The Dislocating Perspective of Assemblages

During a visit to Canada in April 2008, American Homeland Security Secretary Michael Chertoff announced to his audience that fingerprints are not part of a person's personal data: 'A fingerprint is hardly personal data because you leave it on glasses and silverware and articles all over the world; they're like footprints. They're not particularly private.' 1  A reaction was not long in coming. It came from Canada's privacy commissioner, Jennifer Stoddart. 'Fingerprints constitute extremely personal information for which there is clearly a high expectation of privacy.'

The debate about where the private begins and the public ends has a long history. It goes back to the French Revolution. The end of the Ancien Régime, symbolized by the beheading of Louis xvi in 1793, ensured that the sovereignty of the monarch made way for the will of the people. No one had the exclusive right to rule in their own name any longer. Two spheres were created to express what was understood by 'life'. In the private domain, the state was to leave the individual in peace. Beyond the threshold of the home, everyone was free to espouse his or her own desires and opinions. In the public domain – the agoras of the cities – however, the individual was a citizen who was to set aside his desires and opinions for the common good.

The separation between public and private worked quite nicely for a couple of centuries. Now, however, it seems its best days are over. Municipal intervention teams, made up of inspectors from social services, energy suppliers, representatives of housing corporations and other organizations, show up unannounced at the homes of residents with problems. This campaign is called 'beyond the front door'. Various technologies (security cameras, data mining, rfid chips in clothing, dna tests) are employed to increase the perception of security in the broadest sense of the word. 'Police-like' responsibilities, such as the security of semi-public areas like shopping centres, airports and residential areas, are increasingly being carried out by commercial actors. These practices and measures seem very diverse, yet they have a lot in common. They are all employed in the same processes of the prevention of perceived risks. Sadly, debate on this new method of governance has been hijacked by the catch-all word 'privacy'. When we look at the changes in the issue of security without bias, however, we see a more fundamental problem emerge. In all sorts of areas, there is a certain overlap between public and private practices. These overlaps or convergences are never stable or static. They are not sharply demarcated and they are constantly changing: in form, in reach, in composition. In order to uncover the 'ground' of this mobility or fluidity, we need a different ontological and epistemological premise than the private / public dichotomy outlined above, upon which modern society is supposedly based.

In this article I intend to approach the fluid character of social reality from two directions. In the first place I want to make a contribution to the explication of the relationship between governance and social reality. Relying on Michel Foucault's analyses of power I shall first attempt to go one step further into his conceptual world. His analyses of disciplinarian practices in which the individual is shaped by all manner of power operations are my starting point. Via the work of Gilles Deleuze, and in particular his concept of 'assemblage', I wish to give greater depth to the link between the social and governance. By relating this concept to security regimes in our immediate environment, I shall show that this mobility should not be confused with 'chaos' or a 'new disorder'.2 And this leads directly to the second objective of this article. When we approach the social based on the concept of assemblage, we see countless hybrid connections emerge, which enter into unexpected relationships with one another. Which relationships are we then talking about? How do these attain a certain consistency or coherence? Through these questions I ultimately aim to outline a number of rough characteristics of how the issue of governance has come to circle ever closer around social reality.

Unlike in the sovereign society, the realization that the individual can be socially engineered emerges. Building on the humanist insights of the Enlightenment, various techniques are applied in the separate institutions to teach socially desirable behaviour. The consequences of this are most visible in the army. In the seventeenth century, the soldier is still described as someone one recognizes by his courage or fighting spirit. This changes, however, in the eighteenth century. From a meaningful body that radiates energy and honour, the body of the soldier is reduced to a cog in the machine. The soldier is shaped by exercises in which he learns to hold his head high and his back straight and to move in a uniform manner. Through corrective exercises, which are aimed at generating specific and measurable effects, the soldier is furnished with a coherent identity. This disciplining of the body does not take place only in army camps. Discipline-oriented techniques are also applied in other societal institutions: the prison, the hospital, the school and the workplace. And simply because its disciplinarian effect is equivalent to those of a series of other institutions with which the individual is confronted throughout his life, the army can be compared to the factory, which in turn has everything in common with a prison. Without interruption, the individual in fact moves from one institution to the other: from the family to the school, from the school to the factory, and so forth. We are dealing with a continuous progression in a sequence of separate spaces through which the institutions continually refer to one another. At school you are told you are no longer at home. At work you hear 'you're not at school anymore'.4

