CHAPTER ONE

INTRODUCTION

The aim of this book is critically to discuss some of the fundamental problems in the social sciences emanating from the classical tradition referring back to ancient luminaries like Marx, Durkheim and Weber. As shorthand I shall call the point of reference of my discussion “sociology,” although this classical line of social theorizing is also present in disciplines like anthropology, history and political science. Another aim of the book is to identify certain conceptualizations that might lead us in a new and more fruitful direction. These objectives have grown out of a feeling that much of the thinking in sociology is marked an intellectual exhaustion of sorts. As I see it, one of the main problems hampering these disciplines is that our expositions of all sorts of phenomena are given in advance. We have a certain stock of explanations that can be adapted to anything. This explan-atory uniformity makes the whole enterprise of social science not only repetitive and boring, but also essentially predictable, preventing our find-ing anything really new. The diversity of the world is not allowed to im-pinge on us; instead, it is tamed and safely placed within our neatly fash-ioned categories.

All lines of scientific research are supposed to question their own foun-dations now and then, but one of the persistent findings from the sociology of science is that this happens quite rarely. Sociology is no better: the tra-dition has been somewhat sacralized. It is about time to break this spell and to reassess the presupposition underlying this activity. The argument of this book is that the social sciences can be renewed by the incorporation of a few insights taken from continental philosophy or, to be more precise, the theories of Bruno Latour and of Gilles Deleuze and Felix Guattari. The theories of the latter were presented (albeit in a transformed way) to the English-speaking world as a part of the phenomenon of “French Theory” in the 1980s, and Latour was one of the protagonists of the “Science Wars” in the mid-1990s. This background may prejudice some readers against the enterprise of this book, but this form of thinking nonetheless offers us important lessons, and it would be a pity not to take advantage of them.

Chapter One

2

Bruno Latour (b.1947) has been a significant voice in social theory in recent decades, and if this voice were to be taken seriously, it would lead to a significant change of “terrain” compared with that of the sociological tradition. However, Latour’s voice has to some degree been restricted to the field of science studies1 but even here he is often treated as a bête noire. Latour is trained as a philosopher, but turned to sociology and social anthropology in the mid-1970s. In France sociology and anthropology are institutionally located within the humanities and Latour’s career has been somewhat uncommon in this respect, because he has most of the time worked at prestigious, “great schools” like the École des Mines in Paris and later the Institut d’études politiques in Paris. Latour’s work belongs to the broad movement of constructivism which came into prominence in the 1970s. Within science studies this line of inquiry involves foremost a methodological orientation whereby scientific facts are seen not as a set of ideas or doctrines, but as outcomes of local practices within a certain com-munity. Latour began his work in science studies in the mid-1970s with anthropological fieldwork at the Salk Institute in La Jolla, California. This study was presented in the now seminal book Laboratory Life (Latour and Woolgar 1986). The original 1978 subtitle of this book was The Social Construction of Scientific Facts, but in the second edition of 1986 the word social had been taken out. This seemingly minor change was actually a telling sign of Latour’s critique of the social constructivist movement. This movement had, he held, only shifted the weight from a traditional view whereby scientific facts are discovered in nature to a view whereby society explains everything. He has since the mid-1980s promoted a di-vergent form of constructivism identified under the label “anthropology of science” or “actor-network theory” (ANT). The latter label has become very popular in the English-speaking world as a catchphrase, yet Latour himself has alternately embraced and rejected it (2005). I shall here use “anthropology of science”, with which he seems to be most comfortable.

One of Latour’s main ideas is that humans and objects must be studied in conjunction and not separately. Society and technical artefacts presup-pose each other and each can only be seen in isolation after a process of purification by researchers and intellectuals. Most of what is labelled social science, however, separates humans from their technical objects. Social relations – that is, relations between humans – have so far made up the foundations of the social sciences.

Gilles Deleuze (1925 – 1995) and Felix Guattari (1930 – 1992) may perhaps be characterized as one of the most remarkable pairings in philo-sophy. Deleuze was a professional philosopher who taught at universities and from 1970 at the new University of Paris-VIII at Vincennes. This uni-

Introduction

3

versity became one of the centres for the post-1968 activism and where Deleuze’s friend, Michel Foucault (1926 – 1984), for a time chaired the department of philosophy. Guattari worked at the experimental psychiatric clinic of La Borde and had passed his exam as a psychoanalyst at Lacan’s École Freudienne in 1969. He was also a political activist on the left in France and part of the anti-psychiatry movement in the 1960s. He was, moreover, a central figure in groups promoting a wider application of the “institutional analysis” of La Borde. From 1965 the collective FGERI2 organized some of this work and in the early 1970s the French state fi-nanced some members of this group as contract researchers. In the 1980s Guattari was involved in establishing Radio Tomate, an activist radio station, and in this period he worked with environmental activists and other groups (Dosse 2007).

The main focus of my interest in their collective work concerns the two books Anti-Oedipus (1984) and A Thousand Plateaus (1987), published in 1972 and 1980, respectively. These two books were the fruit of a radical political conjuncture in which it was possible to question everything. This background still imbues a certain freshness to the ideas promoted in them, although they are, like all books, marked by the debates of their time. Deleuze and Guattari were critical of the structuralism of the 1950s and 1960s, which emphasized the invariant traits in any given condition, while at the same time they shared with structuralism the critique of the freely acting subject that existentialism promoted. They also followed the struc-turalists’ interest in semiotics, but they applied it differently and were criti-cal of linguistics as a discipline. Their philosophy can therefore be labelled as “post-structuralist”, but what I think is the most interesting aspect of their philosophy for sociological thinking is the conception of a primordial energeticism that is present in all sorts of fixed social entities and contin-ually threatens to destabilize them. It is this fundamental vision and the concepts linked to it that make up some of the lessons that sociology can learn from Deleuze and Guattari.

I shall, of course, go into more detail on both the theories of Latour (see Chapter 5) and those of Deleuze and Guattari (see Chapter 6), but one of the aims of this book is also to promote a critique of and dialogue with the sociological tradition.

The Problem

What is the main problem with the way we do social science? A common way of discussing sociological theories is to view them as different ways of studying society. We all know there are different perspectives within

Chapter One

4

these disciplines: some take the bird’s-eye view while others take a worm’s-eye view; some emphasize conflict while others emphasize coop-eration; some emphasize a close study of the subjective world of actors while others insist on collecting quantifiable data in order to make possible a generalization of the findings to a population. Even though they might give us different “societies” as a result, they are mainly understood in terms of epistemology, the way we extract knowledge about our object of study.

The critique levelled at sociology in this book is not so much concer-ned with how we can best study human societies, but rather with the way sociology tacitly subscribes to certain ontological assumptions. This book focuses on the entities that sociologists regard as important and how they define the relations between them – in other words, the make-up of soci-eties. My main contention is that the manner in which sociology distri-butes properties to the different entities that we might find in a society de-termines how we experience, talk about and write about them to a great degree. By redistributing these properties we can experience, talk and write about a society in new ways, and hence the title of the book: we might arrive at a new society altogether. My aim is not to provide theore-tical meta-comments on human societies, but by changing the conceptual framework I seek to contribute to new ways of doing sociological re-search. The aim is always to facilitate empirical research, and I do so here by analysing concepts and how they operate.

