January 21, 1986
Just as modern cinema breaks with the complementarity of seeing and speaking in favor of a seeing-speaking disjunction, so, too, Foucault, for his part, breaks with the seeing-speaking complementarity, such as it develops in classical philosophy and such as phenomenology still conserves it, where, in a certain manner … speaking becomes linked to seeing. … With Blanchot, with Foucault, there is no more linkage between speaking and seeing; there is a disjunction between seeing and speaking. So, in this sense, Foucault’s encounter with modern cinema seems very normal to me: same problem, same concern.
Seminar Introduction
After Michel Foucault’s death from AIDS on June 25, 1984, Deleuze decided to devote an entire year of his seminar to a study of Foucault’s writings. Deleuze analyses in detail what he took to be the three “axes” of Foucault’s thought: knowledge, power, and subjectivation. Parts of the seminar contributed to the publication of Deleuze’s book Foucault (Paris: Minuit, 1986), which subsequently appeared in an English translation by Seán Hand (Minneapolis: University of Minnesota Press, 1988).
For archival purposes, the English translations are based on the original transcripts from Paris 8, all of which have been revised with reference to the BNF recordings available thanks to Hidenobu Suzuki, and with the generous assistance of Marc Haas.
English Translation
After a brief consideration of Foucault’s relationship with cinema in terms of his problem of the visible and the statable (cf. René Allio’s film on Pierre Rivière), Deleuze returns to power as presented in a diagram exposing a relation of forces, distinguishing characteristics of power in contrast to knowledge, and how power reveals active as well as reactive affects. Addressing the plurality of forces, Deleuze returns to the example of the rise of pastoral power, then to the diagram of power with two different definitions, and to diagrams in different societies, hence indicating mutations from one diagram to another. Deleuze raises the question of origins of a diagram with different responses (the Outside, the dice throw as emission of singularities), and such relinkages of diagrams correspond to “Markov chains”, but also to Nietzsche’s Zarathustra and “the iron hand of necessity” shaking the cup of chance. As regards the diagram, Deleuze describes it as informal, non-stratified, perpetually mutating, abstract without being general, and in fact, virtual yet real, then provides details of how mathematical concepts correspond the knowledge-power distinctions. Moreover, since institutions are forms that integrate micro-relations of power, Foucault resuscitates the word “government” to designate relations of forces most generally, discussing how institutions as molar agencies actualize and integrate molecular relations of forces (cf. Disipline and Punish, The History of Sexuality). As an example of molecular sexuality, Deleuze considers how Proust describes sexuality in Sodom and Gomorrah through numerous variables of the body and pleasures.

Gilles Deleuze
Seminar on Foucault, 1985-1986
Part II: Power
Lecture 11, 21 January 1986
Transcribed by Annabelle Dufourcq; time stamp and additional revisions, Charles J. Stivale
Translated by Christopher Penfield; additional revisions, Charles J. Stivale
Part 1
… So, you see… well, let us begin by taking a step backwards. We had left open a void, and at last we have the means to fill this void. We are returning to the moment when we…, we were trying… to see clearly the seeing/speaking duality, or the visible/statable duality. And if you recall, we said that it was in fact very interesting because it intersects with an adventure in contemporary cinema. And we said: yes, perhaps this explains, among other reasons, why Foucault was so interested in contemporary cinema.
And in what did this shared adventure consist? This shared adventure consisted in the following: that contemporary cinema, in certain of its forms, had broken with a complementarity that characterized so-called “classical” cinema. The complementarity that characterized so-called “classical” cinema, to give a very rough summary, is the complementarity of the field of vision, on the one hand, and on the other, of the off-screen and voice-over. And if contemporary cinema, through its most advanced figures, has broken with this complementarity, it is in favor of what? There is no longer complementarity between a field of vision and an off-screen doubled by a voice-over, but instead there is a disjunction between two fields: a sonorous field and a visual field.
