The Concepts of Anti-Oedipus in Relation to Historical Materialism

Marxism and Anti-Oedipus Part 2

INTRODUCTION

In comparison with Thousand Plateaus, the concepts of Anti-Oedipus can be quite vague. For example, when Deleuze and Guattari write that “the territorial machine is therefore the first form of the Socius, the machine of primitive inscription, the ‘megamachine’ that covers the social field,” should this be taken as saying that Social Machine is simply another way of saying Socius, or merely that in this specific historical scenario the two are identical? Or maybe they’re just being hyperbolic? Slip of the pen? But then what about the several instances of phrases like “Social Machine or Socius?” It does not help that these two concepts—which are central to the argument of Anti-Oedipus—are almost entirely absent from Thousand Plateaus, a book in which Deleuze and Guattari were more open to explicitly defining their concepts. The fact that Socius is more absent than Social Machine, which was in large part folded into the more general concept “abstract machine,” could be taken to indicate a certain level of identity. But! A close look at Deleuze’s seminars, Anti-Oedipus, and other texts reveals that these two concepts are distinct, as will be shown below. 

There is a further difficulty in accurately assessing the status and function of these concepts. In the presentation of these concepts, Deleuze and Guattari often resort to two types of analogies or parallels: on one hand, there is the analogy of social production to desiring production, of the Socius to the Body without Organs, of labor to libido, and so on; on the other hand, there is the analogy, often left implicit, of Deleuze and Guattari’s concepts to those of Marxism. Sometimes it even seems like ‘schizoanalysis’ is a sort of meta-discourse on Marxism. (Deleuze and Guattari were aware of this and both lamented that Anti-Oedipus was too heavily conditioned by psychoanalysis on one hand and Marxism on the other for them to be able to really develop a thoroughgoing set of new and self-sufficient concepts). Neither of these analogies is truly a one-to-one correspondence (“biunivocal,” as they call it), but part of my strategy for showing the departure of Deleuze and Guattari from the typical concepts of Marxism consists in attempting to establish a correspondence between their concepts and those of the Althusserian version of historical materialism, which, as I demonstrate below, is the version most closely studied by Deleuze, who was, moreover, an acquaintance (“friend” seems too strong) of Althusser. I will try to deal with the relation between desiring production and social production as it pertains to this in a later essay.

It has been difficult to write this essay on a third count because, as far as I can tell, very little has been published on it. Most readers of Deleuze and Guattari simply take Thousand Plateaus as the authoritative source and ignore Anti-Oedipus. The archons of Deleuze Studies, Manuel DeLanda and Brian Massumi, openly reject Marxism and mistakenly believe that they have surpassed it. There are three monographs on Anti-Oedipus, by Buchanan, Holland, and Sibertin-Blanc; only Buchanan’s attempts to tie the concepts of Anti-Oedipus to Marxism in a real albeit sometimes questionable way, although all these authors are friendly towards Marxism. The collection Ateliers sur L’Anti-Œdipe (2008) also has nothing on Marxism. The Deleuze and Guattari Dictionary is also lacking when it comes to Anti-Oedipus, to say nothing of its relation to Marxism. There are, however, a handful of essays on the topic: Culp’s “A Method to the Madness: The Revolutionary Marxist Method of Deleuze and Guattari;” Tynan’s “The Marx of Anti-Oedipus.” Also useful were Donzelot’s review essay “An Antisociology” (originally published in Esprit, nouvelle série no. 419 in 1972; English translation in Semiotext(e) vol II, no. 3), Daniel W. Smith’s various essays on Deleuze which make excellent use of Deleuze’s lectures, Zourabichvili’s The Vocabulary of Deleuze, and parts of Sibertin-Blanc’s State and Politics: Deleuze and Guattari on Marx (specifically Chapter 2). Guattari’s Psychoanalysis and Transversality also remains a crucial point of reference, though its pronouncements cannot be taken as authoritative on the concepts or arguments of Anti-Oedipus. Of all the texts cited, however, I must rate Culp’s as the best. His summaries of Anti-Oedipus in that essay, though very brief, cut through all the bullshit of Deleuze Studies and clearly demonstrate the relationship of Anti-Oedipus to Marxism in a profound way. Highly recommended.

