I would like to pursue the problem of the economy of flows; last time, someone wanted a more precise definition of flows, more precise, that is, than something which flows upon the 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. It is not yet important for us to have a real definition of flows, but it is important, as a starting point, to have a nominal definition and this nominal definition must provide us with an initial system of concepts. As a point of departure for our search for a nominal definition of flows, I'll take a recent study by a specialist in the flows of political economy: "Flows and stocks," by Daniel ENTIER. Stocks and flows are two primary notions in modern political economy, remarked upon by Keynes, such that we find in Keynesian economy the first great theory of flows in his "General theory of employment, interest and money." Entier informs us that, "from the economic point of view, we can call flows the values of the quantities of goods and services or money that are transmitted from one pole to another"; the first concept to be placed in relation with that of flows is that of pole: a flow, inasmuch as it flows on the socius, enters by one pole and exits by another. At our last session, we had tried to show that flows implicated codes, in the sense that a flow could be called economic insofar as something passed, and where something else blocked it and made it pass; the example given was that of the rules of alliance in so-called primitive societies, where taboos represent a blockage of the flow of possible marriages; the first permitted marriages, i.e. the first permitted incests, called preferential unions, which are, in fact, hardly ever realized, represent something like the first modes of passage: something passes, something is blocked (this blockage taking the form of incest taboos), something passes, the preferential unions, something blocks it and makes it pass, for example the maternal [utérine] uncle. There is, in any case, the determination of an incoming and of an outgoing flow; the notion of a pole implicates or is implicated in the movement of flows, and this takes us back to the idea that something flows, that something is blocked, that something makes it flow, that something blocks the flow. Entier continues: "In the knowledge that the term pole will be given to an individual or a firm, or alternately a group of individuals or group of firms, or even a fraction of a firm..." "We are thereby defining the interceptors of flows... when the operations carried out by these -- the interceptions of flows -- can be described by a coherent accounting system..." Thus, a correlative of the notion of flow is that of an accounting system; when the operation carried out - such as the passage of a flow from one pole to another - may be described by a coherent system, this evidently being expressed in terms of capitalism, by which I mean that in this context we are in the framework of capitalism at the level of abstract quantities, as the final residue of something that has an entirely different scope in pre-capitalist societies, to wit, what in pre-capitalist societies present themselves as veritable codes. It's when a society is completely decoded that flows are prone to incorporation into an accounting system, that is, into an axiomatic of abstract quantities instead of referring to qualified codes; the accounting system under capitalism is the residue of quantities, abstracted from the coding of flows, and capitalism functions on the basis of decoded flows, from which point on these flows are taken up by a system based on accounting. Entier continues: "...we can consider all goods arriving at the same level of material or legal transformation at the precise moment at which they arrive as constituting one and the same flow..." Here you have a third correlative notion: material or legal transformation, "and if one is speaking of flows that are exchanged between industrial sectors, one must specify the notion of sector, if one is interested in precisely determining the flow of production, the flow of revenues, the flow of consumption, then one has to determine these terms carefully; take for example the flow of monetary revenue reached by calculating the totals of all liquid assets..." The question is, what is meant by 'all liquid assets' [tous les biens en monnaie]. This is what economists refer to as nominal salaries - a concept that covers real wages, as well as management salaries and dividends (interest on assets). Take for example the flow of cash revenues [flux de revenus monétaires]: this is determined by the total of all liquid assets at the disposal of all of the individuals making up the collectivity, where the revenue of a large number of individuals can be precisely evaluated since it is paid out by other individuals, such as state contractors, and since it has been precisely calculated; but for many kinds of revenue the importance of which must not be undervalued, one can't give an exact definition; well, well, so there's a sphere of indetermination in the sector? This is doubtless related to something very profound in what is, as we shall see, the accounting sector; but, for all that, we now have a triple reference : the flows refer for one thing to poles and for another to codes or accounting systems, then - in fourfold fashion - to rates of transformation, next to sectors and finally to stocks. This amounts to five correlative notions. >From an economic viewpoint we will refer to stocks of goods and monetary stocks, as the goods held and the money held by only one pole; the flow, then, is what flows from one pole to another, what goes in and what goes out, and the stock is what is brought back as the material and legal possession of one of the poles considered; this clearly shows the correlative character of the two notions; we can thus define stocks in the following way: the utility of stocks changes from case to case, but is linked in one way or another, at one time or another, to the existence of flows; we will, nevertheless, receive the distinct impression that stocks and flows are effectively the same thing as they relate to two different units: first, the passage from one pole to the other, and secondly, attribution to one of the two poles, as two units of measure for one and the same thing. The utility of stocks varies with every given case, but is linked in one way or another, and at one time or another, to the existence of flows; and yet, whereas flows allow one to extract the movements of values between different poles, stocks represent a sum of values that are at the disposal of a given pole; there are no goods related to a stock which are not, at a specified time, also correlated to a flow; this is, in fact, one of the fundamental principles of accounting, because the influx and output of a stock constitute flows; only the study of flows allows one to realize the role of the incoming and outgoing movements involved in stock variations...