And in very different forms, you will find this in auteurs like the Straubs [Jean-Marie Straub and Danièle Huillet], Marguerite Duras, and [Hans-Jürgen] Syberberg. In this respect, it seems not to be arbitrary that these auteurs are often grouped together. They have in common that they have broken with a certain form of classical cinema in order to introduce a new mode of exploration, a properly audiovisual exploration, taking the form of a disjunction of the visual image and the sound image. In the Straubs, this disjunction appears in the form of an image that, when we were occupied with cinema last year, I proposed to call stratigraphic, that is, an image of empty earthen space, which can be taken as what the earth covers over [recouvre] and will not be shown. But what will be shown is the sedimentary character, the stratified character of tectonic space. And the voice, for its part, evokes what the earth covers over, that is, raises up what the earth buries.[1]
As a result, you find, in the words of one commentator on Straub and Duras: a story that has no place, for a place that has no story. The visual image shows you an empty place, an empty space, a place that has no story. The sound image tells you a story that has no place. And it’s the whole cavernous space, Straub’s space of fissures, the telluric space of the Straubs, during which the voice elevates what the earth takes down. In Duras, if you take “India Song”, it is another approach, but… it is another form, but one which can be grouped under the same rubric: this great fault line, this great audiovisual disjunction. “India Song”, for example, which is the purest expression of this, though I no longer recall the date … no one remembers the date of “India Song”? 1974, that’s very… we will see why that interests me. Indeed, “India Song”, in the visual image, presents us with a ball – which has no other function than to cover a silent ball, since, even when the characters speak, they keep their mouths closed, ensuring the visual-sound disjunction – thus, presents us with a silent ball that stands in for another ball that took place elsewhere and at another time.
So, there is something analogous; it is not the telluric method of the Straubs, but a dramatic method by which what is shown to us covers [recouvre] the old ball, and this old ball is what the voices will be discussing. As a result, you have there a pure disjunction of the sound image, which evokes an old ball, and the visual image, the current ball that we are shown, the silent ball, which has no other function than to cover over and sink the old ball that the voices by contrast exhume. Thus, a seeing-speaking disjunction. And in Syberberg, you have still other methods that will culminate with Hitler, but which begin already with Ludwig … or even with Theodor Hierneis …, and which are also distributed according to a great disjunction between visual image and sound image.
So, I can say, indeed, that what had been in a certain way the exploration of classical cinema changes completely; it is no longer the same exploration at all. Classical cinema was not really audiovisual because it still subordinated the sound image to the visual image. If it is only recently that cinema has become truly audiovisual, it is because the audiovisual can posit a strict equality between the sound image and the visual image only by eliminating the subordination of one to the other, and thus proceeding by way of a disjunction between the sonorous and the visual. There is no longer an off-screen, there is no longer a voice-over. There are two fields that rival each other, each of which is sufficient. It is the great seeing-speaking disjunction.
Now, I was saying that just as modern cinema breaks with the complementarity of seeing and speaking in favor of a seeing-speaking disjunction, so, too, Foucault, for his part, breaks with the seeing-speaking complementarity, such as it develops in classical philosophy and such as phenomenology still conserves it, where, in a certain manner – we will see this when will have occasion to speak of [Martin] Heidegger or [Maurice] Merleau-Ponty –, speaking becomes linked to seeing. With Foucault, who in this regard follows [Maurice] Blanchot… with Blanchot, with Foucault, there is no more linkage between speaking and seeing; there is a disjunction between seeing and speaking. So, in this sense, Foucault’s encounter with modern cinema seems very normal to me: same problem, same concern.
Hence, this is where we left open a void, where I was talking about the interest, if we could, in re-watching the film that René Allio, in 1976 – this is where the dates are important, I believe it was 1976 –, drew from Foucault’s work I, Pierre Rivière, Having Slaughtered My Mother, My Sister, and My Brother. Why? Because it was central to our problem. Why was Foucault interested in this film? Well, first, why was he interested in the case of Pierre Rivière, murderer of his parents from the nineteenth century? Why did it appeal to him so much? No doubt for a reason that we have already seen: because Pierre Rivière is an example of what Foucault calls, in a very specific sense, infamous men. And it is a young infamous man. Good, we have looked at this, so I will not return to it. But more profoundly, because the murder case of Pierre Rivière poses a very particular problem; one could find something equivalent, but this case poses, in a particularly acute way, a problem of the relationship between the act and the story, between the action and the narration. We might say just as well: between seeing and saying, between the visible and the statement.