One more thing. Insofar as the virtual/actual distinction is useful for interpreting Anti-Oedipus, all the concepts I analyze here are virtual. Let’s take flows as an example. Consider a flow of cars. When an auto plant produces cars, there is no physical flow; we can’t use the physics of fluid dynamics to understand what’s happening. Cars are obviously discrete objects and do not actually ‘flow’ unless you get them REALLY hot. (But, at that point it would not be a car anymore.) However, if we take the point of view of an economics of flows, we should consider cars and all other products as flows. In the economic structure of society, cars function not just as actual, physical objects, but also as a virtual flow which goes from one point (the point of production) to another (points of distribution, consumption, etc.). For those in need of a quick and dirty definition of virtuality, just take it to mean “as if.” It’s “as if” cars form a flow. But! For Deleuze and Guattari, virtual properties and objects are not simply fake or mere matters of “as if.” The virtual is just as real as the actual. The virtual property of forming a flow is just as real as the actual property of having mass, for example.

Machines

Instead, we have a Social Machine (Delezue sometimes says large or large-scale Social Machine) and a Socius. Social Machines should not be confused with Desiring Machines or Technical Machines. The former exist at a different socio-ontological level and the latter are fundamentally a different type of thing. Deleuze and Guattari strongly reject technological determinism and instead argue that Technical Machines are always determined in their functioning by Social Machines. However, certain Technical Machines can also function as Social Machines; thus, a clock both technically measures time and socially organizes human beings (“The Social Machine, by contrast, has men for its parts…”). Besides organizing human beings, social machines also process flows and codes. The different ways in which flows and codes are processed are explained below in the section on flows.

At its most basic, then, a Social Machine is just an abstract (yet literal, non-metaphorical) machine for organizing human beings and processing flows. This might be starting to resemble structuralism with a new coat of paint, but it isn’t. In his seminars (14.11.71; 3.7.72; AO 187), Deleuze explained the problem of structure in anthropology by comparing the stances of two anthropologists: Levi-Strauss and Edmund Leach. When we examine primitive societies, do we see a system of exchange which aims at structural equilibrium yet never reaches it, or one in which disequilibrium (≈ debt) is simply part of how society functions? It makes more sense to Deleuze to say that these ‘structures’ (kinship, exchange, etc.) are social practices which are transformed through being enacted and do not tend towards equilibrium. That’s what “functioning only by breaking down” means. (Deleuze and Guattari also distinguishe between breaking down and wearing out: Social Machines and Desiring Machines only function by breaking down, whereas Technical Machines must not break down and typically perish by wearing out).

The Social Machine must be considered more general than the Socius. The text “Bilan-programme pour les machines-désirantes” defines the social machine as the conjunction of the Socius as the determinant instance and the human beings and technical machines or tools as ‘machined’ or engineered by the Socius. (I discuss the problems of this text further below).

Flows, Cuts, Poles, Stocks, Chains, and Codes

“Every code in relation to flows implies that we prevent something of this flow from passing through, we block it, we let something pass: there will be people having a key position as interceptors, i.e. so as to prevent passage or, on the contrary, to effect passage, and when we take note that these characters are such that, according to the code, certain prestations [i.e., counter-gifts -BB] return to them, we better understand how the whole system works.” (Deleuze, Seminar 16.11.71)

Flows flow from one pole or point to another. A Stock is a flow or a part siphoned off of a flow as possessed as an object by one of the poles. The term cut (coupure, usually translated as “break”) qua coupure-prélevement designates a mode of flowing, e.g. whether the flow is continuous or perhaps rhythmically interrupted, whether it flows into one pole or several, etc. Prélever/prélèvement, a word which means “deduct,” “withdraw,” or, more figuratively, “siphon off,” is the concept which designates breaks or cuts in flows. Basically, cuts are the opening and closing of flows. The combination of flows and cuts forms the concept of cut-flow or break-flow (coupure-flux). A flow of money passes through the pole of the merchant, but part of it is retained (siphoned, cut) by this pole as a stock of money or hoard, which only under certain conditions can then become capital. “Blocks of debt” are also a type of stock, namely ones with a negative quantity.