The picture of society that is presented here is a succession of separate spaces, whereby the individual moves from point to point as though there were constantly something new to be added to his life. To emphasize this transformation, Deleuze and Guattari speak of moving in a segmented or striated space.5 With this they indicate that space in a disciplinarian society was above all an oriented space, that is to say an expression of a progressive perception of time in which the individual constituted himself as a subject and emancipated himself with an eye towards a final state to be attained. In reading Discipline and Punish, however, one is immediately struck by the fact that nowhere in it does Foucault address the question of which power relationship acts on the bodies in the spaces 'between' the institutions of modernity. In other words, what forms or categories of power continue to operate in the open space of cities? For this we must go back to two texts by Foucault from the first half of the 1970s. In them he takes a cautious step towards an explanation in which the public space increasingly becomes the domain of an effort towards regulation or control of life. In these texts he refers to biopolitics, a form of power that emerges in the second half of the eighteenth century and regulates social life from the inside out. With the concept of biopower he derives from this, Foucault has a different type of power operation in mind than disciplinarian power. Whereas discipline is directed at the individual body, biopower concentrates on the populations. The object of political strategies is not the social engineering of the individual body, but the body as a type. The term 'biopolitics', which would be addressed in greater detail in The Will to Knowledge (1976), first appears in the lecture 'La naissance de la médecine sociale' which Foucault delivered at the State University of Rio de Janeiro in October 1974. 'For capitalist society, it was bio-politics, the biological, the somatic, the corporal, that mattered more than anything else. The body is a bio-political reality; medicine is a bio-political strategy.'6 Through population control biopolitics has a direct relationship with bare life itself. The population is no longer an abstract quantity, nor does it coincide with the number of inhabitants in relation to a habitable territory. On the contrary, it manifests itself, in Foucault's words, 'as an object of surveillance, analysis, intervention, modifications, and so on'.7

In the process, the conditions under which people live and the way their bodies function as the bearers of biological processes (public health, births and deaths, average lifespan, population growth, education) become part of the 'governance' of society. Foucault expresses this method of governance with the neologism gouvernementalité. In it the ratio is not predicated on the 'control' of the population of which Machiavelli's The Prince (1532) was exemplary, but on the 'management' of relations among people. That is to say, the objective is the optimization of all those aspects of life that promote the welfare of the population as a whole.

Deleuze's argument that control defines the relationships of the social sphere leads to the objection that it is insufficiently clear in what way this form of power genuinely differs from the two eighteenth-century poles of discipline and biopower. Aside from the fact that control also played a fundamental role in the sovereign and disciplinarian societies, the examples in 'Post-scriptum sur les sociétés de controle' do not provide a picture different from that of Foucault's disciplinarian analyses of power. We are still dealing with techniques that turn individual bodies into productive, efficient and obedient labourers. All things considered, nowadays the method used on motorways to indicate that a driver has committed a violation ('You are driving too fast', 'Maintain sufficient distance') has no other purpose than the immediate correction of the driving behaviour. Yet Deleuze undeniably has a point when he links spatial transformations with changes in social reality itself. Whether we define this development in terms of 'risk' (Ulrich Beck in Risk Society, 1992), 'security' (David Garland in The Culture of Control, 2001) or 'ict' (Manuel Castells in his network trilogy The Information Age, 1996), it is clear that the term 'environment' has become a very broad concept in our present society. In particular, Deleuze shows that the striated space of the disciplinarian society is making way for a smooth or open space. Whereas the disciplinarians techniques operated in closed and fixed spaces (walls, borders, gates), each with its specific function, the control society operates through constantly changing networks or open spaces. Mobility, flexibility and acceleration are the new qualities of these environments.

