe alteritywhichunderlies identity-otherthancallingit as to un(or protoindustrial feudalandsemi-feudal opposed capitalist), precapitalist, to and freelabor opposed freelabor), traditional modern). strategy This cannot (as (not the of Indian historicize emergence a modem,colonial-capitalist nation because does it framed andby thathistory.4 in not displacethe categories These approachescan only in the end legitimate the structuresof capitalist modernitythey describe; for, Prakashasks, how is it possible to understand Indianhistory in termsof the developmentof capitalism, "butalso contest, at the same time, the homogenization of the contemporary world by capitalism?"5 In the last part of his discussion, Prakashconsiders what he calls "postOrientalist" histories, which try to move towards postfoundational approaches. These utilise the insights of EdwardSaid and Michel Foucaultand draw furtheron themes from postmodernism,feminism, minoritydiscourses and other advocates of the "politics of difference." These approachesshare Prakash'sconcern to show how knowledge about the third world is historically produced.They seek "to make culturalforms and even historicalevents to contingent, above all, on power relations."6Avoidingthe temptation return to essential identities, they work insteadto displace foundationalsubjectsand essences, to break up notions of a unitary India into a multiplicity of contingent and unstable identities which are the effects of changing power relationships. They refuse the privileged themes of global capitalist modernization and focus instead off-centre on what those themes exclude: histories of the subordinatewhose identity, like all identity, resides in difference. Postmodernistperspectives are importantin shaping these approaches,with their "blurred genres and off-centredidentities"and theirhostility to systematizing theories: Fashioned denialsof grand defiesandrefuses theories, totalizing by postmodernism list can definition.Only a laundry of conditions be offered-TV images,fashion to Salman and Rushdie, Heads,challenges universalist essentialist Talking magazines, irreverence playfulness, and transnational theories,architectural capitalism.7 Nor do the new histories limit their vision to India or other third world others in Westerncontexts, with societies. They forge links with subordinate radicals, feminists, ethnic and other minorities, in a common challenge to and teleologies of modernization theirconstituentthemes of Reason and Prog4 5 6 7 Ibid., Ibid., Ibid., Ibid., 399. 398. 401. 404.

144

ROSALIND O'HANLON,

DAVID WASHBROOK

ress. Above all, they do not draw back from political engagement. They of identify with the subject-position the subordinate,concernthemselves with relationshipsof domination, and self-consciously make their own historical accounts into contestatoryacts. In these respects they differ from the often depoliticized perspectivesof postmodernism,while at the same time sharing its emphasis on the provisionalityof all identities, its resistanceto all systematizing or totalizing theory and its refusal to set up "new foundations in history, culture and knowledge."8 Prakashpoints to examples of these new approaches.Although he notes their limitations, he commends RanajitGuha and the Subaltern Studies project for deploying poststructuralist arguments This has enabledthem to get away from the and the concept of "subalternity." older frameworksof colonialism and nationalismwithin which Indianhistory was studiedand to breakup theirassociatedfoundational categories, revealing Indiainsteadas "a multiplicityof changingpositionswhich are then treatedas effects of power relations."9The work of BernardCohn and Nicholas Dirks reveals in differentways how colonial rule createdand froze social institutions which the British took to be immutable features of India as a primarily religious society. In common with postmodernists,Ashis Nandy'swork on the culture and psychology of colonialism has repudiatedthe "post-Enlightenment ideology of Reason and Progress," in which "Descartes defined rationality and Marxdefined social criticism."'0 To escape these tyrannies,we must turn to "mythographies,"the hidden stories of colonialism's victims, of which will "expose the mythic character colonial and postcolonialfables of modernity." Salman Rushdie likewise shares postmodernism'shostility to "grandtotalizing theories," disclosing in Midnight'sChildrenthe "fable-like characterof real history."' But we see many problems here. The critique of foundationalcategories derives in large part from the work of Jacques Derrida, although Derrida's work contains very little to indicate how we should go about the basic, Derinescapably active, and interventionisttask of historical interpretation. of the conventionaland nonobjective rida'sparticular approachto the problem natureof our categories and schemes of interpretation may actually represent something of an intellectual cul-de-sac, at least for those who would offer forms of historical understanding.As John Searle has argued, but sees Derrida correctly thattherearen'tany suchfoundations, he thenmakesthe of The him that mistake marks as a classicalmetaphysician. realmistake theclassical but was foundations, rather metaphysician notthebeliefthatthereweremetaphysical
8 Ibid., 406. 9 Ibid., 400. 10 Ibid., 404-5. 1 Ibid., 405.

THE THIRD WORLD AFTER ORIENTALISM

145

were necessary, belief that the the belief that somehowor othersuch foundations is or unlesstherearefoundations something lost or threatened undermined.12 In the absence of such foundations, Derridacan do little more than reveal, natureof our categories and over and over again, the subjective and arbitrary the uncertaintyof the knowledge derived from them. He misses, in effect, the crucial point that we cannot actually do without some categories and some means of evaluatingordersof certainty,in orderto comprehend,to explain, to elucidate and to do. That these categories are conventions, Searle further argues, is no bar to our continuingto use them providedwe recognize them for what they are, inventionsof our own necessity. However, this recognition involves a change in the way that we conceive and test them-not against metaphysicallyconceived standardsof objectivity but against their adequacy in serving the purposes for which we want and need to use them. Such considerationsof course include ourselves and the reasons why we require kinds of knowledge. Preoccupiedas he is with the non-problemof particular objectivity at the expense of questions of purposive adequacy, Derrida has aim were ratherlittle to offer us on these key questionsof method. If Prakash's to render our existing knowledge of Indian and other third world simply societies uncertainand unstable,therewould indeedbe a point in his invoking Derrida's attack on foundationalforms of knowledge. Because he actually intends a highly purposive agenda of historical reconstructionand political engagement, however, this invocation seems to us starkly inappropriate. and Prakash'scritiqueof Indianhistoriography his prognoses for its future reflect these contradictions.Most who fall into his category of Marxist and social historians of India have long recognised the irreducibly subjective element in their interpretations, seeing that the historianis inescapablya part 3 of what they study as a constantprocess of movement and transformation. Most would be thoroughly mystified by the charge that they operate with reified and ahistorical categories of class, individual, and structure. Such categories are usually contextualizedin terms of their makingand unmaking,
12 John Searle, "The WordTurned Upside Down," NewvYorkReview of Books (27 October 1983), 78. A good introductionto this debate is in JurgenHabermas,The Philosophical Discourse of Modernitv(Cambridge:Polity Press, 1990), 194-9. Ajay Skariakindly providedthis reference. 13 Prakash'snotion of what constitutes Marxisthistory is problematic,for neitherof the two examples which he provides fall easily into the category.The first, concerningBengali histories of the Bengali renaissancewould seem most influencedby Bengali nationalistideology, as it is cultureand a bourgeoissocial order not clear why Bengal's failureto generatea secularrationalist priorto the developmentof industrialcapitalismis a problemfor Marxism.The second, concernalso ill fits the category,for the ing usage of AndreGunderFrank'sconcept of underdevelopment, concept derives from neo-Smithianratherthan Marxisteconomic theory:see R. Brenner, "The Origins of Capitalist Development: A Critiqueof Neo-Smithian Marxism,"New Left Review, 104:4 (1977), 25-92.