At this point it may be useful to ask what is meant by society. The defi-nition of this very imprecise concept depends, however, on the way a gi-ven sociological theory distribute properties between entities and how the relations between them are pictured. In this context I shall focus on the common strategy in sociological analyses to advance two perspectives at the same time: one taken from without the topic at hand (transcendence) and the other emerging within (immanence). The latter involves following the actors and what they do, say or write. The former strategy, however, stipulates that the actors have to be situated in a proper social context and in Latour’s words it gives sociology “fixed frames of reference” (1996a, 169). This strategy is easily recognizable by the way entities are given clear identities and properties at the outset and these are applied to ana-lyses without much reflection. Latour and Deleuze and Guattari follow a completely different practice, emphasizing absolute immanence giving different definitions of “society” as a result.3 Latour’s sociology is a soci-ology “which has fluctuating referents” (1996a, 169) and his starting point is always in the “middle of things” (2005, 25) with few preconceptions. Deleuze and Guattari give primacy to connections between undetermined

Introduction

5

entities before they are given any essential characteristics. They withdraw to what is logically prior to the fixation of identities. Another facet that distinguishes their approach is that their concepts are completely empty. They are purely functions and have to be applied to concrete empirical ma-terial in order to be meaningful (1987, 40-57). Hence, every new case must involve a new effort and one cannot apply the same categories from one case to the next. One of the consequences of these immanentist ap-proaches is that there is, in principle, no society with fixed properties. Latour even rejects the whole concept of society because it involves a preference for humans over non-humans. He has instead coined the concept of the “socio-technical collective” or “collective of humans and non-humans” (Latour 1993b, 2005). The Latourian theory is thus a socio-logy without society. Deleuze and Guattari, on their part, make use of con-cepts like social plane, socius, social machine and society (1984, 1987), but the way they use these concepts in their analyses may vary from topic to topic.

This book will apply the concept of society in the same way as it is used in the literature that I discuss, and this will, unavoidably, lead to changed meanings of this concept here and there. One of the topics of this book will also be to show how traditional sociology constructs its outside and how the change of perspective between a transcendent and an imma-nent viewpoint takes place in different theories.

I shall now turn to what I hold to be some of the fundamental presup-positions in the sociological tradition. The aim is to pinpoint the properties that this tradition prefers and the relations it leaves out. Below I sum-marize in a set of theses what I find to be the foundations for not only so-ciology, but also other branches of the social sciences. These theses will form the basis for much of the discussion in this book:

1. The social sciences study the relationship between people or col-lective entities made up of people. 2. Either people or collective entities, or both, are granted agency in the form of potentialities residing in them, 3. or the actions of people or collective entities, or both, can be linked to some sort of meaning that makes them do the acts they do.

The “either-or” properties of Theses 2 and 3 are linked to discussions about modes of inquiry in sociology, and they will be treated more tho-roughly later in this chapter. Thesis 1 states the preference for relations be-tween humans, and it is the most fundamental one because it makes a claim about which entities are most important to study, and it is commonly held to be the most self-evident for many sociologists. However, one of

Chapter One

6

the main critiques formulated in this book is that much social scientific thinking is not able fully to integrate two of the most characteristic fea-tures of human social life: that we have languages and that we form an intimate relationship to technical objects. We must, then, formulate two more points to characterize how sociology commonly treats these features:

4. Objects are treated as intermediaries between humans, as neutral tools, as something over which we struggle, or as separate entities with their own potential to determine human life. 5. Signs in the form of speech, texts, images or bodily postures are understood as communication.

To avoid potential misunderstandings I should state that sociology ignores neither technical objects nor signs, but my claim is that they are not con-stitutive parts of the ontology. In the case of technical objects, they can change status from being faithful and neutral tools to being an overwhel-ming “technology” or a “technical system” that might determine our fate. Or, perhaps, a given technical artefact has the inherent potential to change our social relations for the better or worse (see Thesis 4). These latter posi-tions are all examples of a technical determinism whereby potentiality is ascribed to techniques or technical objects while the human part of the equation is diminished. In this way techniques become self-acting – which, I shall argue, is just as problematic as the main claim in Thesis 1. Further-more, according to Thesis 4, technical objects are not only portrayed in so-ciology as neutral tools or part of an abstract “development” or force that changes our society without our having any say in it, but they are also ob-jects worthy of our admiration and competition. This is well known in the common struggle for all sorts of vanity objects, ranging from powerful cars to the latest electronic gadgets, and it is well portrayed in studies of different lifestyles (e.g., Bourdieu 1986).

Thesis 5 may perhaps be met by headshaking disbelief: shouldn’t lang-uage be a question of communication? What could be more self-evident? The answer that Deleuze and Guattari might give would be that language can be communicational in specific situations, but we must make clear what is meant by communication and in this context it is understood accor-ding to the Schema of Linguistic Communication. This schema is closely related to information theory and presupposes a transmission of a message between an addresser and an addressee. These two have a physical channel through which the message is transmitted and there is a psychological connection between them. Finally, they share a common code that makes it possible for the addressee to understand the message (cf. Greimas and Courtés 1982, 37-38). For Deleuze and Guattari this situation is much

Introduction

7

more controlled and streamlined than what is common in linguistic ex-changes. It is a situation where nothing new can happen because there is no opening for new implicit meanings. This is so because there is more to a statement than its message. Language orders the world for us and im-poses a kind of functionality on it. If I make the statement: “all swans on the lake are white”, I do not impart this freely to some interlocutor, but I oblige the other party to believe that this is the case. In other words, I indi-rectly impose a social obligation. This implicit presupposition of my claim forces us to ask questions about the situation where the statement was made and one might ask on what grounds it was made, etc. It might be an example from the theory of science in a classroom, a claim about a state of affairs while walking along the shore of the lake or it might be a malapro-pos like a line from a play by Eugène Ionesco.4 This is the function of what Deleuze and Guattari call order-words (mots d’ordres) which they hold to be the foundation of language and where what is indirect or implicit in discourse is of major importance (Deleuze and Guattari 1987, 75-80). The status of the statement depends on what they call a collective assemblage of enunciation. I shall discuss this concept in more detail later in this chapter.

Bruno Latour formulated another way to distance himself from the communicational model. In an interview he said that “what I am interested in is the radical paradigm where you can’t be the human receiver because even that is a semiotic production” (Crawford 1993, 264). We cannot, therefore, give preference to any model of language that assumes indi-vidual human actors because these actors are themselves part of a common field of semiotic production. Human beings do make interpretations of their situation, of other people’s conduct, etc., but that is part of the same semiotic production: signs that are reacting to other signs. In Latour’s view semiotics is not constituted in the interaction of human actors, but is much more than what actors can interpret.

The aim of this book is to discuss all five theses above and their status as presuppositions in sociology. As mentioned above, my main theoretical supports in this undertaking will be the works of Latour and of Deleuze and Guattari. It is necessary, however, to raise a fundamental question be-fore we start: why on earth would a well-established and institutionalized activity such as modern sociology want to revamp its theoretical foun-dations completely? I believe there are three reasons: there are new in-sights to be had; old dilemmas and problems might be solved; and new strategies for empirical research can be outlined. In the following I shall consider one of the seemingly unsolvable problems in sociology: the re-

Chapter One

8

lationship between structure and agency, which is commonly accepted as one of the fundamental problems in the field.

Structure and Agency – The Impossible Problem In this section I shall argue that the problem of structure and agency

can be connected to the presuppositions of a preferential treatment of hu-mans in social science. This anthropocentrism, summarized in Thesis 1, presupposes as a corollary the subordinate status of technical objects de-scribed in Thesis 4. Only those approaches to social science that seek to redress this imbalance between humans and objects can have any hope of solving the dilemma, or, rather, dissolving it as a badly posed question.

