The usefulness of Deleuzian theory for social transformation will vary with the selection of which conceptual contributions one chooses to appropriate. Studying Deleuzian theory is complicated by characteristics of Deleuze and Guattari’s philosophical method. In What is Philosophy?, they define the function of theory in terms of proliferating concepts – inventing new conceptual categories which construct new ways of seeing. In common with many constructivists, they take the view that our relationship to the world is filtered through our conceptual categories. Distinctively, they also view agency in terms of differentiation – each person or group creates itself, not by selecting among available alternatives, but by splitting existing totalities through the creation of new differences. This approach leads to a proliferation of different concepts which, across Deleuze and Guattari’s collaborative and individual works, total in the hundreds.

Instead of seeking to trim their conceptual innovations and neologisms (new words) for simplicity and necessity (an efficiency model of theory – “just in time”, like modern production), they multiply concepts as tools for use, which, although possibly redundant in some analyses, may be useful for others (a resilience model of theory – “just in case”, like indigenous and autonomous cultures). They encourage readers to pick and choose from their concepts, selecting those which are useful and simply passing by those which are not. This has contributed to the spread of diverse Deleuzian approaches which draw on different aspects of their work, but also makes it easy for people to make incomplete readings of their theories, appropriating certain concepts for incompatible theoretical projects while rejecting the revolutionary dynamic of the theory itself. As a result, a large proportion of what passes for Deleuzian theory has limited resonance with the general gist of Deleuze and Guattari’s work, which is not at all about reconciling oneself to the dominant system, but rather, is about constructing other kinds of social relations impossible within the dominant frame. The proliferation of concepts is intended to support such constructions of other ways of being. Another effect of the proliferation of concepts is to make Deleuzian theory difficult to explain or express in its entirety.

In this article, I have chosen to concentrate on the conceptual pairing of states and war-machines as a way of understanding the differences between autonomous social networks and hierarchical, repressive formations. Deleuze and Guattari view the ‘state’ as a particular kind of institutional regime derived from a set of social relations which can be traced to a way of seeing focused on the construction of fixities and representation. There is thus a basic form of the state (a “state-form”) in spite of the differences among specific states. Since Deleuze and Guattari’s theory is primarily relational and processual, the state exists primarily as a process rather than a thing. The state-form is defined by the processes or practices of ‘overcoding’, ‘despotic signification’ and ‘machinic enslavement’. These attributes can be explained one at a time. The concept of despotic signification, derived from Lacan’s idea of the master-signifier, suggests that, in statist thought, a particular signifier is elevated to the status of standing for the whole, and the other of this signifier (remembering that signification is necessarily differential) is defined as radically excluded. ‘Overcoding’ consists in the imposition of the regime of meanings arising from this fixing of representations on the various processes through which social life and desire operate. In contrast to the deep penetration which occurs in capitalism, states often do this fairly lightly, but with brutality around the edges. Hence for instance, in historical despotic states, the inclusion of peripheral areas only required their symbolic subordination, and not any real impact on everyday life in these areas. Overcoding also, however, entails the destruction of anything which cannot be represented or encoded.

‘Machinic enslavement’ occurs when assembled groups of social relations and desires, known in Deleuzian theory as ‘machines’, are rendered subordinate to the regulatory function of the despotic signifier and hence incorporated in an overarching totality. This process identifies Deleuze and Guattari’s view of the state-form with Mumford’s idea of the megamachine, with the state operating as a kind of absorbing and enclosing totality, a bit like the Borg in Star Trek, eating up and assimilating the social networks with which it comes into contact. Crucially, while these relations it absorbs often start out as horizontal, or as hierarchical only at a local level, their absorption rearranges them as vertical and hierarchical aggregates. It tends to destroy or reduce the intensity of horizontal connections, instead increasing the intensity of vertical subordination. Take, for instance, the formation of the colonial state in Africa: loose social identities were rigidly reclassified as exclusive ethnicities, and these ethnicities were arranged in hierarchies (for instance, Tutsi as superior to Hutu) in ways which created rigid boundaries and oppressive relations culminating in today’s conflicts.

