4. The social factory

Machines, work, control

Capitalism is a system of relationships, which go from inside to out, from outside to in, from above to below, and from below to above. Everything is relative, everything is in chains. Capitalism is a condition both of the world and of the soul.

(Kafka, in Janouch 1971: 151-2)

If in its beginning the factory came out of the social body and tended to separate itself from it in order to elaborate its own rules of operation, it must now reincorporate this social body in order more than ever to dominate it.

(de Gaudemar 1985: 285)

The injunction of Marx's proletarian unnamable is an ever renewed engagement with the social plane of capitalized life - a plane that is at once manifold and mutating, cramped and constraining. In Chapter 3 this plane of capital was presented in general terms. This chapter now turns to consider the specificity of the contemporary capitalist socius. It does this not through a general mapping of Deleuze's and Marx's position, but, following the methodological logic of the minor and the proletarian unnamable, by exploring one manifestation of a political critique of capital. It follows a thread through a particular current in Italian Marxist research and politics - a current known in the 1960s as operaismo ('workerism')1 and in the 70s as autonomia ('autonomy'). This current can be seen as performing Kafka's 'double flux' (K: 41) inasmuch as it analysed capital as an open system which configures around lines of flight, and sought to take these lines elsewhere, whilst - as I explore in Chapter 5 - situating this politics in a cramped space without a delineated people. A central figure in the development of this current is Antonio Negri, and this chapter considers his work in some detail. Negri's recent Empire, co-written with Michael Hardt, has been the subject of much intellectual and political interest, being described by Frederic Jameson as 'The first great new theoretical synthesis of the new millennium', and by Zizek as 'ring[ing] the death-bell not only for the complacent liberal advocates of the "end of history", but also for pseudo-radical Cultural Studies which avoid the full confrontation with today's capitalism' (Hardt and Negri 2000: hardback dust-jacket).2 In the context of this book's argument Negri is a particularly interesting figure. Negri's later work plays heavily on the possi-bilities of a conjunction between Deleuze and Marx,3 and Negri's emergence in the English-speaking academy has had, as Wright (2002: 2) points out, much to do with a certain Deleuzianism, following Deleuze and Guattari's own relation with Negri.4 The work of Cleaver (1979), Dyer-Witheford (2000), Red Notes (1978, 1979, 1981), Ryan (1989), and Wright (2002) notwithstanding, Negri has recently come to prominence rather shorn of a critical sense of his relations to the movements, researchers, and theorists of operaismo and autonomia. This has been especially evident in the reception of Empire. Operaismo and autonomia are usually mentioned as a background to Negri's recent work, but this only seems to reinforce an idea that Negri has synthesized and transcended this current and, as such, is now most usefully discussed outside of this context. This is problematic not because a certain political current is not given its due, but because it masks both the complexity of operaismo and autonomia, and encourages a foreclosure on the possibility of a continued engagement with their insights. This problem is particularly important in the context of Negri because, despite the common alignment of operaismo and autonomia with his trajectory, his more recent work actually breaks with a number of the more important methodological and theoretical concerns of operaismo and autonomia.

Whilst Negri's engagement with Deleuze and Foucault marks this break in Negri's work, Negri's reading of Deleuze actually displays a number of problems. If we follow the minor imperative to consider an author as part of their minorities, and hence draw Negri back into relation to operaismo and autonomia, we actually find not only that operaismo and autonomia provide considerable insight into a minor theory and politics adequate to contemporary capital, but also that in a number of ways Deleuze's understanding of capital resonates more with operaismo and autonomia than it does with the apparently more Deleuzian Negri.5 Whilst Negri develops the important analytic categories of 'socialized', 'affective', and 'immaterial' labour, this chapter argues that he breaks with operaismo's and Deleuze's cramped and minor interrogation of the intricacies of capitalized production to develop a problematic understanding of an emerging autonomy-in-production. In this Negri makes a strange return to the orthodox Marxian and - in an inverted way "” neo-Gramscian positions which operaismo had sought to undermine.

The chapter begins with a brief introduction to operaismo and autonomia before outlining Marx's theory of machines and his 'real subsumption' thesis. The chapter then considers Raniero Panzieri's and Mario Tronti's elaboration of the real subsumption thesis and the 'social factory'. Through this I show how operaismo developed a very different position to both orthodox and neo-Gramscian Marxism, in that technical forces and social democracy were seen not as enabling lines of political mobility, but as creating a complex productive socius which left no room for an autonomous self-defined 'people' or even subject of politics. I suggest that it was the recognition of this very cramped condition and the refusal to designate a coherent and autonomous people that was one of the core strengths of operaismo, as its cramped position compelled an intricate analysis of the new arrangements of production. The chapter then focuses on Marx's 'Fragment on Machines', a text which is of great importance to Negri and to contemporary understandings of work, and also has a place in Deleuze and Guattari's understanding of capital. I then show how Negri develops a problematic reading of the 'Fragment' in his understanding of socialized and affective labour and control. The chapter then moves to consider Deleuze and Guattari's understanding of the capitalist socius through an exploration of the capitalist abstract machine, axiomatics, machinic surplus value, and post-disciplinary control. The discussion of Deleuze and Guattari shows how they overcome the limits of Negri's analysis in a fashion that is more in keeping with operaismo's understanding of the social factory.6 The chapter concludes with a sketch of the contemporary forms of machinic work and production.

Introduction to operaismo and autonomia

Italian Marxism has been known to Anglo-American cultural studies almost exclusively through Gramsci.7 This is no doubt partly an effect of the central role neo-Gramscian thought played in the development of the discipline away from Marxism in general. Neo-Gramscian work on 'hegemony' marked the passage from apparently orthodox concerns with class, capital, and the economy, into a post-Marxist concern with the possibilities of agency, popular practices, and new social movements, in a struggle for inclusion in the 'chain of equivalences' of social democratic political space. Here was a politics adequate to the fluidity of postmodern culture which could exorcize deter-minist Marxism, and indeed much of Marx, 'without apologies' (cf. Laclau and Mouffe 1985). The historical support for this development was not unrelated to the Italian Communist Party's (PCI) formation of its own version of post-Marxism "” 'eurocommunism' "” where neo-Gramscian thought played a central role. As Abse (1985) has suggested, eurocommunism seemed for many on the British left (most notably around the influential Marxism Today) to mark the possibility of a popular radical social democracy which could overcome Marxian orthodoxy and the limits of labourism; the PCI was, after all, the biggest Communist Party in Europe, and was rapidly approaching a place in government.

Behind this formidable post-Marxist trajectory lay another current in Italian Marxism, known in the 1960s as operaismo and in the 1970s as autonomia. Though it emerged from some relation to the PCI and the PSI (Italian Socialist Party), and maintained a complex relation with the orthodox left at least until 1968,8 operaismo and autonomia developed a profound critique of the PCI and the neo-Gramscian politics of hegemony. Contrary to the dominant leftist interpretation of the PCI found in Britain during the 1970s and 1980s, for operaismo and autonomia the PCI was not only an efficient mechanism for curtailing radical energies and disrupting progressive political development as it sought to bind workers' struggle to the development of capital (Partridge 1996: 77), but was also, through its implementation of austerity measures, the agent of pernicious cuts in the standards of living of the Italian working class. In what may now appear as grim humour, the PCI general secretary, Enrico Berlinguer, even went so far as to put forward austerity as a communist moral ideal (cf. Abse 1985: 27). Aside from the critique of the PCI, however, and despite my tendency in this and the following chapter to group operaismo and autonomia together, the current, as Wright (2002) has shown, develops in a number of different directions, comprises many divergent perspectives, and was far from a coherent movement. Operaismo and autonomia were always characterized by small groups, schools, and magazines, and though there were national organizations, they never took the form of pervasive and strictly organized parries. There was not even coherence of position between key figures such as Panzieri, Tronti, Negri, and Piperno.9 Though Potere Operaio (the group fantastically built up by the prosecutors of the '7 April' case as the origin and base of a mysterious 'O' which was seen to orchestrate autonomia and the Red Brigades) had considerable importance, it certainly never characterized more than a small aspect of the workers' movement, and relatively quickly dissolved into the emerging 'area of autonomia'.10 As the area of autonomia developed in the 1970s, things got more complex. Though it has sometimes been described as a flowering of post-political potential, autonomia was comprised of such diverse political figures and perspectives (with organized autonomy or autonomia operaia "” which never fully escapes a vanguardist and militarist understanding of politics "” on one side, and the more countercultural autonomia creativa and aspects of the feminist movement on the other) that it would be problematic indeed to represent it as a coherent whole. Wright conveys the complexity well when he writes:

Making sense of Autonomia as a whole is no simple matter. Ideologically heterogeneous, territorially dispersed, organisationally fluid, politically marginalised: Giorgio Bocca's . . . analogy of an archipelago is an apt one. Never a single national organisation, much less the mass wing of the armed groups, as certain judges would later charge, the 'Area' of autonomist organisations and collectives would begin to disintegrate almost as soon as it had attained hegemony within the Italian far left.

(Wright 2002: 152)

And if operaismo and autonomia developed through the specific situation in Italy, the movement drew much from abroad: from Martin Glaberman, George Rawick and C. L. R. James to Foucault and Deleuze and Guattari; from the Industrial Workers of the World to Socialisme ou Barbarie and American counterculture. Marazzi thus writes:

What can be considered as the most original theoretical contribution to Italian workerism originated abroad . . . There is nothing 'Italian;' about the class warfare in Italy ... To erect a monument to Italy is to play the game of the Italian State; to misrepresent as specific ('the production of certain intellectuals') what is in fact rooted in the worker's history, rooted, above all, in its international dimension.

(Marazzi, in Semiotext(e) 1980: 12"”13)

One can, nevertheless, describe certain central theoretical and methodological tenets of this current. Emerging in the early 1960s in the writings of Raniero Panzieri, Mario Tronti, Romano Alquati, Sergio Bologna, and Antonio Negri amongst others, and in the journals Quaderni Rossi ('Red Notebooks' 1961-4), Classe Operaia ('Working Class' 1964-7), and later Potere Operaio ('Workers' Power' 1969"”74), and Primo Maggio ('May Day', 1973"”86), operaismo was based on a dual strategy of concrete interpretation of particular and new forms of work and new technological paradigms (following the dramatic changes of the Italian post-war 'economic miracle'), and emergent forms of struggle in tension with, or outside of, the organs of the official labour movement. It also involved an intensive rereading of Marx in a rather heretical focus on the Grundrisse and 'missing sixth chapter' of volume I (Marx 1976: 948-1084),11 as well as volumes II and III of Capital. Operaismo and autonomia followed a methodological insistence on the primacy of (changing forms of) political antagonism - what Tronti called 'the reversal of perspective' - in a dynamic 'class composition', and brought everything from absenteeism and housework to developments in the petrochemical industry into consideration.12 This approach was to remain central to the development and mutation of autonomia, and "” whilst it was always in tension with a tendency, as Wright (2002) argues, to theoretical generalization and a certain political impatience "” proved to be a practical, situated, and politically productive research paradigm.13

The next chapter will pursue the political configuration of the reversal of perspective further. Here I want to consider the reading of Marx. Surrounded by the disabling culture of orthodox, and then eurocommunist Marxism that permeated the Italian left, operaismo chose not to break with, but to return to Marx. Despairing of the social democratic trajectory of the orthodox left, but mindful of the dangers of factionalism, Panzieri was to say in 1960, 'I see all paths blocked, the "return to the private" leaves me cold, the possible fate of the small sect terrifies me' (cited in Wright 2002: 33). The re-engagement with Marx in the new journal Quaderni Rossi seems to have offered a 'way out'. A central concern was with the question of technology and social relations in what, in the 'missing sixth chapter' to Capital, Marx had called 'real subsumption'. Before considering Panzieri, it is useful to present Marx's argument.

Marx's theory of machines and the 'real subsumption' thesis

In the spectrum of apparent 'determinisms' with which Marx's work has been charged (economic determinism, labour essentialism, teleological historicism, and so on), the charge of 'technological determinism' is not uncommon.14 However, Marx's understanding of technical machines, as theorists like Panzieri (1976, 1980) and Rosenberg (1982) have argued, is actually rather sophisticated. I want to present it here in relation to Foucault's and Deleuze's understanding of the 'diagram' and 'abstract machine'. We can start with Foucault's (1991) now familiar analysis of Bentham's Panopticon.

The Panopticon is most visibly an architectural technology which uses the interplay of visibility and invisibility to produce internalized self-government. Each cell is arranged in a circle, with one side open to observation by a central tower with an overseer who remains unseen by the occupant of the cell. Not knowing whether s/he is being watched by another or not, the prisoner begins to check her own practice in a process of self-surveillance and self-control. But this architectural device does not stand alone, or emerge from the blue. It only functions effectively within the social environment (what Foucault calls the 'diagram', and Deleuze and Guattari call the 'abstract machine') of 'discipline' (a regime which seeks to both individualize and massify social groups in the pursuit of 'docility-utility'). The similarity of the Panopticon (which, in a strict sense, remained unbuilt) with actual prisons, schools, hospitals, barracks, and so on, is not in the detail of their physical forms as such, but in the way subjects and masses are assembled together or formed in similar fashion in each space. That is, the Panopticon's diagram of discipline is imminent to each space, even though in varying scales and degrees of intensity. It is not that the physical technology determines the practice, but that the technology is the solidification of a social practice.15 As Deleuze puts it:

the machines are social before being technical. Or, rather, there is a human technology which exists before a material technology. No doubt the latter develops its effects within the whole social field; but in order for it to be even possible, the tools or material machines have to be chosen by a diagram and taken up by the assemblages.