In the same way as many other philosophical problems, the question of structure and agency (or determinism and contingency) has a theological origin: if God is not only omniscient, but also omnipotent, how is human free will possible? If one holds that humans have free will, then God is more distant, but if he forms my actions, he somehow intervenes in my conduct. The problem of structure versus agency is thus something handed down to us from Western history and has much deeper roots than does the comparatively short history of the social sciences. The ascent of the na-tural sciences in the seventeenth and eighteenth centuries transformed this theological problem into the determinism of natural systems, whereby previous conditions of a system necessitate its subsequent conditions. Al-though a higher degree of contingency had to be acknowledged in human affairs, a similar determinism can be found in Marx’s formulations on the first pages of The Eighteenth Brumaire of Louis Bonaparte:

Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past. (1954b, 15)

Thus the previous actions of other humans constitute the conditions of agents here and now and, “the tradition of all the dead generations weighs like a nightmare on the brain of the living” (15). Marx then goes on to list how the French revolutions have repeated historical models, quipping famously, “the first time as tragedy, the second as farce” (15). The histori-ical circumstances in this context not only frame a set of possibilities, but tradition also passes on to us probable ways of action and we easily repeat previous patterns. Therefore, by being both the provider of the possibility of action and the originator of why we do what we do, history is given the same role as God. Nevertheless, the emphasis remains on the human

Introduction

9

element. A conversion of Marx’s remarks into contemporary social con-structivist language would amount to the statement that previous humans have constructed the social institutions that we have inherited. These insti-tutions provide both the circumstances in which we live and many of the motives for our conduct. We can, however, change these institutions by changing our habits and our social relations.

Anthropocentrism is very much at the forefront in this way of under-standing society and this view stands in contrast to descriptions of society from the eighteenth century and earlier which emphasized the interaction of humans with nature. The ascent of the human and social sciences from the early nineteenth century was also an ascent of anthropocentrism (cf. Foucault 1966, chs.9 and 10). The problem of structure and agency was transformed in the late nineteenth century and early twentieth century at the hands of Durkheim and Weber. In Durkheim’s work this transforma-tion took place with the notion of the social as a moral order that con-strains the individual, although the social is maintained and carried by the individuals themselves (1982, 51). This is, however, not enough: this col-lective moral order is a product of social life itself. In other words, the social as transcendent is formed by the immanent processes among agents themselves (2001). Weber, on the other hand, emphasized that sociology should be based on understanding, and, unlike Durkheim, he did not take the collective level as his starting point. For Weber, collective entities, like states and organizations, were a kind of shorthand for the numerous actions governed by an identifiable “average” meaning (1978). These two forefathers of sociology indicated two different notions of the social order: either it comes from a collective level that constrains individuals or it comes from the patterns of probable, but meaningful, actions. The epi-stemological choice of starting our social analyses either from a collective level or from the actions of individuals has been with us ever since, and our disciplines are shot through with it.

Contemporary sociologists have in different ways tried to bridge the divide between agency and structure. Pierre Bourdieu has attempted to solve this problem with the concept of habitus, which he defined as: “…systems of durable, transposable dispositions, structured structures predisposed to function as structuring structures” (1977, 72, emphasis in orig.). Habitus makes the body the meeting point between objective condi-tions (structures) and subjective experience (agency). Objective structures are transformed into embodied experiences and become a basis for the production of new practices. Anthony Giddens has claimed to have found a solution in what he called structuration, in which individuals reflect over what they do, but their actions have unintended consequences. The main

Chapter One

10

problem in sociology, he holds, lies in the antinomies between a sociology of understanding and a sociology of explanation, with the former identi-fied easily as the Weberian tradition and the latter as the Durkheimian. In Giddens’ view the sociology of understanding has been marked by an “im-perialism of the subject”, while the sociology of explanation by an “im-perialism of the social object” (1993, 2). In the process of structuration, agents reflect over their actions and make their own choices, but simul-taneously these actions form structural patterns on a social level. In this way Giddens reproduced a dual world view, but without giving priority to an abstract level of structure over the more primary one of action. Among the possible outcomes of processes of structuration might be the unknow-ing reproduction of structural relations. A paradigmatic example for Giddens was Paul Willis’ Learning to Labour (1980), in which the renow-ned Birmingham sociologist followed a group of working-class boys from their last year in the classroom to their time in the workplace. These boys brought with them from their background a form of banter and aggressive humour which they directed against the teachers in the face of school discipline. This opposition was at the same time a sort of acceptance that their school results would be poor. With dismal grades, they were funnel-led into subordinated and boring work – not unlike their school experi-ence. Nonetheless, they found this transition easy because they found the same banter and aggressive humour among the workmates. An unintended consequence of the whole process was a reproduction of subordination and class divisions (Giddens 1993, ch.6).

Both Bourdieu and Giddens have worked hard in their attempts to overcome this central problem, but neither solution will do because they both reproduce, one way or the other, the distinction between transcend-ence (determinism) and immanence (free will, human responsibility). In the example above the reproduction of class relations is of a different order than the behaviours of the schoolboys. The challenge of bridging or tran-scending the question of structure and agency seems insurmountable. And so it is. It cannot be solved with the presuppositions of the sociological tradition; in Latour’s view, this is so because the play between a tran-scendent and an immanent viewpoint in the discourse is fundamental to modernity. Society is seen both as stronger than human beings (tran-scendence) and as formed by them (immanence), but instead of formu-lating a dilemma, he views it as a self-contradiction (Latour 1993b).The presuppositions summarized in Theses 1 and 4 suppose that humans are at the centre of this ontology and technical objects are only part of the peri-phery. What is lacking is an explanation of why we have a certain level of continuity in human societies. This continuity is something that is pro-

Introduction

11

duced every day and cannot be taken for granted. On the other hand, hu-man interactions are ephemeral and they are limited to a certain time and space. Furthermore, humans change their minds and might do something different from what you would like them to do. Latour’s fundamental in-sight is that only our intimate relationship to objects is able provide the foundation for this continuity. His solution is that we must grant agency to non-humans and these (including technical objects) do not change their mind at the blink of an eye. They do what they are prescribed to do: “a profound temporal shift takes place when nonhumans are appealed to: time is folded” (Latour 1988a, 301). This folding of time into objects has the consequence that the work once laid into the house in which I live (and write) remains actualized over 90 years later. This remains true despite re-curring maintenance and periodical refurbishments. This is, of course, a trivial example, one which we would not usually consider worthy of any thought, but the de-trivialization of the trivial is one of Latour’s great strengths. In this way technical objects may be seen as “congealed labor” (1999b, 189), and their tasks would otherwise have been done by humans themselves (or not at all) (1988a, 300-301). The relation between artefacts and humans holds true if people still subscribe5 to them (or are forced to do so). Museums are good places to watch objects that are no longer sub-scribed to in the usual way, and, in that particular context, they are mainly relevant for historical reconstructions or nostalgia.

An obvious answer to this type of critique is that we have known this all along. To be sure, sociologists take for granted that technical artefacts are important, but in what ways do they do so? Usually they are conceived either as neutral tools or as a separate force (see Thesis 4). Marx’s treat-ment of technical artefacts in the first volume of Capital is a telling ex-ample. This book describes objects either as manufactured products or as faithful tools, or – in a reversion of their faithfulness – they replace human labour in the form of machinery and even turn the worker into a living ap-pendage of it (1954a, chs.1 and 13). Yet again, we find the juggling be-tween an immanent and a transcendent viewpoint; who is master and who is slave in the relation between humans and non-humans might change. This switching between viewpoints is the master template of sociological rhetoric: we build our society ourselves, but some social entities are stronger than us and dominate our lives (Latour 1993b). The only way this domination is possible, however, is through technical objects, and one of the key ways of renewing sociological theory is via a new understanding of technical artefacts.