So, we have just seen the correlation between the notion of flux and five other notions: pole, code or accounting system, stage of transformation, sector and stock. If we try to simplify all of this, I think that the notion which I was attempting to launch last time may be able to concentrate or group together all of these five references, this notion being that of the break-flow.

For this notion of the break-flow has to be understood simultaneously in two ways: it is to be understood as the very correlation of flux and code, and if, returning once more to capitalism, we are aware that flows are "accounted for", it is in favour of a movement of decoding such that the accounting system has simply taken the place of codes; it is at this point that we come to realize that it's no longer sufficient to speak of an accounting system, but rather of a financing system or structure.

The strict correlation of flux and code implies, apparently, that in a society -and this is clearly our point of departure - it's impossible to seize flows other than by and through the operation which codes them; the fact is that a non-coded flow is, strictly speaking, an unnameable or merely a thing. This is what I was getting at last time, this is the terror of society - it is the flood, the deluge which is the flow that breaks through the barriers of codes. Societies aren't exactly in mortal dread because everything is coded - the family is coded, death is coded; but what makes societies panic is when something or other breaks down, something that forces the codes to crack. A flow is thus not recognizable as an economic and social flow except by and through the code which encodes it, but this operation of coding implies two simultaneous interruptions, and it is this simultaneity which allows us to define the notion of a break-flow: simultaneously, in an operation of coding the flows, a subtraction [prélèvement] from the flows is produced, due to the code, and this subtraction from the flows defines its poles: it defines a certain entry point and a certain exit, and it is between the two that the break-subtraction takes place; this occurs at the same time as the code itself relates to a break of another sort which is strictly simultaneous, meaning once more that... there is no subtraction from a flow which is not accompanied by a detachment of or from the code which encodes this flow, although it is the simultaneity of the subtraction of the flow and the detachment of a segment of code that permits one to define the flow with reference to poles, sectors, rates and stocks. This notion of a break-flow presents itself in two-fold fashion, for it implies at once a break-subtraction of a flow and a break-detachment of a code. Here you've got the mechanism of delirium: it's this double schizz operation -- it's the schizz which consists both in an operation of subtractions of flow as a function of detachments of code, and vice versa. If, to begin, I take an indeterminate flow as a purely nominal starting point, then what flows on the socius, cannot appear socially speaking as a flow, except in correlation to a code, or at least to an accounting system, and the flow itself is qualified as a function of the code, and the correlation of the two is where you have a break-subtraction on this flow itself (as qualified by the code), at the same time as and in reaction to which you have a break-detachment of the code. A detachment of code is correlative to a subtraction of flow. This is simply a formal description. At first glance, a madman is someone who passes on the unnameable, who carries decoded flows: "a god speaks to me, but it's not your god". The Greeks had a notion of the demon, they had gods and the gods were allotted, everything was very properly allocated, they had powers and spaces; in a way, they couldn't move, they were sedentary, they had their territory and the demons carried out the coding. Religions are not to be understood from an ideological viewpoint, but at the level of their pertaining to a social code. Demons were above all powers which did not respect the codes. In Oedipus there is a text which is badly translated and which reads, "which demon has leapt the longest leap," a text which frankly leaps beyond the limits, it had to do with unnameable powers, with excess, and it is not forcing things to translate this as "decoding". Thus, a demon speaks in a certain way so that a madman receives decoded flows, and in turn emits decoded flows, such that it flows and escapes on all sides, messing up all the codes. Which is why in Oedipus' case, it just won't stick, because literally, Oedipus is a screwed-up code.