(Deleuze 1988: 39)

Once manifested in concrete form, the technology of the Panopticon has great efficacy, but only in so far as it manifests the diagram of discipline. So, to the degree that 'sovereign' societies exist before disciplinary ones, and discipline might be breaking down in 'control' societies (see below), the concrete technology of the prison, 'like a Cartesian diver', rises and falls in prominence and effect 'on a scale gauging the degree to which the disciplinary diagram [or abstract machine] [is] fulfilled' (Deleuze 1988: 41"”2). In this schema, the particular technology is only ever a visible sign of a set of social relations, even as, or because, it has far-reaching effects and functions across social space. That is, the visible technical machine is part of, and selected by a more general or abstract machinic environment.

Marx's works are full of accounts of technical machines in a conceptual framework that resonates with those of Foucault and Deleuze. What in Foucault and Deleuze is the diagram and abstract machine, in Marx is the 'mode of production'. Rather than the work of individual genius or autonomous scientific progress, Marx writes that:

A critical history of technology would show how little any of the inventions of the eighteenth century were the work of a single individual . . . Technology reveals the active relation of man to nature, the direct process of the production of his life, and thereby it also lays bare the process of the production of the social relations of bis life, and of the mental conceptions that flow from those relations.

(Marx 1976: 493; emphasis added)

An example can make the case. In a footnote to Capital Marx reads a particular form of rather rudimentary plough as the visible technology of an abstract machine, or mode of production called slavery. The slave, bought wholesale rather than piecemeal by the hour, is treated, following his definition in antiquity, as little more than an animal, as a 'speaking implement' (who combines with a 'semi-mute implement' of the animal and a 'mute implement' of the plough) (Marx 1976: 303). In this assemblage the plough employed is of a most unsophisticated form, the 'rudest and heaviest [of] implements' which is 'difficult to damage owing to [its] very clumsiness' (303). As Marx says, 'In the slave states bordering on the Gulf of Mexico, down to the date of the Civil War, the only ploughs to be found were those constructed on the old Chinese model, which turned up the earth like a pig or a mole, instead of making furrows' (304). Marx's point is that the instrument's clumsiness is not due to a lack of technological development (it is 'rude' for its time), but rather it is a selected characteristic appropriate to this slave-based mode of production which lacks the intricate device of the wage and complex structuring and ordering machines to prevent the rough treatment, or resistance of the slave in his use of the plough.16

From this basic presentation of the relations between technical machines and social relations we can move to an analysis of machines within what Marx called 'real subsumption'. In the 'missing sixth chapter' to Capital and in a section of the Grundrisse known as the 'Fragment on Machines' Marx develops a thesis (more or less evident in parts of Capital itself, notably Ch. 15) that, with time, work loses any artisanal autonomy and worker control as it is 'subsumed' in an increasingly complex 'automaton' of human parts and concrete technical machines. In 'formal subsumption' capitalist forms of valorization subsume the labour process as it finds it ('on the basis of the technical conditions within which labour has been carried on up to that point in history'; Marx 1976: 425) and extracts surplus value by extending the working day ('absolute surplus value'):

The work may become more intensive, its duration may be extended, it may become more continuous or orderly under the eye of the interested capitalist, but in themselves these changes do not affect the character of the actual labour process, the actual mode of working.

(Marx 1976: 1021)

This form of production has its problems, due both to the limited technical principle of handicraft and the insubordination of workers,17 and hence over time labour becomes increasingly subdivided and mechanized, and con-comitantly 'cooperative' (necessitating a form of overarching management and social plan; cf. 1976: Ch. 13; Panzieri 1976: 6-7).18 This social process with its technical consolidation in machines develops into what Marx called the 'specifically capitalist mode of production' or 'real subsumption' where labour and social life itself become enmeshed or 'subsumed', and hence transformed, in the intricate processes of machinery in large-scale industry.19 It is here that machinery comes into its own as a solution to the need of the social relations of capital to reorient the motive force and unity of production away from the labourer: 'It is machines that abolish the role of the handicraftsman as the regulating principle of social production' (Marx 1976: 491).20 In fully developed machinery the unity of the labourer, already broken down in simple cooperation in manufacture, is radically disrupted and absorbed in a system driven by an 'automaton consisting of numerous mechanical and intellectual organs, so that the workers themselves are cast merely as its conscious linkages' (Marx 1973a: 692). In this 'automaton' - which Deleuze and Guattari would call a 'machinic' relation, in so far as technical, human, and social relations function as an integrated or machinic whole - the governing power or unity ceases to be the rhythms of labour, and becomes the rhythm of capital itself, under the temporality of the machine, which technically embodies the cooperation and socialization of labour and thus 'constitutes the power of the "master"' (Marx 1976: 549).21

Panzieri and capitalist machines

The result of operaismo's return to the Marx of real subsumption "” particularly in the work of Panzieri and Tronti - was a very different understanding of the contemporary socius, and resultant politics, from both orthodox Marxist understandings of a self-moving development of the 'forces of production' (which could be 'planned' by a socialist state) and neo-Gramscian understandings of the relative autonomy of the social (where a leftist democratic movement struggles over 'hegemony'). The first point can be considered through Panzieri.

Panzieri (1976, 1980) posed a direct challenge to the dominant orthodox, or what he called 'objectivist', Marxist positions that posited a technological 'rationality' - as a self-moving development of scientific innovation as part of politically neutral 'forces of production' - distinct from capitalist 'relations of production'. In the objectivist approach, politics is situated externally to the technical process, as a movement towards the eventual assumption of technological processes as they are in a socialist 'planning'. This conjunction of objectivist and planning positions is amply evident in Lenin's 1919 speech Scientific Management and the Dictatorship of the Proletariat:

The possibility of socialism will be determined by our success in combining Soviet rule and Soviet organization or management with the latest progressive measures of capitalism. We must introduce in Russia the study and teaching of the Taylor system and its systematic trial and adoption.

(Lenin, cited in Bell 1956: 41)

For Panzieri, technical forces developed not in a logic of neutral scientific progress, but as a means of consolidating a particular form of the extraction of value. Technological rationality, or the 'machine' (and all the attendant organizational methods and techniques), was the direct manifestation, and naturalization, of capitalist power and control. The forces of production thus had capitalist relations immanent to them in a 'unity of "technical" and "despotic" moments' (1980: 57).

The capitalist objectivity of the productive mechanism with respect to the workers finds its optimal basis in the technical principle of the machine: the technically given speed, the coordination of the various phases and the uninterrupted flow of production are imposed on the will of the workers as a 'scientific necessity' . . . The capitalist social relationship is concealed within the technical demands of machinery and the division of labour seems to be totally independent of the capitalist's will. Rather, it seems to be the simple and necessary results of the means of labour's 'nature'.

(Panzieri 1976: 9)

Any socialist assumption or planning of the forces of production was therefore a misguided approach which failed to recognize (or, rather, actively disguised) the immanence of capitalist relations to technics. Thus, whether or not the Soviet state had abolished private property, the 'collective ownership' of production made no difference to the continued capitalist functioning of the machine:22

Faced by capital's interweaving of technology and power, the prospect of an alternative (working-class) use of machinery can clearly not be based on a pure and simple overturning of the relations of production (property), where these are understood as a sheathing that is destined to fall away at a certain level of productive expansion simply because it has become too small. The relations of production are within the productive forces, and these have been 'moulded' by capital. It is this that enables capitalist development to perpetuate itself even after the expansion of the productive forces has attained its highest level.

The social factory and the general interest of labour

If the real subsumption thesis shows how capitalist relations are immanent to the machine, it also shows how social relations as a whole become increasingly subordinated to capitalist regimes of production. As the compulsion of the machine replaces the need for a human master, the social itself emerges as a vast plane of capitalized activity in the development of what Mario Tronti called the 'Social Factory'. As Tronti put it in 1962:

The more capitalist development advances, that is to say the more the production of relative surplus value penetrates everywhere, the more the circuit production "” distribution "” exchange "” consumption inevitably develops; that is to say that the relationship between capitalist production and bourgeois society, between the factory and society, between society and the state, become [sic] more and more organic. At the highest level of capitalist development social relations become moments of the relations of production, and the whole society becomes an articulation of production. In short, all of society lives as a function of the factory and the factory extends its exclusive domination over all of society.

(Tronti, in Quaderni Rossi no. 2, cited in Cleaver 1992: 137)

The argument needs breaking down a little. The maintenance of circulation on a broad scale (total annual commodity-product) necessitates not the operability of individual capital, or of 'production', 'reproduction', and 'consumption' as distinct spheres, but the maintenance of capitalist relations as a whole across society, such that 'Capital's process of socialization' becomes 'the specific material base upon which [the process of development of capitalism] is founded' (Tronti 1973: 98; emphasis added). Though analysis at the level of individual moments may show the breakdown of one firm, or the composition of the particular exchange value of one commodity, at the level of social capital we see a continuity of circulation as the expansion and maintenance of value, where social capital operates like a 'ramified factory system'. This process is only possible, of course, in so far as tendencies toward competition are matched with a collective ownership, and hence both Panzieri (1976) and Tronti (1973) stress the importance of Marx's understanding of the socialization of ownership of capital in the development of aggregate 'total social capital' (through share holding and credit, as analysed in Capital, volumes II and III), such that profit will be a division of total social surplus value (not the surplus value the individual firm extracts - though it still seeks to extract above average surplus value; cf. Tronti 1973: 106).23 Such collective ownership, Marx writes in a suggestive way, is 'the abolition of capital as private property within the framework of capitalist production itself (1974a: 436), or, as he puts it elsewhere, a 'capitalist communism' (cited in Panzieri 1976: 23):

Here social capital is not just the total capital of society: it is not the simple sum of individual capitals. It is the whole process of socialization of capitalist production: it is capital itself that becomes uncovered, at a certain level of its development, as a social power.

(Tronti 1973: 105)

As this system develops, as I showed in Chapter 3, individual capitalists thus become less owners than managers, that is, functions of capital:

capital comes to represent all capitalists, and the individual capitalist is reduced to an individual personification of this totality: the direct functionary, no longer of his own capital, but of the capitalist class . . . Thus capital raises itself to the level of a 'general social power', while the capitalist is reduced to the level of a simple agent, functionary, or 'emissary' of this power.24

(Tronti 1973: 105, 107)

If the 'objectivist' approach to technology was challenged by the thesis of the immanence of capitalist relations to machines, the social factory thesis posed a direct challenge to neo-Gramscian understandings of the relative autonomy of the political, so central to the PCI's eurocommunism and its 'Historic Compromise' (cf. Negri 1979a: 112). As Bologna (n.d.: n.p.) suggests, the social factory thesis 'eliminate[d] the very bases of the concept of hegemony", for, far from tending to autonomy, the social was seen to be increasingly subordinated to capitalist regimes of production: The process of composition of capitalist society as a unified whole ... no longer tolerates the existence of a political terrain which is even formally independent of the network of social relations' (Tronti, cited in Bologna n.d.: n.p.). Indeed, for operaismo, one of the functions of social democracy, and specifically of socialism, was to naturalize the infusion of productive relations throughout the social, 'representing' "” or even, affirming "” an unproblematized labour in the social democratic political. For Negri writing in 1964 (in Hardt and Negri 1994), the socialist dreams of a 'society of labor' and a 'general social interest' (67) were seen to be actualized "” as the very basis of domination.25 Negri thus describes the centrality of labour to the post-war Italian Constitution26 not as a capitalist ruse, but as the penetration of the 'fundamental ideological principles of socialism . . . [in]to the heart of the Constitution' (56"”7).27 This is Indeed 'a long way from the idyllic image of a continual process of development from democracy to socialism' (80) in that socialism actually affirms the development of the social factory:

The 'democracy of labor' and 'social democracy' . . . consist of the hypothesis of a form of labor-power that negates itself as the working class and autonomously manages itself within the structures of capitalist production as labor-power. At this point, capitalist social interest, which has already eliminated the privatistic and egotistic expressions of single capitalists, attempts to configure itself as a comprehensive, objective social interest . . . The models of humanitarian socialism are assumed as emblems of reunification. The patriotism of common well-being in social production is the ultimate slogan of the capitalist effort at solidarity. Like soldiers, all producers are equally employed in the common sacrifice of production in order to win the battle of accumulation.

(Negri, in Hardt and Negri 1994: 62)

In the elaboration of the social factory thesis, operaismo's political focus was on what they called the 'mass worker' (essentially the Fordist workers of the large industrial plants of the Italian North, notably FIAT, and including a large proportion of Southern migrant workers whose precarious conditions left them excluded from the PCI unions). But though the mass worker always stretched beyond the walls of the factory to include the community (inasmuch as Fordism was a social system), it is arguably not until the 1970s and the development of work and politics around the figure of the 'socialized worker' that the worker of the social factory proper is theorized. The term 'socialized worker' was coined by Alquati in 1974, but it is closely associated with Antonio Negri (from Proletari e Stato in 1975 onwards) (cf. Wright 1988: 306). In Negri's development of this figure, one twenty-page text - Marx's 'Fragment on Machines' - took on central importance.28 Through the 'Fragment' one can discern both a radical enhancement of the social factory thesis and the basis of a number of problems in Negri's later work. It needs to be considered in some detail.

The 'Fragment on Machines'

Since its first publication in Italian in the same issue of Quaderni Rossi (no. 4, 1964) as Panzieri's (1976) essay "Surplus value and planning', the interpretation of the 'Fragment on Machines',29 as Paolo Virno (1996a) suggests, has been akin to biblical exegesis. Such exegesis has taken the form not of a replication of authorial truth, but of an iteration of the text in different socio-historical contexts as parr of the composition of varying political forms:

We have referred back many times to these pages - written in 1858 in a moment of intense concentration "” in order to make some sense out of the unprecedented quality of workers' strikes, of the introduction of robots into the assembly lines and computers into the offices, and of certain kinds of youth behavior. The history of the 'Fragment's' successive interpretations is a history of crises and of new beginnings.

