'Romantic'. 'Noble, but not very realistic'. 'We have to deal with the reality of class struggle, not abstractions about
How can we possibly change the world without taking power? The idea is an attractive dream, and we all like
attractive dreams, but what is their reality? How can we dream after the experience of the twentieth century, when
so many dreams have failed, when so many dreams have ended in misery and disaster?
Where is this anti-power that is the hope of humanity? What is the material reality of anti-power? Because if it has
no material reality, then we are deluding ourselves. We all want to dream that a different type of society is possible,
but is it really? The revolutionaries of the early part of the century built their dreams upon the mass organisations of
the proletariat, but those organisations no longer exist or, if they do, they are not the stuff of dreams.
We have thrown out a lot of bathwater. And how many babies? A defined subject has been replaced by an
indefinable subjectivity. Proletarian power has been replaced by an undefined anti-power. This sort of theoretical
move is often associated with disillusion, with abandoning the idea of revolution in favour of theoretical
sophistication. That is not the intention here. But where, then, is this anti-power?
I scream. But am I alone? Some of the readers scream as well. We scream. But what indication is there of the
material force of the scream?
The first point is that anti-power is ubiquitous.
The television, the newspapers, the speeches of politicians give little indication of the existence of anti-power. For
them, politics is the politics of power, political conflict is about winning power, political reality is the reality of power.
For them, anti-power is invisible.
Look more closely, however. Look at the world around us, look beyond the newspapers, look beyond the political
parties, beyond the institutions of the labour movement and you can see a world of struggle: the autonomous
municipalities in Chiapas, the students in the UNAM, the Liverpool dockers, the wave of international demonstrations against the power of money capital, the struggles of migrant workers, the struggles of the workers
in all the world against privatisation. There is a whole world of struggle that does not aim at all at winning power, a
whole world of struggle against power-over. There is a whole world of struggle that sometimes goes no farther than
saying 'No!' (sabotage, for example) but that often, in the course of saying 'No!', develops forms of selfdetermination and articulates alternative conceptions of how the world should be. Such struggles, if they are reported at all in the mainstream media, are filtered through the spectacles of power, visible only in so far as they are considered to impinge upon power politics.
The first problem in talking of anti-power is its invisibility. It is invisible not because it is imaginary, but because our
concepts for seeing the world are concepts of power (of identity, of the indicative). To see anti-power, we need
different concepts (of non-identity, of the Not Yet, of the subjunctive).
All rebellious movements are movements against invisibility. Perhaps the clearest example of that is the feminist
movement, where much of the struggle has been to make visible that which was invisible: to make visible the
exploitation and oppression of women, but more than that, to make visible the presence of women in this world, to
rewrite a history from which their presence had been largely eliminated. The struggle for visibility is also central to
the current indigenous movement, expressed most forcefully in the Zapatista wearing of the balaclava: we cover
our face so that we can be seen, our struggle is the struggle of those without face.
Yet there is an important distinction to be made here. The problem of anti-power is not to emancipate an oppressed
identity (women, indigenous) but to emancipate an oppressed non-identity, the ordinary, everyday, invisible no, the
rumblings of subversion as we walk in the street, the silent volcano of sitting in a chair. By giving discontent an
identity, 'we are women', 'we are indigenous', we are already imposing a new limitation upon it, we are already
defining it. Hence the importance of the Zapatista balaclava, which says not just 'we are the indigenous struggling
for our identity to be recognised', but, much more profoundly, 'ours is the struggle of non-identity, ours is the
struggle of the invisible, of those without voice and without face'.
The first step in struggling against invisibility is to turn the world upside down, to think from the perspective of
struggle, to take sides. The work of radical sociologists, historians, social anthropologists and so on has made us aware of the ubiquity of opposition to power, in the workplace, in the home, on the streets. At its best, such work
opens a new sensitivity, often associated with struggles against invisibility and consciously starting from those
struggles (the feminist movement, the gay movement, the indigenous movement and so on). The issue of
sensitivity is well posed by an Ethiopian proverb quoted by Scott: 'When the great lord passes the wise peasant
bows deeply and silently farts'. In the eyes, ears and nose of the lord, the peasant's fart is completely
imperceptible. For the peasant herself and for other peasants, and for those who start from the peasant's
antagonism towards the lord, the fart is, however, all too evident. It is part of a hidden world of insubordination:
hidden, however, only to those who exercise power and to those who, by training or for convenience, accept the
blinkers of power.