When something goes bad, you always have to go back up to a higher level to see where things start to go bad (cf. the USSR), and psychoanalysis goes bad, but why and how so? Derrida has seen very clearly in what way psychoanalysis, at least at the level of its first intentions, is opposed to the code; it is in fact a system of decoding, and this is why this affair just couldn't help going bad. Because decoding means either to read a code, to penetrate the secret of a code, or else it means to decode in an absolute sense, i.e. to destroy the codes in order to make the flows flow freely; a major part of the project of psychoanalysis was intended as an absolute decoding of the flows of desire and not as a relative decoding, to make the walls of the code cave in, and to make the flows of desire run wild. It is in this sense that psychoanalysis was very close to a desiring economy, and properly speaking, to desiring-machines, the producers of the flows of desire; and this is seen very clearly in Freud's writings, such as The Interpretation of Dreams, where he says: what distinguishes my method from the method of the key of dreams? The major difference is that the key of dreams proposes a code of desire; Freud says that they have seen everything, but that they propose a systematic coding : this means that, that is the key of dreams; and from the perspective of a key of dreams, if one decodes the dream, one decodes it in a relative sense, i.e. one discovers the cipher of its code. Now Freud says that psychoanalysis has nothing to do with all of that, that it does not interpret. And Derrida, in his article on Freud in Writing and Difference shows this very well. It carries out an absolute decoding, rendering the codes as flows in the raw state, and thereby psychoanalysis is opposed to codes. It goes without saying that, at the same time, and from the outset, they are inventing a new code, that is, the oedipal code which is even more of a code than any other code; thus, the flows of desire pass into the oedipal code, or else, whatever the flow of desire, it is stuffed into the oedipal grid. At this stage, psychoanalysis proves less and less capable of understanding madness, for the madman is really the being of decoded flows.

And who has shown this in a most vital and convincing way if not Beckett, whose strange creatures spend their time decoding things, they make non-codable flows pass. Social processes can't capture flows except in relation to codes which operate on them, and which are simultaneously a detachment of a flow and a subtraction of chains or codes, and the madman makes flows pass on it, flows from which it is no longer possible to detach anything; there are no more codes, there is a chain of decoded flows, but one can't cut into it. There is a sort of deluge or failure of the body, maybe that's it, after all, the body-without-organs, when on the body, or from the body flows enter and exit by way of poles, flows on which one can no longer carry out any subtraction because there are no longer any codes from which to detach anything.

The state of the body of someone who has undergone a fairly severe operation, the eyes of this patient are the eyes of someone who has not been very far from death, who has not been very far from madness, the eyes are elsewhere in a certain sense - he has gone through the wall. It is interesting that what we call convalescence is a kind of return. He's had a brush with death, it is an experience of the body, very strange, psychoanalysis: why does Freud cling so strongly to the notion of a death instinct? He tells his secret in "Inhibition, Symptom and Anxiety." It seems that if there is a death instinct, it's because there is neither a model nor an experience of death. When pressed, he admits that there is a model of birth but not of death, thus, all the more reason to make of it a transcendent instinct. Bizarre. Perhaps the model of death could be something like the body-without-organs.