(Virno 1996a: 265)

The 'Fragment' itself is a particularly complex and provocative text that raises a number of possibilities for understanding the trajectories of capitalist production "” projecting, as it does, an information capital from the heart of the social factory; machines, work, control, manufacture - and the possible processes and forms of communism that are rarely, if ever, so evident in Marx's work. The difficulty of the text, and its varied deployment make a general presentation of the thesis of the 'Fragment' difficult. I will start with the general argument, and then show two variations that it takes.

The complex reconfiguration of labour and machines in the machinism of real subsumption (the point made so far) is made especially clear in this famous passage from the 'Fragment':

The production process has ceased to be a labour process in the sense of a process dominated by labour as its governing unity. Labour appears, rather, merely as a conscious organ, scattered among the individual living workers at numerous points in the mechanical system; subsumed under the total process of the machinery itself, as itself only a link of the system, whose unity exists not in the living workers, but rather in the living (active) machinery, which confronts his individual, insignificant doings as a mighty organism.

(Marx 1973a: 693)

The radical thesis of the 'Fragment' is that in this machinic 'automaton' or 'organism' it is no longer the distinct individual entities of the productive workers that are useful for capitalist production, nor even their 'work' in a conventional sense of the word, but the whole ensemble of sciences, languages, knowledges, activities, and skills that circulate through society that Marx seeks to describe with the terms general intellect (706), social brain (694), and social individual (705). This is a Marx that points to a very different understanding of productive labour from Marxian orthodoxy, and indeed the thesis is challenging enough for Virno (1996a: 265) to suggest that it is 'not at all very "marxist"'. There are, however, two different ways of reading the thesis, which, if they are not wholly at variance in Marx's text, can certainly lead to very different interpretations. The following discussion of these two interpretations is based around two very similar citations (which I have called [A] and [B] to help references to these passages throughout this chapter and the next):

[A]

(Marx 1973a: 705; emphasis added)

Both these citations make the 'Fragment's' general argument that labour time and direct labour diminish in importance in relation to a new force, but they offer slightly different inflections on this force. The first, [A], emphasizes the productive power of 'science' and 'technology', whilst the second, [B], proposes the 'social individual' as the new productive force. The resultant arguments need pursuing through Marx's text.

[A] Contradiction? General intellect outside of work, and the 'watchman'

As we know, Marx sees a narrative in the development of work toward ever greater simplification and abstraction, where the dissection of the division of labour 'gradually transforms the workers' operations into more and more mechanical ones, so that at a certain point a mechanism can step into their places' (1973a: 704). In the 'Fragment' this leads him to introduce something of a dichotomy between the worker on one side and general intellect and the machine on the other. The dichotomy is signalled in [A], but he also puts it more firmly: 'The accumulation of knowledge and of skill, of the general productive forces of the social brain, is thus absorbed into capital, as opposed to labour, and hence appears as an attribute of capital, and more specifically of fixed capital' (694). As the 'social brain' or 'general intellect' is absorbed into machines, 'the human being comes to relate more as a watchman and regulator to the production process itself (705). Contrary to what we might think, this relegation to 'watchman' function is less important as a sign that work has become tedious and alienated than as a manifestation of a new and fatal contradiction for capital, and an indication of the possibilities for a communism without work. Inasmuch as the productive force comes from general intellect embodied in machines and not workers, productivity seems to bypass work, and hence the capitalist valuation of life in terms of work done becomes increasingly anachronistic: 'The theft of alien labour time, on which the present wealth is based, appears a miserable foundation in face of this new one, created by large-scale industry itself (705). An explosive 'contradiction' arises (705"”6) because capitalism continues to measure these forces in terms of (increasingly unproductive) labour and labour time, and the possibility emerges for the valuation and creation of life based on the needs of the 'social individual' and 'free time'. Thus we see in the forces of capital the potential for a communism where:

on one side, necessary labour time will be measured by the needs of the social individual, and, on the other, the development of the power of social production will grow so rapidly that, even though production is now calculated for the wealth of all, disposable time will grow for all.

(Marx 1973a: 708)

As such, the social individual will experience:

not the reduction of necessary labour time so as to posit surplus labour, but rather the general reduction of the necessary labour of society to a minimum, which then corresponds to the artistic, scientific etc. development of the individuals in the time set free, and with the means created, for all of them.

(Marx 1973a: 706; emphasis added)

This 'contradiction' thesis has been common in interpretations of the 'Fragment'. Leaving Negri until later, it is worth mentioning a few examples. Montano (1975) cites these sections of the 'Fragment' to argue that 'we are witnessing . . . the abolition of productive work within the capitalist mode of production itself (54) such that labour is no longer a form of production but of control (58). Andre Gorz similarly (though without a class struggle perspective) uses the 'Fragment' to argue that the majority of the population belong to a 'post-industrial neo-proletariat' whose precarious work 'will [in the not too distant future] be largely eliminated by automation' (1982: 69), that the 'micro-electronic revolution heralds the abolition of work' (1985: 32), and that already 'the amount of time spent working and the relatively high level of employment have been artificially maintained' (1982: 72) in a capital that has moved from production to domination (1985: 39)- Even Jeremy Rifkin (1995: 16"”17) uses the 'Fragment' - if rather superficially - to make his version of Gorz's 'end of work' thesis. Finally, Virno (1996b), whose interpretation of the realization of the 'Fragment's' emancipatory projections within capitalism is similar to the argument of this chapter, still writes of the 'vanishing of labour society'.

The contradiction thesis is in many ways a crucial moment in under-standing Marx's politics, for it posits communism not on a militarization of work, or an unalienated work, but on the destruction of the category of work enabled through complex mechanical processes, and a life of expansive creativity, art, and science beyond the drudgery of repetitive manual labour, or, indeed, work at all. But inasmuch as Marx presents it as a 'contradiction' it is problematic.

[B] The social individual in real subsumption

Marx's potential communism of general intellect-rich production outside work has not materialized, even with a massive expansion in the use of machines and the proliferation to a now axiomatic position of third-generation information machines.30 We can point to other parts of the 'Fragment' which, in conjunction with the real subsumption thesis, explain why. As we have seen, the contradiction is based on a disjunction between work and general intellect/ machines, with an increasing diminution of the productive force of the former (both quantitatively and qualitatively (Marx 1973a: 700) - shrunk to mere 'watchman') vis-Ã -vis the latter. The contradiction only holds in so far as this disjunction holds: in so far as the new productive potential of general intellect lies outside of work in some kind of 'pure science'.31 Given the movement towards the ever greater simplification of factory work that Marx was witnessing, the presentation of this disjunction is understandable. But it goes against the logic of the real subsumption thesis. As we have seen, the essence of real subsumption is that technical and social relations become enmeshed or subsumed within a machinic 'automaton'. As Panzieri and Tronti emphasized, this leaves no autonomous sphere of the technical or the social; everything is infused with capitalist relations. Rather than think of science or general intellect as an autonomous sphere of pure invention, the teal subsumption thesis should thus encourage us to think of it as a product of human activity conditioned within, and functional to this social machinic system "” something called forth by the automaton of capital.

The possibilities for exploring the interrelation of general intellect and work are more apparent when Marx writes of the 'social individual'. In section [B] Marx says not that science embodied in machinery is the productive force, but that 'the social individual appears as the great foundation-stone of production and of wealth'. Marx uses general intellect and the social individual largely interchangeably, but when he talks of the social individual we see a much richer idea of social rather than scientific and technological productivity. The social individual still seems to free-float outside of work, but if we follow the real subsumption thesis we could imagine that the automaton that subsumes the manual worker would also subsume the social individual. Thus, the productivity of the social individual - which could include a wealth of knowledge-based and affective relations and attributes "” would emerge always already in a work relation. When Marx writes in the 'Fragment' that the worker is 'regulated on all sides by the movement of the machinery' (1973a: 693) such that 'The most developed machinery thus forces the worker to work, longer than the savage does, or that he himself did with the simplest, crudest tools' (708"”9), what we need to add is that this is not just because general intellect invents machines that are used to make more manual work, but that general intellect and the practices of the social individual emerge as work "” as forces immanent to a social machinic system. The individual worker is still increasingly irrelevant (in her particularity as against the social whole she contributes to), but this time it is because general intellect signifies the extraction of surplus value not only from repetitive manual labour, but from all sorts of different, more complex forces in the social individual's 'combination of social activity' across society (not just within, but including work time). It is not, then, that a pure science becomes productive, but that a whole series of capacities and knowledges are productive and exploitable; work is not emptied of content, but filled with different content.

The productivity of general intellect, then, signifies a process not towards an increased unproductivity and irrelevance of work, but to the greater expansion of the content of life that can count as work. We can thus understand Marx's (1976: 532) other, rather tragic conclusion concerning the 'paradox that the most powerful instrument for reducing labour-time suffers a dialectical inversion and becomes the most unfailing means for turning the whole lifetime of the worker and his family into labour-time at capital's disposal for its own valorisation'.

Negri's socialized and affective workers

Both of these readings of the 'Fragment' - as well as a strange involution of the two - are evident in Negri's writings on the socialized worker (a term itself derived from the 'Fragment's' 'social individual').32 I will trace his argument through two interrelated points: first, that the content of socialized work has a tendency to become increasingly 'communicational' and 'immaterial', and second, that this form of work tends towards autonomy, becoming almost a majoritarian communist collectivity. This discussion combines Negri's later sole authored work with his work with Michael Hardt (1994, 2000). There is no doubt that his most recent work, Empire, stretches to overcome some of the conceptual problems I will identify, but it does not really manage it. That Negri himself does not seem to see the account of the socialized workers of Empire as a break from his previous work is marked by his reluctance to include this work in his critique of the 'immaterial labour' theorists (Hardt and Negri 2000: 29).33

Before moving into the detail of the argument it is worth noting the historical points of emergence and the general framework of the category of the socialized worker. If the theory of the mass worker marked the emergence of a class of generalized abstract labour, the socialized worker thesis seeks to describe the class composition of fully socialized capital. Negri (1988b: 217) suggests that the mass worker was a stage in the movement of real subsumption between the skilled worker and the fully socialized worker. He links the emergence of the socialized worker with the struggles of 1968, and suggests that 'For a large part of Europe, the mass worker had been conceptualized and had become a reality just when its period of existence was in fact about to end' (Negri 1989: 75). Negri (1988b) argues that in the recomposition of capital away from the large factory-cities, the increasing diffusion of workers across social space, and the regime of austerity measures in the 1970s, the power of the mass worker to extend demands beyond the factory was effectively curtailed.34 This necessitated an expansion of the content of class composition from the mass worker, and thus Negri argues the need for

the broadest definition of class unity, to modify and extend the concept of working-class productive labour, and to eliminate the theoretical isolation of the concept of mass worker (insofar as this concept had inevitably become tied to an empirical notion of the factory "” a simplified factoryism "” due to the impact of the bosses' counter-offensive, the corporatism of the unions, and the historical and theoretical limitations of the concept itself).

(Negri 1988b: 208)

The new class composition emerged as that of the fully diffuse proletariat "” the younger generations in the factories who were less schooled in the traditions of the orthodox communist movement, but also the emarginati (youth, women, sexual minorities, the unemployed, countercultural groupings), whose productive centrality was related to the expansion of casual, part-time, and non-guaranteed work and the underground economy, as well as housework and non-remunerated work. For the PCI this was the terrain of the non-disciplined class, almost the lumpenproletariat (as I show in Chapter 5, 'plague bearers' and 'parasitic strata'), but for Negri and autonomia, this diffuse proletariat was a new central force of production. Thus he suggests that this class composition might be better seen not so much as a 'working class', but as 'social labour-power', to reflect 'the potentiality of a new working class now extended throughout the entire span of production and reproduction "” a conception more adequate to the wider and more searching dimensions of capitalist control over society and social labour as a whole' (Negri 1988b: 209).

Communication and affective labour

From this general background we can move to consider the derail and subsequent development of the socialized worker thesis. The core of Negri's thesis follows the essence of the 'Fragment's' projections that socialized work is extremely rich in techno-scientific knowledge, becoming the living collective of general intellect. Thus, in The Politics of Subversion, Negri (1989; 116) writes that the 'taw material on which the very high level of productivity of the socialized worker is based ... is science, communication and the communication of knowledge'. Communication becomes central because it is the form of cooperation of the social whole: 'intellectual work reveals the mechanism of interaction for all social labour ... it produces a specific social constitution - that of cooperation, or rather, that of intellectual cooperation, i.e. communication "” a basis without which society is no longer conceivable' (Negri 1989: 51). Negri (1.989: 117) thus employs Habermas's theory of 'communicative action' to say that 'It is on the basis of the interaction of communicative acts that the horizon of reality comes to be constituted.' Two contradictory arguments seem to develop from this, and are no more apparent than in Empire.

On one side Negri recognizes that this communicative labour is not just a 'linguistic' but also a 'subjective', and later a 'biopolitical' and 'affective', interrelation (Hardt and Negri 2000), which, following Haraway, Hardt and Negri (1994, 2000) describe as a 'cyborg' condition of a complex assemblage of technical, organic, material, and immaterial processes.35 Hardt and Negri even pose a critique of the post-autonomia immaterial labour theorists (such as those collected in Virno and Hardt 1996), for presenting the new forces of production in 'angelic' fashion, 'almost exclusively on the horizon of language and communication' (Hardt and Negri 2000: 30, 29). Empire suggests that this immaterial and affective labour is not a distinct plane of production (though there are new forms of labour which involve the manipulation of information, code, and sign), but is imminent to the various regimes of production as a whole. Manufacture, for example, does not vanish, but is 'informationalized', as it is increasingly orchestrated through information technologies (Hardt and Negri 2000: 293). Further, largely following my argument in the discussion of section [B] of the 'Fragment', as Hardt and Negri's emphasis on biopower and the cyborg would necessitate, communicative and affective labour is seen as enmeshed in capitalist regimes of control, such that 'constant capital tends to be constituted and represented within variable capital, in the brains, bodies, and cooperation of productive subjects' (2000: 385).

