Some of the most radical criticism coming out of the West today is the result of an interested desire to conserve the subject of the West, or the West as Subject. The theory of pluralised ‘subject-effects’ gives an illusion of undermining subjective sovereignty while often providing a cover for this subject of knowledge. Although the history of Europe as Subject is narrativised by the law, political economy and ideology of the West, this concealed Subject pretends it has ‘no geo-political determinations’. The much-publicised critique of the sovereign subject thus actually inaugurates a Subject.[1] will argue for this conclusion by considering a text by two great practitioners of the critique: ‘Intellectuals and power: a conversation between Michel Foucault and Gilles Deleuze’.

I have chosen this friendly exchange between two activist philosophers of history because it undoes the opposition between authoritative theoretical production and the unguarded practice of conversation, enabling one to glimpse the track of ideology. The participants in this conversation emphasise the most important contributions of French poststructuralist theory: first, that the networks of power/desire/interest are so heterogeneous, that their reduction to a coherent narrative is counterproductive - a persistent critique is needed; and second, that intellectuals must attempt to disclose and know the discourse of society’s Other. Yet the two systematically ignore the question of ideology and their own implication in intellectual and economic history.

Although one of its chief presuppositions is the critique of the sovereign subject, the conversation between Foucault and Deleuze is framed by two monolithic and anonymous subjects-in-revolution: ‘A Maoist’ (FD, p.205) and ‘the workers’ struggle’ (FD, p.217). Intellectuals, however, are named and differentiated; moreover, a Chinese Maoism is nowhere operative. Maoism here simply creates an aura of narrative specificity, which would be a harmless rhetorical banality were it not that the innocent appropriation of the proper name ‘Maoism’ for the eccentric phenomenon of French intellectual ‘Maoism’ and subsequent ‘New Philosophy’ symptomatically renders ‘Asia’ transparent.[2]

Deleuze’s reference to the workers’ struggle is equally problematic; it is obviously a genuflection: ‘We are unable to touch [power] in any point of its application without finding ourselves confronted by this diffuse mass, so that we are necessarily led ... to the desire to blow it up completely. Every partial revolutionary attack or defence is linked in this way to the workers’ struggle’ (FD, p.217). The apparent banality signals a disavowal. The statement ignores the international division of labour, a gesture that often marks poststructuralist political theory.[3] The invocation of the workers’ struggle Is baleful in its very innocence; it is incapable of dealing with global capitalism: the subject-production of worker and unemployed within nation-state ideologies in its Centre; the increasing subtraction of the working class in the Periphery from the realization of surplus value and thus from ‘humanistic’ training in consumerism; and the large-scale presence of paracapitalist labour as well as the heterogeneous structural status of agriculture in the Periphery. Ignoring the international division of labour; rendering ‘Asia’ (and on occasion ‘Africa’) transparent (unless the subject is ostensibly the ‘Third World’); re-establishing the legal subject of socialised capital - these are problems as common to much poststructuralist as to structuralist theory. Why should such occlusions be sanctioned in precisely those intellectuals who are our best prophets of heterogeneity and the Other?

The link to the workers’ struggle is located in the desire to blow up power at any point of its application. This site is apparently based on a simple valorisation of any desire destructive of any power. Walter Benjamin comments on Baudelaire’s comparable politics by way of quotations from Marx:

Marx continues in his description of the conspirateurs de profession as follows: ‘They have no other aim but the immediate one of overthrowing the existing government, and they profoundly despise the more theoretical enlightenment of the workers as to their class interests. Thus their anger - not proletarian but plebian - at the habits noirs (black coats), the more or less educated people who represent [vertreten] that side of the movement and of whom they can never become entirely independent, as they cannot of the official representatives [Repräsentanten] of the party.’ Baudelaire’s political insights do not go fundamentally beyond the insights of these professional conspirators... . He could perhaps have made Flaubert’s statement, ‘Of all of politics I understand only one thing: the revolt’, his own.[4]

The link to the workers’ struggle is located, simply, in desire. Elsewhere, Deleuze and Guattari have attempted an alternative definition of desire, revising the one offered by psychoanalysis: ‘Desire does not lack anything; it does not lack its object. It is, rather, the subject that is lacking desire, or desire that lacks a fixed subject; there is no fixed subject except by repression. Desire and its object are a unity: it is the machine, as a machine of a machine. Desire is machine, the object of desire also a connected machine, so that the product is lifted from the process of producing and something detaches itself from producing to product and gives a leftover to the vagabond, nomad subject.’[5]

This definition does not alter the specificity of the desiring subject (or leftover subject-effect) that attaches to specific instances of desire or to production of the desiring machine. Moreover, when the connection between desire and the subject is taken as irrelevant or merely reversed, the subject-effect that surreptitiously emerges is much like the generalised ideological subject of the theorist. This may be the legal subject of socialised capital, neither labour nor management, holding a ‘strong’ passport, using a ‘strong’ or ‘hard’ currency, with supposedly unquestioned access to due process. It is certainly not the desiring subject as Other.

The failure of Deleuze and Guattari to consider the relations between desire, power and subjectivity renders them incapable of articulating a theory of interests. In this context, their indifference to ideology (a theory of which is necessary for an understanding of interests) is striking but consistent. Foucault’s commitment to ,genealogical’ speculation prevents him from locating, in ‘great names’ like Marx and Freud, watersheds in some continuous stream of intellectual history.6 This commitment has created an unfortunate resistance in Foucault’s work to ‘mere’ ideological critique. Western speculations on the ideological reproduction of social relations belong to that mainstream, and it is within this tradition that Althusser writes: ‘The reproduction of labour power requires not only a reproduction of its skills, but also at the same time, a reproduction of its submission to the ruling ideology for the workers, and a reproduction of the ability to manipulate the ruling ideology correctly for the agents of exploitation and repression, so that they, too, will provide for the domination of the ruling class “in and by words” [par la Parole].’[7]

When Foucault considers the pervasive heterogeneity of power, he does not ignore the immense institutional heterogeneity that Althusser here attempts to schematise. Similarly, in speaking of alliances and systems of signs, the state and war-machines (mille plateaux), Deleuze and Guattari are opening up that very field. Foucault cannot, however, admit that a developed theory of ideology recognises its own material production in institutionality, as well as in the ‘effective instruments for the formation and accumulation of knowledge’ (PK, p.102). Because these philosophers seem obliged to reject all arguments naming the concept of ideology as only schematic rather than textual, they are equally obliged to produce a mechanically schematic opposition between interest and desire. Thus they align themselves with bourgeois sociologists who fill the place of ideology with a continuistic ‘unconscious’ or a parasubjective ‘culture’. The mechanical relation between desire and interest is clear in such sentences as: ‘We never desire against our interests, because interest always follows and finds itself where desire has placed it’ (FD, p.215). An undifferentiated desire is the agent, and power slips in to create the effects of desire: ‘power ... produces positive effects at the level of desire - and also at the level of knowledge’ (PK, p.59).