I46

ROSALIND

O'HANLON,

DAVID WASHBROOK

their emergence and decline. Bayly, for example, presentseighteenth-century India in terms of the making and unmakingof a particular contingentset and of relations, which threwup a distinctiveand ultimatelytransientstructure of class, on the basis of which colonial rule was initiallyestablished.He plainly sees class, along with other forms of structureand identity, as historically contingent, unstable, and given to change-certainly not as immutable in some way. It is also not obvious that these historiansunderstandcapitalist transitionmerely in terms of Westerndevelopmentand Asian underdevelopment. Within the Marxist discourse, debates around the themes of comparativefeudalism, the articulationof modes of production,and the work of Robert Brennerhave all explored the specific dynamics of non-Britishand relationsof productionand social formation.14 Equally,a wider non-Western major thrustof researchon the Indianpast has for a considerabletime now been precisely to breakdown East-West dichotomiesby exploringthe indigenous forms of capitalismand their associatedmilitaryand mercantileinstitutions that were developing in India from the late seventeenthcentury.This researchdescribeshow these indigenousdynamicspowerfullyand importantly shaped the East India Company's initial engagement with the economies and societies of the subcontinentand its own subsequentdevelopment as a colonial state.15Bayly himself sets this against a sharplyredrawnpictureof Britishsociety designed to reveal the precise ways in early nineteenth-century which its forms of modernitywere not only partialand limitedbut createdout of and sustainedby wider imperialrelationships.16 Prakashalso contendsthat any historianwho writes aboutIndia'shistoryin terms of capitalism's development must in the end be complicit in the very hegemony so described. Rather,we must aim for a "refusalof foundational The categories that constructthe theme of global modernity."17 implications of this seem somewhat unclear. If the complicity arises from a tendency to present the world of capitalismas homogeneous, it must be pointed out that most Marxistsocial historycritiquescapitalistmodernityprecisely in orderto challenge the self-images and pretensionsto the universalityof Westernsocial theories of modernization.Lumpingthe two togetherbecause both appearto
14 See, for example, T. H. Aston and C. H. E. Philpin,eds., TheBrennerDebate (Cambridge: Mukhia,eds., "FeudalismandNonCambridgeUniversityPress, 1985); T. J. Byres and Harbans EuropeanSocieties," Journal of Peasant Studies (Special Issue), 12:2, 3 (January, April 1985). 15 Forthese argumentsin Bayly, see TheLocal Roots of IndianPolitics: Allahabad1880-1920 and Bazaars: North Indian Society (Oxford:Oxford University Press, 1975); Rulers, Townsmen in the Age of British Expansion (Cambridge:CambridgeUniversity Press, 1983) and Indian Society and the Making of the British Empire (Cambridge: CambridgeUniversityPress, 1988). For a general guide to recent research in this field, see D. A. Washbrook, "Progress and Problems: South Asian Economic and Social History c. 1720-1860," Modern Asian Studies, 22:1 (1988), 57-96. 16 These argumentsare developed in C. A. Bayly, ImperialMeridian:TheBritishEmpireand The World1780-1830 (London:Longman, 1989). 17 Prakash, "WritingPost-Orientalist Histories," 398.

THE THIRD WORLD AFTER ORIENTALISM

147

address the same problem of the forms and forces of capitalist modernityis theoristson whose deeply misconceived. Prakashand the otherpostmodernist work he draws apparentlyhave the view that merely engaging the question of determinesour understanding it so that we ought actuallyto assume that it does not really exist in any systematic form. What his position leaves quite obscureis what statusexactly this categoryof "capitalistmodernity"occupies for him. If our strategy should be to "refuse" it in favourof marginalhistories, of multiple and heterogeneous identities, this suggests that capitalist modernityis nothing more than a potentiallydisposable fiction, held in place simply by our acceptance of its cognitive categories and values. Indeed, Prakashis particularlydisparagingof Marxist and social historians'concern with capitalism as a "system" of political economy and coercive instrumentalities. Yet in other moments Prakashtells us that history's propertask is to challenge precisely this "homogenizationof the world by contemporary capitalism." 8 If this is so, and there is indeed a graspablelogic to the way in which moder capitalism has spread itself globally, how are we to go about the central task of comprehending this logic in the terms that Prakash suggests? These problems seem furthercompoundedif we turnto the work of historians whom Prakash recommends as exemplars of postfoundationalistapproaches. What is puzzling is that many of these historiansthemselves put forwardtimeless or undifferentiated conceptionsof the Indianpast, often in a particularlyglaring way. BernardCohn has undoubtedlydone much to disassemble monolithic notions of a traditionalIndiaadvancedin colonial social theory.Yet in his accountof how these notions were fabricated,he describesa clash between European and Indian forms of knowledge which are both undifferentiated,the former located in time somewhere between the seventeenth and the nineteenth centuries and the latter not at all.19 Ashis Nandy identifies the psychological damage and "loss" associated with the colonial experience. Yet his strategyfor the recoveryof an "Indianself" seems merely to invert a range of what were originally Orientalistconceptions about India and to generalizethe culturalexperienceof Bengali literatito thatof the whole nation.20RanajitGuha may well criticize "bourgeois"Indiannationalismfor its failureto identify with the very differentneeds of subalternclasses, but he does take the central question of modern Indian history to be the "historic failureof the nation to come to its own," a questionthat plainly derives from the nationalistparadigmthat Prakashcondemns so strongly.21 Many theories
Ix Idem. 19 See, for example, Cohn's "The Commandof Languageand the Languageof Command,"in R. Guha, ed., SubalternStudies IV (Delhi: Oxford University Press, 1985), 279-80. 20 See especially Ashis Nandy,The IntimateEnemy:Loss and Recoveryof Self Under Colonialism (Delhi: Oxford University Press, 1983). 21 RanajitGuha, SubalternStudies I (Delhi: Oxford University Press, 1982), 7.

148

ROSALIND

O'HANLON,

DAVID WASHBROOK

about Indian personality and social structurewhich Guha uses to test the consequences of colonial dominationbear a strong resemblanceto those of elsewheredeems to be "refiguredessenLouis Dumont, whose ideas Prakash tialisms." Indeed, Guhahas of late takento referringto the (undifferentiated) dismisses "totalizing"understandings the of Indiannationas "us."22Prakash Indian past in favour of the alternative and the marginal and commends Nicholas Dirks' attackon Dumont's ahistoricaltheories of caste for making of this possible. Yet Dirks himself presentsus with a countertheory caste that is scarcelyless generalizingthan Dumont'sown. He erects it, moreover,very largely on the basis of the worldview and self-images of locally dominant groups.23Prakashhimself does what he tells us not to. He warns us against writing history aroundthe majorthemes of global transitionbut then writes about Indian historiographical development in precisely these terms, seeing to of the determinants its progressionpassing from imperialism nationalismto a liberal hegemony centered on the United States. of This all makes it very difficult to graspthe character postfoundationalist of understandings the past or to see what they are meant to achieve. These confusions seem to us to arise out of a wish to generatean historicalpraxis from Derrideanand postmodernist perspectivesthatare inherentlyinimical to it. These perspectivesunderminepossibilities for such a praxis in two ways. in First, because they regardany interventionby the historianor interpreter the past as inherentlyillegitimate, a kind of complicity,they fail to acknowland edge the particular specific means by which that scholaracquiresknowlof the past. Prakashobjects to our giving some analytical categories edge privilege on the groundsthatthis "occludesthe historiesthatlie outside of the themes which are privileged in history."But this suggests that the themes of history are or should be given in the materialof historyitself, exposed or not exposed by the historian, whose cognitive relationto them is passive. What this objection reflects is actually a rather old-fashioned, even positivistic assumptionaboutthe sourcesof historicalknowledge, but one also which may not surpriseus. For as Searle has argued, Derrida'sown obsession with the of non-problem objectivityand his failureto recognizeour subjectiveneed for as primary and legitimate, leaves his concerns also laden with knowledge

22 Guha's latest contributionto SubalternStudies, "Dominancewithout Hegemony and its Historiography,"distinguishes between a British and a precolonial Indian form of political He and authority,the latterorganizedaroundprinciplesof Brahmanic kingly authority. concludes by describing his argumentas "a critiqueof our own approachto the Indianpast and our own in performance writingabout it," designed to "assist in the self-criticismof our own historiograof phy-the historiography a colonized people" (Guha'semphasis;SubalternStudies IV [Delhi: Oxford University Press, 1989], 306-7). 23 In this case, the royal and dominantKallarcaste in Pudukottai.See Nicholas B. Dirks, The Hollow Crown: Ethnohistoryof an Indian Kingdom(Cambridge:CambridgeUniversity Press, 1987).