That which is oppressed and resists is not only a who but a what. It is not only particular groups of people who are
oppressed (women, indigenous, peasants, factory workers and so on), but also (and perhaps especially) particular
aspects of the personality of all of us: our self-confidence, our sexuality, our playfulness, our creativity. The
theoretical challenge is to be able to look at the person walking next to us in the street or sitting next to us in a bus
and see the stifled volcano inside them. Living in capitalist society does not necessarily make us into an
insubordinate, but it does inevitably mean that our existence is torn by the antagonism between subordination and
insubordination. Living in capitalism means that we are self-divided, not just that we stand on one side of the
antagonism between the classes, but that the class antagonism tears each of us apart. We may not be rebellious,
but inevitably rebellion exists within us, as stifled volcano, as projection towards a possible future, as the present
existence of that which does Not Yet exist, as frustration, as neurosis, as repressed Pleasure Principle, as the nonidentity which, in the face of the repeated insistence of capital that we are workers, students, husbands, wives,
Mexicans, Irish, French, says 'we are not, we are not, we are not, we are not what we are, and we are what we are
not (or not yet)'. That is surely what the zapatistas mean when they say they are 'ordinary people, that is to say,
rebels'; that is surely what they mean by dignity: the rebellion that is in all of us, the struggle for a humanity that is
denied us, the struggle against the crippling of the humanity that we are. Dignity is an intensely lived struggle that
fills the detail of our everyday lives. Often the struggle of dignity is non-subordinate rather than openly
insubordinate, often it is seen as private rather than in any sense political or anti-capitalist. Yet the non-subordinate
struggle for dignity is the material substratum of hope. That is the point of departure, politically and theoretically.
Probably no one has been as sensitive to the force and ubiquity of the suppressed dream as Ernst Bloch, who in
the three volumes of the Principle of Hope traces the multiple forms of projection towards a better future, the
present existence of the Not Yet, in dreams, fairy tales, music, painting, political and social utopias, architecture,
philosophy, religion: all testimony to the presence in all of us of a negation of the present, a pushing towards a
radically different world, a struggle to walk erect.
Anti-power does not exist only in the overt, visible struggles of those who are insubordinate, the world of the 'Left'.
It exists also - problematically, contradictorily (but then the world of the Left is no less problematic or contradictory)
- in the everyday frustrations of all of us, the everyday struggle to maintain our dignity in the face of power, the
everyday struggle to retain or regain control over our lives. Anti-power is in the dignity of everyday existence. Antipower is in the relations that we form all the time, relations of love, friendship, comradeship, community,
cooperation. Obviously, such relations are traversed by power because of the nature of the society in which we
live, yet the element of love, friendship, comradeship lies in the constant struggle which we wage against power, to
establish those relations on a basis of mutual recognition, the mutual recognition of one another's dignity.
The invisibility of resistance is an ineradicable aspect of domination. Domination always implies not that resistance
is overcome but that resistance (some of it at least) is underground, invisible. Oppression always implies the
invisibility of the oppressed. For one group to become visible does not overcome the general problem of visibility.
To the extent that the invisible becomes visible, to the extent that the stifled volcano becomes overt militancy, it is
already confronted with its own limits and the need to overcome them. To think of opposition to capitalism simply in
terms of overt militancy is to see only the smoke rising from the volcano.
Dignity (anti-power) exists wherever humans live. Oppression implies the opposite, the struggle to live as humans.
In all that we live every day, illness, the educational system, sex, children, friendship, poverty, whatever, there is a
struggle to do things with dignity, to do things right. Of course our ideas of what is right are permeated by power,
but the permeation is contradictory; of course we are damaged subjectivities, but not destroyed. The struggle to do
right, to live morally, is one that preoccupies most people much of the time. Of course, the morality is a privatised,
immoral morality which generally steers clear of such questions as private property and therefore the nature of
relations between people, a morality which defines itself as 'do right to those who are close to me and leave the
rest of the world to sort itself out', a morality which, by being private, identifies, distinguishing between 'those who
are close to me' (family, nation, women, men, whites, blacks, decent-looking, 'people like us') and the rest of the
world, those living beyond my particular moral pale. And yet: in the daily struggle to 'do right', there is a struggle to
recognise and be recognised and not just to identify, to emancipate power-to and not just bow to power-over, an
anger against that which dehumanises, a shared (if fragmented) resistance, a non-subordination at least. It may be
objected that it is quite wrong to see this as anti-power since, in so far as it is fragmented and privatised, such 'morality' functionally reproduces power-over. Unless there is consciousness of the interconnections, unless there
is political (class) consciousness, it may be argued, such private morality is totally harmless to capital, or actually
contributes actively to the reproduction of capital by providing the basis for order and good behaviour. All this is so,
and yet: any form of non-subordination, any process of saying 'we are more than the objectified machines that
capital requires', leaves a residue. Ideas of what is right, however privatised, are part of the 'hidden transcript' of
opposition, part of the substratum of resistance that exists in any oppressive society. The Ethiopian peasant's fart
certainly does not blow the passing lord off his horse, and yet: it is part of the substratum of negativity which,
though generally invisible, can flare up in moments of acute social tension. This substratum of negativity is the stuff
that social volcanoes are made of. This layer of inarticulate non-subordination, without face, without voice, so often
despised by the 'Left', is the materiality of anti-power, the basis of hope.