This parasubjective matrix, cross-hatched with heterogeneity, ushers in the unnamed Subject, at least for those intellectual workers influenced by the new hegemony of desire. The race for ‘the last instance’ is now between economics and power. Because desire is tacitly defined on an orthodox model, it is unitarily opposed to ‘being deceived’. Ideology as ‘false consciousness’ (being deceived) has been called into question by Althusser. Even Reich implied notions of collective will rather than a dichotomy of deception and undeceived desire: ‘We must accept the scream of Reich: no, the masses were not deceived; at a particular moment, they actually desired a fascist regime’ (FD, p.215).

These philosophers will not entertain the thought of constitutive contradiction that is where they admittedly part company from the Left. In the name of desire, they reintroduce the undivided subject into the discourse of power. Foucault often seems to conflate ‘individual’ and ‘subject’;[8] and the impact on his own metaphors is perhaps intensified in his followers. Because of the power of the word ‘power’, Foucault admits to using the ‘metaphor of the point which progressively irradiates its surroundings’. Such slips become the rule rather than the exception in less careful hands. And that radiating point, animating an effectively heliocentric discourse, fills the empty place of the agent with the historical sun of theory, the Subject of Europe.[9]

Foucault articulates another corollary of the disavowal of the role of ideology in reproducing the social relations of production: an unquestioned valorisation of the oppressed as subject, the ‘object being’, as Deleuze admiringly remarks, ‘to establish conditions where the prisoners themselves would be able to speak’. Foucault adds that ‘the masses know perfectly well, clearly’ - once again the thematic of being undeceived - ‘they know far better than [the intellectual] and they certainly say it very well’ (FD, pp.206, 207).

What happens to the critique of the sovereign subject in these pronouncements? The limits of this representationalist realism are reached with Deleuze: ‘Reality is what actually happens in a factory, in a school, in barracks, in a prison, in a police station’ (FD, p.212). This foreclosing of the necessity of the difficult task of counter-hegemonic ideological production has not been salutary. It has helped positivist empiricism - the justifying foundation of advanced capitalist neo-colonialism - to define its own arena as ‘concrete experience’, ‘what actually happens’. Indeed, the concrete experience that is the guarantor of the political appeal of prisoners, soldiers and schoolchildren is disclosed through the concrete experience of the intellectual, the one who diagnoses the episteme.[10] Neither Deleuze nor Foucault seems aware that the intellectual within socialised capital, brandishing concrete experience, can help consolidate the international division of labour.

The unrecognised contradiction within a position that valorises the concrete experience of the oppressed, while being so uncritical about the historical role of the intellectual, is maintained by a verbal slippage. Thus Deleuze makes this remarkable pronouncement: ‘A theory is like a box of tools. Nothing to do with the signifier’ (FD, p.208). Considering that the verbalism of the theoretical world and its access to any world defined against it as ‘practical’ is irreducible, such a declaration helps only the intellectual anxious to prove that intellectual labour is just like manual labour. It is when signifiers are left to look after themselves that verbal slippages happen. The signifier ‘representation’ is a case in point. In the same dismissive tone that severs theory’s link to the signifier, Deleuze declares, ‘There is no more representation; there’s nothing but action’ - ‘action of theory and action of practice which relate to each other as relays and form networks’ (FD, pp.206-7). Yet an important point is being made here: the production of theory is also a practice; the opposition between abstract ‘pure’ theory and concrete ‘applied’ practice is too quick and easy.[11]

If this is, indeed, Deleuze’s argument, his articulation of it is problematic. Two senses of representation are being run together: representation as ‘speaking for’, as in politics, and representation as ‘re-presentation’, as in art or philosophy. Since theory is also only ‘action’, the theoretician does not represent (speak for) the oppressed group. Indeed, the subject is not seen as a representative consciousness (one re-presenting reality adequately). These two senses of representation - within state formation and the law, on the one hand, and in subject-predication, on the other - are related but irreducibly discontinuous. To cover over the discontinuity with an analogy that is presented as a proof reflects again a paradoxical subject-privileging.[12] Because ‘the person who speaks and acts ... is always a multiplicity’, no ‘theorizing intellectual ... [or] party or ... union’ can represent ‘those who act and struggle’ (FD, p.206). Are those who act and struggle mute, as opposed to those who act and speak (FD, p.206)? These immense problems are buried in the differences between the ‘same’ words: consciousness and conscience (both conscience in French), representation and re-presentation. The critique of ideological subject-constitution within state formations and systems of political economy can now be effaced, as can the active theoretical practice of the ‘transformation of consciousness’. The banality of leftist intellectuals’ lists of self-knowing, politically canny subalterns stands revealed; representing them, the intellectuals represent themselves as transparent.

If such a critique and such a project are not to be given up, the shifting distinctions between representation within the state and political economy, on the one hand, and within the theory of the Subject, on the other, must not be obliterated. Let us consider the play of vertreten (‘represent’ in the first sense) and darstellen (‘represent’ in the second sense) in a famous passage in The Eighteenth Brumaire of Louis Bonaparte, where Marx touches on ‘class’ as a descriptive and transformative concept in a manner somewhat more complex than Althusser’s distinction between class instinct and class position would allow.

Marx’s contention here is that the descriptive definition of a class can be a differential one - its cutting off and difference from all other classes: ‘In so far as millions of families live under economic conditions of existence that cut off their mode of life, their interest, and their formation from those of the other classes and place them in inimical confrontation [feindlich gegenüberstellen], they form a class’.[13] There is no such thing as a ‘class instinct’ at work here. In fact, the collectivity of familial existence, which might be considered the arena of ‘Instinct’, is discontinuous with, though operated by, the differential isolation of classes. In this context, one far more pertinent to the France of the 1970s than it can be to the international periphery, the formation of a class is artificial and economic, and the economic agency or interest is impersonal because it is systematic and heterogeneous. This agency or interest is tied to the Hegelian critique of the individual subject, for it marks the subject’s empty place in that process without a subject which is history and political economy. Here the capitalist is defined as ‘the conscious bearer [Träger] of the limitless movement of capital’.[14] My point is that Marx is not working to create an undivided subject where desire and interest coincide. Class consciousness does not operate toward that goal. Both in the economic area (capitalist) and in the political (world-historical agent), Marx is obliged to construct models of a divided and dislocated subject whose parts are not continuous or coherent with each other. A celebrated passage like the description of capital as the Faustian monster brings this home vividly.[15]

The following passage, continuing the quotation from The Eighteenth Brumaire, is also working on the structural principle of a dispersed and dislocated class subject: the (absent collective) consciousness of the small peasant proprietor class finds its ‘bearer’ in a ‘representative’ who appears to work in another’s interest. The word ‘representative’ here is not ‘darstellen’; this sharpens the contrast Foucault and Deleuze slide over, the contrast, say, between a proxy and a portrait. There is, of course, a relationship between them, one that has received political and ideological exacerbation in the European tradition at least since the poet and the sophist, the actor and the orator, have both been seen as harmful. In the guise of a post-Marxist description of the scene of power, we thus encounter a much older debate: between representation or rhetoric as tropology and as persuasion. Darstellen belongs to the first constellation, vertreten - with stronger suggestions of substitution - to the second. Again, they are related, but running them together, especially in order to say that beyond both is where oppressed subjects speak, act and know for themselves, leads to an essentialist, utopian politics.