What is visible? The criminal act. What is stated? The notebook written by Pierre Rivière. What is the link between the murder and the story, that is, between the visible action and the statable text? [Pause] And, long before being a question about the film, Foucault had asked the question thusly, because, in his course and in his own commentary on Rivière’s notebook, the publication of which he thus assured, in his short commentary, he says: the relationship between the text and the murder is very complicated. The text does not recount the action. And you see why this problem interests him: it is truly an exemplary case of the relation between the visible and the statable. The text does not recount the action, just as the statement does not recount the visible. The text does not recount the action. But, from one to the other, there is a whole weave of relations. They support each other, they propel one another in relations that have not stopped changing. And in effect, it is a mobile relation, a modifiable relation, a variable relation between the stated text and the visible action. In effect, for, initially, Rivière the murderer had conceived a kind of intertwining of the notebook and the murder. One end of the notebook would be written before the murder, the other end afterward, and then the notebook would be sent off. You see: the visible action and the written notebook, the scriptable notebook, would intertwine… [Interruption of the recording] [14:02]
Part 2
… But in any case, the other projects persist. For Rivière says that he thought about his notebook so much that the words were practically ready-made, the sentences already written in their being. Good.
You see that I am saying that what could interest both a director and what interested Foucault in the idea of a cinematographic staging of Pierre Rivière was essentially the following: how would cinema distribute the notebook and the murder? Which is to say, the visible and the statable? In this sense, the undertaking to film Pierre Rivière could have two concerns: either a commercial concern that would be of little interest, namely, the history of a murder; or a real cinematographic concern that could only arise at a time when cinema had become aware of a possible disjunction between seeing and speaking, which would find in Pierre Rivière a privileged subject matter, since the whole document Pierre Rivière took its interest in the division between the notebook written by the murderer and the murder committed. You see? Now, this is in 1976, I believe. What happens in 1976 from the point of view of cinema? This is why I asked about the date.
Let us assume in effect – it would be necessary to verify, but in any case, it does not change anything – the three auteurs with whom we are concerned, Straub, Marguerite Duras, and Syberberg, had already begun. Further, I believe – but my apologies, I should have verified this, but I did not have a reference book on hand, so I was unable to do so – I believe that Moses und Aron, which already appears late in Straub’s oeuvre, is from 1975. In any case, Straub’s important films, not the essential ones, but some of Straub’s important films had already appeared before 1975. Duras, in my opinion…, you are saying “India Song”, as you recall, would be from 1974, so a bit earlier. Syberberg had already made… 1974 must also be the year of Syberberg’s Ludwig. So, this new audiovisual cinema is actually very recent. René Allio, who up until then was known for, it seems to me, a very classical form of cinema, very interesting but very classical, found himself confronting a type of problem that seems new to me. It is technically very simple… I mean that it is a technical question: how will one distribute the story of the notebook and the story of the murder? How will one distribute the statable and the visible, once recourse to a voice-over has been excluded?
You will say to me: why is recourse to a voice-over excluded? Well, … as you like, if you resort to a voice-over, you provide this problem with a ready-made solution. What is the disadvantage? It is that you do not account for the specificity of the notebook; you do not prevent the voice-over, at that moment, from having no other function than to guide the vision of the visual image, or in other words, you subordinate what is given in the notebook to what is given in visibility. Thus, strictly speaking, you could do it, but it would be of no cinematographic interest. And Allio, though a director who had, again, remained very classical until then, understood this very well, and since it is certain that Foucault took a great deal of interest in the film adaptation of Pierre Rivière, it is very likely and even certain, I believe Allio said so, it is certain that Allio and Foucault spoke at length about the solution.