This updating of the social factory thesis to explore the capitalization of affective production and general intellect is one of the most important aspects of Negri's work. But it does not emerge unproblematically; there is another side to the argument. At one level, Negri continues at times to conflate affective biopolitical processes with communication, suggesting, for example, that 'communication has increasingly become the fabric of production' (Hardt and Negri 2000: 404). But, more radically, Negri suggests that affective and immaterial labour tend towards increasing autonomy outside of capitalist relations.

Autonomous production and the communist multitude

Apparently ignoring the radical divergence between Foucauldian frameworks (where language is always enmeshed in power/knowledge regimes, and is hence never 'autonomous') and Habermassian autonomous communicative action, Negri seems to equate a tendency toward the productivity of communication with an emerging freedom - as if the more fluid and immaterial production becomes, the more it escapes control36 "” and perceives a rather pure linguistic 'activity' coming to the fore in 'communicational society' (1992: 105).37 Even when in Empire a more biopolitical slant is offered, biopolitical and immaterial labour still tend toward autonomy. Thus, in direct opposition to their comment about variable capital cited above, Hardt and Negri make a strange return to the orthodox dichotomy between forces and relations of production, and write - in the same work - that biopolitical labour

calls into question the old notion ... by which labor power is conceived as 'variable capital', that is, a force that is activated and made coherent only by capital, because the cooperative powers of labor power (particularly immaterial labor power) afford labor the possibility of valorizing itself.

(Hardt and Negri 2000: 294)

The reasons Negri tends to see an emerging autonomy of immaterial labour, even as he uses Foucauldian and Deleuzian conceptions of the immanence of power to all social relations, are not unrelated to Marx's desire in the 'Fragment' to witness an emerging contradiction and the basis for communist sociality. Just as Marx proposed that the new content of productive activity (general intellect) would emerge outside of work, and hence tends toward communism and the abolition of work, Negri similarly sees this increasingly autonomous plane of immaterial, communicative, and affective labour as a communist essence "” what Umpire calls the 'multitude'. Thus, in one reading of the 'Fragment' (Negri 1988c: 115-16), he uses the section noted [A] above to argue that the quantitative contradiction (mass socialized production measured in individual terms) is 'brought to a head', as labour time is indeed a 'dissolving factor', and science is 'immediately incorporated into production'. However, unlike Montano's and Gorz's interpretations of the 'Fragment', as Negri's work develops he tends not to follow Marx in seeing this going on outside work, in a pure productive science. Rather, as the last comment about variable capital suggests, he sees socialized work itself as tending toward autonomy: increasingly operating not in terms dominated by numeration, equivalence, and the value-form ('work' determined by capital), but in terms of 'free individualities' labouring in a self-determined fashion and driven by their own needs ('activities').38 Negri writes that:

The exchange of labour-power is no longer something that occurs, in determinate quantity and specific quality, within the process of capital; rather, an interchange of activities determined by social needs and goals is now the precondition, the premise of social production . . . Work is now an immediate participation in the world of social wealth.

(Negri 1988c: 117-18; emphasis added)

Opening the terrain for the politics of the multitude, Negri argues that this 'interchange of activities' tends to autonomous self-organization where 'cooperation is posed prior to the capitalist machine, as a condition independent of industry', such that 'the entrepreneurial power of productive labor is henceforth completely in the hands of the post-Fordist proletariat', and 'The socialized worker is a kind of actualization of communism, its developed condition. The boss, by contrast, is no longer even a necessary condition for capitalism' (Negri 1992: 78; 1996: 216; 1989: 81; cf. also Hardt and Negri 2000: 294).39

We can see now how Negri at once continues, and radically departs from, operaismo's project. Panzieri and Tronti removed the possibility of thinking the relative autonomy of technical, social, or political spheres, and instead described a universal plane of capitalized production throughout the social factory. Negri continues operaismo's concern with a universal plane of production, and is not shy of showing his disdain for the neo-Gramscian thesis of the relative autonomy of the socio-political (cf. Hardt and Negri 2000: 451). At the same time, however, the essence of the social factory thesis "” the immanence of capital to all social relations "” seems to vanish, as Negri both reintroduces the orthodox separation between forces and relations of production which Panzieri had been so keen to undermine, and begins to produce a strange inversion of the neo-Gramscian thesis whereby it is the realm of production which becomes autonomous. Thus, though Negri oscillates between seeing the communist multitude in forms of work and in forms of resistance, essentially the resistance becomes nor. so much a refusal of work (for 'work' has in a sense been overcome), but an affirmation of the collective embodiment of immaterial and affective labour: 'In effect, by working, the multitude produces itself as singularity' (Hardt and Negri 2000: 395; emphasis added).40 At his extremes Negri (1989: 79) even favours labour-market deregulation (as if 'deregulation' was not always a process of intricate regulation) to enable the development of this potential, and turns away from the refusal of work in a variant of the old council communist theme of 'self-management' (cf. Hardt 1994: 227), as a 'reappropriation of the social essence of production ... to ensure an ever-richer reproduction of accumulated immaterial labor' (Negri 1996: 221).41

This is not to say that Negri dismisses the category of exploitation. He writes that this socialized work is 'inextricably and emotionally linked to the principle characteristics (exclusion, selection, hierarchy) of the labour market' (1989: 47), and that this 'does not mean mocking the reality of exploitation' (Negri 1994: 235).42 But in so far as the multitude tends toward autonomy, exploitation becomes increasingly 'external' and 'empty' (238): 'capitalist power dramatically controls the new configurations of living labor, but it can only control them from the outside because it is not allowed to invade them in a disciplinary way' (235). It thus becomes increasingly unclear what exactly exploitation is.

The minor as majority

This problem of an autonomous multitude working its way to communism is highlighted most starkly in Negri's approach to Deleuze and Guattari's figure of the minor. In conversation with Deleuze (N: 169"”76), Negri asks if in 'communication society'43 the communism of the 'Fragment' as the 'transversal organization of free individuals built on a technology that makes it possible' is 'less Utopian than it used to be' (174). He also raises the possibility that, though domination becomes more perfect, perhaps 'any man, any minority, any singularity, is more than ever before potentially able to speak out and thereby recover a greater degree of freedom' (in N: 174). Though posed as a question, this is clearly a presentation of Negri's general argument.

Deleuze responds, however, by making a very different point. He suggests that instant communication is less concomitant with communism than with the intricate feedback mechanisms of the open spaces of 'control' (see below), and says that speech and communication are 'thoroughly permeated by money - and not by accident but by their very nature' (175), such that 'The quest for "universals of communication" ought to make us shudder'.

Despite this father stark difference in position, Negri (1998: n.p.) elsewhere suggests that the politics of the socialized worker is related to Deleuze's understanding of the minor. But whilst in Deleuze the minor is premised on cramped, impossible, minority positions where social forces constrain movement, Negri reads it as a figure of plenitude and majority. Perhaps recognizing the difference in their interpretations, Negri (1998: n.p.) says that Deleuze and Guattari's concept of the minor was a recognition of the socialized worker and the multitude, but that 'from the point of view of phenomenological analysis' the 'sociopolitical definition given in A Thousand Plateaus does not really go much further than this'.44 Then, taking the minor in a very different direction to Deleuze, he suggests that it contributed to a 'new concept of the majority' of the autonomous multitude. And, most strangely, in another essay Negri links this multitude to Deleuze's (and Foucault's) typologies of abstract machines and diagrams as a seemingly inevitable mode of communist democracy arising out of post-disciplinary control society:

According to Foucault and Deleuze, around this final paradigm [control/ communication] there is determined a qualitative leap which allows thinking a new, radically new, order of possibility: communism. If in the society of sovereignty democracy is republican, if in the disciplinary society democracy is socialist, then in the society of communication democracy cannot but be communist. Historically, the passage which is determined between disciplinary society and the society of communication is the final possible dialectical passage. Afterwards, the ontological constitution cannot but be the product of the multitude of free individuals.

(Negri 1992: 105)

Abstract machines and the capitalist BwO

I showed in Chapters 2 and 3 how Deleuze's minor politics operates in a very different way to Negri's presentation. Now I want to turn to consider how Deleuze's understanding of capital presents a different plane of production to Negri and his understanding of autonomy-in-production. For Deleuze and Guattari, the capitalist socius operates as an ever mutating 'abstract machine', 'megamachine', or 'Body without Organs' (BwO). Anti-Oedipus provides a 'universal history' of three types of abstract social machine: primitive/savage territorial, barbarian despotic, and civilized capitalist. What defines each social machine is its mode of composition through three syntheses (connective, disjunctive, and conjunctive) (cf. Ch. 2 note 12) by which the whole and its parts operate as a socius (AÅ’: 33). The question, following Nietzsche's Genealogy of Morals, is one of the territories and codes by which each social machine engineers its material flows in specific relations to fashion a 'memory' (though 'memory' does not have to be particularly 'deep'; cf. Ansell Pearson 1999: 217"”18) of corporeal, incorporeal, technical relations for the human:

The social machine is literally a machine, irrespective of any metaphor, inasmuch as it exhibits an immobile motor and undertakes a variety of interventions: flows are set apart, elements are detached from a chain, and portions of the task to be performed are distributed.

(AOE: 141)

Despite the clear intonations of linearity, Deleuze and Guattari's universal history describes abstract social machines not by their temporality, but by their mode of operation. That is, they 'define social formations by machinic processes and not by modes of production (these on the contrary depend on the processes)' (ATP: 435), where 'modes of production' ate dated concrete configurations (though I would suggest that this should be seen as an addition to, rather than negation of, Marx's method, which is equally more concerned with modes of composition than dated histories). And further, each concrete form is always a composite of different abstract social machines "” the abstract machines are coexistent 'extrinsically' (they all interrelate "” even the primitive socius, following Clastres (1989), has to 'ward off the state) and 'intrinsically' (each machine can be taken up into another machinic form, like, for example, the return of the despotic Urstaat or 'empire' in the capitalist socius) (cf. ATP: 435"”7, 460). Thus, in a sense, the abstract comes before the concrete, and within the concrete we can always find a coextensive functioning of different abstract machines.45

The capitalist abstract social machine is fundamentally different from the 'primitive' and 'despotic' abstract social machines in that it functions not by codes (coding and overcoding material flows) but on codes (decoding and deterritorialization) - this is its 'most characteristic' and 'most important tendency' (AÅ’: 34). The two principal flows that are brought into conjunction are the deterritorialized and unqualified worker 'free' to sell his labour capacity (no longer coded as slave or serf), and decoded and unqualified money (no longer determined as merchant or landed wealth) capable of buying labour power. But this in itself is not sufficient a description. After all, all social machines operate some form of decoding and deterritorialization. There are two marked differences with the capitalist socius. First, it is characterized by a generalized and continuous process of decoding and deterritorialization. '[C]apitalism has a very particular character: its lines of escape are not just difficulties that arise' - as they are in other social machines - 'they are the conditions of its own production' (Deleuze, In Guattari 1995a: 66-7).46 This is because there is no particular structural regime, authority, or configuration of life to maintain, but a single objective of 'production for production's sake'. The 'essence of wealth' is no longer a concrete objective thing, but 'the activity of production in general (AÅ’: 270).47 Second, concomitant with this deterritorialization and decoding is a simultaneous and continuously reconfiguring process of reterritorialization and recoding, for 'production in general' does have a purpose "” the self-expansion of capital, the maximization of 'surplus value' from the expansive potential of life.48 For the creation and realization of value (utilization of existing capital, commodity consumption, reinvestment in new capital, and profit), there needs to be a form of control, measurement, and organization that determines and creates particular forms (such as 'the worker', 'the capitalist', 'the consumer') immanent to this abstract production. As I showed in Chapter 3, the capitalist socius is thus necessarily populated by, or it 'miraculates' (AÅ’: 144) at every moment, particular determined forms or identities. We could call these identities 'codes', as in previous social machines, except that through the continual process of de/reterritorialization and de/recoding they are forever changing and only exist immanently to their function (they have no 'coded' objective predetermination). Instead, they are the product of a new means of drawing relations as 'conjunctions' or 'axioms'. In the axiomatic process, intrinsic (more 'internalized') codes are replaced by a plethora of immanent (more 'surface') abstract relations and resonances which traverse the socius, but which have no essence, rules, or meaning beyond their immediate relation, and what is functional to them: 'the axiomatic deals directly with purely functional elements and relations whose nature is not specified, and which are immediately realized in highly varied domains simultaneously' (ATP: 454). The capitalist social machine, then, unlike the other abstract social machines,

is constituted by a generalized decoding of all flux, fluctuations of wealth, fluctuations of work, fluctuations of language, fluctuations of art, etc. It did not create any code, it has set up a sort of accountability, an axiomatic of decoded fluxes as the basis of its economy. It ligatures the points of escape and leaps forward.

(Deleuze, in Guattari 1995a: 67)

This axiomatic process is enabled through the transformation of particular wealth-creating practices, forces, and forms into an abstract or universal form of wealth "” 'abstract labour' "” through the medium of money.49 Money is the general equivalent that enables the commensurability of all activity, and, because it can be accumulated, the potentiality of boundless surplus and production beyond that immediately necessary (AÅ’: 258-9). Any flow of labour (as an abstract quality) can then conjoin in an axiomatic 'cash nexus' in any relation with a flow of capital in ever new ways and always beget money in a fashion that is not determined by its current concrete form and is independent of any formal rules beyond simply the begetting of wealth (cf. ATP: 453).