Horror-story writers have understood, after Edgar Allen Poe, that death wasn't the model for schizophrenic catatonia, but that the contrary was true, and that the catatonic was the one who made of his body a body-without-organs, a decoded body, and that on such a body there is a kind of nullification of the organs. On this decoded body, flows can flow under conditions where they can no longer be decoded. This is why we fear decoded flows - the deluge; because once flows have been decoded, you can no longer subtract anything or break into them, no more than you can detach segments from any codes in order to dominate, orient or direct the flows. And the experience of one who has been operated on, of her body-without-organs, is that, on this body, there are literally noncodable flows which constitute a thing, an unnameable thing. At the very moment that she breathes, there is great confusion of the flows that form one great indivisible flow, no longer susceptible to subtractions, one can no longer interrupt it. One long stream that cannot be tamed, where all of the flows that are usually distinguished by their codes are united in one and the same indivisible flow all flowing on one and the same non-differentiated body, the body without organs. And as for the mad patient who has undergone an operation, every breath of air he takes is at the same time a breath of spittle, a flow of air and spit that tend to get mixed up together, so that there are no longer any distinctions. Moreover, each time that he breathes and spits, he feels a vague desire to defecate, a vague erection: it's the body-without-organs that escapes on every side. It is sad, but then again, it has moments of great joy, mixing up all the codes, it has its great moments, and this is what makes Beckett a comic writer.

Though here again, one can't help saying, and then, and then, though this is what constitutes the madman and his place in society, as the place of one through whom all of the decoded flows pass, which is why he is perceived as the fundamental danger. The madman doesn't decode in the sense that he would harbour a secret the meaning of which ordinary people have forgotten, but rather he decodes in the sense that he sits in his little corner and makes little machines which make the flows pass and which make social codes break down. The schizophrenic process as such, of which a schizo is only the schizophrenic continuation, well that schizophrenic process is a revolutionary potential in itself, in opposition to paranoid investments which are fundamentally of a fascist type.

This leads to a first result, namely, that the economic operation of coding flows involves a double break-flow, a break-detachment and a break-subtraction; and on the socius of a society one encounters these strange creatures, the mad, who make all of these decoded flows carry on. The strangest phenomenon of world history is the formation of capitalism because, in a certain sense, capitalism is madness in its pure state, and in another sense, it is likewise the opposite of madness. Capitalism is the only social formation which presupposes, as its condition of emergence, the breakdown of all preceding codes. In this sense, the flows of capitalism are decoded flows, and this poses the following problem: how could a society, with of all of its repressive formations, create itself on the basis of what constituted the terror of all other social formations: namely, the decoding of flows.

The intimate bond between capitalism and schizophrenia consists in their common basis and installation on decoded flows (insofar as they are decoded). How then was this decoding carried out? One has to keep the following two requirements clearly in mind: the basic affinity between schizophrenia and capitalism, but, at the same time, find in this basic affinity the reason for which the repression of madness under capitalism is conducted with much greater harshness and specificity than under pre-capitalist formations. In the one case there is a political economy and a libidinal economy, and, in the other, an economy of decoded flows. I'd like to show that, historically, this shift took place over a long period of time: there are synchronic social machines and there are diachronic social machines; despotic Asiatic social machines are of a truly synchronic nature - Marx's Asiatic state emerges in one blow, and all of the cogs and wheels of its state apparatus appear synchronically. The formation of the capitalist machine extends over many centuries. It's a diachronic machine and has taken two long periods to come about; it isn't capitalism that decodes the flows, rather, they are decoded on what we call the ruin and decline of great empires, and feudalism is only one of the forms of this ruin and decline. Capitalism does not proceed from the decoding of flows because it presupposes them, it presupposes flows that have lost their codes.