An open space is no simple concept. The word 'open' can give rise to all sorts of misunderstandings, misunderstandings related to form, trajectory and unity. An open space differs from a striated space in three particulars. In the first place in its form: the surface of a striated space is delimited and enclosed; special spaces are assigned to categories of persons (school pupils, patients, prisoners). An open space has no definite boundaries or a privileged form. It can be extended in any direction and is confined only by a horizon that shifts as the audience moves. For this reason, we can no longer speak of an absolute 'inside' or 'outside'. Even concepts like 'distance' or 'opposite' lose their classical meanings here. In the second place, the relationship between point and line is inverted. In a striated space a line lies between two separate points. As we have seen, each of these points (school, factory, hospital) has its own customs. In an open space the point lies between two lines, which implies that the separate points are subordinate to the trajectory that continues on a horizontal plane or field. An open space stimulates and orders separate dimensions without turning them into a totalizing whole. There is only a continual variation of form and size. In the third place, the nature of the line differs. Whereas in a striated space dimensional lines and closed intervals can be distinguished, in an open space we are dealing with directional lines and open intervals. An open space realizes itself in what it causes to disappear. That does not make it a homogeneous or undivided space, as though there were no segments or ruptures within it. Multiple spaces can be present in an open space, just as multiple languages exist in one language. We should only understand that the ruptures between the spaces are no longer absolute, as they are in a striated space in which one must pass through all sorts of physical barriers (gates, booms) in order to enter. An open space, in and of itself, always has multiple meanings. Or to put it another way, you can be private in a public space and public in a private.

What does this mean in terms of governance? Or expressed another way, in what way do all manner of 'hybrids', to use one of Bruno Latour's terms, emerge in our environment, whose objective is the prevention of potential risks? If we look at recent writings on the imbedding of the issue of security, we find discussions of 'surveillance assemblages'.13 This term expresses the fact that surveillance is driven by an uncontrollable need to bring together actors, practices, technologies and information systems and to integrate them into larger entities. These can be insurance companies, national security, multinationals, social security, shopping centres, and so on. All these separate practices have a distinct style of operation, use their own information systems, apply specific definitions of normality and deviating truths, and all these characteristics are aimed at making a specific public (or to put it a better way, 'publics') visible. It would therefore be inaccurate to identify this public with an individual or a population.14 Each medium creates its own users. This is about the 'public of an insurance plan', the 'public of a shopping centre', the 'public of a policy measure'. Because of the growing influence of information and communication technologies on contemporary society and the organization of the urban space in particular, this new entity does not manifest itself in a demarcated space ('school' or 'national state'), but rather actualizes itself in an open environment in which people encounter one another differently and are monitored in a different way. To put it a better way, surveillance is incorporated into the movement of a public through an open space.

Take the example of a Sunday football match. At 1:29 p. m. I close my front door behind me. The lady who lives across the street looks at me inquisitively. To increase local security she's signed up with Burgernet ('Citizen Net'), a police initiative to enrol citizens in the investigation of crimes. The police left a message on her answering machine yesterday with the description of a man who has broken into several cars in the area. If my neighbour notices anything she can call a direct number, whereupon the dispatcher sends the nearest police officers to the location. On the way to the neighbourhood shop to quickly buy a pack of gum, I am watched by a network of intelligent cameras that link my face to a database of photos of recidivists, comparing me to millions of people in 60 seconds. It is now 1:35 p. m. The neighbourhood shop, in turn, is part of the Collectieve Winkelontzegging ('Collective Shop Ban') project. This is an initiative of shop owners and shop-owners' associations to combat trouble on their own. If someone behaves inappropriately in the shop, be it shoplifting, or being rude to the staff, this person can be banned. This ban applies not just to the neighbourhood shop, but to all the other shops in the city centre. By now it is 1:41 p. m. With a pack of Sportlife in my pocket I press my public transport chip card against the scanner of the turnstile at the metro station at 1:47 p. m. 'Easy, fast and secure' – these are the marketing terms printed on the chip card. Thanks to a unique identification code, all my travel details are recorded in a central database. This provides a complete picture of the distances I travel by metro, bus, tram and train. When I arrive at the stadium I show my season ticket to the stewards who are responsible for order and security in the stands. It is now 1:56 p. m.