THE THIRD WORLD AFTER ORIENTALISM

I49

residues of positivism.24The objection entirely misses the fact that the past, including its historical subjects, comes to the historianthroughfragmentary and fracturedempirical sources, which possess no inherentthemes and express no unequivocalvoices. In and of themselves, these sources and voices are just noise: "Other"histories uncovereddo not speak for themselves any more than the "facts" of history do. To state the obvious, the historianmust undertake prior,and in partsubjective, tasks thatonly the historiancan do: the to turnthe noise into coherentvoices throughwhich the past may speakto the presentand to constructthe questions to which the past may give the present intelligible answers. Prakashseems to refuse to acknowledgethe inevitability (and the responsibility)of this task. Indeed, he offers us a methodologythat would seem to rule out even the refusalsof which he speaks. He enjoins us to refuse particularthemes and categories, most notably those pertainingto the global transitionto moderncapitalism, lest simply by engaging with them we become implicatedin and so reproducethe hegemonies which they represent. But how can we refuse certain themes if we do not know what they are and how can we know what they are if we are not permittedto engage and study them? Second, and in common with others who have drawn on postmodernist perspectives, Prakashseems to think that it is not possible to recognise differences or resistance under the rubric of general or totalizing systems and theories of transition. There are fundamentalmisconceptionshere. As Raymond Williams and FredricJamesonhave arguedin theirdifferentways, it is unclearwhy a system or process should by definitionbe incapableof generatMarxing differenceor raising resistances. Capitalismas most contemporary ist historianssee it indeed constitutesa system or process but one inherently conflictual and changeful, incapable of realizing or of stabilizing itself. It producesand operatesthrougha wide varietyof social relationsof production and exploitation, which are themselves in constanttransformation. Although its forces may shape forms of resistance, they do not predetermineits outcomes, for no hegemonic system can pervade and exhaust all social experience, least of all one which fails to meet so many humanand social needs.25 Indeed, it is only in the light of some conceptionof a dominantculturallogic or hegemonic system that resistance, emancipation, or difference can be meaningfullyidentifiedor measuredat all.26 It is also difficultto take Prakash seriously when he recommends postmodernistperspectives on the grounds that they avoid totalizing forms of theoryor explanation.As Jamesonhas also pointed out, postmodernistapproachesare themselves built arounda form of
24 Searle, "The WorldTurnedUpside Down," 78-9. 25 RaymondWilliams, Marxismand Literature(Oxford:Oxford UniversityPress, 1977), 125; Fredric Jameson, "Marxism and Postmodernism,"New Left Review, no. 176 (1989), 34-9. 26 FredricJameson, "Postmodernism,or The CulturalLogic of Late Capitalism,"New Left Review, no. 146 (1984), 57.

I50

ROSALIND O'HANLON,

DAVID WASHBROOK

totalizing abstractionthat distinguishes postmodernculture by its logic of difference and its sustainedproductionof randomand unrelatedsubsystems of all kinds.27 In these ways, then, postfoundationalist history and the wider from which it derives seem to us to offer an uncertainand deeply perspectives inconsistent premise from which to conceive our relationshipto the past.
REPRESENTATION, SELF-REPRESENTATION, AND POLITICS

If these practicalexamples of a postfoundationalist approachseem beset with problems, what of the theoretical arguments, the combination of cultural critiques, styled after Said, Foucaultianperspectiveson power, engaged politics of difference, and aspects of postmodernisttheory that Prakashsees as animatingthese new directions in history? The core of his argumentis that these perspectives can be combined and employed both to emancipateother histories and to develop new approachesto the largerquestionof representaas tion and its politics. But there are critical questions here too, in particular issue of self-representation minority these argumentsrelate to the wider by and marginalgroups themselves and in contexts involving the developed as well as undevelopednations. As we shall argue, we need to look rathermore carefully here at what we are emancipatingourselves into. Prakashclearly wishes to retainsome notionof an emancipatory politics for view of the the dispossessed, as against, for example, an extremeFoucauldian inescapabilityof relationsof power and domination.If we do wish to hold to some view of political struggle as potentially emancipatory, yet simultaneously refuse to define what the larger structuresand trajectoriesof such struggle might be, on the groundsthat this would constitutea totalizing form such a politics and its trajectoof analysis, we push the burdenof representing ries onto those who are in strugglethemselves. This is notjust by default.The is, principle of self-representation as we shall see, enshrinedand positively recommended in much explicit postmodernisttheory as the very means to recovering suppressed histories and identities. The obvious problem here, though, is that self-representation,the idea that there can be unitary and centredsubjectswho are able to speakfor themselvesandpresenttheirexperience in their own authentic voices, is precisely what postmodernisttheory attacks in the Westernhumanisttradition. A numberof critics have tried to blurthis problemby talking in termsof a kind of rainbow alliance shared among a range of oppositionalvoices. This may, indeed, be Prakash'sattemptedsolution to this dilemma. He describes how "the new post-Orientalist scholarship'sattemptto release the thirdworld forms a part of the movement that advocatesthe from its marginalposition 'politics of difference'-racial, class, gender, ethnic, national and so
27 Jameson, "Marxismand Postmodernism,"34.

THE THIRD WORLD AFTER ORIENTALISM

151

forth."28This appears at first to resolve the difficulties in privileging selfrepresentation,for what is offered instead is a common platformsharedbetween a variety of dissenting groups, who can speak to and for others and for themselves. In some respects, resistances from the point of view of class, gender, ethnicityor thirdworld nationhoodindeed sharecommon ground;but assumingthat these share the same agenda in some more generaland positive way simplifies what are actually very complex and sometimes fiercely antagonistic positions.29 It is also very difficult, from any set of Foucauldian perspectives at least, to generate a common platformor a fusion of struggle them for these localised oppositionalgroups. Doing so means subordinating to a transcendent totalizingform of political logic. If it is hardto generatea or common agenda for these oppositionalgroups, we are led back to some form to of privileged self-representation. Veryclearly, it is tremendouslyimportant attendto the experiencesand self-accountsof marginalgroups;but this is very different from the nativist view, implicit here, that they have some kind of inherentlysuperiorvalidity. Prakashdisassociateshimself stronglyfrom such a view, but it is hard to see how he can avoid it, given the contradictions described above. This leads on to a furtherset of problems. We are invited to see these new critiquesof Orientalistand otherforms of privilegedknowledgeas contestatory acts, to commend their concern with relationshipsof dominationand their efforts to unlock and release histories, cultures and identities frozen by the essentialisations of the past. This implies not only that subjects can and do representthemselves on the basis of theirexperience;it suggests also thattheir resistances eventuate in forms of knowledge which are emancipatory,transcending relationshipsof domination,in some senses at least. The problemis that these assumptionsare not consonant with the kind of Foucauldianperspective on power and identitythat Prakashcommendselsewhere. As a range of critics have pointed out, including Said himself, it is difficult to see how any concerted political engagement, let alone one with the processes of capitalist modernization,is possible on the basis of Foucault'sdeliberatelyamorphous and dispersed vision of power.30Such an engagementlooks even less promising when we are told that postfoundationalism'smajor virtue is its intellectualrefusal to accept the very analyticaltheme of capitalistmodernity,
28 Prakash, "WritingPost-OrientalistHistories," 406. 29 On the issue of sati in India, for Ashis

example, compare Nandy,At the Edge of Psychology: Essays in Politics and Culture (Delhi: Oxford University Press, 1980), 1-31, with Sharada Jain, Nirja Misra, and Kavita Shrivastava, "Deorala Episode: Women's Protest in Rajasthan," Economic and Political Weekly,nos. 7, 11 (1987), 1891-4. See also the very interestingdiscussion of Nandy'sposition on the Roop Kanwarcase in LataMani, "MultipleMediations:Feminist Scholarshipin the Age of MultinationalReception," Inscriptions, no. 5 (1989), 15-16. 30 Edward Said, The World, The Textand The Critic (London:Faberand Faber, 1984), 245.