Here is Marx’s passage, using ‘vertreten’ where the English use ‘represent’, discussing a social ‘subject’ whose consciousness and Vertretung (as much a substitution as a representation) are dislocated and incoherent: The small peasant proprietors ‘cannot represent themselves; they must be represented. Their representative must appear simultaneously as their master, as an authority over them, as unrestricted governmental power that protects them from the other classes and sends them rain and sunshine from above. The political influence [in the place of the class interest, since there is no unified class subject] of the all peasant proprietors therefore finds its last expression [the implication of a chain of substitutions - Vertretungen - is strong here] in the executive force [Exekutivgewalt - less personal in German] subordinating society to itself.’

Not only does such a model of social indirection - necessary gaps between the source of ‘influence’ (in this case the small peasant proprietors), the ‘representative’ (Louis Napoleon), and the historical-political phenomenon (executive control) imply a critique of the subject its individual agent but a critique even of the subjectivity of a collective agency. The necessarily dislocated machine of history moves because ‘the identity of the interests’ of these proprietors ‘fails to produce a feeling of community, national links, or a political organization’. The event of representation as Vertretung (in the constellation of rhetoric-as-persuasion) behaves like a Darstellung (or rhetoric-as-trope), taking its place in the gap between the formation of a (descriptive) class and the non-formation of a (transformative) class: ‘In so far as millions of families live under economic conditions of existence that separate their mode of life ... they form a class. In so far as ... the identity of their interests fails to produce a feeling of community ... they do not form a class.’ The complicity of vertreten and Darstellen, their identity-in-difference as the place of practice - since this complicity is precisely what Marxists must expose, as Marx does in The Eighteenth Brumaire - can only be appreciated if they are not conflated by a sleight of word.

It would be merely tendentious to argue that this texualises Marx too much, making him inaccessible to the common ‘man’, who, a victim of common sense, is so deeply placed in a heritage of positivism that Marx’s irreducible emphasis on the work of the negative, on the necessity for de-fetishising the concrete, is persistently wrested from him by the strongest adversary, ‘the historical tradition’ in the air.[16] I have been trying to point out that the uncommon ‘man’, the contemporary philosopher of practice, sometimes exhibits the same positivism.

The gravity of the problem is apparent if one agrees that the development of a transformative class ‘consciousness’ from a descriptive class ‘position’ is not in Marx a task engaging the ground level of consciousness. Class consciousness remains with the feeling of community that belongs to national links and political organizations, not to that other feeling of community whose structural model is the family. Although not identified with nature, the family here is constellated with what Marx calls ‘natural exchange’, which is, philosophically speaking, a ‘placeholder’ for use value.[17] ‘Natural exchange’ is contrasted to ‘intercourse with society’, where the word ‘intercourse’ (Verkehr) is Marx’s usual word for ‘commerce’. This ‘intercourse’ thus holds the place of the exchange leading to the production of surplus value, and it is in the area of this intercourse that the feeling of community leading to class agency must be developed. Full class agency (if there were such a thing) is not an ideological transformation of consciousness on the ground level, a desiring identity of the agents and their interest - the identity whose absence troubles Foucault and Deleuze. It is a contestatory replacement as well as an appropriation (a supplementation) of something that is ‘artificial’ to begin with - ‘economic conditions of existence that separate their mode of life’. Marx’s formulations show a cautious respect for the nascent critique of individual and collective subjective agency. The projects of class consciousness and of the transformation of consciousness are discontinuous issues for him. Conversely, contemporary invocations of ‘libidinal economy’ and desire as the determining interest, combine with the practical politics of the oppressed (under socialised capital) ‘speaking for themselves’, restore the category of the sovereign subject within the theory that seems most to question it.

No doubt the exclusion of the family, albeit a family belonging to a specific class formation, is part of the masculine frame within which Marxism marks its birth.[18] Historically as well as in today’s global political economy, the family’s role in patriarchal social relations is so heterogeneous and contested that merely replacing the family in this problematic is not going to break the frame. Nor does the solution lie in the positivist inclusion of a monolithic collectivity of ‘women’ in the list of the oppressed whose unfractured subjectivity allows them to speak for themselves against an equally monolithic ‘same system’.

In the context of type development of a strategic, artificial and second-level ‘consciousness’, Marx uses the concept of the patronymic, always within the broader concept of representation as Vertretung: the small peasant proprietors are therefore incapable of making their class interest valid in their proper name [imeigenen Namen], whether through a parliament or through a convention’. The absence of the non-familiar artificial collective proper name is supplied by the only proper name ‘historical tradition’ can offer - the patronymic itself - the Name of the Father: Historical tradition produced the French peasants’ belief that a miracle would occur, that a man named Napoleon would restore all their glory. And an individual turned up’ - the untranslatable ‘es fand sich’ (there found itself an individual?) demolishes all questions of agency or the agent’s connection with his interest - ‘who gave himself out to be that man’ (this pretence is by contrast, his only proper agency) because he carried [trägt - the word used for the capitalist’s relationship to capital] the Napoleonic Code, which commands’ that ‘inquiry into paternity is forbidden’. While Marx here seems to be working within a patriarchal metaphorics, one should note the textual subtlety of the passage. It is the Law of the Father (the Napoleonic) that paradoxically prohibits the search for the natural father. Thus, it is according to a strict observance of the historical Law of the Father that the formed yet unformed class’s faith in the natural father is gainsaid.

A have dwelt so long on this passage in Marx because it spells out the inner dynamics of Vertretung, or representation in the political context. Representation in the economic context is Darstellung, the philosophical concept of representation as staging or, indeed, signification, which relates to the divided subject in an indirect way. The most obvious passage is well known: ‘In the exchange relationship [Austauschverhältnis] of commodities their exchange-value appeared to us totally independent of their use-value. But if we subtract their use-value from the product of labour, we obtain their value, as it was just determined [bestimmt]. The common element which represents itself [sich darstellt] in the exchange relation, or the exchange value of the commodity, is thus its value.’[19]

According to Marx, under capitalism, value, as produced in necessary and surplus labour, is computed as the representation/sign of objectified labour (which is rigorously distinguished from human activity). Conversely, in the absence of a theory of exploitation as the extraction (production), appropriation and realization of (surplus) value as representation of labour power, capitalist exploitation must be seen as a variety of domination (the mechanics of power as such). ‘The thrust of Marxism’, Deleuze suggests, ‘was to determine the problem [that power is more diffuse than the structure of exploitation and state formation] essentially in terms of interests (power is held by a ruling class defined by its interests)’ (FD, p.214).