In less than half an hour, from my front door to the football stadium, I have passed five difference surveillance assemblages. At first glance we move autonomously and without friction through the same open space. Yet while this environment gives the suggestion of being continuous, it is actually populated by so many different assemblages that any openness or smoothness is merely illusion. Most of the time the unique interplay of concealments and revelations remains invisible to the moving public. This changes only when the public transport chip card is blocked, facial markers match details in the shop-owners' association register, or the stadium stewards have been notified of the rather turbulent football history of a particular person. While each 'island' has its own values, its own logic and principles, we should not imagine that these assemblages have nothing to do with one another. These environments can just as easily ignore or exclude one another – sometimes they even turn against one another, but more often they reinforce one another, overlap or converge into new assemblages. I have confined myself to a few examples. Private institutions are getting more and more access to information from government departments, and vice versa. Organizations and institutions such as internal revenue departments, police, social services, supermarkets and hospitals also exchange information in order to chart life. In addition, government personnel are increasingly working for private parties. The largest shopping centre in Europe, the MetroCentre in Gateshead, England ('If we don't have it, you don't want it'), is equipped with the latest surveillance electronics, but that has not kept its management from increasing security within its walls by hiring police officers from the Northumbria Police. Not only does the police still enjoy great symbolic power and authority, but this also gives the shopping centre's security personnel access to the information sources and intelligence (crime-related data) of the police force.15

In short, information travels back and forth between practices over all sorts of complex networks; in one assemblage citizens turn out to be policemen, in another assemblage policemen are in the employ of private security firms. Unfortunately, research into the splintering of security measures usually focus on one environment, for example Burgernet, camera surveillance or private security. Research that is not limited to a single environment, but rather outlines how separate elements affect different practices, is scarcely undertaken. As a result, too little attention is paid to the fact that a surveillance is never a starting point or an end point, but always a middle, literally a medium in which elements from all sorts of heterogeneous practices interconnect. Instead of seeing in these assemblages a simple curtailment of the freedom of movement or an invasion of privacy, we must try to understand its ontological and epistemological premise. For one element of an assemblage can break away, to a relative extent, and go on to function in another assemblage. It can be taken out of one assemblage, concludes Manuel DeLanda in A New Philosophy of Society, and be incorporated in another context.16 In turn, this context is formed by new variables, unforeseen interactions and other outcomes. Order and unity are not provided a priori; they form at a secondary level, from the relationships within the assemblages. This still does not answer the question of the consistency of an assemblage. In other words, in what way are heterogeneous elements kept together in a surveillance assemblage? Is there a specific 'causality', and if so, how can we explain it?

Deleuze and Guattari distinguish two dimensions that give an assemblage order and cohesion, in other words a basis from which to operate: the horizontal and the vertical dimension. The horizontal dimension is formed by the relationship between expression and content. By the content aspect of an assemblage, Deleuze and Guattari mean the interaction or organization of qualities among objects, bodies and animals in a concrete practice. They call these practices non-discursive formations. These can be institutions like a school or a prison, but also political events (the French Revolution, 9 / 11), economic practices (insurance systems) and (social) processes (exclusion). By the expressive aspect they mean the totality of signs that links these formations. This can include linguistic expressions (symbols, words) and non-linguistic expressions, such as the bodily postures or clothing of persons. For clothing is more than simply something to keep the body warm. It is also used to express a particular function (police officer, steward), indicate a social status (a three-piece suit) or works as a form of self-styling (football supporter).