Still, the theoretical powers of sociology are yet to be exhausted, for, as I have mentioned in Thesis 2, collective entities are sometimes ascribed

Chapter One

12

potentialities for action and the agency of technical artefacts is usually hid-den or taken for granted within entities like “groups”, “interests”, “states”, “institutions”, “forces”, and so on. In other words, the agency of objects is hidden by being sublimated into abstractions. Similarly, the agency of hu-man collectives is hidden by abstraction in formulations that assume that techniques have their own momentum, which is the case when we speak of “the development”, “productive forces”, “technology”, “the system”, and so forth. Objects and humans are still seen as being apart. A major strength in Latour’s analyses is his ability to describe humans and non-humans symmetrically. One of his examples in the book Pandora’s Hope concerns how murder by gunshot is explained. The slogan, “Guns kill people”, pro-moted by those campaigning against gun ownership in North America, in-volves a technical determinism as an explanation of murder. This slogan emphasizes matter over humans in the way that the mere spread of guns leads to more murders. The actions of the humans are only assumed, and humans are made into intermediaries. To this slogan the American National Rifle Association has answered with, “Guns don’t kill people. People kill people.” The second slogan exculpates the gun and makes the human the only responsible actor. In this social explanation the object turns into a helper; in other words, the gun is only an intermediary. Latour’s twist to this dilemma is to claim that a citizen holding a gun is mediating action: he or she makes a detour via the gun to kill (or to threaten, maim, etc.) while the gun “needs” to go via a human to unleash its built-in programme (1999b, 176-180). Both the human and the gun have their own programmes of action and they have to become part of each other’s joint programmes. In other words, both gun and citizen ex-change properties in the gun-wielding situation. The person is different with the gun than without it, and the gun is different in a human hand than in its case. It is the gun-human or human-gun assemblage that kills (178-180). It is possible to kill by other means, but the gun adds some qualities that other weapons might not have. It extends human action in certain ways, but at the same time it constrains it and may displace it in unfore-seen directions. A new uncertainty is involved. Similarly, seen from the gun’s point of view, its action is extended by the human, but the human might use it in ways that are prohibited by the built-in programmes and may even harm it. The criss-crossing interaction between humans and non-humans forms hybrids that make our world possible. These quasi-subjects or quasi-objects cannot be classified as clearly belonging to either society or nature unless there has been a prolonged work of purification through which we tend to overlook some qualities and to emphasize others (Latour

Introduction

13

1993b). In the Latourian perspective the age-old question of structure and agency is transformed and dissolved.

Signs and Language We have seen above that we cannot give a preferential status to re-

lations between humans because it excludes objects. What about signs and language? Thesis 5 states that signs are commonly understood as com-munication and an objection that might be raised is if this claim is still relevant with the emergence of discourse analysis. This is a multidisci-plinary field which has been thriving since the 1970s and some parts of it has been influenced by French post-structuralism. The relevant question to ask is whether the practitioners in this field really break with the com-municational model. Or, phrased differently: are they able to make signs and language relatively independent of interpreting actors? A look at dif-ferent handbooks in the field reveals that the phrase “discourse analysis” covers an enormous smorgasbord ranging from rhetoric, conversation analysis, narratives and storytelling, analysis of tropes, socio-linguistics, social semiotics, etc. (Grant et al. 2004, Dijk 1997). I do not think I can answer this question definitely and I will leave it open. Like the linguist Émile Benveniste I shall argue that language provides a sort of funda-mental condition for human societies (1974, 62) and that language and signs are something more than intermediaries between humans. How we speak and write about ourselves and the world in which we live is not only important for the way we understand this world, but also for the way humans experience themselves and change their conduct, for these depend on social concepts and categories. We write and speak all the time, and the effects of our words and deeds escape us and they might be forgotten the next moment or they might begin a life of their own by being repeated by others. This phenomenon is, of course, connected to the technical means by which we distribute writing and images in our societies, but language is such a fundamental faculty that we can hardly think without something psychologists call “mentalese” (Pinker 2009).

I shall try to develop an example of the sort of status that signs in the form of text and language might have in the type of social analyses pro-moted in this book. In his book Mythologies, Roland Barthes’ commented on the trial against the farmer Gaston Dominici in 1954. Dominici was ac-cused of murdering three English tourists who were camping near his pro-perty in the Alpine region of Southern France. The whole affair loomed large in this period of French history owing to the gruesome details of the murder and the ensuing speculations. These speculations were nourished

Chapter One

14

by the lack of evidence against Dominici and by the absence of a motive for his committing a triple murder. In order to build the case against him, the Public Prosecutor and the Presiding Judge had to make a psychological profile of a cunning and scheming farmer. But where did they take this psychology from? Barthes’ answer was that it was taken from bourgeois literature from the nineteenth century, and it was highly questionable whether it was in any way representative of the accused. In Barthes’ view the real triumph in this case was that of literature:

… it is in the name of a “universal” psychology that old Dominici has been condemned: descending from the charming empyrean of bourgeois novels and essentialist psychology, Literature has just condemned a man to the guillotine. (Barthes 1972, 43)

French bourgeois literature of a different epoch thus manifested itself as the way the magistrates and prosecutors understood certain categories of people.

This example shows how a certain genre of literature can liberate itself from its origins and enter into what Deleuze and Guattari call a collective assemblage of enunciation. Enunciation may be understood as the context that makes a specific utterance possible, and Barthes’ observation was to link the prosecutors’ and magistrates’ discourse to its possible nineteenth-century source. This was, probably, part of a wider phenomenon of how particular sections of the French bourgeoisie saw other sections of the po-pulation in the middle of the twentieth century. The case thus evoked some of the social significance of class relations. This particular flow of dis-course about peasant life and how “these people” behave had evidently taken on a life of its own. In Deleuze and Guattari’s parlance it had been deterritorialized and then reterritorialized in the situation of the court pro-ceedings. This essentialist psychology was, of course, not the only form of discourse in the court: there are all sorts of rules and regulations that gov-ern such proceedings, which ascribe strict roles for the participants and set the parameters both for assessing evidence and for the punishments to be meted out. Furthermore, Dominici was neither the first nor the last to be accused in a criminal court. Thus a flow of those accused and a flow of discourse meet at court and they are all conjugated at this specific site. In this process both discourse and persons are transformed. The accused can be either condemned or exonerated, and these linguistic transformations have consequences for the treatment of the body of the accused, since he or she is either liberated or punished.

This example illustrates the concept of assemblage (agencement) and its two segments: one pertaining to signs, called assemblages of enuncia-

Introduction

15

tion, and the other to bodies (in the widest possible terms), called machinic assemblages. This latter concept concern how humans and non-humans re-late to each other, engage with each other or part from each other. The rea-son it is called machinic is the parallel Deleuze and Guattari made between technical machines and processes in society. These processes operate with regular input and output and they have components functioning in con-junction with each other (Deleuze and Guattari 1987, 456-458, 1995, 120-121). The fundamental unit of analysis is thus the assemblage and in our example it includes judges, lawyers, the accused, bailiffs, the public, buildings, legislation and other legal documents, the evidence, the verbal argumentation of the parties, etc. The advantage of the concept of as-semblage is that it permits the analysis in a single grasp to recognize how diverse entities function together (Deleuze and Parnet 1996, 65-66). The transformation of signs is, nevertheless, something different from the transformations of humans and objects. Moreover, the flow of discourses and the flow of people and objects might be deterritorialized or perhaps conjugated (reterritorialized) in other ways at other sites. The notions of flows and assemblage were formulated to avoid the fixation of specific attributes into “actors” – whether individual or collective – and to indicate that such actors, which are commonly assumed in the social sciences, to be mere results to be explained and not something with which to start an analysis.