The axiomatic process is the application or creation of ever changing 'images', 'organs', or determinate relations across the capitalist socius, which in itself is a wholly virtual and imageless Body without Organs.50 As such, it Is both the means to conjugate an infinite series of relations, and to formalize these relations at each instant so as to extract a surplus. This is how the process is 'directly economic':

The socius as full body has become directly economic as capital-money; it does not tolerate any other preconditions. What is inscribed or marked is no longer the producers or non-producers, but the forces and means of production as abstract quantities that become effectively concrete in their becoming related or their conjunction.

(AÅ’: 263)

Following Marx, Deleuze and Guattari argue that money-as-general equivalent enables not only the commensurability of all activity, but also the extraction of surplus value because it operates on two intersecting planes. 'The true economic force', the full body of capital, is the total social productivity of this process (where surplus value emerges), as money begets money in the realm of financing. The other plane is the reterritorialization in (at any one time) axiomatized subjects that receive 'impotent' money as payment for work done in an individualized quantitative valuation. The two planes necessarily function in tandem because capital needs to exceed itself in a continual maximization of surplus value, and realize itself at any given moment in the maintenance of existing value. To realize itself, everybody must be invested in the system, receiving some form of 'wage' (impotent money) and concomitant identity from their contribution to the total process (be this from work done in a field or factory, from managing a substation of a business, from share ownership, or from 'indirect wages' "” state benefits, 'family wages', and so on). Given the way all are formed by, and invested in, the socius, Deleuze and Guattari (AÅ’: 253) argue that there are not two classes which face each other, but 'only one class, a class with a universalist vocation' (a class they name the 'bourgeoisie',51 but it is easier to think of it as a generalized capitalist class):

there are no longer even any masters, but only slaves commanding other slaves; there is no longer any need to burden the animal from the outside, it shoulders its own burden. Not that man is ever the slave of technical machines; he is rather the slave of the social machine . . . [T]here is only one machine, that of the great mutant decoded flow - cut off from goods -and one class of servants, the decoding bourgeoisie, the class that decodes the castes and the statuses, and that draws from the machine an undivided flow of income convertible into consumer and production goods, a flow on which profits and wages are based.

(AÅ’: 254-5)52

'Class', then, signifies the decoding and deterritorialization of castes and status groups in a fashion that is functional to capitalist expansion (cf. also AÅ’: 344). Inasmuch as capitalism functions across the social whole, it continually breaks down any fixed identity or group: 'the very notion of class, insofar as it designates the "negative" of codes .. . implies that there is only one class' (255). Deleuze and Guattari arc, of course, not saying that we are ail equal. It is not difficult to demarcate groups of people on a global scale in terms of how they accrue money for their practices, with super-exploitative and poverty wages (or no wages) on one side and profit derived from surplus value on the other. Indeed, a fundamental of capitalist axiomatization and accumulation is the intimate striation and segmentation of social groups "” a process that the ever smoother flow of global capital serves, by intent, to proliferate and maximize (cf. ATP: plateaus 13 and 14). Essentially, however, all are axiomatized manifestations of the abstract process (and hence politics, as I argued in Chapters 2 and 3, emerges with minority problematizations rather than distinct groups, and the proletariat is a mode of composition immanent to, and against, capitalist configurations, rather than a subject which 'faces' the bourgeoisie).

Inasmuch as all art miraculated from the plane of the cash nexus and are, hence, invested in the project of the maximization of surplus value, capital has no need for an overall belief system, or, to use Marxian terms, 'ideology' "” the capitalist socius is strictly amoral (AÅ’: 250; cf. Holland 1999: 21, 80). It is not a question of how the populace is tricked at the level of the superstructure into investing their interests in the system, but how they are composed, axiomatized, or inscribed in the system as a whole.53 This is not to say that the capitalist socius does not produce ideas which mystify its workings (Deleuze 1994a: 208), that the functional integration of markets, geopolitical governance, exploitation and death does not tend to be 'hidden' from view (cf. Bordiga 2001), or that it is not populated by the most inane and oppressive systems of belief. The point is, rather, that the generalized functioning of the system, with its naked and open cash nexus is without secrets: 'nothing is secret, at least in principle and according to the code (this is why capitalism is "democratic" and can "publicize" itself, even in a juridical sense)' (Deleuze, in Guattari 1995a: 55).

To say that the capitalist axiomatic system operates on the level of abstract quantities and is composed of one class is not to say that it does not produce subjects. Deleuze and Guattari distinguish two subjective forms, both of which operate simultaneously in the capitalist axiomatic "” machinic enslavement and social subjection. Machinic enslavement produces an integrated machine of human, animal, and tool subject to a higher unity (the despotic state-form is the first example, but Marx's productive 'automaton' could be another), whereas social subjection isolates the human from the machine to become itself the higher unity ('The human being is no longer a component of the machine but a worker, a user. He or she is subjected to the machine and no longer enslaved by the machine'; ATP: 457). In the capitalist socius, the functioning of the axiomatic through abstract quanta (turning a force into a determined comparable conjunction) is the element of machinic enslavement, and the production of the molar aggregate out of this (the personified capitalist, the worker) is the social subjection. But at each moment one simultaneously experiences subjection and enslavement. An adaptation of Deleuze and Guattari's (ATP: 458) example of television can exemplify this. The worker is subjected inasmuch as s/he is subject to the statements 'you must work ... to survive/for the good of your soul/to display your working class nobility/to contribute to society' (where the statement is a material compulsion) and enfolds this as a subjective core with the enunciation 'I am a worker and it is good for me.' And the worker is enslaved inasmuch as s/he is a series of component quanta reconfiguring in the machinic automaton of capital. Spivak (1996: 122) illustrates the co-functioning of these two forms when she writes: 'It is a paradox that capitalist humanism does indeed tacitly make its plans by the "materialist" predication of Value' "” what I am describing as the machinic enslavement of labour power "” 'even as its official ideology offers the discourse of humanism as such.'

Societies of control

If this is the general axiomatic process, in 'Postscript on control societies' (AT: 177"”82) Deleuze makes some specific comments about the operation of contemporary axiomatic processes in the time of real subsumption, or what Deleuze calls "” following William Burroughs54 "” 'control'.55 Deleuze argues that we are witness to the breakdown of the relatively distinct spaces of Foucault's (1991) disciplinary enclosure.56 Discipline is based on the double figure of individual and mass, where each site of disciplinary enclosure both disciplines and maximizes collective energies and produces individual identities appropriate to that enclosure. Though discipline has a general consistency, each confinement has its own type of mass and individuality. The subject traverses different sites of enclosure, being a subject of the function of worker, prisoner, patient, student, and so on, in series. With the emergence of control, there is a movement away from this thermodynamic model of the ordered dispensation of energy in discrete spaces of enclosure "” family, school, army, factory "” to a more general cybernetic model of what Massumi (1998: 56) calls 'unleashed production', with a varying overlay of each disciplinary technique across social space.57 Rather than discrete 'moulds' (in each enclosure), there is a continuous variation or 'modulation' of activity. Discrete and coherent analogical individuals and masses are thus replaced with much more fluid and digital 'dividuals' which are in a 'superposition', caught in overlapping series of different 'self-transforming' and metastable configurations, and subject simultaneously and in varying ways to a multiplicity of controlling and productive mechanisms, such that, as Joseph K testified in Kafka's (1953) The Trial, one is never 'done' with anything. The expression 'dividual' is important in emphasizing that the self-autonomy of the individual (the 'subjection' of discipline) is breaking down into a subdivided series of changing capacities, possibilities, and limits in each modulation ('enslavement'). At any one moment, of course, there are precise mechanisms of dividualized 'identity'. In these metastable configurations, the contours of the dividual are modulated through continuous absorption and feedback of information across 'data banks' - including agencies such as police, social work, and psychiatry, as well as consumer profiling, and credit assessment processes (cf. Rose 1999a: 260).

For Deleuze, control is both an extension of discipline, a kind of permeation - Massumi (1998: 56) describes it is a 'release' of discipline across the social -and also something new that is directly related to a post-Second World War 'mutation of capitalism' (N: 180). Rose (1999a: 234) has warned against reading control in epochal terms since, like all Deleuze's abstract machines, it is a mode of 'configuration' rather than a specific spatio-temporal system, and hence always operates in conjunction with other configurations. Indeed, Deleuze sees Kafka's work, at the turn of the twentieth century, as straddling discipline and control (hence the superpositions of The Trial "” the self-transforming labyrinths that emerge within apparently distinct disciplinary territories such as the court-house, and the endless postponement of the verdict "” are control experiences).58 Nevertheless, Deleuze does also specifically link control to some pervasive features of post-war capitalism: the end of the gold standard and the emergence of floating exchange rates (N: 180), and a form of capital based not on production and proprietorship, but on businesses, services, administrators, and computers. In many ways 'business' becomes the societal-wide technology, much as the Panopticon was the visible technology of the abstract machine of discipline. Deleuze is suggesting not so much a 'social factory', but a 'social business'.59 'Capitalism in its present form' is

essentially dispersive, with factories giving way to businesses. Family, school, army, and factory are no longer so many analogous but different sites converging in an owner, whether the state or some private power, but transmutable or transformable coded configurations of a single business where the only people left are administrators.

(N: 181)

Machinic surplus value

I will return to the 'social business' below, but first I want to consider the nature of the labour and 'value' of control. I have shown how Marx raised the question of a different content of activity as the general intellect and the social individual, and how Negri tried to see it as the near actualization of communism within the regimes of an affective and biopolitical labour that had escaped the law of value. Deleuze and Guattari explicitly address this question around what they call 'machinic surplus value' but, unlike Negri, they firmly situate it within the capitalist framework of axiomatics and control.

In a fashion that at times seems to tie in with Negri's thesis chat we are heading beyond the labour theory of value, Deleuze and Guattari write that the capitalist socius depends increasingly less on the extraction of a surplus of labour time and quantity than on a 'complex qualitative process' (ATP: 492; emphasis added). On closer inspection, however, they are more 'Marxist'. In Anti-Oedipus Deleuze and Guattari suggest, much like section [A] of the 'Fragment' (and they reference the text at this point), that alongside conventional 'human' surplus value, machine-rich production sees the emergence of a 'machinic surplus value' of constant capital (AÅ’: 232) that is the product of an 'intellectual labor distinct from the manual labor of the worker' (233). But this is a mistake in their own terms, in that it seems to present machines and humans as distinct entities rather than, as they always insist, products of a social machinic process, and it makes a split between intellectual labour ('machinic surplus value') and manual labour ('human surplus value') that makes little conceptual sense, in that it seems to exclude intellectual labour from the realm of the human.60 This, however, is not fundamental to Deleuze and Guattari's argument, and it is later rectified. The fundamental point of Anti-Oedipus's discussion of surplus value is that the aggregate of the two surplus values is a 'surplus value of flux'. In A Thousand Plateaus surplus value of flux is replaced with 'machinic surplus value' (this time defining the product of the machinic whole rather than just constant capital and intellectual labour), where it has two (fully interrelated) senses. First, it signifies not the differential between the value the worker accrues for her work and the value created by labour capacity, but the break between the two planes of capital "” the flow of the full BwO and the axiomatized identities that are its reterritorialization. Thus, the 'exploitation' of machinic surplus value is, in part, the very formation of axiomatized subjects (though, of course, immanent to this formation is the continual extraction of surplus from labour).61 Second, machinic surplus value signifies the societal-wide production of the complex, qualitative, affective, and machinic processes that the socialized worker thesis sought to describe, and the very diffuse and unlocatable nature of value in this system (ATP: 45S, 491-2).62

I will consider this second point (having developed the first in the explanation of axiomatics). Guattari's (1996a) essay 'Capital as the integral of power formations' is a useful point of focus, for it is in many ways a reading of the 'Fragment's' general intellect and social individual as categories of capitalist productivity.63 The explicit aim of the essay is to use the concept of machinic surplus value to 'lift' Marx's 'collective worker' from a category based on average labour with a generalized dispensation of energy (which can be quantitatively calculated) to one based on the qualitative intensity and variation of work. The essay has a few problems,64 but it emphasizes the important point that a qualitative variation exists in the content of value-productive activity beyond a simple definition of work and work time. Guattari argues that 'it is complex arrangements "” training, innovation, internal structures, union relations etc. "” which circumscribe the magnitude of capitalist zones of profit, and not simply a levy on work-time' (1996a: 205), and that capitalism actualizes productive socio-economic forms in varied desires, aesthetics, ecologies, and so on (1995b: 55). Unlike Negri, who sees this increasingly complex and varied activity as tending towards an autonomy from capital, Guattari argues that capital still operates as the universal plane "” the 'integral' "” of these different 'universes of value'. That is, each of the different universes of value is subsumed in capitalist general equivalence (Guattari 1995b: 54-5) and valorized through a 'machinic phylum which traverses, bypasses, disperses, miniaturizes, and co-opts all human activities' (Guattari 1996a: 207). Guattari does detect a move towards a tremendous multiplication of activities, but, if subsumed in the integral of general equivalence, these activities and their machinic surplus value are not produced autonomously in society. Rather, machinic surplus value is 'miraculated', or created by the socius as it is needed (AÅ’: 144), becoming a 'required' machinic surplus value (cf. Guattari 1996a: 208). There is no play of autonomous creativity and capitalist recuperation (cf. AÅ’: 337-8). Instead, as Massumi (1998: 57) puts it, 'Control involves the assimilation of powers of existence, at the moment of their emergence', and at each stage there is the axiomatic normalization of each new form, where '"Normal" is now free-standing.' As Camatte (1995: 43) argues, capitalized human activity has escaped any fixed base value such that 'human beings are fixed to its movement, which can take off from the normal or abnormal, moral or immoral human being'. Capital is thus dependent on an increasing degree of differentiation, innovation, and variation in social practice. Dividuals are not just 'normalized', but maintained with a certain degree of what we could call functional difference such that the thresholds of knowledge and practice are rather radically open, and are always being reconfigured (cf. Rose 1999a). The lines of flight, which might be experienced as entropy in disciplinary space, here become the driving force of production. New aspects of social productivity might escape for a little while. Indeed the capitalist socius has many little lines of flight, even autonomous zones where creation is allowed to operate outside of capitalist relations of productivity (Anti-Oedipus offers an image of the mad scientist creating on the fringes and, we could add, at a different level, new milieux of subcultural and countercultural innovation, and social and political 'danger' as examples)65 before they are generalized as a new productive activity; but such spaces (or lines of flight) enrich rather than contradict capital "” at least in normal functioning.