152

ROSALIND O'HANLON,

DAVID WASHBROOK

lest we take on its ideologies by admittingto any of its realities. The principal casualtyof this inadequacymustbe politics, for whatkindof resistancecan be raised to capitalism's systemic coercions if that resistanceapparentlydenies their existence? Indeed, it is even less clear that one can generate what is ultimately a politics of emancipationfrom a set of Foucauldianassumptionsabout power and social relations. Prakashand many who sharehis approachesvigorously and virtuously assert the presence of struggle in all social relations whilst saying very little aboutthe systematicpolitical meansby which emancipation is to be pursuedor what indeed it might look like if it were ever achieved. becomes a strugglepurelyinternalto the Accordingto this view, emancipation consciousness of those who resist andonly representable them. The precise by effect of this reading of emancipationback into Foucaultis to returnthese areas of his argumentto their sources in Nietzsche. Emancipation becomes a Nietzschean act of pure autonomouswill. This might seem an ironic position for a theory concerning itself with the struggles of underclasses,31but as Prakashhimself notes, this has been precisely the approachof the Subaltern Studies group, which he then commendsto us for its creativeappropriation of poststructuralist perspectives! There are furtherdifficultiesconcerningquestionsof subjectivityand hence of history and agency. Prakashdraws on Foucaultto argue that subalternity, indeed the multiplicityof changing positions within Indiansociety, are to be of regardedas "effects of power relations."The subject-position the subaltern likewise is an effect, contingentand unstable, which "residesin difference." Questions of subjectivity are discussed in terms of the discourses which construct it. Thus,
the identificationwith the subordinated's subject-position,ratherthan nationalorigin, has been the crucial element in formulating critical third-worldperspectives. Of course, as subordinatedsubjects, Indian historians have obviously developed and embracedthe victim's subject-positionmore readily. But because the experience and are expression of subordination discursively formulated,we are led back to the processes and forces that organise the subordinate'ssubject-position.32 The difficulty here is that it is hard to see how this approach can have room for any theory about experience as the medium through which resistances emerge and are crystallised or about the conditions under which the subordinate can become active agents of their own emancipation on the basis of this experience. Some conception of experience and agency are absolutely required by the dispossessed's call for a politics of contest, for it is not clear how a
31 The wider and deeply conservativeimplicationsof post-Nietzscheanprojectsfor emancipation outside any frameworkof instrumentalreason are discussed in JurgenHabermas'sclassic article, "Modernityversus Postmoderity," New German Critique, 22 (Winter 1981). 32 Prakash, "WritingPost-Orientalist Histories," 402-3.

THE THIRD WORLD AFTER ORIENTALISM

153

dispersed effect of power relations can at the same time be an agent whose experience and reflectionform the basis of a strivingfor change. To arguethat we need these categories in some form does not at all imply a returnto the liberal humanundifferentiatedand static conceptions of nineteenth-century ism. Our present challenge lies precisely in understandinghow the underclasses we wish to study are at once constructedin conflictual ways as subjects yet also find the means throughstruggleto realizethemselves in coherent and subjectively centred ways as agents.33 The question of historical understandingis still more crucial. As Fredric Jamesonand Andreas Huyssen have argued, and we have tried in a different way to suggest above, postmodernistapproachesdesperatelylack a sense of and through history, a capacity for that labourof remembrance understanding which agents become able to experience history in an active way, to orient themselves individuallyand collectively in the present, and so to act. Indeed, this capacity must lie at the very centre of what Prakashand many otherscall for-in the recovery of frozen and silenced histories as part of a conscious relationsof domination,as political strategydesigned to engage contemporary these have affected third-worldsocieties. The problem, though, is that it is extremely difficult to see how we can actuallyhave a postmodernperspective which possesses any kind of strong historical sense. On present definitions, the two would seem to be a ratherstrong contradictionin terms. What distinguishes the former is precisely its sense of depthlessness, of the past's disassembly into a vast collection of images and fragmentsavailable in the present only for the purposes of nostalgia or pastiche.34 Whilst acknowledging the extent to which he and others have drawn on these perspectives, Prakash certainly emphasises the very significant differences in their approachesto issues of politics and power. The concerns of postmodernismhave in the end been differentin two ways. First, they have of tendedto take pleasurein a Bakhtinianproliferation voices for its own sake and in a way more aesthetic than political. Second, their own efforts to fragment Western proceduresof representationrun the risk of using thirdworld voices and culturesmerely as others. Yet Prakashdoes not really tell us how his more politically engaged stance is substantiallydifferent from the politics of postmodernism.In fact, it is striking how much the two have in common. Both are caught between the critiqueof objectivist forms of representationon the one hand and what becomes a slide towardsself-representation on the other. Likewise, postfoundationalhistory tries to dissolve the
33 This question of how we might conceptualise the presence of the subalternis discussed furtherand with differentemphases in R. O'Hanlon, "Recoveringthe Subject:SubalternStudies and Histories of Resistance in Colonial South Asia," ModernAsian Studies, 22:1 (1988), 218. 34 For these arguments in Jameson and Huyssen, see Jameson, "Postmodernism,or the CulturalLogic of Late Capitalism,"especially pp. 64-71; and AndreasHuyssen, Afterthe Great Divide: Modernism, Mass Culture and Postmodernism(London: Macmillan, 1988).

154

ROSALIND

O'HANLON,

DAVID WASHBROOK

concepts of experience and identity and to question the use of any historical Like postcategory "which resists furtherdecompositioninto heterogeneity." moder theory itself, this tends to inhibit ratherthan to promote an active politics. Ironically, in fact, not all feminist and black criticism, which Prakash would draw into alliance, is actually so hostile to founding categories or concepts of experience, identity,or political agency. Withinfeminist criticism there is, of course, an immensely wide rangeof positions and approaches; but as Denise Riley has argued, if feminism abandonsthe categoryof women and the propositionthat they have a differenthistory,it dissolves its own subject. Although feminists contend strongly amongst themselves as to whether the concept of woman constitutes a universalcategory, they must for some purposes and at some levels continue to act as if such a category indeed exists, precisely for the reasonthatthe world continuesto behaveand treatwomen as though one does.35 Not all feminists have foreclosed on questionsof agency, experience, and identity. Both feminism and postmodernismstrive to reveal the implication of many forms of knowledge in power, but many feminists argue that they cannot limit themselves to dissection or to the fundamental cultural relativism that underlies postmodernism'srefusal to do more than proliferatedeconstructivequestions. Showing how certain kinds of knowlitself edge are privilegeddoes not in itself change very much. Postmodernism cannot provide a theory for or make the move to agency, precisely because it regardsall knowledge as taintedand complicit. Because its ultimateconcern is with real social change feminism can and must make this move, which also keeps open the possibility that there may be some forms of knowledge which are emancipatoryratherthan tainted and complicit and which are measured against their usefulness for feminist purposesratherthan againstthe inverted positivist standardsof postmodernistepistemology. Likewise, questions of experience and identityremainopen ones for many feminists. In the Western tradition,as LindaHutcheonsuggests, women have not been identifiedhistorically with origins, authority,or ego. On the contrary,they envisage themas selves as lacking these attributesalready.Their task must be to reconstruct well as question concepts of self and experience, for as emphasised above, political action becomes impossible if women as subjectssee themselves and their experience only in terms of dispersal.36 If feminists have made these differences very clear, so too have at least some critics writingfrom otherminoritybackgrounds,certainlysome of those to which Prakashrefers. In an article on these minority discourses in their relation to the Western intellectual tradition and its academic institutions,
35 Denise Riley, Am I ThatName? Feminismand the Category of 'Women'in History (London: Macmillan, 1988), 112-4. 36 LindaHutcheon,ThePolitics of Postmodernism (London:Routledge, 1989), 39 and 167-8.