In Chapter 5 I shall show how Latour has a somewhat different ap-proach to the question of signs and language. He does not distance himself from theories of agency in the same way as Deleuze and Guattari, but the agency of humans and non-humans is possible to describe within lan-guage. His so-called “Translational Model” is a way to explain how texts can refer to states in the world.

A Non-Starter The previous, perhaps hesitant, introduction of concepts both from

Latour and from Deleuze and Guattari leads us to consider how similar forms of thinking have been introduced to the English-speaking world. French social thinking has previously been applied in several ways, and one of the most debated is called “French Theory” which is the North American application of the philosophies of Foucault, Derrida, Deleuze and Baudrillard in the 1980s and 1990s. French Theory had its main insti-tutional foundation within literary studies and cultural studies. The chron-icler of the upheavals of French Theory, the intellectual historian François Cusset, was astonished when he first came to the United States and found

Chapter One

16

out what kind of impact French social thinking had had on American intel-lectual life. What was often called La pensée de ‘68 in France had in North America been transformed into something else: French Theory. In this ap-plication the theory was used to deconstruct texts and to analyse all sorts of cultural expressions in relation to the identity of groups. In Cusset’s view, the main blind spot in French Theory has been the denial of market forces:

[It was] detailing clothing styles and coded lingos as forms of rebellious expressions with little or no consideration of social positions and contexts; debating sex wars and gender norms with hardly a mention of the profi-table commodification of femininity as today’s ultimate existential pro-duct. (2008, xvi)

Yet the social and political values promoted by this line of cultural studies were the opposite of the values promoted by the philosophers they claimed to be building on. With French Theory, French social thinking was shorn of its political and historical context, and all types of semiotic expressions were made autonomous in a way that could only feed a culturally oriented and flexible capitalism (xvi-xviii). Cusset’s answer is not to renounce theory, but he stresses that theory needs to be linked to historical and social contexts in order to be able to form something that might also link up the experiences outside of academia. One of the main problems, as I see it, is that French Theory has become a major article of export from the United States and it has to a high degree formed the reception of French social thinking in the rest of the English-speaking (and English-reading) world. This has been, and probably still is, one of the main obstacles for a reasonable reception of Deleuze and Guattari’s philosophy in the social sciences.

In recent years a new line of scholarship has emerged in the Anglo-American field that claims to be “Deleuzian” or, at least, claims to have some theoretical pedigree from Deleuze. Under the headings of an “af-fective turn” (Protevi 2009, Gregg and Seigworth 2010), a “non-representational theory” (Thrift 2008) or a “political ecology” (Bennett 2010) etc., this literature has the laudable ambition of applying French philosophy to particular analyses. They do not want to limit themselves to making comments on Deleuze and Guattari’s work, and in this regard their aim is in line with the present book. The main problem I find with this form of application of their theories is the lack of precision in the treat-ment of concepts. The political scientist Jane Bennett wants to formulate what she calls a “vital materialism” moving through and across bodies. In

Introduction

17

a chapter called “The Agency of Assemblages” she gives the following de-finition of assemblage:

Assemblages are ad hoc groupings of diverse elements, of vibrant mate-rials of all sorts. Assemblages are living, throbbing confederations that are able to function despite the persistent presence of energies that confound them from within. They have uneven topographies, because some of the points at which the various affects and bodies cross paths are more heavily trafficked than others, and so power is not distributed equally across its surface. (Bennett 2010, 23-24)

A few lines further down the page she writes:

Each member and proto-member of the assemblage has a certain vital force, but there is an effectivity proper to the grouping as such: an agency of the assemblage. (24; emphasis in orig.)

With some poetic license the first sentences of the first quotation may be acceptable, but the rest is difficult to square with the definitions given by Deleuze and Guattari themselves (Deleuze and Guattari 1975, 112, 1987, 88). Another Deleuzo-Guattarian concept, which is presented in more detail in Chapter 6, is that of abstract machine which, put simply, is a macro-assemblage uniting some the functions of the particular (concrete) assemblages in a given empirical material. It is a concept that gives a synoptic grasp of very different processes (Deleuze and Guattari 1987, 141-146). The political scientist William E. Connolly has a similar project to Bennett and in the book A World of Becoming he writes the following about this concept:

… an “abstract machine” is a cluster of energized elements of multiple types that enter into loose, re-enforcing conjugations as the whole complex both consolidates and continues to morph (2011, 134)

So far, so good. Then he writes the following a few lines further down the page:

A lava flow constitutes a simple abstract machine; the flow of molten lava, the melted rocks of different types carried by it, the uneven terrain over which it flows, the differential cooling rates of each type of rock when the lava meets water and open air. Each lava flow congeals into a granite for-mation, the pattern of which is not predictable in advance. (135)

Chapter One

18

This sounds very concrete to me and, much like Bennett’s, it is equally difficult to square Connolly’s adaptation with the original approach. These quotations are not picked haphazardly and more examples could be pro-vided.

Maybe the main problem in this context lies in my expectations of how theories should be treated. I expect a focus both on the concepts and on their relations to each other and how they can be involved in specific empirical materials. In the philosopher Arnaud Bouaniche’s words, the American reception of Deleuze and Guattari’s work is too “mimetic and incantatory” (2007, 300; my trans.). The reception has yet to achieve a phase that is properly critical and liberated from all forms of aesthetic readings. This scholarly situation is pretty sad and may overshadow work of serious interest. The present book is, hopefully, a modest contribution that can help to redress this situation.

Two Modes of Inquiry The aim of this section is to introduce some initial distinctions in order to organize the argument of the book. These distinctions will serve as an ana-lytical model for the analysis of the different sociological theories discus-sed in the coming chapters. This model is summarized in Figure 1.1. Above I presented five theses summarizing the presuppositions of the so-ciological tradition, but I have so far mainly discussed Theses 1, 4 and 5 and it is time to have a closer look at the remaining two. Thesis 2 states that the social sciences grant agency to people or collectives in the form of potentialities, and Thesis 3 that the acts of people or collectives can be linked to some sort of subjective meaning. These two theses can be seen as opposites, but they may as well be present in some theories as a mixture. In this context I shall discuss them in relation to a distinction made by the French philosopher Paul Ricoeur between an energetic and a hermeneutic mode of inquiry (1978, 62-64). He made this distinction into a starting point for his reading of Freud’s work, but we can generalize it as a way of distinguishing between different types of sociological theories. In this dis-tinction there is, on the one hand, “an explanation of physical phenomena through conflicts of forces”, hence energetics. On the other hand, there is “an exegesis of apparent meaning through a latent meaning” (1978, 62), in a word, hermeneutics. Agency understood as forces working on each other may be the paradigm of the energetic model in the social sciences, but agency has often been subordinated to interpretation. Interpretation is, for example, a major point in Weber’s sociological theory when he claims that social action must be understood in relation to its subjective meaning

(1978, 4-33action. Furthown, but is“typical” cotype. The spstudy of the tain historictwo seeminthese ideal ta particular quiry is tha“looking thrprivileged v

Figure 1.1 A Stengers (see

Energeti

action. Agenand the effeAmerican GDecision, thferent modeof energeticagencies proby standard proximatelythey will doleadership cstructure. Th