According to this theory of the state-form, states are at once ‘isomorphic’, sharing a basic structure and function, and heterogeneous, differing in how they express this structure. In particular, states vary in terms of the relative balance between ‘adding’ and ‘subtracting axioms’ (capitalism is also seen as performing these two operations). An axiom here refers to the inclusion of a particular group or social logic or set of desires as something recognised by a state: examples of addition of axioms would be the recognition of minority rights (e.g. gay rights), the recognition and systematic inclusion of minority groups in formal multiculturalism (e.g. Indian ‘scheduled castes’), the creation of niche markets for particular groups (e.g. ‘ethnic food’ sections in supermarkets), and the provision of inclusive services (e.g. support for independent living for people with disabilities). It is most marked in social-democratic kinds of states. The subtraction of axioms consists in the encoding of differences as problems to be suppressed, for example in the classification of differences as crimes, the institutionalisation of unwanted minorities (e.g. ‘sectioning’ people who are psychologically different), or the restriction of services to members of an in-group (excluding ‘disruptive’ children, denying council housing to migrants). This process reaches its culmination in totalitarian states. It is important to realise that in both cases, the state is expressing the logic of the state-form, finding ways to encode and represent differences; but that the effects of the two strategies on the freedom and social power of marginalised groups are very different.

The state is also viewed as a force of ‘antiproduction’. This term is defined against the ‘productive’ or creative power Deleuze and Guattari believe resides in processes of desiring-production (the process through which desires are formed and connected to objects or others) and social production (the process of constructing social ‘assemblages’ or networks). Desiring-production tends to proliferate differences, because desire operates through fluxes and breaks, overflowing particular boundaries. The state as machine of antiproduction operates to restrict, prevent or channel these flows of creative energy so as to preserve fixed social forms and restrict the extent of difference which is able to exist, or the connections it is able to form. Hence, states try to restrict and break down the coming-together of social networks by prohibiting or making difficult the formation of hierarchical assemblages; it operates to block ‘subject-formation’ in terms of social groups, or the emergence of subjectivities which are not already encoded in dominant terms. Take for instance the laws on ‘dispersal’, in which the British state allows police to break up groups (often of young people) congregating in public spaces. Absurdly, the state defines the social act of coming-together as anti-social, because it creates a space in which different kinds of social relations can be formed. The state wishes to have a monopoly on how people interrelate, and so acts to prevent people from associating horizontally. Another example of antiproduction is the way that participation in imposed activities such as the requirement to work and the unpaid reproductive labour involved in families, leaves little time for other kinds of relationships – people don’t have time to form other assemblages either with other people or with other objects of desire. Hakim Bey has argued that this pressure to restrict connections is so strong that simply finding time and space for other forms of belonging – regardless of the goal of these other connections – is already a victory against the system.

So what, in Deleuzian theory, is the alternative to the state? Deleuze and Guattari argue for a type of assemblage (social group or cluster of relations) which they refer to as the ‘war-machine’, though with the proviso that certain kinds of ‘war-machines’ can also be captured and used by states. This should not be considered a militarist theory, and the term ‘war-machine’ is in many respects misleading. It is used because Deleuze and Guattari derive their theory from Pierre Clastres’ theory of the role of ritualised (often non-lethal) warfare among indigenous groups. Paul Patton has suggested that the war-machine would be better called a metamorphosis-machine, others have used the term ‘difference engine’, a machine of differentiation, and there is a lot of overlap with the idea of autonomous groups or movements in how the war-machine is theorised. We should also remember that ‘machine’ in Deleuze and Guattari simply refers to a combination of forces or elements; it does not have overtones of instrumentalism or of mindless mechanisms – a social group, an ecosystem, a knight on horseback are all ‘machines’. The term ‘war-machine’ has the unfortunate connotations of brutal military machinery and of uncontrollable militarist apparatuses such as NATO, which operate with a machine-like rigidity and inhumanity (c.f. the phrase ‘military-industrial complex’).

For Deleuze and Guattari, these kinds of statist war-machines are also war-machines of a sort, because they descend from a historical process through which states ‘captured’ or incorporated autonomous social movements (particularly those of nomadic indigenous societies) and made them part of the state so as to contain their subversive power. Early states learned to capture war-machines because they were previously vulnerable to being destroyed by the war-machines of nomadic stateless societies, having no similar means of response. Hence, armies are a kind of hybrid social form, containing some of the power of autonomous war-machines but contained in such a way as to harness it to state instrumentalism and inhumanity. Captured in this way, war-machines lose their affirmative force, becoming simply machines of purposeless destruction – having lost the purpose of deterritorialisation (see below), they take on the purpose of pure war as a goal in itself. Deleuze and Guattari argue that state-captured war-machines are regaining their autonomy in a dangerous way, tending to replace limited war in the service of a state’s goals with a drive to total war. This drive is expressed for instance in the ‘war on terror’ as permanent state of emergency. There was a recent controversy about Israeli strategists adopting Deleuzian ideas, which reflects the continuities between state war-machines and autonomous war-machines, but depends on a selective conceptual misreading in which the drive to total war denounced by Deleuze and Guattari is explicitly valorised. The Israeli army is a captured war-machine in the worst possible sense, pursuing the destruction of others’ existential territories in order to accumulate destructive power for a state. For Deleuze and Guattari, it is not the Israeli army but the Palestinian resistance which is a war-machine in the full sense.