THE THIRD WORLD AFTER ORIENTALISM

155

and Abdul JanMohamed David Lloyd do not hesitateto use privilegedcategories or totalizing forms of analysis. For them, the problems of minority intellectualsspring "as inevitablyfrom the modes of late capitalistsociety as do the systematic exploitationof the less privileged minoritygroups and the feminization of poverty."37They are very clear, moreover, that for all the importanceof changes at the level of discourse, emancipationdepends ultimately on "radical transformationsof the material structuresof exploitation."38The questionof identityalso remainsan open one, significantonly in the end for issues of practice and struggle. Fragmentedidentity a is for minorities givenof theirsocialexistence.Butas sucha givenit is notyet by and liberation not which anymeansanindexof liberation, evenof thatformal abstract from in is all thatpoststructuralism, itself and disarticulated any actualprocessof of remains sign the the couldoffer.On thecontrary, non-identity minorities struggle, of material damage,to which the only coherentresponseis struggle,not ironic distanciation.39
EDWARD SAID: PROBLEMS OF A PARADIGM

That Prakash'sposition should be so shot throughwith inconsistencies is in He some senses understandable. takes his definitions and many of his premises from Said, whose text also has many of these same contradictions.It is worth returningto these aspects of Said's work, because Prakashis only one of a great numberof historianswho seem to us to have based themselves on Said's positions without attendingadequatelyto the problems in them. It is well known thatSaid drawsheavily on a rangeof Foucauldian perspectives, both for the analysis of Orientalismas a form of discourseand for his own repudiationof Europe's "universalisinghistoricism." He brings these themes together to press home one of his centralarguments:Orientalistconstructionsare not merely inaccurate,biased, or in need of replacingwith more is representation adequateones. Rather,Orientalismas a style of authoritative of an epistemology and an intellectualtraditionin itself the tainted product which "the one human history uniting humanityeither culminatedin or was observed from the vantage point of Europe."40 Said's continuingcommitmentat other levels both to conventionalhumanand to an implicitly universalistdiscourse of ist techniques of representation freedom is often less well appreciated.Despite his criticismof Orientalismas a style of representation,he makes it clear that his concern is not to reject the Knowledge for Said clearpossibility of any kind of objective representation. is not just the endlessly self-referentialproduct of all-pervasive power ly
37 Abdul R. JanMohamed and David Lloyd, "Introduction: MinorityDiscourse-What is to Be Done?," Cultural Critique, Fall (1987), 12. 38 Ibid., 15. 39 Ibid., 16. 40 Edward Said, "Orientalism Reconsidered," in Francis Barker et al., eds., Literature, Politics and Theory (London: Methuen, 1986), 223.

156

ROSALIND

O'HANLON,

DAVID WASHBROOK

relations. On the contrary,his interestlies in developing forms of representation and knowledge which are emancipatoryin their effects and which can serve as a basis for active political commitmentand intervention.As he says, unless intellectuals are interestedin changing political relations, in dismantling systems of dominationas well as defining them, the critiqueof Orientalism is merely "an ephemeralpastime."4' He sees any worthwhilecultural criticism as "constitutionallyopposed to every form of tyranny,domination, and abuse; its social goals are non-coercive knowledge producedin the interests of humanfreedom."42This pursuitof criticism's active emancipatory potential is "a fundamentalhuman and intellectualobligation."43He differs sharplyhere from Derridaand Foucault, whom he sees as having abandoned the critic's propertask of an engagement which is ultimatelypolitical in its culture. Derridaelected of naturewith the dominantstructures contemporary illustratewhat is undecidablewithin texts, ratherthan to investigate their to worldly power;and Foucaultforgot thatultimately "the fascinateddescription of exercised power is never a substitutefor trying to change power relations within society."44 Said also reserves a place and a significance for individual agents and individual experience in the shaping of Orientalistdiscourse: "Yet unlike Michel Foucault, to whose work I am greatly indebted, I do believe in the determiningimprintof individualwritersupon the otherwiseanonymouscollective body of texts constitutinga discursive formationlike Orientalism."45 This position is wildly at odds with Foucault'sown unremittingattemptsto fragment these categories on the grounds of their humanistand essentialist character.In contrast, Said refers to his own and similarprojectsas humanist in a broadsense and in an interviewin 1986 referred very explicitlyboth to the contradictions in his own position and to his radical disagreement with and power: "Orientalismis theFoucauldianperspectives on representation inconsistent, and I designed it that way: I didn't want Foucault's oretically method, or anybody's method, to overridewhat I was trying to put forward. The notion of a non-coercive knowledge, which I come to at the end of the book, was deliberatelyanti-Foucault."46 How, then, is the critic to go about the universalmoraland political tasks, which Said commends, without appearingto invoke the tainted authorityof Europeanor any other single and dominatingintellectualtradition?He notes that a whole range of intellectual projects, just like his own, have already
41 42 43 44 45
46

Ibid., 229.

Said, The World, 29.

Ibid., 30. Ibid., 222.

Said, Orientalism(London: PeregrineBooks, 1985), 23. See the interview with Said in Imre Salusinszky,Criticism in Society (London:Methuen, 1987), 137.

THE

THIRD

WORLD

AFTER

ORIENTALISM

157

begun to breakup old objects of knowledge ruledby Orientalismand to form new fields of investigation. These projects are local and self-convicted but form a common endeavour.Their methods deliberatelyavoid totalizing and systematizing;rather,they strive consciously to be secular, marginal,oppositional. They work out of a decenteredconsciousness, intending the end of dominating, coercive systems of knowledge; but they do not seek common methodologicalconsistenunity by appealsto any kind of sovereign authority, cy, canonicity, science.47 The point aboutconsistency is certainlytrue, for what comes out of all this is a very strainedand contradictory position. Said recommendsthat we abandon totalizationand systematizationin favourof the off-centreandthe marginal. But what view could have been more centrallyfocussed and systematising than that which he presentedin Orientalism?What gave the latter its power was precisely its ability to reinterpret,within a single analyticalframework, core elements in the Europeanintellectual and political traditionfor a very them in ways that obscured internal long period and, indeed, to reinterpret relations of contestationand resistance in Westerncultures. If Said had followed his own injunctions,now echoed in Prakash,Orientalismwould never have been written, with much loss to the whole scholarlycommunity.Again, Said advocates humanist values and a set of universal moral imperatives regardingpolitics and human freedom, the fundamentalobligations of intellectuals, the proper role of culturalcriticism. But how are these strong and central normativethemes reconciled with the secular and marginalposition, the extremerelativist "pluralityof terrains,multipleexperiencesand different constituencies" which Said commends elsewhere?48Ambiguity also marks He Said's position on representation. repudiatesthe view thatonly women can write aboutwomen, blacks aboutblacks, thatonly criticismwhich treatsthem well is good criticism. But as he himself says, the kind of local and selfcommittedintellectualprojectshe commendsare always in dangerof slipping into a kind of "possessive exclusivism," which holds that the only valid kind of of representationis the self-representation insiders.49 it is true that such contradictionscan be very fruitful, particuOf course, larly in hands as deft as Said's. But their fruitfulnesslies surely in prompting us to recognize and go beyond them. Moreover,theredo seem to be levels in Said's wider position at which creative tensions begin to look like submerged self-contradictions.This was perhapsmost interestinglyso, for our purposes, in what he said early in 1989 duringthe battles over SalmanRushdie'swork. Rushdie's "fundamentalrights" should be protected, Said argued, because the contemporaryworld, for all its particularities,must be regardedas one
47 Said, "OrientalismReconsidered,"228. 48 Idem. 49 Ibid., 229.