Again the 'business' is the archetype. Deleuze writes that in disciplinary production discrete amounts of energy were extracted in the factory and costs were reduced, but in control we see a buying of 'activities' and a 'fixing of rates'. Deleuze thus writes that

the factory was a body of men whose internal forces reached an equilibrium between possible production and the lowest possible wages; but in a control society businesses take over from factories, and a business is a soul, a gas. There were of course bonus systems in factories, but businesses strive to introduce a deeper level of modulation into wages, bringing them into a state of constant metastability punctuated by ludicrous challenges, competitions, and seminars.

(N:179)

'Marketing' lies at the centre of the business's 'soul'.66 But Deleuze is not proposing that marketing is a distinct practice circumscribed in a single social group. Rather, marketing is a sign of the business's free floating ability to discern and require a wealth of activities through its permeation and intimate control of social life, and its understanding of the variation and potential of activity that its 'data banks' provide. Inasmuch as we all become part of the business, marketing can also be seen to be a generalized feature of social activity "” a necessary attribute of the 'dividual'. Making this case Lazzarato (1996: 142) writes that what he calls 'communication' (marketing, production of cultural content such as fashion and taste, consumer feedback mechanisms, public opinion) is enmeshed in 'the post-industrial commodity' such that it 'is the result of a creative process that involves both the producer and the consumer'. Capital still operates through the enforced splitting, or axiomatization, of producers and consumers (134), bur the flows and relations of production are continually enriched through processes outside of the immediate sphere of work: 'the product is enriched through the intervention of the consumer, and is therefore in permanent evolution' (142). This 'communication', Lazzarato argues, emerges in a condition of intimate axiomatization where 'one has to express oneself, one has to speak, communicate, cooperate, and so forth' (135) and hence leads to a situation where every aspect of subjectivity - as it is expressed in frameworks which must be 'clear and free of ambiguity' (135) "”becomes productive of value (143). But this is not only a 'subjective' phenomenon. As Massumi (1996) and Morris (1998) have indicated, this process of 'communication' or 'work' also occurs in a subhuman fashion, as the communication and axiomatization of 'affect' or intensity immanent to particular machinic environments (witness the use of bio-feedback mechanisms in focus-group research).67

Contemporary machinic work

To draw this chapter to a close I want to return to the question of machine/ human relations in a brief sketch of the general plane of contemporary forms of work. If we follow Marx's emphasis on the 'organism' and 'automaton' of the capitalist socius upon which technical and human parts are engineered (rather than the dichotomy he presents in section [A] of the 'Fragment' between machines and humans), we can conceive the process of production in terms of a series of machinic assemblages which traverse global social space and the, increasingly fluid, division of 'work time' and 'free time'. Each would be composed of varying quantities of technical and human parts, where in each instance the societal-wide competences, languages, knowledges, physical forces, affects, interactions, skills, expertises are present in different degrees in the worker and the technical machine, and where each would maintain a 'social' productivity, regardless of whether they immediately contribute to what we conventionally call work. As Guattari (1996a: 209) puts it, in this framework fixed capital, variable capital, and free time are interlaced in particular 'machinic environments', where the whole ensemble of forces, relations, and affects in each environment are axiomatized and produce a machinic surplus value: 'exploitation concerns machinic arrangements at first - man and his faculties having become an integral part of these arrangements.'68 And even when 'Machines in the factory seem to be working all by themselves ... in fact it is the whole of society which is adjacent to them' (212). Lazzarato thus proposes that production under the general intellect functions not as a machine-based automated system but more as a societal-wide machinic system, ever coming into being, and dispersing again:

This immaterial labor constitutes itself in forms that are immediately collective, and we might say that it exists in the form of networks and flows. The organization of the cycle of production of immaterial labor . . . is not obviously apparent to the eye, because it is not defined by the four walls of the factory. The location in which it operates is outside in the society at large . . . The cycle of production comes into operation only when it is required by the capitalist; once the job has been done, the cycle dissolves back into the networks and flows that make possible the reproduction and enrichment of its productive capacities.

(Lazzarato 1996:137)

This framework has the advantage of accounting for the great mutability and flexibility of the plane of contemporary work. We can envisage examples of these contemporary machinic work regimes (of the most modern and traditional kinds), from a fully automated car plant at one end, fitting well with Marx's 'watchman' thesis, to an advertising industry brain-storming session, an Export Processing Zone garment-making sweatshop, the key-tap-regulated keyboard, the hourly labour contract, the zero-hours contract, the Research Assessment Exercise and 'vocationalism' in higher education, workplace drug tests, office telephone call-time monitoring and e-mail regulation software, housework, 'jobseeking', career opportunity maximization, work-based self-actualization workshops, neo-Puritan ethics, sheer poverty-driven overwork, to, indeed, consumer-feedback mechanisms and correctly competent fashion-conscious consumption. Rather than one general portfolio of skills that may be employed in the narrative of a single career or job over a lifetime, these myriad machinic work regimes would pick up, incorporate, and manifest a whole series of different competences and attributes at different times. In these regimes, one's lifestyles, ethics, even rebellious identities, and one's consumption and reproduction patterns become directly productive as generalized potential, actualized in varying specific enactments of work. And within work time, or the quantitative basis for a wage, vastly different, varying, and expansive qualitative skills, knowledges, competences, relations, interactions, disciplines, languages, and skills may be actualized. Here, it matters as much that workers work on themselves (optimize their skills, and deploy and feedback their knowledges and capacities in each axiomatized work relation) to enable the productivity of vastly complex assemblages, as they 'put in their time'69 (even though labour time retains a continued role as the "” albeit modulating rather than fixed - measurement of the ('impotent') 'value' of the system that the worker accrues to herself). Such work thus requires a population to be perpetually able to reskill, self-scrutinize, and modulate its demeanour, skill, aptitude, and competence. As an example, this process is particularly evident in the emphasis on training and pursuit of jobs for the British unemployed, now receiving their 'Jobseekers Allowance' and their 'New Deal' on the basis that they are always ready, prepared, and preparing to be propelled into productive arrangements.70

We should not infer from all this that we have left behind the extremes of workplace enclosure and control. Whilst, as Guattari and Alliez (in Guattari 1984: 286) argue, the 'management of productive space now becomes the arrangement of its optimal fluidity', in the age of multinational subcontract-ing and outsourcing the extremes of 'post-industrial' infotech employment and the nineteenth-century sweatshop are fully interfaced "” often in one and the same 'subject' (cf. Caffentzis 1997; Lazzarato 1996: 137; Ross 1997). It is the relative stability of enclosure that is seen to be disappearing in a more fluid, axiomatic, and socialized model of work that is characterized by 'Precariousness, hyperexploitation, mobility, and hierarchy' (Lazzarato 1996: 137). At the same time - and in the midst of the increasing pervasiveness of the discourse of 'pleasure in work' (Donzelot 1991; cf. Leadbeater 2000; Reeves 2001) - the compulsion to varied and self-optimizing activity in the 'basin' of contemporary labour, and the, for many, chronic lack of access to a regular and sufficient wage, induces affective conditions of anxiety and competitiveness in what Bifo (n.d.) has described as the contemporary 'factory of unhappiness'.

Conclusion

Taking the model of production that emerges in operaismo and autonomia as its focus, this chapter has explored the operation of the capitalist socius as it is developed in Marx, Panzieri, Tronti, Negri, and Deleuze and Guattari. I started by showing Marx's machinic understanding of the relation between technical machines and humans in his 'real subsumption' thesis "” a position that is central to the work of Panzieri and Tronti. The chapter then showed that against orthodox Marxist understanding of a neutral force of production and the neo-Gramscian presentation of the relative autonomy of the sociopolitical, operaismo developed a rather cramped, minor knowledge of the plane of production in that it allowed no space for a coherent and autonomous people that could exist within the models of socialist planning or social democracy, but compelled an intensive investigation of the productive forces of the social factory. I then explored Marx's 'Fragment on Machines' - a text of great import in the development of operaismo and autonomia "” and showed how it stretched to understand the development of production into the realm of 'general intellect' and the 'social individual'. At one level this is a Marx that points to an increasingly complex machinic form of production that helps extend the social factory thesis to include general intellect-rich production, and, to use Foucault's and Deleuze's figures, seems - in the midst of disciplinary society "” to discern the coming diabolical powers of the society of control. On the other hand, perhaps evidencing some of the constraints of thinking beyond one's own social regime, Marx suggests "” in a fashion which actually goes against his analysis of the tendencies of real subsumption "” that the powers of general intellect may emerge outside of work in a productive autonomy that presents a fatal contradiction for capital. In Negri's analysis of the 'Fragment' and his development of the socialized worker thesis this tension remains. On one side there is a concern with the intricacies of a capitalized affective and immaterial labour, such that the politics of hegemony is still dismissed as a misrecognition of capitalist regimes of control. But, on the other, Negri breaks with operaismo's, Marx's, and Deleuze's understanding of the immanence of controlling regimes to productive forces in a certain inversion of the neo-Gramscian thesis whereby it is the realm of production which tends toward autonomy. In this 'self-determined production' Negri is right to return to the question of the centrality of work and production, and he is careful to elaborate a potential 'multitude' rather than a present people, but this does not prevent him from discerning an emerging communist subject which has overcome the law of value, and seems to produce its singularity through its work, in an almost inevitable process which 'cannot help revealing a telos, a material affirmation of liberation' (Hardt and Negri 2000: 395).

The problematic aspect of Negri's thesis was seen to be particularly apparent in his interpretation of Deleuze and Guattari's concept of the minor, which was presented not as a cramped and complex mode of engagement, but as an emerging plenitude. The chapter then used Deleuze and Guattari's understanding of the capitalist socius, axiomatics, class, control, and machinic surplus value to see how one can understand the socialized worker and the new attributes and networks of production as emerging within capitalist relations, and being intricately controlled by them. Contrary to Hardt and Negri's (2000: 28, 368) proposition that Empire has overcome the last vestiges of metaphysical thinking in Deleuze, Guattari, and Foucault, this section has sought to show that Deleuze offers a complex and productive conception of contemporary capital, control, and production, and one that actually resonates more with the conceptual constellation of operaismo and Marx than it does with Negri's tendencies to present an emerging autonomy-in-production.