One cannot object to this minimalist summary of Marx’s project, just as one cannot ignore that, in parts of the Anti-Oedipus, Deleuze and Guattari build their case on a brilliant if ‘poetic’ grasp of Marx’s theory of the money form. Yet we might consolidate our critique in the following way: the relationship between global capitalism (exploitation in economics) and nation-state alliances (domination in geopolitics) is so macrological that it cannot account for the micrological texture of power. To move toward such an accounting one must move toward theories of ideology - of subject formations that micrologically and often erratically operate the interests that congeal the macrologies. Such theories cannot afford to overlook the category of representation in its two senses. They must note how the staging of the world in representation - its scene of writing, its Darstellung - dissimulates the choice of and need for ‘heroes’, paternal proxies, agents of power - Vertretung.

My view is that radical practice should attend to this double session of representations rather than reintroduce the individual subject through totalizing concepts of power and desire. It is also my view that, in keeping the area of class practice on a second level of abstraction, Marx was in effect keeping open the (Kantian and) Hegelian critique of the individual subject as agent.[20] This view does not oblige me to ignore that, by implicitly defining the family and the mother tongue as the ground level where culture and convention seem nature’s own way of organizing ‘her’ own subversion, Marx himself rehearses an ancient subterfuge.[21] In the context of poststructuralist claims to critical practice, this seems more recuperable than the clandestine restoration of subjective essentialism.

The reduction of Marx to a benevolent but dated figure most often serves the interest of launching a new theory of interpretation. In the Foucault-Deleuze conversation, the issue seems to be that there is no representation, no signifier (Is it to be presumed that the signifier has already been dispatched? There is, then, no sign-structure operating experience, and thus might one lay semiotics to rest?); theory is a relay of practice (thus laying problems of theoretical practice to rest) and the oppressed can know and speak for themselves. This reintroduces the constitutive subject on at least two levels: the Subject of desire and power as an irreducible methodological presupposition; and the self-proximate, if not self-identical, subject of the oppressed. Further, the intellectuals, who are neither of these S/subjects, become transparent in the relay race, for they merely report on the non-represented subject and analyze (without analyzing) the workings of (the unnamed Subject irreducibly presupposed by) power and desire. The produced ‘transparency’ marks the place of ‘Interest’; It is maintained by vehement denegation: ‘Now this role of referee, judge, and universal witness is one which I absolutely refuse to adopt.’ One responsibility of the critic might be to read and write so that the impossibility of such interested individualistic refusals of the institutional privileges of power bestowed on the subject is taken seriously. The refusal of the sign-system blocks the way to a developed theory of ideology. Here, too, the peculiar tone of denegation is beard. To Jacques-Alain Miller’s suggestion that ‘the institution is itself discursive’, Foucault responds, ‘Yes, if you like, but it doesn’t much matter for my notion of the apparatus to be able to say that this is discursive and that isn’t ... given that my problem isn’t a linguistic one’ (PK, p.198). Why this conflation of language and discourse from the master of discourse analysis?

Edward W. Said’s critique of power in Foucault as a captivating and mystifying category that allows him ‘to obliterate the role of classes, the role of economics, the role of insurgency and rebellion’, is most pertinent here.[22] I add to Said’s analysis the notion of the surreptitious subject of power and desire marked by the transparency of the intellectual. Curiously enough, Paul Bové faults Said for emphasizing the importance of the intellectual, whereas Foucault’s project essentially is a challenge to the leading role of both hegemonic and oppositional intellectuals’[23]. I have suggested that this ‘challenge’ is deceptive precisely because it ignores what Said emphasises - the critic’s institutional responsibility.

This S/subject, curiously sewn together into a transparency by denegations, belongs to the exploiters’ side of the international division of labour. It is impossible for contemporary French intellectuals to imagine the kind of Power and Desire that would inhabit the unnamed subject of the Other of Europe. It is not only that everything they read, critical or uncritical, is caught within the debate of the production of that Other, supporting or critiquing the constitution of the Subject as Europe. It is also that, in the constitution of that Other of Europe, great care was taken to obliterate the textual ingredients with which such a subject could cathect, could occupy (invest?) its itinerary - not only by ideological and scientific production, but also by the institution of the law. However reductionistic an economic analysis might seem, the French intellectuals forget at their peril that this entire over-determined enterprise was in the interest of a dynamic economic situation requiring that interests, motives (desires) and power (of knowledge) be ruthlessly located. To invoke that dislocation now as a radical discovery that should make diagnose the economic (conditions of existence that separate out ‘classes’ descriptively) as a piece of dated analytic machinery may well be to continue the work of that dislocation and unwittingly to help in securing ‘a new balance of hegemonic relations’.[24]

I shall return to this argument shortly. In the face of the possibility that the intellectual is complicit in the persistent constitution of Other as the Self’s shadow, a possibility of political practice for the intellectual would be to put the economic ‘under erasure’, to see the economic factor as irreducibly as it inscribes the social text, even as it is erased, however imperfectly, when it claims to be the final determinant of the transcendental signified.[25]

The clearest available example of such epistemic violence is the remotely orchestrated, far-flung, and heterogeneous project to constitute the colonial subject as Other. This project is also the asymmetrical obliteration of the trace of that Other in its precarious Subjectivity. It is well known that Foucault locates epistemic violence, a complete overhaul of the episteme, in the redefinition of sanity at the end of the European eighteenth century.[26] But what if that particular redefinition was only a part of the narrative of history in Europe as well as in the colonies? What if the two projects of epistemic overhaul worked as dislocated and unacknowledged parts of a vast two-handed engine? Perhaps it is no more than to ask that the subtext of the palimpsestic narrative of imperialism be recognised as ‘subjugated knowledge’, ‘a whole set of knowledges that have been disqualified as inadequate to their task or insufficiently elaborated: naive knowledges, located low down on the hierarchy, beneath the required level of cognition or scientificity’ (PK, p.82).

This is not to describe ‘the way things really were’ or to privilege the narrative of history as imperialism as the best version of history.[27] It is, rather, to offer an account of how an explanation and narrative of reality was established as the normative one. To elaborate on this, let us consider briefly the underpinnings of the British codification of Hindu Law.