And it is a solution, for those who saw the film the other night, a solution in fact very close to …, perhaps we could even say that if there is something reproachable about the film, it is perhaps that there is something pedantic in the application of a principle that is much more complex than the voice-over. But Allio, with, I believe, a great deal of cinematographic intelligence, grasped and inserted himself well into this current – which is, again, quite surprising if one considers Allio’s prior works – this current of Duras and Syberberg, solely on the occasion of this film, because this film demanded it. What do I mean? Well, the first image – you are going to immediately understand, I believe, even those who did not watch the film – the first image seemed striking to me, yet precisely it showed almost the limits of the film; it is too perfect, that is, one has the impression of something systematic. It is even so systematic that the same method will not be able to appear with such crudeness. What does the foreground of the film show? It shows a central tree that divides a Norman meadow. It takes place in Normandy, there is a meadow and then a tree in the middle. It is an empty space. … It is the only empty space of the film. It is like an homage to this new cinema; an empty space à la Straub. Only, what makes…? Okay, it doesn’t matter.
And at the same time, what do we hear? We hear the sounds and the formulations of the circuit court. Understand what I mean: it is not a voice-over, these are not off-screen sounds. It is: contamination, the confrontation of heterogeneous visual and sound images. You have the three strikes of the gavel [Deleuze knocks three times], we begin by hearing the three gavel strikes at the same time as we see the tree and the Norman meadow; we hear the three strikes [Deleuze knocks three times], about which we say, look, it announces the opening of the play, yet it does not announce the opening of the play, but rather the opening of the trial, as is indicated by the following… “Gentlemen…” um… no, “the court is now in session” or else “please rise,” I no longer recall… at any rate, a legal formulation.
This, I am saying, is prototypically the disjunction of the sound image; it is not a voice-over, but the confrontation of two fields: sound field and visual field. I am saying: here is such a pure image of the visual-sound disjunction that it is too perfect. And personally, I do not know… if those who were there… so it bothered me, it bothered me because I said to myself: there is something that doesn’t work, it is anthological, as they say, it is a piece of bravura, one cannot maintain like that for long, and in fact never did Straub or Marguerite Duras or Syberberg ever proceed like that. It is a little bit forced. Okay, good. And next, happily, it cannot continue like that with the double sound-visual source, it’s not tenable, it won’t work, everyone would leave. But it will play out a whole series of discrepancies [décalages] between the sound image and the visual image, in such a manner that the notebook will never function as voice-over. The notebook must inspire and fill a sufficient sound image while the gaze invests a visual image that is itself sufficient, the two never coinciding.
As a result, I took some notes, very poorly because I could not see anything. So… Well, … I will go quickly so that this becomes clearer. What appears to be a voice-over says at a certain point something like “my mother never stopped talking and speaking…”, and the whole visual image shows the mother all alone saying not a word with her mouth closed. At one point … I noted some things down, I already no longer remember… “At one point my parents were arguing all the time” says the voice that seems, again, to be a voice-over, and the image shows the kid, Pierre Rivière, who is working a field, with no parent and no sound of dispute. Then, at one point, the visual image, we are told that Pierre Rivière was bizarre, that he did grotesque things to animals, that he hurt them badly and that it made him laugh; and, among other things, we are shown Pierre Rivière forcing a draft horse to climb onto a manure heap, and then the horse loses its balance, it falls over… all this is a catastrophe, and Rivière laughs hard. But it is much later in the film that the sound image will describe this scene that we are first shown visually. You see: a whole system of discrepancies [décalage].
So, to summarize, I would say: I believe that what is interesting about the film is the way that Allio knew to be aware of this problem – the disjunction between seeing and speaking, between the visible and the statable – and distributed it throughout the whole film using different means, the purest of which being that of the foreground. And surely this responded to the intention of Foucault; I imagine that in the discussions between Allio and Foucault, Foucault would say to Allio, well, you are the filmmaker, what are you going to find in order to take into account this visual-sound double game? And that Foucault, for his part, had in mind solutions along the lines of Duras, Syberberg, and Straub, as, no doubt, did Allio as well.