Notes

1. The translation of operaismo as 'workerism' is, as Hardt (1990a: 249) points out, problematic: 'The English usage of "workerism" and the French "ouvrièrisme" correspond to the Italian "fabrichismo" in that they are used pejoratively to designate those who cannot or will not recognize the power of social struggles outside the factory. The characteristic of "operaismo" is that it has been able to transform itself in step with the changing nature of work.'
2. Zizek's (2001) more recent comments on Empire have been more critical.
3. In this context it is noteworthy that Negri (1998: n.p.) indicates that he has some knowledge of the intended focus of Deleuze's unfinished book on Marx. Whatever access Negri had to Deleuze's ideas in progress, he presents Deleuze's argument, I think problematically, in terms very similar to his own, as a communism of the multitude: 'Here there is the multitude that constitutes the common. And this is the concept of communism that, from what I have understood, was constructed in the "Grandeur de Marx", Deleuze's unfinished book.'
4. An earlier English language engagement with Negri, operaismo and autonomia emerged in the more expressly political milieu of the Red Notes group (cf. Red Notes 1977, 1979, 1981; Italy 79 Committee 1982; Negri 1988a), and in the US journals Zerowork and Midnight Notes, and the work of Harry Cleaver (1979).
5. Thus, whilst I would agree with Wright (2002) that a simplification of the complexity of operaismo and autonomia and the over-alignment of this current with Negri owes something to Negri's reception through Deleuze and Guattari, a 'melange', as Wright (2002: 2) puts it, of Deleuze and operaismo can also function in a productive way, and without either simplification or subsumption of complex positions to Negri's perspective.
6. Operaismo and autonomia maintain a persistent presence in Deleuze and Guattari's work. A Thousand Plateaus, for example, cites Yann Moulier, Tronti, and Negri in the context of new forms of socialized work, the emarginati, the problem of the Subject of orthodox Marxism, and the refusal of work (cf. ATP: 469, 571"”2). Guattari wrote an essay with Negri (1990) and had some involvement with autonomia (cf. Guattari 1980a, 1980b; and Semiotext(e) (1980: 133) for a photograph of Guattari in Radio Alice's studio). He also wrote the preface to Collectif A/traverso (1977). As well as signing the petition against the repression of autonomia, along with Sartre, Barthes, and Foucault, amongst others (cf. Red Notes 1978: 36"”7), Deleuze (1980) wrote a letter against Negri's imprisonment in 1979, a preface to the French edition of Negri's The Savage Anomaly, and a review of Marx beyond Marx (Deleuze n.d.c) that was forwarded as a proof of Negri's innocence. As apiece of anecdotal evidence of the influence of Deleuze and Guattari in autonomia, Liberation reported that a student questioned in France about Franco Piperno (who had fled Italy to escape imprisonment) was asked if he had read Anti-Oedipus (Massumi 1987: 71).
7. There are, of course, exceptions. It is notable that Meaghan Morris, who is generally critical of cultural studies' tendency to populism and the neo-Gramscian politics of hegemony, wrote an excellent account of autonomia and its relations with the PCI as early as 1978. Paul Gilroy (1982), equally critical of the neo-Gramscian vein in cultural studies, also draws on some of the insights of autonomia.
8. Whilst the theoretical and political tendencies of operaismo pushed well beyond the PCI, the current maintained a relation to the orthodox left, due both to a reluctance to develop as an independent faction, and to a sense of the possibility of radicalizing the rank and file, and even the party itself. For Tronti, in particular, the party was of central political importance, and his struggle to save it from social democracy saw him eventually return to the fold (cf. Wright 2002: 68"”75; Piotte 1987: 28). If the critique of the functionality of the PCI to the incorporation of working-class struggle was a while coming, the struggles around '68 saw a change of position, from where the PCI was to have no place in operaismo's and, later, autonomia's politics (cf. Wright 2002: 110-14).
9. See Bologna (1980a) for a short account of the variations of position, Piotte (1986) for relations between Tronti and Negri, Wright (2002: 58"”62) for the split between Panzieri and Tronti, and Wright (2002: 141-51) for the tension in Potere Operaio between Negri, Piperno, and Scalzone.
10. Despite a continued flirtation with Leninism, Potere Operaio dissolved following a meeting in Padua in 1973 saying: 'We have rejected the logic of the political group in order to be within the real movement, in order to be within organised class autonomy' (in Red Notes 1979: 32). Bifo (1980: 151-2) suggests that following the big FIAT Mirafiori occupation earlier that year (cf. Negri 1979b), within which the revolutionary groups only had a marginal presence, Potere Operaio's dissolution showed that it was the only group to recognize the changes taking place in the movement.
11. 'Results of the immediate process of production' (sometimes known as 'the missing sixth chapter') was first published in 1933 in German and Russian, but took on particular importance - especially for the Italian and French extra-parliamentary communists "” when it was republished in other European languages in the late 1960s (1976 in English).
12. Pioneered by Romano Alquati operaismo adopted Marx's method of the 'Workers' Inquiry' (cf. Marx 1973d) as a means of 'hot investigation' into the conditions and forms of resistance in the factories (cf. Bologna 1991). The workers' enquiry enabled the operaists to develop analysis from close attention to a social sphere which itself embodied a considerable degree of political, tactical, and organizational sophistication developed through the collective experience of the workers' movement since the Resistance (cf. Bologna, in Cuninghame 2001). Ironically, as Moulier (1989: 14) reports, these 'hot investigations' were the object of considerable interest from the employers who found they gave more insight to the functioning of their factories than conventional studies.
13. The practicality of operaismo's position is evidenced by Moulier's (1989: 13) anecdote that the bedroom walls of activists saw the substitution of diagrammatic maps of the FIAT Mirafiori factory for the epinal figures of Mao and Che Guevara.
14. As Rosenberg (1982: 36) points out, this accusation usually follows a citation from The Poverty of Philosophy, where Marx writes 'The handmill gives you society with the feudal lord; the steam-mill, society with the industrial capitalist'.
15. This is not to deny the possibility of specifically technological innovation, but it is to say that it is only as an expression of particular social problematizations, possibilities, and lines of flight that a technological innovation could be possible, and maintain any consistency. To cite a passage from Marx (1970: 21) that Deleuze is fond of using at these moments, 'Mankind . . . inevitably sets itself only such tasks as it is able to solve, since closer examination will always show that the problem itself arises only when the material conditions for its solution are already present or at least in the course of formation' (cf. Deleuze 1994a: 186).
16. Deleuze and Guattari (1977) at one point suggest that Marx does not always present such a machinic conception of the relation between the technical machine and the human, but, rather, that he sees the machine as a succession to the tool in an evolutionary understanding of the human biological organism. I am arguing, however, that the presentation of machines in Capital (Chs 14 and 15) and in the Grundrisse "” though occasionally displaying humanist errors "” does not present the machine in such a fashion, but rather, as Deleuze and Guattari (1977: 131) suggest elsewhere, presents 'man and the tool [as] already components of a machine constituted by a full body acting as an engineering agency'. For, as I argue below, Marx's distinction between tools and machines is not based on the notion that machines are more complex tools, but that the technical machine "” as it emerges in capitalist manufacture "” is created by, and is functional to the social configuration of capital, machining the humans and tools within itself for the maximization of surplus value.
17. Marx (1976: 490) makes these points as follows. First: 'manufacture was unable either to seize upon the production of society to its full extent, or to revolutionise that production to its very core. It towered up as an artificial economic construction, on the broad foundation of the town handicrafts and the domestic industries of the countryside. At a certain stage of its development, the narrow technical basis on which manufacture rested came into contradiction with requirements of production which it had itself created.' Second: 'the complaint that the workers lack discipline runs through the whole of the period of manufacture'.
18. 'Cooperation in its capitalist form is ... the first and basic expression of the law of (surplus) value' (Panzieri 1976: 7). This is the directly 'capitalist' process where the super-adequate power of collective labour is manifested after the sale of individual labour at its necessary price (cf. Marx 1976: 451).
19. 'The specifically capitalist mode of production not only transforms the situations of the various agents of production, it also revolutionises their actual mode of labour and the real nature of the labour process as a whole' (Marx 1976: 1021).
20. Marx (1976: 563) thus writes: 'It would be possible to write a whole history of the inventions made since 1830 for the sole purpose of providing capital with weapons against working-class revolt.' Other crucial elements of this new machinic environment, as Linebaugh (1991) and Thompson (1967) have emphasized, are the wage and the clock.
21. In the terms of Anti-Oedipus, the recording surface of the Body without Organs of capital sets its disjunctions in and through the technical machines such that they become the quasi-cause of production, and the productive desiring machines are left circulating around, and constrained within them. With the development of real subsumption, the 'productive powers and the social interrelations of labour in the direct labour-process seem transferred from labour to capital. Capital thus becomes a very mystic being since all of labour's social productive forces appear to be due to capital, rather than labour as such, and seem to issue from the womb of capital' (Marx, cited in ACE: 11).
22. Bearing in mind that there is little debate about the possible relations that could be drawn between aspects of operaismo and the earlier Italian left, it should be noted that this point was made by Bordiga in his 1953 essay 'The spirit of horse power' (in Bordiga 2001). In this trenchant critique of the Russian and Chinese states' claims to be an existent socialism, Bordiga rehearses Marx's discussion of machines to show that with the development towards production driven by 'the sinister steel automatons', 'The physical person of the individual master is ... not required, and bit by bit he disappears into the pores of share capital, of management boards, of state-run boards, of the political state, which has become (since a long time ago) entrepreneur and manufacturer, and into the very latest vile form of the state which pretends to be "the workers themselves'" (82).
23. In the second and third volumes of Capital Marx explains how, through credit and finance, initially through the formation of stock companies, capital develops into a social system that is in a sense 'social' in its ownership. Individual capitalists and separate spheres of society, all competing with each other, and necessarily not supporting an 'unproductive' (reproduction) sphere, are replaced by a mutually self-supporting system of 'social capital'. Competition is no less important, but it increasingly becomes a mechanism internal to the social whole (rather than a game between distinct players).
24. Marx (1974a: 388) puts it like this: 'But since, on the one hand, the mere owner of capital, the money-capitalist, has to face the functioning capitalist, while money-capital itself assumes a social character with the advance of credit, being concentrated in banks and loaned out by them instead of its original owners, and since, on the other hand, the mere manager who has no title whatever to the capital, whether through borrowing it or otherwise, performs all the real functions pertaining to the functioning of capitalist as such, only the functionary remains and the capitalist disappears as superfluous from the production process.'
25. The PCI is a central example. Building on its earlier politics of an 'anti-fascist' cross-class alliance, the post-war PCI was to develop with an explicit focus on formal democratic politics and working class participation in the development of national capital (cf. Partridge 1996: 76-7; Wright 2002: 8-9).
26. The first article of the 1948 Italian Constitution reads: 'Italy is a democratic republic founded on labor' (cited in Hardt and Negri 1994: 55).
27. The American journal Zerowork (1975: 6) neatly summarizes the case against socialism: 'Our analysis of the crisis implies a rejection of the basic proposal of the Left: socialism . . . [Socialism] can mean only one of two dubious things. Either, as the ideology of the libertarian Left, it finds in small-scale production the solution to the "degradation of work", or it is a capitalist strategy of economic planning. In the first respect socialism is romantic and quaintly useless. In the second respect, however, socialism means primarily disciplining the working class ... In both cases the demand for socialism clashes with the working class demands against work.'
28. Passages from the 'Fragment' return throughout Negri's work from his essays in Potere Operaio up until Empire. The importance he attributes to the 'Fragment' is clear when he writes that it is 'without doubt, the highest example of the use of the antagonistic and constituting dialectic that we can find, certainly in the Grundrisse, but perhaps also in the whole of Marx's work' (Negri 1991a: 139).
29. The 'Fragment on Machines' covers the end of Notebook VI and the beginning of VII of the Grundrisse, but the exact page references vary a little between commentators. I use Negri's (1991a) inclusion of pages 690"”712 in Marx (1973a).
30. In his excellent critique of the 'end of work' thesis, Caffentzis (1997: 30) cites a range of sources to show that in the US the work day, the work year, and the number of waged workers have all significantly increased since the 1973"”4 energy crisis (and that OECD figures are similar for the advanced capitalist world).
31. There are, thus, sections in the 'Fragment', notably at the point where Marx uses the expression 'general intellect' (706), which seern to present technology more as a generic human creation, an almost pure knowledge "” the product of the 'human hand' and the 'human brain' "” than as a functional product of specific (and, in capital, exploitative) social relations.
32. 'Socialized worker' is a translation of operaio sociale, sometimes also translated as 'diffuse worker' and 'social worker'.
33. The broader argument of Empire concerning the history and contemporary forms of global governance is beyond the scope of this chapter.
34. Negri traces this development as a direct response by capital to the effective power of the mass worker (cf. 1988b: 212-16).
35. Hardt and Negri (1994: 280"”1) give a fuller definition: 'living labor is manifest above all as abstract and immaterial labor (with regard to quality), as complex and cooperative labor (with regard to quantity), and as labor that is continually more intellectual and scientific (with regard to form). This is not reducible to simple labor "” on the contrary, there is a continually greater convergence in techno-scientific labor of artificial languages, complex articulations of cybernetics and systems theory, new epistemological paradigms, immaterial determinations, and communicative machines. This labor is social because the general conditions of the vital process (of production and reproduction) pass under its control and are remodelled in conformity with it.'
36. '[T]he more production becomes immaterial and the more it is socialized, the more labour becomes autonomous from capitalist command' (Hardt and Negri, in Brown et al. 2002:205).
37. Negri (1989: 78) describes this communicational network of activity/work as both a Foucauldian 'spatial universe' and a site of Habermassian 'communicative action'. The premise of Foucault's work is of course that micro-powers infuse the social as its very basis of constitution (cf. Foucault 1980: 94). The degrees of intensity and complexity of this are such that, contra Habermas, any talk of pure communication is a theoretical fiction (or, put another way, itself a product of a particular conjunction of power/knowledge). For Negri to utilize Foucault's image of proliferating networks as constituting a possibility for communism as an equality in communication is thus, to say the least, problematic. The idea that communism is collective control over a purified language resurfaces in Empire where Habermas is again deployed, only this time he is seen as presenting the possibilities of communicative action in a too limited fashion: '[Habermas] grants the liberated functions of language and communication only to individual and isolated segments of society' (Hardt and Negri 2000: 404).
38. Negri sees this process as an overcoming of the law of value, interpreted as a quantitative relation between labour time and price, and its replacement with a law of command' (Negri 1991a: 172; cf. also Hardt and Negri 2000: 357-8, 401). This is a reductive interpretation of the law of value, which, as Elson (1979) argues, should not be seen as a question of the price of a commodity, but of the form labour takes in capital. However, in so far as Negri suggests that production becomes determined by social needs (rather than the capitalist need for productive work) he seems to have dropped both a limited and a full concept of the law of value.