158

ROSALIND O'HANLON,

DAVID WASHBROOK

world and human history as one history. (But not, to paraphrase earlier his remarks,a humanhistory seen from Europe'svantagepoint.) This meantthat there was no pure unsullied essence to which Muslims or anyone else could this return; single world was irredeemably heterogeneous,and Rushdie'swork was a part of that. At the same time, one featureof his work that made it legitimate was that "Rushdie, from the communityof Islam, has writtenfor the WestaboutIslam. TheSatanic Versesis thus a self-representation."50 This its brings Said very close to what he rejectedearlieraboutself-representation: tendency merely to invertthe essential categoriesof Orientalism.It is simply very difficult to combine argumentsconcerningfundamental rights and possibilities for emancipationwith a postmodernist refusalof any kind of unitary or systematizingperspectiveas to what these rights might be or what emancipation is from or into. Consequently,rights, dominance, and emancipation are defined only from the extreme relativist perspective of the multifarious struggles of oppositional groups. And when one version of emancipation conflicts with another,the naturaldefence for both becomes the principleof as self-representation such.
HISTORICIZING POSTMODERNISM? ON A LIBERAL CULTURE PERSPECTIVES

Why, then, have these perspectives achieved such widespreadpopularityin American, academiccircles? There is now, of course, a Western,particularly and influential body of postmodernist writing in history and anlarge thropology,mostly publishedin the UnitedStates.5' This writingdoes notjust and embracepostmodernist poststructuralist strategiespartiallyand contradicas Said and Prakashdo but advocatesthem wholeheartedlyas the very torily means to fashion new possibilities for writing and representationin a here, postcolonial world. There have been a range of prominentcontributors but perhapsthe most influentialhas been JamesClifford,both in the collection edited with George Marcusin 1986, WritingCulture,andhis own more recent
volume of essays, The Predicament of Culture.52 We would like to turn now

of to look at Clifford's more thoroughgoingrecommendation postmodernist to discuss what we see to be its extremelyconservativepolitical perspectives, implications, implications which Prakashcannot logically disassociate himself from.
50 This short article appearedin the Observernewspaper(26 February1989, 14). 51 Useful introductionsto this literatureare AndreasHuyssen, After the Great Divide: Modernism, Mass Cultureand Postmodernism;and D. Kellner,ed., Postmodernism,Jameson, Critique (Washington,D.C.: MaisonneuvePress, 1989). 52 James Clifford and George E. Marcus, eds., WritingCulture:The Poetics and Politics of Ethnography(Berkeley: University of CaliforniaPress, 1986); JamesClifford, The Predicament of Culture: Twentieth-Century Ethnography,Literature, and Art (Cambridge,Mass.: Harvard University Press, 1988).

THE THIRD WORLD AFTER ORIENTALISM

I59

Clifford himself notes that Said remains "ambivalentlyenmeshed in the totalizing habits of Western humanism."53 For him, the relativist and features of Said's work make it important; humanistand its poststructuralist universalistelements are merely an unfortunatehangoverfrom an outmoded intellectual tradition: the privilege standing of abovecultural of to particularism, appealing the universalist for for of etc., is powerthatspeaks humanity, universal experiences love, work,death, a privilegeinvented totalizing Western liberalism.54 by Clifford's critique of Said flows out of a set of clear postmodernistand commitments.New possibilities for postcolonialethnography poststructuralist are best opened up througha rejectionof all universalforms of understanding culture or the past. Ethnographyshould focus instead on the ways in which cultures, as forms of "collectively constituteddifference," are in a constant process of local invention, carriedout in relationto recent colonial histories and new national identities.55 In this mobile postcolonial world, in which exotic others returnthe ethnographer's gaze, new ways must also be found of talking about relations between cultureswhich emphasise that these are relationships of power. This does not mean, however, that we can devise new theories about global homogenization or the transformation postcolonial of societies in the image of Europe. Certainly, Clifford concedes, there are increasingly pervasive processes of economic and cultural centralisationat work. But these do not tell the whole or the only story. What emerges constantly at the level of local societies are new and inventive ordersof cultural difference and of subversion, mockery, syncretism and revival, which challenge all efforts to construct any single master narrativeof global historical change: "Indeed, modern ethnographichistories are perhapscondemned to oscillate between two metanarratives: one of homogenization, the other of one of loss, the other of invention."56Here, then, postmodernist emergence; hostility to any kind of universal history, and what is in effect a position of extreme cultural relativism, feed into and reinforce one another. From this perspective, one can see why Cliffordis anxiousto hold on to some conceptof culture itself, for its "differentialand relativist"functions are precisely what is important.57 Whatwe thereforeneed, he argues, are new ways of constructand authorisingknowledge about others. Insteadof the ethnographer as ing the privileged purveyorof such knowledge, we must learnto envisage a world of generalized ethnographyand texts which are franklythe productof many
53 Clifford, The Predicament of Culture, 271. 54 Ibid., 263.

55 Ibid., 274. For a good summaryof the argumentsaboutcultureas collectively constituted, see Roger M. Keesing, "Anthropologyas Interpretative Quest," CurrentAnthropology, 28:2 (April 1987). 56 Clifford, The Predicament of Culture, 17.
57 Ibid.,

274.

I60

ROSALIND

O'HANLON,

DAVID WASHBROOK

voices. This means going beyond methods which make the writer into an omniscient authorityand spokesman, which screen off the whole business of researchand writing, and which deal with abstractcollectivities and typifying processes, such as "the Nuer think .. ." It means having ethnographies which are open about their status as "a constructivenegotiationinvolving at least two, and usually more, conscious, politically significant subjects."58 These new dialogical approachesnot only strive to create texts which are an open-ended interplayof many voices, along the lines that Mikhail Bakhtin envisaged. They also seek to returncontrol over knowledge to its indigenous sources, to representadequatelythe authorityof informants,and to open real textual spaces for a multitudeof indigenous voices whose perspectives and agendas are not imposed on them from outside: "If accordedan autonomous textual space, transcribedat sufficient length, indigenous statements make sense in terms differentfrom those of the arrangingethnographer. Ethnography is invaded by heteroglossia."59 Although these aims are in some senses still utopian, Cliffordpoints to a rangeof recent studies that have triedto accordto particularly knowledgeable or sophisticated informantsthe status not merely "of independentenunciators, but of writers."60Anthropologistswriting from this perspective "have described the indigenous 'ethnographers'with whom they shared, to some degree, a distanced, analytic, even ironic view of custom. These individuals became valued informantsbecause they understood,often with real subtlety, what an ethnographic attitude toward culture entailed."61In this way, anthropology has been able not only to move towards a world of plural auin participation the actualinventionof thorshipbut to recognizeethnography's culture, as in the collectively producedstudy,Piman Shamanismand Staying Donald Bahr, appearson the title page with three Sickness. The ethnographer, otherauthors,who are PapagoIndians.The book is intended"to transferto a shaman as many as possible of the functions normallyassociated with auThe thorship."62 shaman,Gregorio, is thus the main sourcefor the "theoryof disease" describedin the book. The audiencesto which the book is addressed are also multiple. Gregorio's commentariesare in Piman, with translations David Lopez; and the linguist, Albert Alvarez; and made by the interpreter, Thus the book not only keeps accompaniedby Bahr's own interpretations. distinct the contributionsof each but provides materialfor qualified Papagos as well as for Westernaudiences. Indeed, Alvarez himself designed the trans58 Ibid., 41. 59 Ibid., 51. 60 Idem. 61 Ibid., 49. 62 Ibid., 51.