The second point is that anti-power is not only ubiquitous: it is also the motor force of power.
This has not been the predominant emphasis either in the Marxist tradition or in left thought in general. On the
whole Marxism has focussed its analysis on capital and its development, and left thought in general usually prefers
to highlight oppression, to stir up indignation against the evils of capitalism. There is a tendency to treat the
oppressed as just that, victims of oppression. This emphasis may stir us to indignant action, but it tends to leave
open completely the question of how oppressed victims can possibly liberate themselves – other, of course, than
through the enlightened intervention of saviours like ourselves.
Within the Marxist tradition, this emphasis on domination rather than struggle has been attacked most articulately
by the current which developed, initially in Italy, from the 1960s onwards, variously referred to as 'autonomist
Marxism' or 'operaismo'. The point was sharply formulated in an article by Mario Tronti first published in 1964,
"Lenin in England", that was to do much to shape the approach of 'autonomist' Marxism: "We too have worked with
a concept that puts capitalist development first, and workers second. This is a mistake. And now we have to turn
the problem on its head, reverse the polarity and start again from the beginning: and the beginning is the class
struggle of the working class" (1979, p. 1).
Tronti immediately takes the reversal of the polarity a step further. Starting from the struggle of the working class
means not simply adopting a working class perspective, but, in complete reversal of the traditional Marxist
approach, seeing working class struggle as determining capitalist development. "At the level of socially developed
capital, capitalist development becomes subordinated to the working class struggles; it follows behind them and
they set the pace to which the political mechanisms of capital's own reproduction must be tuned" (1979, p. 1).
This is the core of what Moulier refers to as "operaismo's ... Copernican inversion of Marxism" (1989, p. 19). This,
according to Asor Rosa, "can be summed up in a formula which makes the working class the dynamic motor of
capital and which makes capital a function of the working class ... a formula which in itself gives an idea of the
magnitude of the inversion of perspectives which such a position implies politically" (quoted by Moulier 1989, p.
The attraction of the inversion of the traditional approach is obvious, but how is the working class to be understood
as the ‘dynamic motor’ of capitalism? As Tronti himself says in the same article ‘this is not a rhetorical proposition.
Nor is it intended just to restore our confidence... an urgent practical need is never sufficient basis for a scientific
thesis.’ (1979, p. 1)
The autonomist re-interpretation of Marxism has its roots in the upsurge of factory struggle in Italy in the 1960s,
which led to a re-reading of Capital, putting particular emphasis on a part which generally been neglected by
‘Marxist economists’, namely the long analysis in Volume I of the development of the labour process in the
factories. In this discussion, Marx shows that capital is constantly forced to struggle with the ‘refractory hand of
labour’ and that it is this struggle which determines changes in factory organisation and technical innovation. Thus,
for Marx, automation is ‘animated by the longing to reduce to a minimum the resistance offered by that repellent yet
elastic natural barrier, man’. (1965, p. 403) Consequently, ‘it would be possible to write quite a history of the
inventions, made since 1830, for the sole purpose of supplying capital with weapons against the revolts of the
working class’. (1965, p. 436)
Taking as its focus first the struggles in the factories, the autonomist analyses show how all the organisational and
technical innovations introduced by management can be understood as a response designed to overcome the
force of insubordination on the part of the workers. Labour insubordination can thus be seen as the driving force of
This provides a way of analysing the history of struggle. The workers develop a form of struggle; management
introduce a new form of organisation or new machinery in order to re-impose order; this in turn gives rise to new
forms of insubordination, new forms of struggle and so on. One can speak of the struggle as having a certain
composition. By analogy with Marx’s idea that capital at any point is characterised by a certain technical and value
composition, depending on the relation between constant capital (that part of the capital represented by machinery
and raw materials) and variable capital (that part of the capital which corresponds to wages), the autonomists
developed the concept of class composition to denote the relation between labour and capital at any particular
moment. The movement of struggle can thus be seen as a movement of class composition. The forms of struggle
at any particular time are expressions of the composition of the working class; when management introduce
changes to restore order, they aim to bring about a de-composition of the class; this de-composition gives rise in
turn to the development of new forms of struggle, or a re-composition of the class. The history of struggle can thus
be described in terms of the movement of composition, de-composition and re-composition.