The autonomous war-machine, as opposed to the state-captured war-machine, is a form of social assemblage directed against the state, and against the coalescence of sovereignty. The way such machines undermine the state is by exercising diffuse power to break down concentrated power, and through the replacement of ‘striated’ (regulated, marked) space with ‘smooth’ space (although the war-machine is the ‘constituent element of smooth space’, I shall save discussion of smooth space for some other time). In Clastres’ account of Amazonian societies, on which Deleuze and Guattari’s theory is based, this is done by means of each band defending its own autonomy, and reacting to any potential accumulation of power by other bands. One could similarly think of how neighbourhood gangs resist subordination by rival gangs, or how autonomous social movements resist concentrations of political power. Autonomous social movements, such as the European squatters’ movement, the Zapatistas, and networks of protest against summits, are the principal example Deleuze and Guattari have in mind of war-machines in the global North, though they also use the concept in relation to Southern guerrilla and popular movements such as the Palestinian intifada and the Vietnamese resistance to American occupation, and also in relation to everyday practices of indigenous groups resisting state control.

One could also argue that the ‘war-machine’ is implicit in practices of everyday resistance of the kind studies by James Scott. Marginal groups, termed ‘minorities’ in Deleuzian theory, often coalesce as war-machines because the state-form is inappropriate for them.

According to Deleuze and Guattari, war is not the aim of the war-machine (except when it is captured by the state); rather, war-machines tend to end up in a situation of war with states because of the incommensurability of the war-machine with the state and with striated space. War-machines end up in conflict with states because their goal is the ‘deterritorialisation’ of the rigid fixities of state space, often to create space for difference or for particular ways of life. Think for instance of squatters’ movements: in themselves they do not aim for conflict, but rather, seek different kinds of arrangements of space by forming new combinations of unused geographical spaces with otherwise ‘spaceless’ social groups. Yet such movements are often forced into conflict with the machinery of state repression because the state ignores, or refuses to recognise these new articulations. As I write this, the JB Spray squat in Nottingham is continuing a campaign of resistance to reoccupation by state forces acting on behalf of capitalist owners who have no intent of putting the space to use. This is a struggle I would very much encourage readers to support (see this article and related links for details; contact 07817493824 or email jbspray[at]hotmail.co.uk). It is also a clear, local example of how autonomous social movements are forced into conflict by the state’s drive to repress difference.

War-machines are also associated with the formation of special types of groups which are variously termed ‘bands’, ‘packs’ and ‘multiplicities’. These groups are seen as operating as dense local clusters of emotionally-intense connections, strongly differentiated from the ‘mass’, which is a type of group based on large scale, lack of intensity and vertical integration. ‘Packs’ or ‘bands’ instead form as unstable groups, avoiding fixed hierarchies (any leaders emerging are subject to rapid succession), usually with small numbers, and dispersed through space rather than concentrated in particular sites. Their diffusion is enabled by a multiplicity of objectives which resonate through horizontal, molecular connections rather than being represented in overarching structures. They tend to detach materials from the connections in which they are inserted in the dominant system, instead reconstructing different ‘universes’ or perspectives around other ways of seeing and relating. One can think for instance of the way groups of children reconstruct urban spaces as spaces of play, finding new, dissident uses for objects such as shopping trollies. For Deleuze and Guattari, the process of forming ‘bands’ or ‘packs’ is necessarily dangerous, risking the self-destructive implosion of small groups, but also offering hopeful possibilities of forming ways of relating which are more open to difference than those prevalent in the dominant system.

Deleuze and Guattari’s usefulness for radical activism is by no means limited to this particular pair of concepts, but this way of thinking about social transformation raises useful questions and provides insights into how autonomous groups differ from dominant hierarchical forms of social collectivity. For instance, this theory points towards the need to avoid duplicating statist ways of relating within autonomous spaces, and to avoid coalescing in formal organisations which ultimately lead back into the state-form (albeit usually through the addition of axioms). It also suggests the inevitability of antagonism between radical movements and the state, even when the goal of a radical movement is simply to defend or express its own difference. Strategically, therefore, autonomous activists need to be prepared to ‘ward off’ the state, both within movements (by challenging statist ways of thinking and acting) and in relation to the wider context (by resisting state repression). According to Deleuze and Guattari, there is a basic incompatibility between state ‘antiproduction’ and the flourishing of difference, and this requires overcoming the former. This requires attention to the creation and defence of autonomous spaces, in full awareness of their underlying transformative potential.