THE

THIRD

WORLD

AFTER

ORIENTALISM

I6I

lations so that the,book could be used in languageteaching, thus contributing to the developmentof Piman as a writtenlanguage: "Thus the book contributes to the Papagos' literaryinvention of their culture."63 What, then, are the broaderimplicationsof this approach? Certainly,issues of power are taken to be central to the relation between ethnographerand and writer-informant; a very large effortis made to change the termson which they conducttheirexchanges. However, we need to look more closely at these terms of exchange and to ask how far they manage to avoid the problems identifiedearlier.We would like to arguenot only that these problemsare not avoided, but that there is actually anotherand much more disturbingpolitical logic in these argumentsas presentedby Clifford. is Here certainly, the principle of self-representation pushed to its logical of conclusion, which is the self-representation individuals. This is precisely what is implied in the new dialogical approachto ethnographythat Clifford and others advocate as the means to supercedeolder styles of representation, with their questionable assumptions about authorship,their typifying procedures, and their referencesto abstractcollectivities. If we are not to employ the latter,indeed, it certainlyis very difficultto see what othercategoriesand accounts ethnographerscould work with except for direct indigenous statements, quotations and translations, such as those of Gregorio the shaman, who have a sophisticatedknowledge of the culture and an understanding of what a properlyethnographicattitudeentails. But because it privileges only the voices of authoritativeindigenous individuals, this approachpresents a clear problem. It is hard to see how such an approachcan recognise or give adequate place to conflict within social contexts thus examined or to those groups or communities who may dissent very strongly from these authoritativeaccounts. It is not clear how such relationshipsof power are discussed at all if the analytical means of abstractionand typification are eschewed in favourof a dialogue between individuals. Indeed, the strategies proposed here look disturbinglysimilar to those of East Indian Company officials, who also thoughtof cultureas "collectively constituteddifference"in early colonial India. When they wished to elucidate the major principles of what they assumed to be a composite Hindu culture, they turned to the Brahman pandits who were deemed to be experts and authoritiesin the matter.The resultof this privilegingof particular informants was the longer-termemergence of an all-IndiaHindu traditionvery much in the image of Brahmanicreligious values. These values, now embodied in written legal codes and disseminated in a wide range of social contexts, graduallyeroded what had previously been a much more heterogeneouscol63 Ibid., 52.

I62

ROSALIND

O'HANLON,

DAVID WASHBROOK

lection of local social and religious practices.64Given the great play that Clifford and others make with their vigorous repudiationof all legacies of colonialism, one would have thought that an especial target of their attack would have been precisely this sort of colonial effort to establish dominance through the textualisationof cultures in collaborationwith carefully chosen that indigenousauthorities.But this is just the kind of intervention he seems to recommendin the example of the jointly producedbook on Papagoculture, in which the shaman Gregorio's translatedaccounts were designed in part to contributeto "the Papagos' literaryinvention of their culture." Postmodernismsupposedly distinguishes this kind of collaborationfrom colonial strategies, of course, with the argument that ethnographic consciousness is now no longer the monopolyof Westernspecialistsbut is shared with a whole rangeof indigenousaudienceswho will scrutinizeethnographic texts and decode them in their own ways. Indigenous as well as Western on voices are now free to negotiate and contest such representations what has become a world-wideculturalstage. Local culturesconstantlyreinventthemselves within and against these new circumstancesof global relationality. Their stories are different.They continuallyundercutand forbidthe construction of any single or totalising narrative. To question these basic suppositions is not to deny that indigenous audiences are sharply alive to the political consequences of novel culturalinterpretationsand interventions.The disseminationof Brahmanicalreligious India and values was consciously and bitterlycontested in nineteenth-century continues to be fought by ratherdifferentgroups at present. But it is quite a different thing to posit, as Clifford appears to here, a shared ethnographic in of consciousness, a common participation the textualization culturesand in what he calls the "distanced, analytic, even ironic view of custom" that ethnographicconsciousness entails.65 Most obvious, it seems unlikely that those amongst indigenous audiences who are neitherpowerholdersnor specialist purveyorsof knowledge will be able to afforda detachedor abstracted view of custom, particularlywhen its terms are being reinterpretedfrom outside as well as from above. Even within the termsof a dialogicalapproach, which focusses much more narrowly on exchanges between ethnographers it and their selected writer-collaborators, is hardto see how we can speak of a
64 Historians have documented this process across a range of fields. See, for example, L. Mani, "ContentiousTraditions:The Debate on Sati in Colonial India," CulturalCritique, Fall (1987); D. Washbrook, "Law, State and AgrarianSociety in Colonial India," Modern Asian Studies, 15:3 1981; R. O'Hanlon, "Cultures of Rule, Communities of Resistance: Gender, Social Analysis, no. 25 (SepDiscourse and Traditionin Recent South Asian Historiographies," tember 1989); C. Bayly, Indian Society and the Makingof the BritishEmpire, 136-68; N. Dirks, "The Invention of Caste: Civil Society in Colonial India," Social Analysis, no. 5 (September 1989); Lucy Carroll, "Law, Custom and StatutarySocial Reform:The Hindu Widows' Remarriage Act of 1856," Indian Economic and Social History Review, 20:4 (1983). 65 Clifford, The Predicamentof Culture, 49.

THE THIRD WORLD AFTER ORIENTALISM

163

terms. The issue is not dialogue or negotiationwhich both shareon near-equal the problemof a text's internalcomposition, which is the chief concern simply of dialogical approaches. It is also, as Bob Scholte has argued, that ethnographic texts are subject to external as well as to internal relations of production, which include a professional academic apparatusof seminars, lectures and conferences, funding bodies, researchcouncils and committees of appointment.66 would be very difficult to deny that this intellectualand It institutional apparatushelps set to a considerable extent the agendas and take with them into the field and that framingquestions which ethnographers it also exerts a large control in shaping professional standards, styles of writing, and access to publication;in awardingrecognition and conferring academic authority;and in approvingand financing furtherresearch. Local writer-collaborators may indeed have long-lasting and intimate connections with individualethnographers. is much less clear what access and influence It they, let alone wider and less privileged indigenous audiences, are able to command in these complex externalcontexts of a text's production. blindness. As we have seen, postmodernist This is an extraordinary writing concern is with relationships in this field repeatedlyinsists that its paramount of power and the immersionof all knowledge withinthem. But this apparently applies to all knowledge and to all forms of historical and social belonging except the postmoderncritic's own. In many ways, such a position is entirely consistent with postmodernism'sbroaderpremises, which deny possibilities and experience in favour of myfor an active historical self-understanding thified and fabulized stories which melt our sense of the past's solidity. They refuse to equip themselves for any kind of wider historical or sociological vision, for to do so would need the rangeof analyticaltools thatboth Clifford and Prakashask us to echew: privileged categories which "occlude" other histories, abstractcollectivities and typifying processes, totalizing and sysWhat follows from this, in termsof posttematizingforms of understanding. modernism'srefusalto examine its own historicalprovenance,may be consistent; but it is none the less disconcerting. It bears a strangeresemblanceto colonial strategies of knowledge, which notoriouslyregardedall indigenous identities and relationsas properobjects for investigation(in consultation,of course, with properindigenousauthorities)whilst veiling its own historyfrom scrutiny. If, as Cliffordsees it, indigenouswritersnow virtuallydefine and represent themselves throughethnographictexts, so too do local culturesthemselves in these new global relationships.In view of postmodernism's hostility to totalities, of course, it is somewhatdifficult to hold onto any concept of a culture as such. The way aroundthis, which Cliffordtakes, is to suggest that cultures
66 Bob Scholte, "The Literary Turn in ContemporaryAnthropology," Critique of Anthropology, 7:1, 38.