Marx is the author who has demonstrated the radical contingency of the formation of capital. Any history of philosophy is either theological, or else it is the history of contingencies and unforeseen encounters. The originary phenomenon of capitalism is this: decoded flows qua decoded flows must enter into conjunction. What then assures that this conjunction will take place? Here, one feels that history can teach us as much about the process of the decoding of flows, as about what ensures the conjunction of the decoded flows as such, and this can be nothing else than the processes of a specific historic sector. This tale of capitalism implies a generalised decoding of flows and at the same time something else, as if what ought to be put in place were an apparatus for conjugating decoded flows; this is what gives capitalism its purely illusory appearance of liberalism. It has in fact never been liberal, it has always been state capitalism. This tale begins in Portugal in the 12th century, this tale of state capitalism. There has never been a time when flows were decoded and when everything was free, bringing recuperation, which is an awful thing. If it's true that capitalism replaces the old ruined codes with conjugating machines, and axiomatic machines which are infinitely more cruel, crueller than the cruellest despot, although of another cruelty, it is at the same time as the decoding takes place [que ça se décode] that it is taken up by another machine which is a machine for conjugating decoded flows; whence capitalism?s affinity with schizophrenia, for it is based on decodings, and its opposition to schizophrenia, because instead of making the decoded flows pass, it blocks them in another way, and it makes them enter into a machine for conjugating decoded flows.

Take an example from the history of painting, the very bizarre history of the Venetian School: it is marked very late by the so-called Byzantine style, while Venice itself was already very advanced in mercantile capitalism, but this merchant and banking capitalism remained entirely nestled in the pores of the old despotic society. And all of Christianity at that time finds something like its pictorial form in these pyramidal structures, literally, in hierarchical mode, which respond to despotic overcoding. These Byzantine paintings of the Venetian school continue until the middle of the 15th century. Here you have this beautiful Byzantine style, and what do you see? - overcoded Christianity, Christianity interpreted according to the style and the manner of overcoding: there is an old despot, there is the father, there's Jesus and there are the tribes of the Apostles. In one of Delphiore's paintings, there are rows of pyramids which are spread in fine rows facing straight ahead. It is not just the people who are coded and overcoded in Byzantine art, it is also their organs which are coded, coded and overcoded, under the great unifying influence of the despot, whether this despot is God or the father or whether he is the great Byzantine Emperor. We get the impression that their organs are the object of a collective hierarchic investment. It would be mad for a Virgin to be looking to the right while baby Jesus was looking the other way. You've got to be mad to invent something like that; you couldn't do something like that under a regime where organs are collectively invested, are coded by the collectivity and are overcoded. Under Christianity, the codes are mixed up, but this is because despotic codes coexist with territorial codes, and the colours themselves enter into the pictorial code. And if, in a museum, you change rooms, you will discover something else altogether, it is a great joy and a great anxiety too, for they are in the process of decoding the flows, a process which doesn't coincide with the explosion of capitalism, but comes quite a bit later. The great decoding of the flows of painting takes place around 1450, right in mid-15th century, and it's a kind of radical break: all of a sudden we see the hierarchy of overcoding breaking down, the ruin of the territorial codes, the flows of painting go insane too, destroying all of the codes, a flow passes. We get the impression that painters - occupying their usual position amongst artists in relation to the social system - create Christs that are totally queer, they are totally mannerist Christs, it's all sexualized, they create Virgins who stand in for all women, and baby boys who have just nursed, little boys pooping, they really play at this process of decoding flows of colour.

And how does this happen? Everything happens as if, for the first time, the characters represented became the owners of their own organs: the collective hierarchic codification of organs, the social investment of organs is done with; from now on the Virgin and every other character will, literally, start to run their own affair; in fact the painting escapes on all sides: the Virgin looks to one side, there are two guys looking at baby Jesus, a third is looking on as if something were happening, there are scenes in the background and the picture explodes in all directions so that every one starts possessing their own organs. They are not insane, there is one member of the Venetian School who makes a creation of the world that is incredible: generally the creation of the world 'a la Byzantine' was done according to a hierarchy; there was a kind of cone or big pyramid of the despotic order and at the very bottom, the territorial codes. The creation of the world that interests me is a departure from this: there is God the Father up in the sky in the position of a runner, and he gives a starting signal; in front of him are ducks and chickens who are racing away as fast as they can, and in the sea there are fish who are also racing away, and God is the one who sends it all away, it is the end of all the codes.