So, if that is what is interesting about the film, at the same time, perhaps, perhaps, I do not know but perhaps, in my opinion, in my impression as a recent viewer, a certain limitation of the film is that the method is flat [plaqué], a bit flattened out. It does not, as in the case of Duras …, it does not emerge from the very exigency of the work. Alright… perhaps. That is what I wanted to say, I wanted thus to fill in what we had left open as an empty category regarding this relationship between Foucault and cinema in terms of his problem of the visible and the statable, or his problem of the seeing-speaking disjunction.
Is there anything to add to what has just been said? [Pause] Among those who saw the film, do you have anything to add? [Pause] No? [Pause] Is there anyone here among the undergrads [premier cycle] who saw the film? [Laughter] You’ll say no, obviously… No undergrads saw it? Fine. Nothing to add, okay. So then, after this detour backwards, we are continuing going forward. Yes?
A student: [Inaudible question]
Deleuze: Did you see the film?
The student: [Inaudible reply]
Deleuze: Ah! [Laughter] Yes, stand up because we will hear you better if you stand. Sorry to ask this of you…
Another student: [Inaudible question]
Deleuze: It is a primacy that implies no homogenization. I had said: it is exactly like in Kant, where there is a duality between what Kant calls understanding and what he calls intuition. There is an understanding-intuition duality. The understanding has primacy, but that does not at all mean that intuition is reducible to the understanding. So, what does that mean, primacy without reducibility? A has primacy over B, and yet B does not allow itself to be reduced at all to A. It is, if you like, a bit like if I were to say, well yes, those who govern have primacy over those who are governed, from the point of view of the government, and yet this does not imply a reduction of the governed to those who govern. So why is there primacy? I believe it is very simple, it was already Kant’s sole reason: there is a primacy of the active over the passive. Or, if you prefer, there is a primacy of the active over the receptive. Why? Because the active is what does something. It is what does something.
So, when Foucault recognizes a primacy of the statement over the visible, it is in the name of the activity of language through which statements produce themselves. Just like, in Kant, there is an activity of the understanding and a receptivity of intuition. So, this is what it means. As for cinema, it has nothing to do with primacy. It has nothing to do with primacy because primacy, it seems to me, belongs to a kind of philosophical reflection on two irreducible terms. Unless… can we say that there is also, in this cinema of the audiovisual, a primacy of the statement? There, I think the answer is no, because, on the other hand, the work of the visible matter makes it so… I suppose we would have to say that either, for cinema, there is no primacy, or else that cinema has nothing to do with the question “which has primacy?”, that this is not a question for it, not a question that interests it. Yes. Good.
So, forget all that, since you do not want to say anything, and finally we are continuing on. Last time, we delved into the question of power in the following form: given that power is a set of relations of forces corresponding to a formation, power is presented in a diagram, and I attached a great deal of importance to the word “diagram,” which Foucault uses one time. Why? Because it gave me a reason, or rather it gave me a convenient word for clearly marking that we were no longer in the domain of the archive. The archive is the archive of knowledge. The archive of knowledge is not opposed to, but distinguished from, the diagram of power. Now, we have seen, and this was our principal objective last time, that there were two characteristics of the diagram, insofar as it is an exposition of relations of forces. Its general definition would be: exposition of relations of forces.
But what are the characteristics of the diagram? The first characteristic of the diagram is that it brings together [met en présence] non-formed matters and non-formalized functions. In this way, it distinguishes itself from the archive of knowledge. And last time we had just made… In what way is power distinguished from knowledge? And there, the answer should thus be my two characteristics of the diagram, it should be the two differences between power and knowledge.
Thus, power will be a set of non-formed matters and non-formalized functions. By contrast to knowledge, which presents formed matters and formalized, finalized functions. Every form of knowledge implies an already formed matter and an already formalized, finalized function. You sense, if you understand this, that every form of knowledge implies a formed matter and formalized function, and you should already understand why the visible and the statable come together at the level of knowledge. [Pause] It is because formed matters are distinguished by visibility, in the sense that Foucault gave to “visibility,” which is not simply sight, as we saw in the first term; I could say that formed matters are distinguished by visibility, and finalized functions are distinguished by statements.