Coding is the qualitative differentiation of flows and, relatedly, the incorporation of flows into social relations. More precisely, coding the qualitative differentiation of flows via the ‘recording’ (enregistrement) or inscription of flows-as-coded on the Socius or social body, which forms the ‘surface’ upon which codes are inscribed and flows flow. Coding is the ‘interpellation’ of flows into the social machine, or the subsumption of flows under socially effective concepts. (Readers of Zizek and Althusser will know about the Lacan-Althusser idea of the interpellation of individuals into subjects by the ISAs or the symbolic order [=signifying chain], etc.). A distinctly Kantian feeling should be detected at this point. Yes, some the crypto-Kantian residue of Freud and Lacan (viz., the absent drive or impossible real) rubbed off on Deleuze and Guattari, the former of which was himself thoroughly coated in Kantian slime. Encountering a decoded flow would be like perceiving a manifold without subsuming it under categories and forms of sensibility. But, unlike in Kant’s case, decoded flows are a fact of history and not a limit-concept. (Although to be sure there are Kantian limit-concepts in Anti-Oedipus.) 

The code ‘in itself’ is a chain or signifying chain. A cut in one of these chains of code is called a détachement. We can think about this as a piece of genetic code or computer code attached to the flow which is coded and which carries ‘instructions.’ In more Marxian terms: the social relations inhere in the things affected by said relations via the fetishism of the socius which consists in the coding of flows just as the commodity fetish consists in the attachment of social relations of exchange value to objects qua use value.

Surplus Value

Deleuze says that surplus value is concomitant with surplus labor: “...il y a plus-value dès qu'il y a surtravail.” (Seminar 2.22.1972). However, there are different forms of surplus value throughout history. There are three (okay, five) different forms of ‘surplus value’ in Anti-Oedipus. In this essay, however, I will not be discussing the niceties of machinic and human surplus value. Two of the other types of surplus value are called “surplus value of code,” but these two are very different. Before describing their unique characteristics, I will first establish the commonalities of any form of surplus value of code. Surplus value of code is different from the capitalist surplus value familiar to readers of Capital because it entails at least two qualitatively different flows/codes, whereas capitalist surplus value is a purely quantitative affair. The appropriating chain (code) intercepts another chain which (here I interpolate) passes through the same machine. Deleuze and Guattari use the example of the wasp and the orchid: each participant gives and receives something fundamentally different. Because there is no equivalence, Deleuze and Guattari prefer to analyze this form of social intercourse as something more akin to debt than exchange. (That they stringently refuse to call it exchange and clearly differentiate it from capitalist surplus value is downstream from Marx’s claim in Capital that commodity exchange requires a universal equivalent, aka money).

Territorialist surplus value of code is the system of prestige, mana, and bride-wealth. It is operative in the type of reciprocal gift economy studied in Marcel Mauss’s famous Essay on The Gift. Certain exchanges—matrimonial ones above all else—will provoke, instead of the return of some prestation of ‘equivalent value,’ the return of a qualitatively different prestation which creates a disequilibrium: excess on one side, lack or deficiency on the other. There are three qualitatively different basic flows in a territorialist society according to Deleuze and Guattari: material consumption goods, prestige or objects of prestige, and women. A man gives a gift, thereby losing material wealth, but gains prestige in return. Marriage or alliance, which is really an exchange of women between men, can result in one side gaining a right to enjoy the goods of the bride’s family. The most extreme form is that of the potlatch, in which a chief may effectively destroy all of his material wealth in order to gain prestige. These two flows (material goods and prestige) are incommensurable, and, needless to say, the flow of women is also incommensurable with the other two. Deleuze and Guattari hypothesize that at least some of the disequilibrium results from the incest prohibition-induced blockage of desire inherent in all structures of alliance (marriage). However, the essentially local character of the different manifestations of territorial surplus value of code prevent the discernment of any definite universal ‘laws.’ 

The other form of surplus value of code is called despotic surplus value of code. I was going to write about it in this essay, but then I decided not to. It will be covered later.