39. Exploring Empire's very confusing sense of the relation between Empire, multitude, and biopolitical production, Moreiras (2001: 225) provocatively asks of the clothes company Zara (which operates in a decentralized, flexible manner in direct relation to consumer desire and without the exploitation of third world labour), 'what keeps Zara from understanding itself as an instance of counter-Empire? And what would keep us from suspecting that there is finally no difference between Empire and counter-Empire, once immanentization has run full course?' One suspects that Hardt and Negri interpret the apparent coming to immanence of production and desire in contemporary biopolitical production ('In the new modes of life, in an ever larger domain, labour becomes desire'; Hardt and Negri, in Brown et al. 2002: 205) as the impending arrival of what Deleuze and Guattari see as the condition of the 'new earth', when desiring-product ion and labour-power finally manifest their unity of substance in 'production in general and without distinction (something capital discovers, but continually realienates) (ACE: 302), hence Hardt and Negri's ambivalence over the affirmation or critique of Empire that is the focus of Moreiras' essay. The emphasis in contemporary work on the drawing-in of desire to work (as a means of overcoming the 'crisis of work' of the 1960s and 70s; cf. Virno 1996c; Heelas 2002) is a crucial site for contemporary research, but, as Moreiras' example highlights, it is problematic indeed to see desire as somehow set free in this arrangement (desire, after all, always invests the socius - and its identities, arrangements, objects, and horrors - in some fashion).
40. Though this discussion focuses on Negri's more recent elaboration of the socialized worker, it is worth noting that in his early exploration of the socialized worker's tendency toward autonomy (in, for example, Negri 1979a) it is less self-management than political violence and 'armed struggle' which becomes the mechanism for shrugging off an external capitalist command.
41. By moving toward an affirmation of the current composition of life as communist, Negri also starts to sound like the 'planning' perspective critiqued by Panzieri. Whilst, no doubt, certain forms of general intellect-rich labour are composed of more diffuse and complex attributes and forces that far exceed the limited form of composition of factory work, as Bifo (1980: 168) writes, 'it would be simplistic to conclude that the revolution . . . needs to substitute a Leninist seizure of Knowledge for a Leninist seizure of the State. The problem is in reality much more complicated, since not only the properties and use of Knowledge, but also its structure, are determined by its capitalist functioning.'
42. See the exchange between Negri and Derrida on this point in Sprinker (1999).
43. 'Communication society' is in this essay Negri's term for Deleuze's model of 'control society' (cf. Negri 1992:105). Deleuze's expression 'control' clearly brings pejorative connotations to an understanding of a system (where communication is indeed prevalent) that Negri would prefer to elide, as is evident in his question. Though in Empire control emerges to an apparently central place, as the book develops it seems to become subsumed in the category of 'Empire', which itself becomes increasingly 'empty'. When Deleuze and Guattari (ATP: 460) write of the return of 'empire' ('modern States of the third age do indeed restore the most absolute of empires'), it is immanent to the most intricate control.
44. In Empire Hardt and Negri (2000: 28) suggest that Deleuze and Guattari 'discover the productivity of social reproduction . . . but manage to articulate it only superficially and ephemerally'.
45. That said, it is only after the arrival of capital that, as was also the case for Marx, the possibility of approaching a 'universal history' emerges, for the deterritorialization actualized by capital (as it discovers and sets free abstract labour) is revealed to be the limit that all previous socii' sought to ward off (ACE: 153). As Holland (1999) considers in detail, it is on this precondition that it is possible for life in capital to perform an autocritique towards the full development of universal or world history, whose subject "” akin to Marx's communist overcoming "” would be molecular life, where 'Nature = Industry = History' (ACE: 25; cf. Holland 1999: 95,111).
46. Thus Deleuze and Guattari are fully in accord with Marx's description of capital in the Manifesto: 'Constant revolutionising of production, uninterrupted disturbance of all social conditions, everlasting uncertainty and agitation distinguish the bourgeois epoch from all earlier ones. All fixed, fast-frozen relations, with their train of ancient and venerable prejudices and opinions, are swept away, all new-formed ones become antiquated before they can ossify. All that is solid melts into air, all that is holy is profaned, and man is at last compelled to face with sober faces his real conditions of life and his relations with his kind' (Marx and Engels 1973: 36-7).
47. For Marx's account of the centrality and novelty of the quest for 'wealth itself in disrupting all previous modes of community (and the denunciation of such practice in antiquity for fear of this very potential) see Marx (1973a: 540"”1).
48. This 'double movement' is presented in Anti-Oedipus (259) thus: 'In Capital Marx analyzes the true reason for the double movement: on the one hand, capitalism can proceed only by continually developing the subjective essence of abstract wealth or production for the sake of production, that is, "production as an end in itself, the absolute development of the social productivity of labor"; but on the other hand and at the same time, it can do so only in the framework of its own limited purpose, as a determinate mode of production, "production of capital", "the self-expansion of existing capital.'"
49. In this Marxian sense, money is first and foremost not a mechanism of exchange, but of command and management of social labour. For an explanation of this point, and a series of analyses of the politics of money that follow from it, see Bonefeld and Holloway (1996).
50. Massumi (1992: 128-9) explains this well: 'Capital functions directly through incorporeal transformation, without having to step down or up to another level . . . Capital can be given an image - in fact it must have one in order to act - but it is imageless as such. It is a body without organs. In other words, a network of virtual relations, a selection of which is immediately actualized at ground level wherever one of capitalism's working images (organs) goes. These images are conveyances (components of passage). They bring to designated bodies at each spatiotemporal coordinate through which they circulate a relation that fundamentally changes those bodies' social and physical reality. That relation is capital as an immanent social agency.'
51. The term 'bourgeoisie' is used because it is the dominant class, or axiomatic model of smooth-running capital "” the mechanism of identity formation that functions to realize and fix the super-adequacy of life in capitalist forms. This notion that the bourgeoisie is not a social group but a particular mode of composition immanent to capital explains Deleuze's argument (in Guattari 1995a: 65) that the bourgeoisie 'has never been revolutionary'. Even in the emergence of capital, it is the name for one side of the double movement of capital "” the immanent control of the forces that the other side "” production for production's sake "” sets loose.
52. This investment in the capitalist socius, as fundamental to identity as it is (since that which is invested produces the identity and its investment in the first place), is the basis for Deleuze and Guattari's crucial assertion that the question of support for, and critique of, the status quo resides not in one's 'interest', but in one's 'desire', or libidinal investment. Since we are all, as Guattari (1996a: 101"”15) puts it, 'machinic junkies', Anti-Oedipus asks, how can one not invest in the great mutant flow of capitalism?: 'a pure joy in feeling oneself a wheel in the machine, traversed by flows, broken by schizzes. Placing oneself in a position where one is thus traversed, broken, fucked by the socius, looking for the right place where, according to the aims and the interests assigned to us, one feels something moving that has neither an interest nor a purpose ... a taste for a job well done' (ACE: 346-7).
53. 'There is no ideology, there are only organizations of power once it is admitted that the organization of power is the unity of desire and the economic infrastructure.' For example, 'The church is perfectly pleased to be treated as an ideology. This can be argued; it feeds ecumenism. But Christianity has never been an ideology; it's a very original, very specific organization of power that has assumed diverse forms since the Roman Empire and the Middle Ages' (Deleuze, in Guattari 1995a: 57"”8).
54. It is Burroughs' delirious fascination with the intricacies of controlling techniques "” from word locks to Nova police, the atom bomb to the Mayan calendar and psychoanalytic and hypnotic suggestion "” and his fully social, even cosmic understanding of the plane of composition of power which, one can imagine, inspires Deleuze.
55. Incidentally, Deleuze (1995a: 51) calls this text 'completely marxist'.
56. Guattari (1984) identified this tendency in the mid-1970s in the context of the break-up of the psychiatric hospital. '[W]hat. . . strikes me', Guattari says, 'is that all the great repressive organizations like schools or army, which used to consist of a single institutional whole, are now tending to become fragmented and scattered all over the place .. . {V}ery soon everyone will become his own mini-instrument of repression, his own school, his own army . . . [T]he policy of community psychiatry and psychoanalysis (and the two are now closely related) corresponds to the most sophisticated technocratic forms of population surveillance and control' (48). It is interesting, given Guattari's complex relation to Lacan, that he identifies 'Lacanism' "” with its 'mathematico-linguistic' model of the unconscious and its subtle and incorporeal mechanisms of analysis "” as 'A testing-ground, an advance technology, the prototype of [these] new forms of power': 'The psychoanalyst of today doesn't say a word to his patient. Such a system of channelling the libido has been achieved that silence is all that is needed' (50).
57. My use of the term 'cybernetic' here is slightly problematic, for in Deleuze and Guattari's (ACE: 251"”2) assessment it is too mechanical a model, too reliant on 'isolated formulas'. The axiomatic generally - and this tendency is only increased in control "” operates as a more subtle series of 'intuitions' and 'resonances' with a plethora of decisions, administrations, predictions, reactions, and inscriptions, for which more technical concretizations are only aids. That said, Guattari and Alliez (in Guattari 1984: 285) describe this model of production as 'cybernetic capital'.
58. It is for this reason that Guattari (1984: 259) writes that 'Kafka is not, as some have said, a nineteenth-century writer imprisoned in family conflicts. He is a twenty-first-century writer describing the earliest stages of a process whose implications we are barely beginning to grasp today.'
59. The importance of business as an abstract form concretely embodied in varied spheres is stressed in a number of examples: TV game-shows are said to be popular 'because they're a perfect reflection of the way businesses are run' (N: 179), continuing education and continuous assessment are 'the surest way of turning education into a business' (179), and 'Even art has moved away from closed sites and into the open circuits of banking' (181). This 'business' or 'enterprise' model is also put forward by Negri as central to what he calls in pit-Empire work, the 'crisis-state'. As with Deleuze, this is a form of control that arises with the collapse of distinct enclosure, and of the normalizing regulation of labour through Keynesian wage/productivity tie-ins. In ever more fluid productive space, the 'enterprise' comes to be the site of productivity across the social as a modulating capture of energies that is able to remove the stabilities of large-scale production and compose forms of identity and self-control in varying and changing fashion: 'The key control mechanism in this transformation is the enterprise, in the sense that it extends the norms of factory-command over work to the whole social labour time' (Negri 1988c: 123).
60. Clearly aware of the centrality to Marx's overall system of his argument that machines cannot create value, Anti-Oedipus' assertion of machinic surplus value is couched in what Deleuze and Guattari rather self-consciously call a deliberate 'indispensable incompetence'. This is a ruse they take from Maurice Clavel's apparent use of 'wilfully incompetent questions' to Marxist economists concerning the credibility of the centrality of human surplus value in the face of the productive power of machines (ACE: 232), but they situate their 'incompetence' around the question of the 'surplus value of flux' which, as I am arguing, is in accord with the essential logic of the labour theory of value.
61. Making a similar point, Diane Elson (1979: 123) has argued that 'the object of Marx's theory of value was labour. It is not a matter of seeking an explanation of why prices are what they are and finding it in labour. But rather of seeking an understanding of why labour takes the forms it does, and what the political consequences are.'
62. Though Deleuze and Guattari (ACE: 492) suggest that machinic surplus value emerges 'less and less by the striation of space-time corresponding to the physicosocial concept of work' we have seen already how 'business' becomes a pervasive model for an increasingly subdivided and diffuse 'productivity', and hence I would suggest that they are here using the word 'work' in a limited, descriptive sense (something like the Keynesian 'job') rather than in a machinic sense. In the sense in which this book defines work in abstract terms as the axiomatized reterritorialization of human practice immanent to the capitalist mission of production for production's sake, the extension of machinic enslavement is simultaneously an extension of work (cf. ATP: 400"”1).
63. Guattari (1996a: 206) thus suggests that 'The recasting of the quantification of value based on work-time won't be, as Marx assumed, the privilege of a classless society.'
64. Notably the essay overplays the reduction of work time as a measure of value (since the quantification of labour, however impossible it is to really measure individual contribution, is still fundamental to the capitalist valuation and axiomatization of life) and suggests that the concept of 'average social labour' is an abstraction inappropriate for an understanding of the concrete practices of labour (when in fact it is central to an understanding of the processes of abstraction necessary for unbounded productivity, as I explored above through Anti-Oedipus).
65. See Terranova (2000) for an examination of internet labour in similar terms.
66. 'The sales department becomes a business centre or "soul" . . . Marketing is now the instrument of social control and produces the arrogant breed who are our masters' (N: 181).
67. Massumi (1996) provides an example of a study of children's experience of a TV film where non-verbal bodily response, even as it contradicted verbal response, was used as the basis for judgement of the affective content of the image.
68. This primacy of machinic arrangements, as against a framework based on the demarcation of human and technical elements, is clear in Paolo Virno's (1996b: 22) characterization of contemporary production: 'In contemporary labor processes there are entire conceptual constellations that function by themselves as productive "machines", without ever having to adopt either a mechanical body or an electronic brain.'
69. As Rose (1999b: 483) argues in his discussion of the 'etho-politics' of the political and governmental imaginary of the 'Third Way', we have a model of the human actor that is 'no longer the nineteenth-century economic subject of interests but an entrepreneur of his- or her-self, striving to maximize his or her own human capital by choices which are, as it were, investments for the purpose of the capitalization of one's own existence.'
70. Fox Piven and Cloward (1972: 6"”7) draw attention to the historical problem of unemployment which the JSA and the New Deal are the latest attempts to overcome: 'The regulation of civil behavior in all societies is intimately dependent on stable occupational arrangements. So long as people are fixed in their work roles, their activities and outlooks are also fixed . . . Each behavior and attitude is shaped by the reward of a good harvest or the penalty of a bad one, by the factory paycheck or the danger of losing it. But mass unemployment breaks that bond, loosening people from the main institution by which they are regulated and controlled. Moreover, mass unemployment that persists for any length of time diminishes the capacity of other institutions to bind and constrain people . . . [W]ithout work, people cannot conform to familial and communal roles; and if the dislocation is widespread, the legitimacy of the social order itself may come to be questioned.' It is with this in mind that Walters (1994) has shown how the 'invention' of unemployment "” loosely fitting with the diagram of discipline "” and its institutional apparatus was a strategy intended to construct a coherent unemployed subject comparable to the employed subject. Contemporary emphasis in neo-liberal governance on an ethically intensive process of 'jobseeking' (where benefit is only paid on the basis that the jobseeker enters into arrangements of self-optimization "” including training and maintenance of acceptable physical appearance "” and continual job application) can be seen as the form of 'unemployment' appropriate to control. And, indeed, the JSA can be seen as a direct response to the breakdown of the disciplinary model of the unemployed subject that became evident "” with the increasing affirmation of unemployment as a space of relative autonomy from work "” under Thatcher and Reagan (cf. Aufheben 1998).