164

ROSALIND O'HANLON,

DAVID WASHBROOK

may not actually be totalities at all but "mobile ensembles" that constantly reinvent themselves, tell their own stories, and create their own variantson global political relationships. We end up with still a totality but one conceived, like postcolonialsubjectsthemselves, in extremelyvolatile and voluntaristic ways. Postcolonial societies are free, it would seem almost, to reinvent global political and economic relationships at will. There are forces through which the world is becoming increasingly homogeneous, but we cannotaccept a unitaryor systematicanalysis of these changes. Ourstories of homogenization are in the end no different from their stories of local and different self-invention. What, then, are we to make of the apparent popularityof this combination of extreme culturalrelativismwith a liberal, almost individualistunderstanding of these postcolonialsocieties' abilityto define andcreatethemselves?For Prakash, as indeed for others who share his approaches,postmodernistperassault, issued along with a spectives help make possible a radical-sounding of cultures, declamatorypublic commitmentto the emancipation marginalised For on all existing frameworksof interpretation. Clifford,just as for Prakash, moder capitalism'sglobal spreadcan produceonly homogenization,just as any history focussing on the theme of capitalisttransitioncan recognize only homogeneityto the detrimentof otherand differenthistories. We see here the misconceptiondescribedabove, thatsystems can only generate postmodernist sameness. This makes it possible, within a culturedeeply antagonisticto any kind of materialisthistoricalexplanation,to dismiss suggestionsthatthe local differenceswe see emerging in postcolonialsocieties might have somethingat intrinsicto modem capitalism,since it least to do with logics of differentiation and in spite of such logics that these local cultures invent themis against selves. But the result brings us strangelyclose to the classic liberalview that culturerepresentssome realm of freedomand choice. Althoughwe can study larger forces of global economic centralizationand the coercions they exert, can culturalrelativism means that this metanarrative do no more than stand these very its opposite, that of local cultures'self-creation.Further, alongside commitmentsto culturalemancipationseem to displace most of the public who intellectualrisk onto writer-collaborators authorisetheir own representations, indigenous audienceswho decode texts in their own ways, and a range of national, ethnic and other marginalizedpeople who are made responsible theirown visions of emancipationand politfor theirown self-representation, ical struggles towards it. CliffordGeertz has identified some of the logics underlyingthis position. All these approaches(he calls them pretensions)try to "get roundthe un-getroundablefact that all ethnographicaldescriptionsare homemade, that they are the describer'sdescriptions, not those of the described."67Although the
67

CliffordGeertz, Worksand Lives: TheAnthropologistas Author(Cambridge: Polity Press,

1988), 144.

THE

THIRD

WORLD

AFTER

ORIENTALISM

165

business of representationhas become infinitely more complex in recent and years, althoughethnographers historiansare more sharplyawarethanever before of its acute moraland political difficulties, these cannotbe shifted onto those whose control over the productionof ethnographictexts is more apparent than real; nor can it be resolved throughtechnique: Theburden authorship of cannot evaded, be however there is heavyit mayhavegrown; no possibility displacing onto "method", of it or "language" (anespecially popular maneuver the moment)"thepeoplethemselves" at redescribed is ("appropriated" the probably betterterm)as co-authors.68 We would go rather furtherthan this. These postmodernistapproaches, particularlyClifford's, actually offer us an epistemology that denies that its own history can be seriously investigated and an analytical preoccupation with a very narrowlydefined set of individualrelationships.Effectively depoliticised by being insulated from their materialand institutionalcontexts, these relationships are presented as an arena in which indigenous collaboratorsand audiences are free, as it were, to invent and be themselves. Such effortsto sever off spheresof activity for free individualsor culturesare a very old device of liberal ideology. The British colonial record is full of them. If all this looks more like a device for legitimationthan any basis for an emancipatoryform of knowledge, what is being legitimized?Said, Huyssen, and others have made the point that French postmodernismand poststructuralismunderwenta peculiar metamorphosiswhen they were domesticated within American liberal culture from the early 1960s. Their rapidgrowth in popularityreflected the degree to which they were evisceratedof their earlier and radical political content by literary and culturalcritics, who converted them into forms of "writerlyconnoisseurismand textualgentrification."69 We see these intellectualpositions sustainingkey aspects of contemporary political culture in the United States. The first concerns the way in which the advance of argumentsaboutthe self-representation thirdworld peoples fits of neatly into its self-consciously multiminorityacademic culture. What marks debate here is, of course, a deep concern with multipleand conflictualidentities. Yet what is striking about these debates, particularlythose employing postmodernistperspectives, is how one particularidentity, that of class or materialrelations, is so often downplayedor screenedoff. Not only do participants in these debates frequently ignore questions of class, but they see themselves also as having to challenge the larger intellectual tradition of historical materialism that establishes those questions as central, on the grounds that its universalistand objectivist pretensionsare really no different to those of liberal modernizationtheory.One consequence of this is that selfdefined minority or subalterncritics are saved from doing what they constantly demand of others, which is to historicise the conditions of their own
68 69

Ibid., 140. Huyssen, Afier the Great Divide, 212; Said, The World, 3-5.

I66

ROSALIND O'HANLON,

DAVID WASHBROOK

emergence as authoritativevoices-conditions which could hardly be described without referenceof some kind to materialor class relations. At other levels, the exclusion of class and of the materialistcritique of capitalism from the agenda of scholarshiphas implicationsthat seem to us of absolutelycritical. Whatit means is thatthe trueunderclasses the world are kinds of only permittedto presentthemselves as victims of the particularistic gender, racial, and nationaloppressionwhich they sharewith preponderantly middle-classAmericanscholarsand critics, who would speak with or in their voices. What such underclassesare denied is the ability to presentthemselves as classes: as victims of the universalistic,systemic and materialdeprivations of capitalismwhich clearly separatethem off from their subalternexpositors. In sum, the deeply unfortunateresult of these radical postmodernistapproachesin the minoritiesdebateis thus to reinforceand to give new credence to the well-known hostility of American political culture to any kind of materialistor class analysis. These approachesalso seem to us to have had important wider implicaand tions in Americanpolitical and academicculture.Anotheranthropologist who employs them, Paul Rabinow,tells us engaginglythathe is "temperamentally more comfortablein an oppositionalstance."70The same seems to be trueof a wide range of currentacademic writing. There runs throughit a desire to be seen on the side of the dispossessed againstpower, workingwith their strange voices and different stories, subverting dominant cultures and intellectual traditions "from within the academy."But in the case of postmodernistapproaches, these commitments can be made with a lightened burden of authorshipand a comfortingsense thatin this volatile new worldof culturalselfinvention, the critic's own history is at best a fable. What all this begins to look very like, in fact, is a new form of that key and enduringfeature of Westerncapitalist and imperialistculture:the bad conscience of liberalism, still strugglingwith the continuingparadoxbetween an ideology of liberty at home and the realityof profoundlyexploitativepolitical relationsabroad,and now striving to salve and reequip itself in a postcolonial world with new But argumentsand bettercamouflagedforms of moralauthority. the solutions it offers-methodological individualism,the depoliticising insulationof social from material domains, a view of social relations that is in practice politics-do extremelyvoluntaristic,the refusalof any kind of programmatic not seem to us radical, subversive, or emancipatory. They are on the contrary as conservative and implicitly authoritarian, they were indeed when recommended more overtly in the heyday of Britain'sown imperialpower. Prakashhimself does not push these perspectivesto theirmost authoritarian conclusions and tries rightly to be critical of their depoliticising effects. But
Are Social Facts: Modernityand Post-Modernityin AnPaul Rabinow, "Representations thropology,"in Clifford and Marcus, WritingCulture, 258.
70

THE THIRD WORLD AFTER ORIENTALISM

167

since he sharesmany of their core assumptions,his effortsresultin ambiguity and contradiction.His is basically an attempt,like that of Said and of many others who try to use his position as a point of departure, ride two horses at to once. But one of these may not be a horse that brooks inconstantriders, and Said himself does at least seem to know which of them in the end he would ratherbe on.