Socius

“What I call the socius is not society, but rather a particular social instance which plays the role of a full body. Every society presents itself as a socius or full body upon which all kinds of flows flow and are interrupted, and the social investment of desire is this basic operation of the break-flow to which we can easily give the name of schizz.” (Deleuze, Seminar 14.12.71)

“Dans une société, codée ou axiomatisée, il y a une instance sociale fondamentale qui est celle du corps sans organes ou de l’improductif ou de l’anti-production. On a vu dans les sociétés dites primitives comment ce qui joue le rôle de corps plein, de corps sans organes, d’instance d’anti-production, c’est la terre comme entité indivisible.” (Deleuze, Seminar 3.7.72)

“These are the two aspects of the full body: an enchanted surface of inscription, the fantastic law, or the apparent objective movement; but also a magical agent or fetish, the quasi cause.” (AO 154)

The quotes assembled above combine to form the closest thing we get in the entire Deleuzoguattarian corpus to a definition of Socius. Before getting into the meat of the argument, I will tentatively propose that a socius is a social machine which forms a certain part or instance of a ‘large-scale’ social machine. The word “instance” here is used in the continental, European sense of Freud’s drei Instanzen. It means something roughly similar to “agent,” “actor,” or “authority.” Because the concept is so obscure, let me propose—forgive me!—a simile: the Socius is something on the order of Marx’s concept of fetishism. This is most clear with the case of the full body of capital, but, as I will show, all three Sociuses are directly taken from Marx. 

The Socius “is not the product of labor, but rather appears as its natural or divine presupposition” (AO 10). In the same passage, we encounter a somewhat odd phrase: “forces and agents of production.” Forces, sure, but agents? The concept of an ‘agent of production’ comes from the Althusserian school. Productive forces are, paradigmatically, means of production (‘instruments of production’ and raw materials) and labor (and maybe more abstract things like scientific knowledge). Agents of production will refer, then, to human beings qua material or physical economic actors in the production process. The capitalist does not only act as the (economic) appropriator of surplus value, but also as the entrepreneur or organizer of production (sometimes also a manager of the immediate labor process itself, etc.), and thus as an agent in the production process. (Balibar calls this ‘real appropriation,’ i.e. material or physical ‘appropriation’ of the means of production or natural materials in order to set them in motion). In general, ruling classes as well as ‘tertiary’ classes such as merchants typically have some hand in organizing the production process. However! Let’s recall what Marx says about the capitalist merely being a ‘character-mask’ for capital (=Socius). Yes! The fetish (Socius) appears as the organizer even of the classes who really put the production process in motion because of their dominant position in the relations of production. It appears as a ‘divine’ instance which is a unity and, as a unity, legitimizes all the historically correlated forms of social domination and economic exploitation as ‘natural.’ Nothing is the result of history, it is all ‘presupposed’ and ‘naturalized’ (Marx in German Ideology: ideology has no history). This is related to what Althusser calls the “absolute guarantee” that everything is just as ideology says it is and everything will work out if the subjects of ideology simply do their duty and follow the rules. For Deleuze and Guattari, this involves affecting desire and the unconscious. 

The Earth appears as the natural and supernatural origin of everything human and inhuman and (virtually) ‘appropriates’ the forces and agents of production. The Despot, as the organizer of grand public works and as a divinity, appears as the ‘unmoved mover’ or motor of production as well as its rightful owner. Capital does the same thing but in a capitalist way. I’m not going to explain the fetishism of capital, money, and commodities. Readers of Marx should be familiar with it.

Besides being the fetish object which virtually appropriates production and acts as the apparent “quasi-cause” of all production, the socius also serves as an “enchanted surface of inscription” upon which codes are written or the “fantastic law” which (presumably) consists in the codes. (Remember, this is all virtual). Deleuze and Guattari do not cease to describe the socius as “divine,” and so we can say with some justification that it is both God and His law. It can further be likened to the ‘Absolute Subject’ (Big Other?) of Althusser’s second theory of ideology, if one likes.

Anti-Production

There are two kinds of anti-production: repressive and non-repressive. Anti-production is not repression, but repression is a form of anti-production. All the social machines studied in Anti-Oedipus make use of repressive anti-production, while non-repressive anti-production is examined only with regard to the (full) body without organs, i.e. the process of desiring production considered in itself. However, it is possible to have non-repressive anti-production in society. If we wanted to make a more elegant terminological distinction, we could refer to non-repressive anti-production as improduction. I may return to the topic in the future, but just know that the important things to read to understand the concept of non-repressive anti-production are in Part 1 of Anti-Oedipus and in Deleuze’s treatment of the Freudian death drive in Difference and Repetition. 