The concept is developed not only in relation to struggles in particular factories or industries but as a way of
understanding the dynamic of struggle in capitalism as a whole. Thus, it is argued, working class struggle in the
period up to the first world war was characterised by the particular place within production of the skilled worker.
This gave to the working class movement a specific form of organisation (skill-based trade unionism) and a
particular ideology (based on the notion of the dignity of labour). The de-composing response by management was
the introduction of Taylorism, designed to de-skill the skilled worker and deprive him of control of the labour
process. This gives rise in turn to a re-composition of the working class as mass worker, with new forms of
struggle, new forms of organisation (the general trade unions) and a new ideology (the rejection of work). The decomposing response by capital is seen by some autonomist theorists (Negri, in particular) as coming now not at the
level of factory management but at the level of the state, with the development of Keynesianism and the Welfare
State (Fordism, as it is often called) as a way of both recognising the growing strength of labour and at the same
time integrating it into the maintenance of order (through social democracy) and into the dynamic of capitalism
(through demand management). This gives rise, in Negri’s analysis, to a socialisation of capital, the transformation
of society into a ‘social factory’ and the emergence of a new class composition, the ‘social worker’ (‘operaio
sociale’). The strength of this new composition is expressed in the struggles of the late 1960’s and 1970’s which go
far beyond the factory to contest all aspects of the capital’s management of society. It is the strength of these
struggles which forces capital to abandon the Keynesian-Fordist form of management and develop new forms of
attack (neo-Liberalism, or what Hardt and Negri now refer to as ‘empire’).
Class composition thus takes us beyond the analysis of factory struggles to become the key concept for
understanding capitalist development. Thus, Moulier characterises the notion in broad terms: "We must remember
that the notion of 'class composition' is a concept which aims to replace the too static, academic and in general
reactionary concept of 'social classes'. Class composition comprises simultaneously the technical composition both
of capital and of waged labour, which refers to the state of development of the productive forces, to the degree of
social cooperation and division of labour. But this level of analysis is not separable from the political composition
which is its ultima ratio. We can find in it all that characterises the collective subjectivity of needs, desires, the
imaginary and their objective translation into the forms of political, cultural and community organisation." (1989, pp
The notion of class composition takes us significantly beyond the mere observation that resistance to capitalism is
ubiquitous. It suggests a basis for speaking of the developing force of this resistance, a basis for trying to
understand the specificity and the force of the current forms of struggle. It proposes a way in which we can see our
scream not just as an ever-present feature of oppression, but as a scream that has a particular historical
resonance. It suggests, for example, that what seems to be the peculiar loneliness of our scream is the result of the
fact that it comes after the defeat of the old labour movement and the old revolutionary movement, movements
which, through their defeat, nevertheless forced capital on to a new terrain of battle, a new terrain of recomposition.
This new terrain is described by Negri in his latest work (together with Michael Hardt: Hardt and Negri 2000) as
Empire. This they see as the new paradigm of rule: ‘In contrast to imperialism, Empire establishes no territorial
centre of power and does not rely on fixed boundaries or barriers. It is a decentred and deterritorialising apparatus
of rule that progressively incorporates the entire global realm within its open, expanding frontiers. Empire manages
hybrid identities, flexible hierarchies, and plural exchanges through modulating networks of command. The distinct
national colours of the imperialist map of the word have merged and blended in the imperial global rainbow.’(2000,
pp. xii-xiii) There is a change in sovereignty, ‘a general passage from the paradigm of modern sovereignty toward
the paradigm of imperial sovereignty’. In the latter, it is no longer possible to locate sovereignty territorially in the
nation state, or indeed in any particular place. Even the United States, although it plays a particularly important part
in the network of power, is not the locus of power in the same way that the imperialist powers of the earlier age
were. One implication of this would seem to be that it no longer makes sense to think of revolutionary
transformation in terms of the taking of state power.