First, a few disclaimers: in the United States the third-worldism currently afloat in humanistic disciplines is often openly ethnic. I was born in India and received my primary, secondary and university education there, including two years of graduate work. My Indian example could thus be seen as a nostalgic investigation of the lost roots of my own identity. Yet even as I know that one cannot freely enter the thickets of ‘motivations’, I would maintain that my chief project is to point out the positivist-idealist variety of such nostalgia. I turn to Indian material because, in the absence of advanced disciplinary training, that accident of birth and education has provided me with a sense of the historical canvas, a hold on some of the pertinent languages that are useful tools for a bricoleur, especially when armed with the Marxist skepticism of concrete experience as the final arbiter and a critique of disciplinary formations. Yet the Indian case cannot be taken as representative of all countries, nations, cultures and the like that may be invoked as the Other of Europe as Self.

Here, then, is a schematic summary of the epistemic violence of the codification of Hindu Law. If it clarifies the notion of epistemic violence, my final discussion of widow-sacrifice may gain added significance.

At the end of the eighteenth century, Hindu law, insofar as it can be described as a unitary system, operated in terms of four texts that ‘staged’ a four-part episteme defined by the subject’s use of memory: sruti (the heard), smriti (the remembered), sastra (the learned-from-another) and vyavabara (the performed-in-exchange). The origins of what had been heard and what was remembered were not necessarily continuous or identical. Every invocation of sruti technically recited (or reopened) the event of originary ‘hearing’ or revelation. The second two texts - the learned and the performed - were seen as dialectically continuous. Legal theorists and practitioners were not in any given case certain if this structure described the body of law or four ways of settling a dispute. The legitimation of the polymorphous structure of legal performance, ‘internally’ non-coherent and open at both ends, through a binary vision, is the narrative of codification I offer as an example of epistemic violence.

The narrative of the stabilization and codification of Hindu law is less well known than the story of Indian education, so it might be well to start there.[28] Consider the often-quoted programmatic lines from Macaulay’s infamous ‘Minute on Indian Education’ (1835): ‘We must at present do our best to form a class who may be interpreters between us and the millions whom we govern; a class of persons, Indian in blood and colour, but English in taste, in opinions, in morals, and in intellect. To that class we may leave it to refine the vernacular dialects of the country, to enrich those dialects with terms of science borrowed from the Western nomenclature, and to render them by degrees fit vehicles for conveying knowledge to the great mass of the population.’[29] The education of colonial subjects complements their production in law. One effect of establishing a version of the British system was the development of an uneasy separation between disciplinary formation in Sanskrit studies and the native, now alternative, tradition of Sanskrit ‘high culture’. Within the former, the cultural explanations generated by authoritative scholars matched the epistemic violence of the legal project.

I locate here the founding of the Asiatic Society of Bengal in 1784, the Indian Institute at Oxford in 1883, and the analytic and taxonomic work of scholars like Anhur Macdonnell and Arthur Berriedale Keith, who were both colonial administrators and organisers of the matter of Sanskrit. From their confident utilitarian-hegemonic plans for students and scholars of Sanskrit, it is impossible to guess at either the aggressive repression of Sanskrit in the general educational framework or the increasing ‘feudalization’ of the performative use of Sanskrit in the everyday life of Brahmanic-hegemonic India.[30] A version of history was gradually established in which the Brahmans were shown to have the same intentions as (thus providing the legitimation for) the codifying British: ‘In order to preserve Hindu society intact [the] successors [of the original Brahmans] had to reduce everything to writing and make them more and more rigid. And that is what has preserved Hindu society in spite of a succession of political upheavals and foreign invasions.’[31] This is the 1925 verdict of Mahamahopadliyaya Haraprasad Shastri, learned Indian Sanskritist, a brilliant representative of the indigenous elite within colonial production, who was asked to write several chapters of a ‘History of Bengal’ projected by the private secretary to the governor general of Bengal in 1916.[32] To signal the asymmetry in the relationship between authority and explanation (depending on the race-class of the authority), compare this 1928 remark by Edward Thompson, English intellectual: ‘Hinduism was what it seemed to be ... It was a higher civilization that won [against it], both with Akbar and the English.33 And this, from a letter by an English soldier-scholar in the 1890s: ‘The study of Sanskrit, “the language of the gods”, has afforded me intense enjoyment during the last 25 years of my life in India, but it has not, I am thankful to say, led me, as it has some, to give up a hearty belief in our own grand religion.’[34]

These authorities are the very best of the sources for the non-specialist French intellectual’s entry into the civilization of the Other.[35] I am, however, not referring to intellectuals and scholars of postcolonial production, like Shastri, when I say that the Other as Subject is inaccessible to Foucault and Deleuze. I am thinking of the general non-specialist, non-academic population across the class spectrum, for whom the episteme operates its silent programming function. Without considering the map of exploitation, on what grid of ‘oppression’ would they place this motley crew?

Let us now move to consider the margins (one can just as well say the silent, silenced centre) of the circuit marked out by this epistemic violence, men and women among the illiterate peasantry, the tribals, the lowest strata of the urban sub-proletariat. According to Foucault and Deleuze (in the First World, under the standardization and regimentation of socialised capital, though they do not seem to recognise this) the oppressed, if given the chance (the problem of representation cannot be by-passed here), and on the way to solidarity through alliance politics (a Marxist thematic is at work here), can speak and know their conditions. We must now confront the following question: on the other side of the international division of labour from socialised capital, inside and outside the circuit of the epistemic violence of imperialist law and education supplementing an earlier economic text, can the subaltern speak?

Antonio Gramsci’s work on the ‘subaltern classes’ extends the class-position/class-consciousness argument isolated in The Eighteenth Brumaire. Perhaps because Gramsci criticises the vanguardistic position of the Leninist intellectual, he is concerned with the intellectual’s role in the subaltern’s cultural and political movement into the hegemony. This movement must be made to determine the production of history as narrative (of truth). In texts such as ‘The Southern Question’, Gramsci considers the movement of historical-political economy in Italy within what can be seen as an allegory of reading taken from or prefiguring an international division of labor.[36] Yet an account of the phased development of the subaltern is thrown out of joint when his cultural macrology is operated, however remotely, by the epistemic interference with legal and disciplinary definitions accompanying the imperialist project. When I move, at the end of this essay, to the question of woman as subaltern, I will suggest that the possibility of collectivity itself is persistently foreclosed through the manipulation of female agency.