For the foundations of the difference between the two aspects, they base their argument loosely on the work of Danish linguist Louis Hjelmslev, who in Deleuze's terms, has developed a Spinozaesque theory of language in which content and expression do not rely on a predominant signifier. Expression, Deleuze and Guattari argue in A Thousand Plateaus, does not coincide with a signifier. At the same time, content is not the same as the signified. There is no equivalence or analogy – in the sense of 'description' or 'correspondence' – between the two. Content and expression function relatively independently from each other. Relatively, because they only exist through the relationships that take place between them. In no way are content and expression directly or absolutely dependent on each other. In this Deleuze and Guattari reject the supposed synthesis between content and expression. Take the statement 'I swear'. This takes on a different meaning when it is spoken by a pupil to a teacher, by a minister taking the oath of office, or by a defendant during a trial. For this reason, it is not enough to observe that only the setting (school, parliament, courtroom) changes. That would suggest that the statement remains essentially the same. Not only do the elements or 'the nature' of the separate settings differ, but the statement itself takes on a different expression.

A rather fundamental distinction, it seems. Yet the attention of the social sciences turns sporadically to everyday interactions among people in divergent formations. Criminology, for instance, seems to nurture a structural distrust of the incidental character of everyday reality. In order to safeguard the sustainability and homogeneity of the social, natural forms of expression (sensation, gossip, frustration, kick) and so-called coincidental elements (the role of women in organized crime) are seldom investigated. These are largely kept outside 'the order of the discourse', to quote Foucault. Criminology prefers to concentrate on patterns or expressions that can be labelled as rational and that are the product of abstract quantities such as 'the economy', 'the culture', or 'the criminal organization'. In this it builds on a structure of general laws that can be applied to individual elements. In this reduction of social reality to a static-free order, there is only room for linear processes and predictable behaviours.17

According to Deleuze, however, expression is in no way the logical consequence of content, in the sense that without content no expression can exist. Or to put it another way, there is no causal link between content and expression. If there are notable similarities, this is only because these are the consequences of the relationships between content and expression in an assemblage. Similarities must therefore not be seen as the cause of production. This confuses process and product, argues Brian Massumi.18 Content and expression are independent processes that operate separately from 'the incident' or 'the case' to which they refer. In this Deleuze and Guattari are going quite far. Not only do they call content and expression two 'non-parallel formalizations', but these also have their own form and substance that are again entirely heterogeneous, and sometimes even multiple forms and substances.19 From this standpoint there is no final form that still ensures a connection between content and expression. Between content and expression there is only a process that links the two forms. This process itself has no form. Deleuze speaks of a zone of indiscernibility, a play of forces, which he characterizes as pure intensity.

This brings us to the last aspect of the vertical dimension. Perhaps the misunderstanding that territorialization only curtails the mobility of an assemblage has been created. The process of territorialization does bring about a unification of a social space, a certain cohesion of the place and identity of the persons present. But a territory like a gated community or deprived neighbourhood cannot always maintain its form; it does not remain a cohesive arrangement of a concrete social field indefinitely. An assemblage is only conceivable against the backdrop of an infinite mobility of social reality. In order to thematize this, Deleuze speaks of a line of deterritorialization, a movement that sets an assemblage adrift. This line escapes every assemblage, which means that it is constantly breaking open the existing field of arrangements. It dismantles every signifying and every formative order by creating new openings and new connections. So an assemblage can break down at any moment. This movement of continual decomposition always corrects itself. Deleuze and Guattari call this reterritorialization. The two movements imply each other. The one does not exist without the other. Every reterritorialization entails a deterritorialization.

So an assemblage consists of four aspects: in addition to content, expression and territory, deterritorialization is also part of an assemblage. This last notion needs further explication. There is always something that escapes an assemblage. Deleuze calls this alternately a line of deterritorialization or a line of flight. In Dialogues he describes this line as follows: 'It liberates a pure matter, it undoes codes, it carries expressions, contents, states of things and utterances along a zigzag broken line of flight, it raises time to the infinitive, it releases a becoming which no longer has any limit, because each term is a stop which must be jumped over.'22