). This meanhermore, this primarily on

omplex of mepirit of capitalemergence ofal period and

ngly antagonitypes make us

time and plaat it always rough” what s

viewpoint and

combination otext)

ic models are nts or collectect is a form

Government’s he political scels of decisionc processes.6 Toduce output aoperating pro

y) how local bo in particulacan produce his is so even

Introduc

ning is then trmeaning is n

ne that is “suaning (11), folism and the Pf capitalism. Hwithin a certastic attitudes

s understand wace. A hallma

involves a msomeone else an ascription

of the distinctio

concerned lestive entities ar

of resultant response to th

cientist Grahamn-making, twoThe first of thas a kind of ro

ocedures makebranches will ar cases. By action over athough the lo

ction

reated as the neither any meubjectively ador instance, wProtestant ethiHe tried to shoain geographic

to life (Webwhy agents didark of the hermoment of adoes or writeof meaning.

ons made by Ri

ss with interprre forces thataction. In hishe Cuban Mism T. Allison o of which arehese emphasizoutinized actie it possible fohandle their making rules

a distance throcal level, info

latent meaninmeaning nor thdequate” or inwhat he calledic are ideal typow the accordcal area, betwber 1992). Sd the things thrmeneutic moan observer oes. There are

icoeur and Prig

retation and mt work on oths seminal boossile Crisis, Esummarizes t

e relevant as zes how goveion. Routines or leaders to kenvironment s and instructrough a kind ormed by thei

19

ng of the he agent’s nvolves a d an ideal pes in his , in a cer-

ween these Somehow, hey did in ode of in-or author always a

gogine and

more with her forces ok on the

Essence of three dif-examples ernmental instituted

know (ap-and what tions, the

of relay ir specific

Chapter One

20

routines, might develop a rather parochial view of the world (1971, 78-96). The second example is a model of governmental decision resulting from political processes among bureaucratic players. The description of agents as “players” is particularly relevant in this model because they play as though they were on a stage, but it is in a play with an open-ended plot. Policies are the outcomes of these “games” and they are not the result of routines, but of the alliances and the forces the different players are able to muster. Not all outcomes are possible, however, and policy has to be re-layed through some channels, and this fact contributes to a certain pre-selection of players. What the players can do is also restricted by rules that the players have to respect, and they form factions that fight for a specific outcome. At the same time, players are also involved in many different games, and this might affect how a player acts in one or the other of the games (162-181). In both of these models there are agents making inter-pretations of each other’s aims and actions, but these interpretations are not of the comprehensive and grand type that Weber prescribes, for exam-ple. They do not involve a “looking through” of a whole discourse or ideo-logy from some privileged vantage point. The emphasis is on the forces of agents working on each other.

Both the hermeneutic and the energetic modes of inquiry are problem-atic in the way they are applied in sociology, but for different reasons. Let us first consider the hermeneutic mode of inquiry. Weber claimed, for in-stance, that some social actions are so simple that we can even understand them by direct observation. For instance, we understand “2 · 2 = 4” when we see it or hear it read out loud (1978, 8). This is, however, to conclude rather too rashly. This mathematical calculation is only intelligible in a society in which natural numbers, arithmetic and schools are common. Without this intimate connection and the scholastic homogenization of young human minds, the expression “2 · 2 = 4” would not be directly understandable. This expression is thus only comprehensible as an effect of education. The expression “2 · 2 = 4” can to some extent liberate itself (a deterritorialization) from this assemblage and enter into other contexts (a reterritorialization), but it is nevertheless a product of social processes and not a latent cause of them. This example shows that even the most self-evident forms of meaning are produced by social processes under spe-cific historical conditions. For actions less standardized than 2 · 2 = 4, the difficulties in establishing a “typical” meaning may mushroom in different directions. In my view, searching for latent meanings for why we do what we do is not a sound way of doing social research, but is instead a way of skipping over the difficult part of describing the mechanisms that produce these meanings.

Introduction

21

Although meaning in itself is thus not excluded from the horizon of in-terest of the social scientist, meaning is a contingent product instead of something anchoring what agents do. Attempting to find the meaning of some pattern of action amounts to letting a result explain the cause (Latour 1993b). Instead, meanings are multiple and agents themselves produce them locally, in a manner akin to the actors in Allison’s models. We might, then, distinguish between grand hermeneutics and those locally constituted by the agents’ own semiotic production. Within this perspec-tive, we might regard the establishment of a comprehensive meaning of social agency more as a social and political problem than as a solution. Any nation will, for example, shelter within its borders an enormous multiplicity of different meanings linked to the actions of individuals, but in times of crisis the governing elites might be able to infuse the whole body politic with their own overarching codifications. This comprehensive meaning is thus a product of political mobilization.

Reversible and Irreversible Processes

The second distinction I have made, following Ricoeur, is the socio-logical theories emphasizing energetic processes. In other words, there are processes in which forces act on other forces. This is a form of thinking which has strong affinities to that of physics and this discipline has from time to time functioned as a model-science for many social scientists. I am not going to trace these influences in any way, but what I want to intro-duce are some discussions from the philosophy of the natural sciences concerning the understanding of processes and time. The historian of science Isabelle Stengers has for a long time emphasized, both in her own work and in cooperation with the physicist Ilya Prigogine, the importance of the distinction between processes that are time-reversible and those which are time-irreversible (Stengers 1995). In their book Order out of Chaos, Prigogine and Stengers pose the following question about natural science:

What are the assumptions of classical science from which we believe science has freed itself today? Generally those centering around the basic conviction that at some level the world is simple and is governed by time-reversible fundamental laws. Today this appears as an excessive simplifi-cation. (1985, 7; emphasis in orig.)

Time-reversibility is the foundation for any repetitive order because every-thing will repeat itself given the same initial conditions. Time-reversibility entails that the processes under study can in theory flow backwards. The

Chapter One

22

classical example is the movement of the planets around the sun. Follow-ing the physical models applied to understand these movements they might theoretically be reversed and time would flow in the opposite direction. These models do not emphasize time as an important element and in clas-sical science “temporality was looked down upon as an illusion” (7), while contemporary science must take it into consideration. A repetitive and re-versible order is, according to Prigogine and Stengers, only possible under stabilized and artificial conditions that rarely occur in nature. The funda-mental laws of dynamics are, for example, formulated for vacuums and other specific conditions and are reversible. “The artificial”, they con-clude, “may be deterministic and reversible. The natural contains essential elements of randomness and irreversibility” (9). In contrast to physics, there is no reversibility in chemistry: When we burn a match or add alco-hol to water there is theoretically no valid way we can think of the proc-esses in reverse. These processes are thus time-irreversible and my view is that social processes share the same characteristic. Nevertheless, socio-logical theories tend to emphasize time-reversibility.

How is time-reversibility conceived in this discipline? In Stengers and Prigogine’s understanding, time-reversible models do not privilege any direc-tion of time and this is what I shall claim is present in many sociological theories. I argued above that discussions about agency and structure in socio-logy involve a play between social agency (immanence) and social structure (transcendence) and this play is an example of time-reversibility in the way these theories acknowledge that agents make their own social conditions and then suddenly these conditions become stronger than their makers. Hence, the direction of cause and effect has changed and this form of reversibility is necessary for the way sociology constructs its transcendence. I shall in Chap-ter 3 show how this way of understanding social processes is important in, for instance, the social constructivism of Peter Berger and Thomas Luckmann who have claimed that: “Society is a human product. Society is an objective reality. Man is a social product” (1984, 79). All bases seem to be covered in this claim that can be taken for a self-contradiction. Bourdieu even shows this dividedness between agency and structure in the course of a couple of sen-tences. In his article Social Space and Symbolic Power he characterized his sociological approach as “constructivist structuralism” or “structuralist con-structivism” although he emphasized that his structuralism was different from the one espoused by semiotics or the anthropology of Claude Lévi-Strauss:

By structuralism or structuralist, I mean that there exist in the social world itself, and not merely in symbolic systems, language, myth, etc., objective structures which are independent of the consciousness, and desires of agents and are capable of guiding or constraining their practices or their representations. By constructivism, I mean that there is social genesis on

Introduction

23

the one hand of the patterns of perception, thought and action which are constitutive of what I call the habitus, and on the other hand of social structures, and in particular of what I call fields and groups, especially of what are usually called social classes. (1990, 123; emphasis added)

In this long and complex quotation Bourdieu in the first sentence argues that the social conditions form and constrain agents (transcendence), while in the second sentence he makes the opposite claim (immanence). The direction of causation has changed.