The first part of my proposition - that the phased development of the subaltern is complicated by the imperialist project - is confronted by a collective of intellectuals who may be called the ‘Subaltern Studies’ group.[37] They must ask, Can the subaltern speak? Here we are within Foucault’s own discipline of history and with people who acknowledge his influence. Their project is to rethink Indian colonial historiography from the perspective of the discontinuous chain of peasant insurgencies during the colonial occupation. This is indeed the problem of ‘the permission to narrate’ discussed by Said.[38] As Ranajit Guha argues,

The historiography of Indian nationalism has for a long time been dominated by elitism colonialist elitism and bourgeois-nationalist elitism ... shar[ing] the prejudice that the making of the Indian nation and the development of the consciousness nationalism - which confirmed this process were exclusively or predominantly elite achievements. In the colonialist and neo-colonialist historiographies these achievements are credited to British colonial rulers, administrators, policies, institutions and culture; in the nationalist and neo-nationalist writings - to Indian elite personalities, institutions, activities and ideas.[39]

Certain varieties of the Indian elite are at best native informants for first-world intellectuals interested in the voice of the Other. But one must nevertheless insist that the colonised subaltern subject is irretrievably heterogeneous.

Against the indigenous elite we may set what Guha calls ‘the politics of the people’, both outside (‘This was an autonomous domain, for it neither originated from elite politics nor did its existence depend on the latter’) and inside (‘it continued to operate vigorously in spite of [colonialism], adjusting itself to the conditions prevailing under the Raj and in many respects developing entirely new strains in both form and content’) the circuit of colonial production.[40] I cannot entirely endorse this insistence on determinate vigour and full autonomy, for practical historiographic exigencies will not allow such endorsements to privilege subaltern consciousness. Against the possible charge that his approach is essentialist, Guha constructs a definition of the people (the place of that essence) that can be only an identity-in-differential. He proposes a dynamic stratification grid describing colonial social production at large. Even the third group on the list, the buffer group, as it were, between the people and the great macrostructural dominant groups, is itself fined as a place of in-betweenness, what Derrida has described as an ‘antre’.[41]

Consider the third item on this list - the antre of situational indeterminacy these careful historians presuppose as they grapple with the question, Can the subaltern speak? ‘Taken as a whole and in the abstract this ... category ... was heterogeneous in its composition and, thanks to the uneven character of regional economic and social developments, differed from area to area. The same class or element which as dominant in one area ... could be among the dominated in another. This could and did create many ambiguities and contradictions in attitudes and alliances, especially among the lowest strata of the rural gentry, impoverished landlords, rich peasants and upper-middle peasants all of whom belonged, ideally speaking, to the category of “people” or “subaltern classes”.[42]

‘The task of research’ projected here is ‘to investigate, identify and measure the specific nature and degree of the deviation of [the] elements [constituting item 3] from the ideal and situate it historically’. ‘Investigate, identify, and measure the specific’: a program could hardly be more essentialist and taxonomic. Yet a curious methodological imperative is at work. I have argued that, in the Foucault-Deleuze conversation, a post-representationalist vocabulary hides an essentialist agenda. In subaltern studies, because of the violence of imperialist epistemic, social and disciplinary inscription, a project understood in essentialist terms must traffic in a radical textual practice of differences. The object of the group’s investigation, in the case not even of the people as such but of the floating buffer zone of the regional elite-subaltern is a deviation from an ideal - the people or subaltern - which is itself defined as a difference from the elite. It is toward this structure that the research is oriented, a predicament rather different from the self-diagnosed transparency of the first-world radical intellectual. What taxonomy can fix such a space? Whether or not they themselves perceive it - in fact Guha sees his definition of ‘the people’ within the master-slave dialectic - their text articulates the difficult task of rewriting its own conditions of impossibility as the conditions of its possibility.

‘At the regional and local levels [the dominant indigenous groups] ... if belonging to social strata hierarchically inferior to those of the dominant all-India groups still acted in the interests of the latter and not in conformity to interests corresponding truly to their own social being’. When these writers speak, in their essentialising language, of a gap between interest and action in the intermediate group, their conclusions are closer to Marx than to the self-conscious naiveté of Deleuze’s pronouncement on the issue. Guha, like Marx, speaks of interest in terms of the social rather than the libidinal being. The Name-of-the-Father imagery in The Eighteenth Brumaire can help to emphasise that, on the level of class or group action, ‘true correspondence to own being’ is as artificial or social as the patronymic.

So much for the intermediate group marked in item 3. For the ‘true’ subaltern group, whose identity is its difference, there is no unrepresentable subaltern subject that can know and speak itself; the intellectual’s solution is not to abstain from representation. The problem is that the subject’s itinerary has not been traced so as to offer an object of seduction to the representing intellectual. In the slightly dated language of the Indian group, the question becomes, How can we touch the consciousness of the people, even as we investigate their politics? With what voice-consciousness can the subaltern speak? Their project, after all, is to rewrite the development of the consciousness of the Indian nation. The planned discontinuity of imperialism rigorously distinguishes this project, however old-fashioned its articulation, from ‘rendering visible the medical and juridical mechanisms that surrounded the story [of Pierre Rivière]’. Foucault is correct in suggesting that ‘to make visible the unseen can also mean a change of level, addressing oneself to a layer of material which had hitherto had no pertinence for history and which had not been recognised as having any moral, aesthetic or historical value’. It is the slippage from rendering visible the mechanism to rendering vocal the individual, both avoiding ‘any kind of analysis of [the subject] whether psychological, psychoanalytical or linguistic’, that is consistently troublesome (PK, pp.49-50).

The critique by Ajit K. Chaudhury, a West Bengali Marxist, of Guha’s search for the subaltern consciousness can be seen as a moment of the production process that includes the subaltern. Chaudhury’s perception that the Marxist view of the transformation of consciousness involves the knowledge of social relations seems to me, in principle, astute. Yet the heritage of the positivist ideology that has appropriated orthodox Marxism obliges him to add this rider: ‘This is not to belittle the importance of understanding peasants’ consciousness or workers’ consciousness in its pure form. This enriches our knowledge of the peasant and the worker and, possibly, throws light on how a particular mode takes on different forms in different regions, which is considered a problem of second-order importance in classical Marxism.’[43]

This variety of ‘internationalist’ Marxism, which believes in a pure, retrievable form of consciousness only to dismiss it, thus closing off what in Marx remain moments of productive bafflement, can at once be the object of Foucault’s and Deleuze’s rejection of Marxism and the source of the critical motivation of the Subaltern Studies group. All three are united in the assumption that there is a pure form of consciousness. On the French scene, there is a shuffling of signifiers: ‘the unconscious’, or ‘the subject-in-oppression’ clandestinely fills the space of ‘the pure form of consciousness’. In orthodox ‘internationalist’ intellectual Marxism, whether in the First World or the Third, the pure form of consciousness remains an idealistic bedrock which, dismissed as a second-order problem, often earns it the reputation of racism and sexism. In the Subaltern Studies group it needs development according to the unacknowledged terms of its own articulation.