I cannot define anti-production more clearly than Andrew Culp has, so I will simply reproduce his definition:

Anti-production here is not the antithesis to production, as in the destruction of circuits of production. Rather, anti-production is the sum of the restrictions put on productive activity to guarantee that production occurs in a certain way. As such, anti-production sets the conditions under which any given mode of production operates. (Culp 2019, p. 169)

I would only add that anti-production as analyzed in Anti-Oedipus particularly pertains to the repression or control of desire or desiring-production.

The Marxist Origin of the Three Sociuses

It seems to me very likely that Deleuze (and it was probably Deleuze here, not Guattari, who detested the ‘Stalinist’ Althusser) got his inspiration for the three Sociuses from Part Five of Reading Capital, namely “On The Basic Concepts of Historical Materialism” by Balibar. In this text, Balibar cites two important passages from the Grundrisse section commonly called “Forms Which Precede Capitalist Production.” In fact, the passage about “natural or divine presupposition,” an uncited borrowing from Marx, is quoted in Balibar’s text. In the full version it refers to a fetishism of the Earth and the community (Gemeinwesen):

The earth is the great laboratory, the arsenal which provides both the means and the materials of labour, and also the location, the basis of the community. Men’s relation to it is naive: they regard themselves as its communal proprietors, and as those of the community which produces and reproduces itself by living labour. Only in so far as the individual is a member—in the literal and figurative sense—of such a community, does he regard himself as an owner or possessor. In reality appropriation by means of the process of labour takes place under these preconditions, which are not the product of labour but appear as its natural or divine preconditions. (Pre-Capitalist Economic Formations, tr. Cohen, p. 69)

The description of the Socius as an immense ‘unity’ is also taken from Marx. The movement from religious or political fetishism (pre-capitalist modes of production) to economic fetishism (capitalism) is also pointed out by Balibar and then later reiterated by Deleuze in a lecture.

Later, in the same section on pre-capitalist forms, Marx also describes a despotic form of fetishism:

For instance, as is the case in most Asiatic fundamental forms it is quite compatible with the fact that the all-embracing unity which stands above all these small common bodies I may appear as the higher or sole proprietor, the real communities only as hereditary possessors. Since the I unity is the real owner, and the real precondition of \ common ownership, it is perfectly possible for it to \ appear as something separate and superior to the j numerous real, particular communities. The individual is then in fact propertyless, or property—i.e. the relationship of the individual to the natural conditions! of labour and reproduction, the inorganic nature which he finds and makes his own, the objective body of his subjectivity—appears to be mediated by means of a grant (Ablassen) from the total unity to the individual through the intermediary of the particular community. The despot here appears as the father of all the numerous lesser communities, thus realising the common unity of all. It therefore follows that the surplus product (which, incidentally, is legally determined in terms of the real appropriation through labour) belong to this highest unity. (ibid, pp. 69-70)

The important thing here is that Marx shows that a specific type of superstructural (political, religious, ideological, etc.) instance or unity or etc. (Marx even goes so far as to say that it “appears as a person” at one point) is required to guarantee the reproduction of pre-capitalist modes of production (maybe even all modes of production, depends on how you interpret the stuff about capital being a subject in the “Results of the Immediate Process of Production” and Grundrisse manuscripts in connection to fetishism). Productive cooperation is not based on the ‘transparent’ relation of human beings, but is instead based either on the fetish of Earth (Territorialism) or that of the Despot and his God as dominator of the local communities (Despotism).

Relating The Concepts of Anti-Oedipus to Historical Materialism chez Althusser

Deleuze and Guattari are broadly in agreement with the thesis of Althusser et al that “on the basis of the existing productive forces and within the limits they set, the relations of production play the determinant role.” (Althusser, On The Reproduction of Capitalism 2015, p. 20). (See also above on Technical and Social Machines). In Anti-Oedipus, however, the role of the relations of production is taken over by the Social Machines, the codes, and so on. The role of the forces of production is played in part by the Technical Machines, in part by the desiring machines (in an obscure way), and in part by the general economy of Flows (“...flows of production, flows of means of production, of producers and consumers…” [AO 142]).