A counter-argument to my claim of time-reversibility in sociological think-ing might be that there is a time-line built into this type of argumentation: in Berger and Luckmann there is first a time of construction by the agents and then, after at threshold has been reached, a time when the structure starts to constrain the agents. In my view, this argument is not tenable because Berger and Luckmann only present the period when the structure is built as a logical exercise. This is a “time” of unconstrained activity, but neither they nor Bourdieu would seriously argue that there are real periods in which agents construct their world unconstrained. I think the aim of their thinking is laud-able because they want to explain how societies change while at the same time remain relatively stable. The problem is theoretical and both approaches fall into the same trap of ascribing potentiality to results (see Thesis 2). As I have shown above Latour argues that this situation has emerged owing to the rele-gation of objects to a minor position in the ontology of sociology. There are, however, other ways that sociologists and other social scientist do the same operation of producing time-reversible models and this happens every time they construct an argument on the basis of something they hold to be a “norm” in social life. This norm can be established in the form of statistical averages or other regularities in a particular milieu. Two examples might suffice to present this point more clearly. With reference to a statistical average a socio-logist might say that “girls tend to have better grades in upper secondary school compared with boys” and an anthropologist might say on the basis of either statistics or observations that “Hindus tend to marry within their own caste”. These statements clearly rely on the work of the social scientist herself and are her own constructions. They may indicate processes in society pro-ducing these repetitive patterns over time, while the nature of the processes themselves may be less obvious. The reversal of causality happens whenever the social scientist formulates a statement in the following way: “since girls do better at school, then...” or “owing to the fact that Hindus marry within their caste, then ...” The “norm” or repetitive pattern is then taken as a structural cause and this form of argumentation is quite common in social science discourse.

The question we arrive at is how can we achieve time-irreversibility and reconcile sociology with a strict immanentist perspective. I think the answer is to show that stability and “normality” is an ongoing production and not some-

Chapter One

24

thing to be taken for granted. Even though a person might experience every day as a repetition of previous ones, he or she might only perceive them as identical owing to the fact that conditions are stable. Willis’ detailed study of how the schoolboys “returned” to their own social class showed that it was the result of a prolonged process of mutual estrangement between boys and teachers (1980). In Allison’s book routines and standards produce through their execution a certain repetitive and predictable order (1971). Therefore, re-petitiveness, and the stability it entails, is a product and not a cause. Time-reversible theories nevertheless tend to alternate between an immanent and a transcendent perspective in a similar way as seen above. The repetitive pat-terns are transformed into potentialities hidden in the “conditions” or in agents (see Thesis 2 above). This way of thinking also marks the entities of the socio-logical tradition: some of them are defined as actors (e.g., humans) while others are merely intermediaries (e.g., technical tools). In a sociological study the social field may be populated by social institutions, status hierarchies, social groups, and so on. In this way an antecedent homogeneity of the field is achieved. In this line of thinking the relevant entities are given specific quali-ties before we study them. Thus all analyses are re-analyses from a pre-given repertoire. This seems to be one of the elements that makes Latour despair over the condition of contemporary sociology:

A reader, asking in what sense our theory of the social could be reconciled with “conventional” sociology, offered as an objection the way AIDS patients mobilized as a group. Looking at traditional “social movements,” it was obvious to her that patients’ organizations corresponded to “conven-tional” definitions of the social because she had entirely forgotten how deeply innovative it was for patients to make politics out of retroviruses. (Latour 2005, 23)

In a traditional context, concepts like “group” or “social movement” are sub-stances with their particular pre-given characteristics pre-defining the concrete instances.

Neither Latour nor Deleuze and Guattari can, without qualification, be put into the camp of time-irreversibility. Latour, for example, protests against the time-irreversible view of the moderns who claim that there is always historical progress. In the modern view we shall always have a su-perior knowledge today compared with earlier periods. Social and tech-nical change is always for something better and brighter (1993b, 73). In the modernist world-view the chronological flow of time itself has become a transcendent schema. He nevertheless argues in favour of an irreversible movement of time, but this movement is connected to his concept of event. In Latour’s anthropology of science the event is identified with the experi-ment and in the experiment something absolutely new may take place. The

Introduction

25

event is something unheard of and it changes the course of time by be-coming the point of reference for what happens next. However, at the same time it changes the way can talk and write about what came before the event. The event has displaced the “coordinates” of our history and in this way the causality of the event is split into two directions (Latour 1999b, 168-173): the past is re-written in light of the present. One of Latour’s examples is the surgical examination of the mummy of Pharaoh Ramses II in Paris in 1976. The scientists found that he died from tubercu-losis. Some might protest that this claim must be an anachronism because Koch’s bacillus was discovered in 1882 while Ramses II died 3000 years ago. Others would accept the realist view that the bacillus was somehow present also at Ramses II’s time although humans didn’t know about it. Latour argues that none of these explanations are any good. Historically it is not difficult to show that there was no bacillus before 1882. However, scientific events do not only open up new possibilities for the future, but they also re-write the past: the past is retrofitted with the new knowledge. Ramses II died of tuberculosis, but he only did so in 1976. Before this year the scientists had few clues to the cause of his death (Latour 2000).

In a way similar to Latour, Deleuze and Guattari regards events as be-ing marked by a rupture. But there is a double causality of another type: on the one hand, there are the transformations of bodies taking place at some moment and, on the other, the sense produced and transmitted in the after-math (Deleuze 1969, 115-117). For them May 68 is an example of a his-torical event in the sense that it took place at a given time and place, but it also became a major point of reference for the whole historical period after it (2003). It opened up new possibilities that were almost unthinkable in the previous period. In this way the event is extended into the future even though May 68 had already become a part of the past in June. May 68 was a historical turning point, but events need not be only those of major im-portance. I shall discuss this concept further in Chapter 6.

The Organization of the Argument In Figure 1.1 I have summarized the interrelationships between the dif-

ferent modes of inquiry by combining the distinctions made by Ricoeur and by Prigogine and Stengers. These distinctions will form an analytical prism for the organization of the chapters in this book. One the one hand, many sociological theories are characterized by what we might call a “Grand Hermeneutic” mode of inquiry. In these theories society is divided into two parts, with one treated as appearances and the other as reality. On the other hand, we have theories that emphasize society as being com-

Chapter One

26

posed by some forces working on other forces, for instance, in the form of actor theories. At this pole we can distinguish between those concerned with repetitive patterns and those that emphasize time-irreversible proces-ses. The division into different chapters will follow these rather rudi-mentary distinctions. It is important to emphasize that some sociological theories involve a mix of the Grand Hermeneutic and the energetic modes of inquiry and that my classification might now and then seem arbitrary.