For such an articulation, a developed theory of ideology can again be most useful. In a critique such as Chaudhury’s, the association of ‘consciousness’ with ‘knowledge’ its the crucial middle term of ‘ideological production’: ‘Consciousness, according to Lenin, is associated with a knowledge of the interrelationships between different classes and groups; i.e., a knowledge of the materials that constitute society. … These definitions acquire a meaning only within the problematic within a definite knowledge object - to understand change in history, or specifically, change from mode to another, keeping the question of the specificity of a particular mode out of the focus.[44]

Pierre Macherey provides the following formula for the interpretation of ideology: What is important in a work is what it does not say. This is not the same as the careless notation “what it refuses to say”, although that would in itself be interesting: a method might be built on it, with the task of measuring silences, whether acknowledged or unacknowledged. But rather this, what the work cannot say is important, because there the elaboration of the utterance is carried out, in a sort of journey to silence.” Macherey’s ideas can be developed in directions he would be unlikely to follow. Even as he writes, ostensibly, of the literariness of the literature of European provenance, he articulates a method applicable to the social text of imperialism, somewhat against the grain of his own argument. Although the notion ‘what it refuses to say’ might be careless for a literary work, something like a collective ideological refusal can be diagnosed for the codifying legal practice of imperialism. This would open the field for a political-economic and multidisciplinary ideological reinscription of the terrain. Because this is a ‘worlding of the world’ on a second level of abstraction, a concept of refusal becomes plausible here. The archival, historiographic, disciplinary-critical and, inevitably, interventionist work involved here is indeed a task of ‘measuring silences’. This can be a description of ‘investigating, identifying, and measuring ... the deviation’ from an ideal that is irreducibly differential.

When we come to the concomitant question of the consciousness of the subaltern, the notion of what the work cannot say becomes important. In the semioses of the social text, elaborations of insurgency stand in the place of ‘the utterance’. The sender - ‘the peasant’ - is marked only as a pointer to an irretrievable consciousness. As for the receiver, we must ask who is ‘the real receiver’ of an ‘insurgency’? The historian, transforming ‘insurgency’ into ‘text for knowledge’, is only one ‘receiver’ of any collectively intended social act. With no possibility of nostalgia for that lost origin, the historian must suspend (as far as possible) the clamour of his or her own consciousness (or consciousness-effect, as operated by disciplinary training), so that the elaboration of the insurgency, packaged with an insurgent-consciousness, does not freeze into an ‘object of investigation’, or, worse yet, a model for imitation. ‘The subject’ implied by the texts of insurgency can only serve as a counter-possibility for the narrative sanctions granted to the colonial subject in the dominant groups. The postcolonial intellectuals learn that their privilege is their loss. In this they are a paradigm of the intellectuals.

It is well known that the notion of the feminine (rather than the subaltern of imperialism) has been used in a similar way within deconstructive criticism and within certain varieties of feminist criticism.[46] In the former case, a figure of ‘woman’ is at issue, one whose minimal predication as indeterminate is already available to the phallocentric tradition. Subaltern historiography raises questions of method that would prevent it from using such a ruse. For the ‘figure’ of woman, the relationship between woman and silence can be plotted by women themselves; race and class differences are subsumed under that charge. Subaltern historiography must confront the impossibility of such gestures. The narrow epistemic violence of imperialism gives us an imperfect allegory of the general violence that is the possibility of an episteme.[47]

Within the effaced itinerary of the subaltern subject, the track of sexual difference is doubly effaced. The question is not of female participation in insurgency, or the ground rules of the sexual division of labour, for both of which there is ‘evidence’. It is, rather, that, both as object of colonialist historiography and as subject of insurgency, the ideological construction of gender keeps the male dominant. If, in the context of colonial production, the subaltern has no history and cannot speak, the subaltern as female is even more deeply in shadow.

The contemporary international division of labour is a displacement of the divided field of nineteenth-century territorial imperialism. Put simply, a group of countries, generally first-world, are in the position of investing capital; another group, generally third-world, provide the field for investment, both through the comprador indigenous capitalists and through their ill-protected and shifting labour force. In the interest of maintaining the circulation and growth of industrial capital (and of the concomitant task of administration within nineteenth-century territorial imperialism), transportation, law and standardised education systems were developed - even as local industries were destroyed, land distribution was rearranged, and raw material transferred to the colonizing country. With so-called decolonization, the growth multinational capital, and the relief of the administrative charge, ‘development’ not now involve wholesale legislation and establishing educational systems in comparable way. This impedes the growth of consumerism in the comprador countries. With modern telecommunications and the emergence of advanced capitalist economies at the two edges of Asia, maintaining the international division of labour serves to keep the supply of cheap labour in the comprador countries.

Human labour is not, of course, intrinsically ‘cheap’ or ‘expensive’. An absence of laws (or a discriminatory enforcement of them), a totalitarian state (often entailed by development and modernization in the periphery), and minimal subsistence requirements on the part of the worker will ensure it. To keep this the crucial item intact, the urban proletariat in comprador countries must not be systematically trained in the ideology of consumerism (parading as the philosophy classless society) that, against all odds, prepares the ground for resistance through the coalition politics Foucault mentions (FD, p.216). This separation from the ideology of consumerism is increasingly exacerbated by the proliferating phenomena of international subcontracting. ‘Under this strategy, manufacturers in developed countries subcontract the most labour intensive stages of production, for example, sewing or assembly, to the Third World nations where is cheap. Once assembled, the multinational re-imports the goods - under generous tariff exemptions - to the developed country instead of selling them to the market.’ Here the link to training in consumerism is almost snapped. ‘While recession has markedly slowed trade and investment worldwide since 1979, international subcontracting has boomed. … In these cases, multinationals are to resist militant workers, revolutionary upheavals, and even economic turns.’[48]

Class mobility is increasingly lethargic in the comprador theatres. Not surprisingly, members of indigenous dominant groups in comprador countries, members of the local bourgeoisie, find the language of alliance politics attractive. Identifying forms of resistance plausible in advanced capitalist countries is often of piece a with that elitist bent of bourgeois historiography described by Ranajit Guha.

Belief in the plausibility of global alliance politics is prevalent among women of dominant social groups interested in ‘international feminism’ in the comprador countries. At the other end of the scale, those most separated from any possibility of an alliance among ‘women, prisoners, conscripted soldiers, hospital patients, and homosexuals’ (FD, p.216) are the females of the urban sub-proletariat. In their case, the denial and withholding of consumerism and the structure of exploitation is compounded by patriarchal social relations. On the other side of the international division of labour, the subject of exploitation cannot know and speak the text of female exploitation even if the absurdity of the non-representing intellectual making space for her to speak is achieved. The woman is doubly in shadow.