In this new paradigm, there is no longer any place of rule, and consequently no longer any inside or any outside, no
longer any possible external standpoint. Empire is an all-embracing system of rule, the latest re-formulation of what
Negri had earlier characterised as the ‘social factory’ or ‘integrated world capitalism (IWC)’. This does not mean
that all possibility of resistance or change has been obliterated. On the contrary, Hardt and Negri insist that Empire
is to be understood as a reaction to the struggles of the multitude. ‘The history of capitalist forms is always
necessarily a reactive history.’(2000, p. 268) Thus, ‘the multitude is the real productive force of our social world,
whereas Empire is a mere apparatus of capture that lives only off the vitality of the multitude – as Marx would say,
a vampire regime of accumulated dead labour that survives only by sucking off the blood of the living.’(2000, p. 62)
Within Empire, the driving force continues to be the multitude. Empire has as its material basis the development of
‘immaterial labour’, the intellectual communicative and affective labour characteristic above all of the development
of the service sector of the informational economy. The important thing about this immaterial labour is the degree to
which it is immanently and immediately co-operative, thus creating a new subjectivity. ‘The immediately social
dimension of the exploitation of living immaterial labour immerses labour in all the relational elements that define
the social but also at the same time activate the critical elements that develop the potential of insubordination and
revolt through the entire set of labouring practices.’(2000, p. 29) The inherently cooperative nature of this type of
labour ‘annuls the title of property’ (2000, p. 410) and creates the basis for an absolute democracy, a communist
It should already be clear that the argument of Negri and Hardt pushes in a direction similar to the argument in this
book in two crucial respects. Firstly, they emphasise the centrality of oppositional struggle (whether we call it the
power of the multitude or anti-power) as the force which shapes social development; and secondly, they argue that
it is important to focus on revolution, but that revolution cannot be conceived in terms of the taking of state power.
Their argument is immensely rich and suggestive, yet, at the same time, their approach is very different indeed
from the approach that has been adopted here. This leaves us with a dilemma. Are we to say that method does not
matter, that there are many different ways of reaching the same conclusion? But if we adopt that position, then
much of the previous argument about fetishism and critique falls. If, however, we say that method does matter,
precisely because method is part of the struggle against capitalist domination, then what are we to say of Hardt and
Let us look at the matter more closely.
The difference in approach can be seen as centred in the issue of paradigm. The argument of Hardt and Negri
focuses on the shift from one paradigm of rule to another. This shift is characterised primarily as a shift from
imperialism to Empire, but it is also variously described as a move from modernity to post-modernity, from
discipline to control, from Fordism to post-Fordism, from an industrial to an informational economy. What interests
us here is not the name, but the assumption that capitalism can be understood in terms of the replacement of one
paradigm of rule by another, one system of order by another. ‘The US world police acts not in imperialist interest
but in imperial interest. In this sense the Gulf War did indeed, as George Bush claimed, announce the birth of a
new world order.’(2000, p.180)
Hardt and Negri are not alone, of course, in this paradigmatic approach. Another approach which relies heavily on
the notion of a shift from one paradigm to another and which has had great influence in recent years is the
regulationist school, which analyses capitalism in terms of a shift from a Fordist to a post-Fordist mode of
regulation. The paradigmatic approach has obvious attractions as a method of trying to understand the current
changes in the world. It permits one to bring together many apparently disparate phenomena into a coherent
whole. It allows one to paint an extremely rich and satisfying picture in which all the millions of pieces of the jigsaw
click into place. This is immensely stimulating, for it suggests a whole series of correspondences that were not
obvious before. It is also very attractive to academics because it suggests a whole world of research projects which
can be completed with no jagged edges.
The problem with the approach, however, is just that, that it paints an orderly world of correspondence. The
negative impulse which is the starting point becomes converted into a positive science. The ‘party of chaos’
(Negri’s phrase) is slotted into a word of order. Although Hardt and Negri insist that order must be understood as
the response to disorder, the tale is told through the account of order, not through disorder. Although they insist
that refusal is the driving force of domination, refusal is in fact relegated to a subordinate place: it is only in the
closing pages of the book (2000, p. 393) that the authors say, ‘Now that we have dealt extensively with Empire, we
should focus directly on the multitude and its potential political power.’
The paradigmatic approach takes classification to extremes. There is an eagerness to capture the new, to classify
it, label it, make it fit into the paradigmatic order. There is almost indecent haste to declare the old order dead and
proclaim the new. ‘The King is dead! Long live the King!’ As soon as one system of rule is in crisis, the new system of rule is proclaimed. ‘At this point the disciplinary system has become completely obsolete and must be left
behind. Capital must accomplish a negative mirroring and an inversion of the new quality of labour power: it must
adjust itself so as to be able to command once again.’ (2000, p. 276) The adjustment to the new command is
assumed as reality, not just seen as a project: this is the substance of the new paradigm, this is Empire.
The desire to make everything fit, to see the new paradigm as established, leads easily to an exaggeration that
often seems quite unreal. Thus, ‘autonomous movement is what defines the place proper to the multitude.
Increasingly less will passports or legal documents be able to regulate our movement across borders.’ (2000, p.