Althusser and his followers distinguish between concrete ‘social formations’ and theoretical ‘modes of production.’ (Deleuze and Guattari use the term “social formation” in an unsystematic way in Anti-Oedipus, but they give an ostensive definition in Thousand Plateaux.) In every historical social formation, at least two modes of production operate, and one of these modes of production is ‘determined as dominant.’ The ‘economy’ or infrastructure (forces + relations) is, in turn, determined as dominant only insofar as it determines which ‘social instance’ ultimately ‘occupies the dominant place.’ Certainly a number of small Social Machines can exist simultaneously. Deleuze and Guattari also describe the historical transition from Despotism to Capitalism as gradual, entailing the co-existence of the Despotic and Capitalist Machines for at least a part of history. Also, even though the transition from Territorialism to Despotism is, in a strange way, instantaneous, Despotism is itself defined by the existence of two systems of code and hence two Machines, i.e. the Territorial Machine and its local codes as subordinated to the Despotic Machine via the operation of Overcoding. 

It is not clear, however, whether the Socius or the Social Machine is determinant, or indeed whether Deleuze and Guattari even accept the idea that one ‘social instance’ can be considered the determinant one. The Socius, as the ‘mode of fetishism,’ is the instance which (partially) guarantees the reproduction of the Social Machine, as the ‘relations of production.’ That makes sense, but it doesn’t explain which one is determinant. In attempting to answer this question strictly on the basis of the text, we again run up against the equivocation of the Socius and the Social Machine. Let’s move forward, however, with the acceptance that they are separate machines/instances. How, then, should we find which instance is determinant-in-the-last-instance?

If we follow the Althusserian line of reasoning, there should be something about the infrastructure (Social Machines + Flows + technical machines or something) which determines (in the last instance) which instance is actually determinant and also whether said instance is economic or superstructural. But the problem here is that, while the Social Machines do indeed take on the role of the relations of production, they are by no means categorically economic. The claim that “desire is part of the infrastructure” is in fact a red herring which distracts from the replacement of the ‘dualist’ concepts of historical materialism (cleaved along infrastructure/superstructure) by the ‘monist’ concept of Social Machine. In some regards, we can see Deleuze and Guattari as taking one step forwards and two steps back here. The Social Machines are not precisely decomposed into definite constituent ‘relations’ and ‘modes’ (= small-scale social machines or something) as the Althusserians attempted to do with the modes of production. Why does the territorial machine have the Earth as its socius? We get some associations, but not much else. They can give a more convincing account of the origins of the capitalist machine only because they are borrowing directly from Marxist histories.

Another problem arises, this time regarding the role of the Socius. Yes, the Socius creates an illusion of unity which nonetheless has a certain ‘truth’ to it. Fetishism. Etc. But in some cases, notably in “Bilan-programme pour les machines-désirantes” (1973), they (Guattari, really—it’s clearly him writing) seem to assert that the fetish/Socius/full body really and always is the prime mover or determinant instance (“instance machinisante,” i.e. the social agency/instance which ‘engineers’ or assembles human beings into Social Machines). The place of ‘determiner in the last instance’ is either vacant or occupied by the socius depending interpretation. The socius determines the social machine, but nothing determines the socius. The indications of the “Bilan-programme” indeed seem to me a regression even from Anti-Oedipus. It is not as if Deleuze and Guattari generally traffick in aporia, as e.g. Adorno sometimes does. Rather, these problems must be taken as indications that when Deleuze and Guattari admit that Anti-Oedipus was a failure we must take them seriously. 

(Perhaps this is a result of the “Bilan-programme” being an attempt to redefine and reassert the concepts of Anti-Oedipus while also shaking off the remaining Marxist and psychoanalytic baggage; perhaps it's a result of Deleuze letting Guattari go buck wild and not examining what exactly his partner had just written).

The changes to their conceptual system in Thousand Plateaus in some ways more resemble a return to Althusserian historical materialism than a rejection of it: a plurality of machines or modes of production, determination of the determinant instance itself being determined by the complex composition of the social formation instead of a single instance or structural ‘place,’ different levels or instances (strata) possessing different roles depending on the specific social formation, etc.