In the analyses of theories subscribing to a Grand Hermeneutic, we shall start with Weber and Durkheim in Chapter 2 and follow in Chapter 3 with an analysis of two well-known syntheses of these two founding fig-ures: Peter Berger and Thomas Luckmann’s The Social Construction of Reality (1984) and Pierre Bourdieu’s sociology. One common trait of Weber’s and Durkheim’s sociology, as well as their followers, is a divi-sion of the world into two separate realms: one where there is the flow of singular instances taking place in the social world and another where these instances are either understood (Weber) or explained (Durkheim). This di-vision emerges as different forms in the two theorists, however. With Durkheim it is connected to the concept of a social constraint that coerces individuals to act and think in certain ways. It is important to note that Durkheim, first of all, is interested in this social level and not in the flow of acts and thoughts in the population. A similar view of singularities can be found in Weber, but his works subordinate these to the meaning an ob-server can give to them. I have sketched above a critique of grand herme-neutics; I shall develop this critique further in this case, and apply it to Berger and Luckmann’s and Bourdieu’s works. We shall see that while the flow of singular instances is not given any status at all in Berger and Luckmann’s book, Bourdieu’s sociology conforms more closely to the pattern of perceiving these singular instances as a screen to be looked through in order to unveil the relations governing them. Durkheim’s, Weber’s and Bourdieu’s theories are shown to conform to what Deleuze and Guattari call the despotic regime of signs, in which everything refers back to a given signification (Deleuze and Guattari 1987, ch.5).

Many social scientists do not conform to the Grand Hermeneutic, but instead work in what can be characterized as an energetic mode of inquiry, where we can observe in their theories how forces work on other forces. However, the chief interest of the majority of sociologists working within this mode of inquiry is to identify repetitive patterns, and to do so, certain theoretical operations tend to take place: first, a focus on certain limited properties of their case and an inflation of the importance of these pro-perties, and then an ascription of agency to them, often in the form of hid-den potentialities. Chapter 4 examines a minor sample of the works of

Introduction

27

Manuel Castells, Robert Putnam and Randall Collins, and identifies these problems in their works, although they appear in different ways. While Castells tries to analyse networks (2000), he harks back to an old sub-stantialism; Putnam’s study of social capital in the United States (2000) re-lies on a simple set of linear causes that cannot really explain the historical changes that he tries to explain; and Collins’ sociology of philosophies (1998) depends on a repetitive and transhistorical model of how intel-lectual life works. The only major figure in the sociological tradition who avoids these pitfalls is Harold Garfinkel because he focuses on how order is produced in specific situations instead of trying to locate a repetitive order (1967). His approach emphasizes how actors cope in a changing world, and it incorporates true time-irreversibility and immanence since what happens at one moment is the foundation for the next. The chapter also demonstrates the striking congruency between Garfinkel’s sociology and Deleuze and Guattari’s pragmatic theory of language (1987, ch.4) and suggests a rudimentary synthesis.

The last chapters of the book are dedicated to the theories of Latour and of Deleuze and Guattari. These theories conform to the energetic mode of inquiry and emphasize irreversible processes. In Chapter 5 I pre-sent the theoretical approach of Bruno Latour and how it grew out of the field of science studies in the 1970s and 1980s. Central in this field is the Strong Programme of the Edinburgh School, and Latour extended this pro-gramme to include the simultaneous changes in society by scientific and technical objects. This view has led to several confrontations with Latour’s British colleagues, and my examination of Latour’s debate with David Bloor in 1999 reveals the major differences between these two conceptions in science studies. The chapter then concentrates on an analysis of Latour’s conception, the so-called “Translational Model,” and its close relation to the linguist A.J. Greimas’ analyses of narratives (cf. Greimas and Courtés 1982). At this point it will also be possible to describe some of the limitations of Latour’s own approach: he does not seem able to in-clude the historically settled conditions into his analyses that are taken for granted by the human actors and he tends to emphasize the forms of action over the material that this action affects.

Chapter 6 follows the discussions from the previous chapters, but considers how Deleuze and Guattari’s philosophy might contribute to a new form of sociological theory. This chapter will involve a presentation and discussion of some of their main concepts: plane of consistency, as-semblage, event and diagram. My main aim is to indicate some of the theoretical context that envelope these concepts in order to show how they are related to other concepts and enter specific theoretical debates. Plane

Chapter One

28

of consistency is, for example, connected to the philosophy of Spinoza and the biologist Jakob von Uexküll (2010) provided examples which were im-portant to Deleuze and Guattari’s formulations. I also show the relation-ship between the model of the sign developed by the linguist Louis Hjelmslev and the concept of assemblage. Assemblage and diagram are also discussed with reference to Deleuze’s reading of Foucault’s Disci-pline and Punish (1991). The relevance of the Deleuzo-Guattarian con-cepts for social analysis is tested with a discussion of how they are applied in the works of a few of their contemporary “inheritors” in France: pri-marily the literary scholar Yves Citton and the sociologist Maurizio Lazzarato. Furthermore, I also take the liberty to discuss Durkheim’s con-temporary, Gabriel Tarde, a precursor of sorts to Deleuze and Guattari in the sociological field.

In the final chapter (Chapter 7) I summarize the debate between the different types of research modes and the different types of social-science theories discussed in the book. In light of the last two chapters, a point-by-point comparison between the sociological tradition and the Deleuzo-Guattarian line of research suggested in this book can be formulated. The main thrust is that sociological theory must both shed itself of all attempts to formulate a grand hermeneutics of society and resist the temptation of referring to inherent potentialities as explanations. A major advantage that would be gained with a less subject-centric social science would be the overcoming of the dilemmas involved in the traditional relegation of objects to a lesser status in the way we understand societies and in the inclusion of texts and other sign vehicles. The concept of assemblage also makes possible analyses of processes on different levels that are free from any assumption of an “organistic” composition of social entities. Further-more, unlike in the Durkheimian-Weberian tradition, there is no privileged point of view from which society can be observed. Nevertheless, it is still possible to produce descriptions of how society works.

CHAPTER TWO

BREAKING THE SPELL: WEBER AND DURKHEIM

The aim of this chapter is to go back to two of the classics in order to show the origins of some of the problems that still mar sociology. Weber and Durkheim continue to be cherished by their inheritors, and their con-cepts still live on in a more or less transformed way in contemporary so-ciology and political science. Within the tradition, the works of these two classic sociologists are not really analysed and discussed; people render their homages and show their respect. Not so this time, and this might lend the chapter a certain “slash-and-burn” character. Instead of adding yet again a number of footnotes to the sacred texts of the tradition, this chapter tries to go straight to the problems. This is why I turn directly to the origi-nal texts and pay less attention to studies of the historical and social cir-cumstances of their origins, although these may be of interest for a dif-ferent problematic. I shall also, to a minor extent, point out the way their works have been received and thereby transformed by the different tradi-tions of the social sciences.

The most troublesome concept in Weber’s sociology is the alleged meaning of social action and the most problematic concept of Durkheim’s is social constraint. Both have such a central position in the discourse on society, among both professionals and the general public, that it may be asked whether we can think of social science at all without them. I think we can, but it will require considerable changes in the theoretical make-up of these disciplines.

Weber’s and Durkheim’s theories are in this chapter linked to what I called a Grand Hermeneutic (cf. Figure 1.1) in the previous chapter. This hermeneutic is marked by a sort of “play” between appearances and the latent reality of the same appearances, and it is this latent reality that is important for the social scientist. This view is connected to the third thesis I formulated in order to describe the presuppositions of sociological theo-ries and this thesis states that the actions of people or collective entities, or both, are linked to something else – for example a subjective meaning –