397) Or: ‘there are no time-clocks to punch on the terrain of biopolitical production; the proletariat produces in all its
generality everywhere all day long.’ (2000, p. 403)
The paradigmatic approach shades into functionalism. In a world of correspondences, everything is functional,
everything contributes to the maintenance of a coherent whole. Thus, for Negri and Hardt (as earlier for Negri),
crisis is not so much a moment of rupture as a force of regeneration in capitalism, a ‘creative destruction’. Thus, ‘as
it is for modernity as a whole, crisis is for capital a normal condition that indicates not its end but its tendency and
mode of operation.’ (2000, p. 222) Or: ‘the crisis of modern sovereignty was not temporary or exceptional (as one
would refer to the stock market crash of 1929 as a crisis), but rather the norm of modernity. In a similar way,
corruption is not an aberration of imperial sovereignty but its very essence and modus operandi.’ (2000, p. 202)
Although the project of the book is very clearly one of rupture, the method adopted seems to absorb the possibility
of rupture, to integrate movement into a photograph. A paradigmatic approach inevitably involves a freezing of
The functionalism extends to the understanding of sovereignty and the state. The authors interpret Marx’s view of
the state as a functionalist one. Referring to Marx and Engels’ characterisation of the state as the executive that
manages the interests of capitalists, they comment: ‘by this they mean that although the action of the state will at
times contradict the immediate interests of individual capitalists, it will always be in the long-term interest of the
collective capitalist, that is, the collective subject of social capital as a whole.’(2000, p. 304) Thus, the system of
modern states succeeded in ‘guaranteeing the interests of total social capital against crises’ (p. 306), while in the
postmodern age of Empire, ‘government and politics come to be completely integrated into the system of
transnational command’. (p. 307) The political and the economic come to form a closed system, an ‘integrated
It is entirely consistent with this paradigmatic approach that Hardt and Negri are very explicitly anti-dialectical and
anti-humanist in their approach. Hegel is repeatedly dismissed as the philosopher of order rather than seeing him
as being also the philosopher who made subversive movement the centre of his thought. Dialectics is understood
as the logic of synthesis rather than as the movement of negation. It is quite consistent with this that the authors
insist on the continuity between animals, humans and machines. They see themselves as carrying on ‘the
antihumanism that was such an important project for Foucault and Althusser in the 1960s’ and quote with approval
Haraway’s insistence upon ‘breaking down the barriers we pose among the human, the animal and the machine’.
(2000, p. 91) Postmodernism gives us the opportunity to ‘recognise our posthuman bodies and minds, [to] see
ourselves for the simians and cyborgs we are’ (2000, p. 92). In the new paradigm ‘interactive and cybernetic
machines become a new prosthesis integrated into our bodies and minds and a lens through which to redefine our
bodies and minds themselves. The anthropology of cyberspace is really a recognition of the new human condition.’
(2000, p. 291) The problem with this approach, surely, is that neither ants nor machines revolt. A theory that is
grounded in revolt has little option but to recognise the distinctive character of humanity.
Surprisingly, perhaps, given their general project, Hardt and Negri have no concept of capital as class struggle.
There is, in other words, a tendency to treat capital as an economic category, reproducing in this (as in other
points) the assumptions of the Marxist orthodoxy which they so rightly attack. Capital does not seem to be
understood as the struggle to appropriate the done and turn it against the doing. Thus, in apparent contradiction of
their insistence on understanding the shift of paradigm as a response to class struggle, they assert that ‘in addition
to looking at the development of capital itself, we must also understand the genealogy from the perspective of class
struggle’ (2000, p. 234 – my emphasis) – thus implying that the development of capital and class struggle are two
separate processes. The actual analysis of ‘the development of capital itself’ is in terms of under-consumptionism
rather than the antagonism between capital and labour. The barriers to capitalist development all ‘flow from a single
barrier defined by the unequal relationship between the worker as producer and the worker as consumer’. (2000, p.
222) In order to explain the movement from imperialism to Empire, they follow Rosa Luxemburg’s underconsumptionist theory that capitalism can survive only through the colonisation of non-capitalist spheres. ‘At this
point we can recognise the fundamental contradiction of capitalist expansion: capital’s reliance on its outside, on
the non-capitalist environment, which satisfies the need to realise surplus value, conflicts with the internalisation of
the non-capitalist environment, which satisfies the need to capitalise that realised surplus value.’ (2000, p. 227 –
my emphasis) According to the authors, capital finds a solution to the exhaustion of the non-capitalist world by
turning from the formal subsumption of the non-capitalist sphere to the real subsumption of the capitalist world. It is after this explanation of the passage from imperialism to Empire that it is pointed out that ‘we must also understand
the genealogy from the perspective of class struggle’ (2000, p. 234 – my emphasis).