And what do they do with the body of Christ? The body of Christ is useful as a body without organs; so they machine it in every way and direction, they make it amorous, suffering, tortured, but you can feel that it is joyous. The perspective, you see, the perspective, that's nothing, a useless trick; those who have done without it, it's because they didn't need it, they had other problems. Perspective is lines of flight, and can only be useful in a painting of decoding, but it is very secondary, it doesn't even count in the organization of a painting. So what are they trying to do, we're going to cut low along Christ's hip, we're going to make a mannerist Christ, all the tortured bodies are good bodies without organs, Saint Sebastian with his arrows sticking out in all directions; again, in this overthrowing of the pictorial system, perspective is only a little thing. The generalized decoding of flows has to be taken up again by something other than a code, and in effect, there is no longer a pictorial code, but instead a strange sort of pictorial machine that conjoins and that will give rise to the unity of the picture, no longer a signifying unity of a code or overcode, but a system of echoes, of repetitions, of oppositions, of symmetries, a veritable conjugating machine, where flows of colours and decoded features are conjugated. There emerges a real pictorial axiomatic that replaces the failing codes.

Capitalism doesn't arise by means of the simple decoding of flows, for it does not appear until such time as flows, inasmuch as they are decoded, enter into conjunction with one another. Marx has said that when this happens it proves his great theory of contingency. In Rome, as at the end of feudalism, the decoding of flows brought in a new kind of slavery and not at all capitalism. For what had to take place was the meeting of the flow of decoded capital and that of deterritorialized labour. Why did this encounter take place? Refer to Marx's notion of primitive accumulation, on one condition and that is that primitive accumulation, this can be a dangerous thing, if we say: oh yes, primitive accumulation, this is what served to fuel the process of accumulation, and we might as well say the formation of stocks at the beginning of capitalism. One has to see that primitive accumulation is called primitive to distinguish it from other types of accumulation, not primitive because it comes first...

The way capitalism functions, even if we are talking about its industrial essence, is the way banking and trading work, and we must hold that capitalism is essentially industrial, although it only functions through its banking system and through its trade circuits. Why? There is a kind of conjunction: capital starts to take control of production, but is it the first time? No. So, we take up Marx's analysis again, an analysis which Marx insists on: the control of production by capital has existed in a certain guise from the beginning, but it appears in another guise with capitalism. I mean to say that even from the perspective of banking and mercantile capitalism, the banks and merchants create a monopoly for themselves: there is at the outset of capitalism, the way in which English mercantile capitalism forbids foreign capitalists from buying wool and woollen cloth; in that case, this exclusive clause is the means whereby the local merchant capitalists ensure that they control production because the producers can only sell through them; so, we have to mark two times: a first time where merchant capitalists, in England for example, make the producers work for them by means of a system of delegation whereby the producer becomes a sort of sub-contractor, and this is where commercial capital directly takes possession of production; it is the great historic moment when merchant capital declared war on the leagues, i.e. the associations of producers. It's a war between producers who looked with trepidation upon their being in the service of merchant capital, and merchant capital which, on the contrary, was trying to obtain greater and greater control of production via this sub-contracting method. But it would take, as Marx declared, a second time ... means whereby the local merchant capitalists ensure that they control production because the producers can only sell through them; so, we have to mark two times: a first time where merchant capitalists, in England for example, make the producers work for them by means of a system of delegation whereby the producer becomes a sort of sub-contractor, and this is where commercial capital directly takes possession of production; it is the great historic moment when merchant capital declared war on the leagues, i.e. the associations of producers. It's a war between producers who looked with trepidation upon their being in the service of merchant capital, and merchant capital which, on the contrary, was trying to obtain greater and greater control of production via this sub-contracting method. But it would take, as Marx declared, a second time ...