More specifically, a line of flight has two characteristics. In the first place it is abstract. Because the line of flight is abstract, it should not be understood in terms of content or expression. It goes much further. It is abstract because it ignores not only the difference between content and expression, but also the distinction between form and substance. A line of flight is therefore not abstract merely because it is immaterial. It is also formless. In the second place, a line of flight is immanent, which means that it is always part of a concrete assemblage.23 The line of flight is incorporated in the organization of an assemblage. In order to emphasize the openness of an assemblage and the mobility of social reality, therefore, there has to be something that breaks through the order and cohesion and establishes a connection to other elements. This does not happen by synthesizing or adding elements, but by removing them from an assemblage and forming a different assemblage by connecting them to new elements. This is how movements of deterritorialization form new assemblages. In a dual movement, the territory is continually being reorganized, and as the principle of a deterritorializing movement, no less. For this reason, Deleuze considers the line of flight primary; it comes before everything else. A line of flight, after all, has no territory. Territories always come second.24

1. The idea of social engineering is based on a distance between an individual and an environment. Without being part of it, the individual faces his immediate environment. From an external position, he can apprehend and comprehend social reality in its entirety. As an answer to its limitations in bringing about effective changes in society, governance should not be seen as a strategy one can deliberately strive for. It is not based on a subject-oriented approach. If we look, for instance, at the technologies described (camera, public transport card, neighbour) in the surveillance assemblages, it would be a mistake to interpret these as neutral instruments that can purposefully be employed to achieve long-term objectives. In reality, technologies are never value-free. Technologies are social before they are technological. Rather than defining this kind of element in isolation, we should therefore look at their context and its effects. In other words, a 'co-functioning' is needed to achieve meaning. Otherwise these technologies remain marginal or they are little used. This takes us far from a traditional subject philosophy ('I think, therefore I am') in which the actions and decisions of a person are the product of a free, autonomous actor who always remains equal to himself. We have to assume that the effects of an assemblage cannot be ascribed to an individual and are not ascribed to an individual. Instead, the point is that the individual himself is an assemblage, a ceaseless process of transformation that, as it were, no longer has a beginning or an end.

2. Behind the distinction part / whole lurks the hypothesis that parts exist because of the whole ('something that already exists'). Not only are they part of the whole, they maintain the whole in existence. Evocative examples include the well-known theories that speak of 'society' ('risk society') or 'culture' ('prevention culture'). Yet when these focus on society as a closed whole, one can no longer speak of a strategy that extends in all directions and operates the same way at all levels of society. If the analysis that society is not an immutable, static quantity, an undifferentiated social space that has a fixed order, is correct, then we must stop studying the all-encompassing whole, that is to say society as a homogeneous entity with an internal cohesion. Instead we should look for the countless different signifying and formative arrangements created by new types of relationships and the categories and meanings that function as a result. In more general terms, we should focus on concrete assemblages, keeping in mind that there are always lines of flight that establish connections with unforeseen elements in other assemblages. On that point we have already observed that an assemblage is never self-contained, but rather always refers to other environments that operate or are yet to operate, with as a result an almost unlimited growth of completely different transformation matrices and productions of social syntheses in social reality.

3. The dynamic in a social constellation has traditionally been described in terms of a causal infrastructure. Behind this idea we find the presumption that the actions of individuals are the product of the whole ('society', 'culture', 'group') these persons are part of. This whole precedes the actions of persons, so that these actions unfold in an identifiable and predictable way. Subsequently, the explanation for these behaviours is sought in rational capacities. The degree of social engineering then coincides with the stubborn view that people select the option that they expect will benefit them most. A characteristic feature of an assemblage, however, is that everything, in principle, has the same potential for meaning. Every connection creates something new. Therefore we should free ourselves of the idea that everyday forms of expression such as emotions, sensations, gossip and frustration play no signifying role, behaviours that in the philosophy of social engineering are still dismissed as irrational and unimportant. An assemblage is a matter of an infinite potential of relationships that continually bring about different connections between things and people. This lends unity to social reality. Not that of an eternal and static substance, but that of an unlimited surface upon which a unique play of interactions unfolds, without these being able to be traced back to fixed characteristics or rational processes.