The consequence of understanding class struggle and capital as being separate, and of seeing the ‘fundamental
contradiction of capitalist expansion’ as being something other than capital’s dependence upon the subordination of
labour, is that there is no understanding of the way in which the insubordination of labour constitutes the weakness
of capital (especially in capitalist crisis). In this book, as in all of Negri’s analyses, there is a clash of Titans: a
powerful, monolithic capital (‘Empire’) confronts a powerful, monolithic ‘multitude’. The power of each side does not
appear to penetrate the other. The relation between the two sides of the capitalist antagonism is treated as external
one, as is indicated, indeed, by the authors’ choice of the word ‘multitude’ to describe the opposition to capital, a
term which has the grave disadvantage of losing all trace of the relation of dependence of capital upon labour.
Negri does not, of course, stand for all autonomist theorists. Other autonomist theorists have criticised the Empire
book for mistaking tendency for reality. Nevertheless, Negri is immensely influential within the autonomist
tendency. Not only that, but Negri and Hardt’s development of the autonomist argument does raise questions about
the whole approach.
The great merit of the autonomist approach is that it insists on seeing the movement of capitalist rule as being
driven by the force of working class struggle, on seeing capital as a ‘function of the working class’. There is,
however, an ambiguity here, two possible ways in which this affirmation can be understood. The weaker version
would be to say that capital can be understood as a function of the working class because its history is a history of
reaction to working class struggle, in much the same manner as one might see, say, the movements of a defending
army at war to be a function of the movements of the attacking army, or, possibly, the development of the police to
be a function of the activities of criminals. The stronger version would be that capital is a function of the working
class for the simple reason that capital is nothing other than the product of the working class and therefore
depends, from one minute to another, upon the working class for its reproduction. In the first case, the relation
between the working class and capital is seen as a relation of opposition, an external relation; in the second case,
the relation is seen in terms of the generation of one pole of the opposition by the other pole, as an internal relation.
In the first case, the working class is seen as existing simply against capital, in the second case it exists againstand-in capital. These two interpretations, the 'reaction' interpretation and the 'product' interpretation, are not
necessarily mutually exclusive, but in so far as the emphasis is placed on one rather than the other, the theoretical
and political implications may be quite different.
Both of these elements are present in the autonomist analysis, but it is the first, the 'reaction' interpretation, which is
more prominent. Typically, the dynamic of capitalist development is understood as a reaction or response to the
power of the working class movement. The development of capital is then understood as the defensive reaction by
capital to the strength of the working class movement revealed in moments of open revolt. Keynesianism, for
example, in Negri's analysis (1988) is a response to the revolution of 1917, which made clear that capital could
survive only by recognising and integrating the working class movement. Empire too is a response to the force of
the multitude. These analyses are immensely suggestive, but the point being made here is that capitalist
development is understood as process of reaction, that the relation between labour and capital is understood as an
The reversal of the polarity between capital and labour, essential though it be as a starting point, ends by
reproducing the polarity in a different form. The traditional Marxist analysis emphasises the logical development of
capital and relegates class struggle to a 'but also' role; autonomist theory liberates class struggle from its
subordinate role, but still leaves it confronting an external logic of capital. The difference is that the logic of capital
is understood now not in terms of 'economic' laws and tendencies, but in terms of a political struggle to defeat the
enemy. It is easy to see how, in the analyses of some autonomists (such as Negri) the law of value, the key
category in the Marxist-economic interpretation of capitalist development, is seen as being redundant
(Negri1988b). In the face of the power of the working class movement, capital develops into Integrated World
Capitalism (Guattari and Negri 1990), and now Empire, and its sole logic is the logic of maintaining power As is
perhaps inevitable, the reaction understanding of the labour-capital relation leads to a mirror-image view of
capitalism: the greater the power of the working class movement, the more monolithic and totalitarian the response
of the capitalist class. Autonomist theory has been crucial in reasserting the nature of Marxist theory as a theory of
struggle, but the real force of Marx's theory of struggle lies not in the reversal of the polarity between capital and
labour, but in its dissolution. As Bonefeld puts it, ‘the difficulty inherent in 'autonomist' approaches is not that
'labour' is seen as being primary but that this notion is not developed to its radical solution.’ (1994, p. 44)
The understanding of the relation between capital and labour as an external one has its consequences for the
central category of ‘class composition’. Instead of analysing particular struggles in terms of the overall movement of
capital’s dependence upon labour (not Lukács’s perspective of totality but certainly his aspiration towards totality),
there is a tendency to project from particular struggles (the struggles in Fiat in the early 1970s, say) and see them as being typical of a certain stage of capitalist development. What is constructed is an ideal type or paradigm, a
heading under which all struggles are to be classified. The struggles in the Italian car factories then become a
measure for other struggles, rather than being understood in terms of their place in the general movement of
capital’s dependence upon labour. This procedure, so notorious in Hardt and Negri’s discussion of Empire, or, even
more so, in the regulationist discussions of Fordism and Post-Fordism, is present also in the discussions of class
composition and leads easily to crude generalisations, to the construction of categories as Procrustean beds into
which struggles arising from very different conditions must be forced to fit.