Yet even this does not encompass the heterogeneous Other. Outside (though not completely so) the circuit of the international division of labour, there are people whose consciousness we cannot grasp if we close off our benevolence by constructing a homogeneous Other referring only to our own place in the scat of the Same or the Self. Here are subsistence farmers, unorganised peasant labour, the tribals and the communities of zero workers on the street or in the countryside. To confront them is not to represent (vertreten) them but to learn to represent (darstellen) ourselves. This argument would take us into a critique of a disciplinary anthropology and the relationship between elementary pedagogy and disciplinary formation. It would also question the implicit demand, made by intellectuals who choose a ‘naturally articulate’ subject of oppression, that such a subject come through history as a foreshortened mode-of-production narrative.

That Deleuze and Foucault ignore both the epistemic violence of imperialism and the international division of labour would matter less if they did not, in closing, touch on third-world issues. But in France it is impossible to ignore the problem of the tiers monde, the inhabitants of the erstwhile French African colonies. Deleuze limits his consideration of the Third World to these old local and regional indigenous elite who are, ideally, subaltern. In this context, references to the maintenance of the surplus army of labour fall into reverse-ethnic sentimentality. Since he is speaking of the heritage of nineteenth- century territorial imperialism, his reference is to the nation-state rather than the globalizing centre: ‘French capitalism needs greatly a floating signifier of unemployment. In this perspective, we begin to see the unity of the forms of repression: restrictions on immigration, once it is acknowledged that the most difficult and thankless jobs go to immigrant workers; repression in the factories, because the French must reacquire the “taste” for increasingly harder work; the struggle against youth and the repression of the educational system’ (FD, pp.211-12). This is an acceptable analysis. Yet it shows again that the Third World can enter the resistance program of an alliance politics directed against a ‘unified repression’ only when it is confined to the third-world groups that are directly accessible to the First World.[49] This benevolent first-world appropriation and reinscription of the Third World as an Other is the founding characteristic of much third-worldism in the US human sciences today.

Foucault continues the critique of Marxism by invoking geographical discontinuity. The real mark of ‘geographical (geopolitical) discontinuity’ is the international division of labour. But Foucault uses the term to distinguish between exploitation (extraction and appropriation of surplus value; read, the field of Marxist analysis) and domination (‘power’ studies) and to suggest the latter’s greater potential for resistance based on alliance politics. He cannot acknowledge that such a monist and unified access to a conception of ‘power’ (methodologically presupposing a Subject-of-power) is made possible by a certain stage in exploitation, for his vision of geographical discontinuity is geopolitically specific to the First World:

This geographical discontinuity of which you speak might mean perhaps the following: as soon as we struggle against exploitation, the proletariat not only leads the struggle but also defines its targets, its methods, its places and its instruments; and to ally oneself with the proletariat is to consolidate with its positions, its ideology, it is to take up again the motives for their combat. This means total immersion [in the Marxist project]. But if it is against power that one struggles, then all those who acknowledge as intolerable can begin the struggle wherever they find themselves and in terms of their own activity (or passivity). In engaging in this struggle that is their own, whose objectives they clearly understand and whose methods they can determine, they enter the revolutionary process. As allies of the proletariat, to be sure, because power is exercised the way it is in order to maintain capitalist exploitation. They genuinely serve the cause of the proletariat by fighting in those places where they find themselves oppressed. Women, prisoners, conscripted soldiers, hospital patients, and homosexuals have now begun a specific struggle against the particular form of power, the constraints controls, that are exercised over them. (FD, p.216)

This an admirable program of localised resistance. Where possible, this model of rice is not an alternative to, but can complement, macrological struggles along ‘Marxist’ lines. Yet if its situation is universalised, it accommodates unacknowledged privileging of the subject. Without a theory of ideology, it can lead to a dangerous utopianism.

Foucault is a brilliant thinker of power-in-spacing, but the awareness of the topographical reinscription of imperialism does not inform his presuppositions. He is taken in by the restricted version of the West produced by that reinscription and thus helps to consolidate its effects. Notice the omission of the fact, in the following passage, that the new mechanism of power in the seventeenth and eighteenth centuries (the extraction of surplus value without extra-economic coercion is its Marxist description) is secured by means of territorial imperialism - the Earth and its products - ‘elsewhere’. The representation of sovereignty is crucial in those theatres: ‘In the seventeenth and eighteenth centuries, we have the production of an important phenomenon, the emergence, or rather the invention, of a new mechanism of power possessed of highly specific procedural techniques ... which so, I believe, absolutely incompatible with the relations of sovereignty. This new mechanism of power is more dependent upon bodies and what they do than the Earth and its products’ (PK, p.104).

Because of a blind spot regarding the first wave of ‘geographical discontinuity’, Foucault can remain impervious to its second wave in the middle decades of our own century, identifying it simply ‘with the collapse of Fascism and the decline of Stalinism’ (PK, p.87). Here is Mike Davis’s alternative view: ‘It was rather the global logic of counter-revolutionary violence which created conditions for the peaceful economic interdependence of a chastened Atlantic imperialism under American leadership. ... It was multi-national military integration under the slogan of collective security against the USSR which preceded and quickened the interpenetration of the major capitalist economies, making possible the new era of commercial liberalism which flowered between 1958 and 1971.’[50]

It is within the emergence of this ‘new mechanism of power’ that we must read the fixation on national scenes, the resistance to economics, and the emphasis on concepts like power and desire that privilege micrology. Davis continues: ‘This quasi-absolutist centralization of strategic military power by the United States was to allow an enlightened and flexible subordinancy for its principal satraps. In particular, it proved highly accommodating to the residual imperialist pretensions of the French and British ... with each keeping up a strident ideological mobilization against communism all the while.’ While taking precautions against such unitary notions as ‘France’, it must be said that such unitary notions as ‘the workers’ struggle’, or such unitary pronouncements as ‘like power, resistance is multiple and can be integrated in global strategies’ (PK, p.142), seem interpretable by way of Davis’s narrative. I am not suggesting, as does Paul Bové, that ‘for a displaced and homeless people [the Palestinians] assaulted militarily and culturally ... a question [such as Foucault’s ‘to engage in politics ... is to try to know with the greatest possible honesty whether the revolution is desirable’] is a foolish luxury of Western wealth’.[51] I am suggesting, rather, that to buy a self-contained version of the West is to ignore its production by the imperialist project.

Sometimes it seems as if the very brilliance of Foucault’s analysis of the centuries of European imperialism produces a miniature version of that heterogeneous phenomenon: management of space - but by doctors; development of administrations - but in asylums; considerations of the periphery - but in terms of the insane, prisoners and children. The clinic, the asylum, the prison, the university - all seem to be screen-allegories that foreclose a reading of the broader narratives of imperialism. (One could open a similar discussion of the ferocious motif of ‘deterritorialisation’ in Deleuze and Guattari.) ‘One can perfectly well not talk about something because one doesn’t know about it’, Foucault might murmur (PK, p.66). Yet we have already spoken of the sanctioned ignorance that every critic of imperialism must chart.