Underlying the tendency for the notion of class composition to provide the basis for a paradigmatic approach is a
more profound problem. The reversal of the polarity undertaken by autonomist theory transfers the positive from
the side of capital to the side of the struggle against capital. In orthodox theory capital is the positive subject of
capitalist development. In autonomist theory the working class becomes the positive subject: that is why the
positive concepts of class composition and class recomposition are on the side of the working class, while the
negative concept of decomposition is placed on the side of capital. In the reversal of the polarity, identity is moved
from the side of capital to the side of labour, but it is not exploded or even challenged. This is wrong. Subjectivity in
capitalism is in the first place negative, the movement against the denial of subjectivity. A truly radical reversal of
the polarity involves not just transferring subjectivity from capital to the working class but also understanding that
subjectivity as negative instead of positive, as the negative subjectivity of the anti-working anti-class. In the
beginning is the scream, not because the scream exhausts itself in negativity, but because the only way in which
we can construct relations of dignity is through the negation of those relations which deny dignity. Our movement,
then, is in the first place a negative movement, a movement against identity. It is we who decompose, we are the
wreckers. It is capital which constantly seeks to compose, to create identities, to create stability (always illusory, but
essential to its existence), to contain and deny our negativity. We are the source of movement, we are the subject:
in that, autonomist theory is right. But our movement is a negative one, one that defies classification. What unites
the zapatista uprising in Chiapas or the Movement of the Landless (MST) in Brazil with the struggle of the internet
workers in Seattle, say, is not a positive common class composition (as ‘immaterial labour’?) but rather the
community of their negative struggle against capitalism. The problem is not to understand our composition in the
present paradigm but to understand our negativity as the substance of capitalist crisis.
Politically, the emphasis on the power of the working class movement has an obvious appeal. Nevertheless, the
understanding of labour and capital in terms of an external relationship leads to a paradoxical (and romantic)
magnification of the power of both. The failure to explore the internal nature of the relation between labour and
capital leads the autonomist analysis to underestimate the degree to which labour exists within capitalist forms. The
existence of labour within capitalist forms, as will be argued more fully later, implies both the subordination of
labour to capital and the internal fragility of capital. To overlook the internal nature of the relation between labour
and capital thus means both to underestimate the containment of labour within capital (and hence overestimate the
power of labour against capital) and to underestimate the power of labour as internal contradiction within capital
(and hence overestimate the power of capital against labour). If the inter-penetration of power and anti-power is
ignored, if the issue of fetishism is forgotten, then we are left with two pure subjects on either side; we are left with
the subject as ‘a strong ego in rational control of all its impulses, the kind taught in the whole tradition of modern
rationalism, notably by Leibniz and Spinoza, who found here, at least, a point they could agree upon.’ On the side
of capital stands Empire, the perfect subject, and on the side of the working class stands: the militant. Autonomism
– and this is both its attraction and its weakness – is a theorisation of the world from the unmediated perspective of
the militant. Appropriately, Hardt and Negri’s discussion of Empire ends with a paean to the militant: ‘the militant is
the one who best expresses the life of the multitude: the agent of biopolitical production and resistance against
empire.’ (2000, p. 411) And the example of communist militancy which they propose in the closing paragraph of the
book (2000, p. 413) is the perfect embodiment of the Pure Subject: Saint Francis of Assisi! An attractive image,
perhaps, for the dedicated militant, but hopelessly out of touch with the experience of those of us who live enmired
in the filthy impurities of daily fetishisation and who, in spite of and precisely because of that, struggle for revolution.
To understand the force of anti-power we must go beyond the figure of the militant. The scream with which we
started the book is not the scream of the militant, but the scream of all the oppressed. It is necessary to go beyond
the force of overt militancy to ask about the force of all who refuse to subordinate themselves, the force of all who
refuse to become capitalist machines. It is only when grounded in the ubiquity of resistance that revolution
becomes a possibility.