But as for power, it speaks to us of something else, speaks to us… but I cannot say otherwise…. Power presents to us something else, it regards non-formed matters and non-formalized functions. You will say: that is so abstract! But at any rate, I hope that it was less so the last time. It is precisely abstract, it is abstract. In fact, non-formed matters and non-formalized functions are pure abstraction; it is matter and then voila, matter qua matter or function qua function. How could I put in any variety? The answer is simple: there is a great amount of variety to the diagram, or to diagrams, for the simple reason that non-formed matters are not inevitably one and the same matter, and non-formalized functions are not necessarily one and the same function but are only distinguished by the space-time variable. Thus, in order to define a non-formed matter or a non-formalized function, I will only be able to take into account the space-time variable, following this first characteristic of the diagram.
So non-formed matter will be: some multiplicity, taken to be human… You will tell me: ah, but human is a form! Yes, yes, yes, we will see why we cannot escape, this amounts to saying that we already in fact cannot escape from determinations of knowledge when we speak of power, it goes without saying, but be patient; we will understand why we cannot speak of power in its pure state and can only speak of power as already incarnate in a form of knowledge. It is unavoidable. But that does not prevent the distinction from being valuable.
Thus, I will say: some human multiplicity, without at all specifying… – in this way it is diagrammatic – without at all specifying what multiplicity it is. Some human multiplicity in a closed space, this multiplicity being small, small multiplicity in a closed space. There you have a diagrammatic trait: to impose a task, to impose some task, to impose some task on a small multiplicity in a closed space. I will say: there you have a category of power. It is a diagrammatic trait or a trait of the diagram. If I now say: to control the principal events – I am not specifying which ones – to control the principal events in a large multiplicity itself situated in an open space, I am defining another diagrammatic trait. Thus, the abstraction of the diagram does not prevent its potential varieties, since I have at my disposal a space-time variable that is largely sufficient. And there are all sorts of types of multiplicities. In effect, at this moment, I will define a multiplicity by the way in which it spreads out in a space. Multiplicities will vary according to the type of space and also according to the way it which they occupy the space; so I will have a great deal of variety in the diagram, or from one diagram to another.
Thus, you see, if I say, “to impose a task on a small multiplicity in a closed space,” or if I say “to control the principal events of a large multiplicity in an open space,” I provide categories of power. On the other hand, if I say, “to punish, to educate, to put to work, to teach”, etc., these are not categories of power, these are categories of power/knowledge. These are categories of knowledge, which of course imply power, but they are categories of knowledge. Why? Notice that all these terms in fact mobilize formalized functions and formed matters. The schoolchild, worker, prisoner are formed matters. To impose some task on a small multiplicity in a closed space was a category of power, but now I am in the concrete, there is no matter that is not formed. What is this generic multiplicity? It is a multiplicity of children, so the function will not be to impose a task, but simply to teach. To teach a multiplicity of children in a closed space that will be called the high school or elementary school. But, next door, to impose some task on a small multiplicity in a closed space will become… the formed matter has become the prisoner and no longer the schoolchild, and to impose some task is not to teach but to punish. Is it surprising that, after all, there are punishments at school and there are lessons in the prison? No, they respond to the same diagrammatic trait. You understand? But, thus, if I consider power in the abstract, it being said that power is an abstraction, if you tell me “in fact, it’s never separated!” Clearly, it’s never separated, Foucault is the first to say it. If I make an analysis, it is to mark the difference in that which is not separated in reality.
What must be called “power” is the diagram, which consists in swirling together [brasser] non-formed matters and non-formalized functions. And I would distinguish it in this way from archives of knowledge, which begin from the moment when the functions are formalized and finalized, that is, when it is no longer a question of imposing some generic task, but of teaching, or punishing, or putting to work in a factory, or the like. It is no longer a question of some generic closed space, but sometimes of a school, sometimes a prison, sometimes a workshop…, sometimes a factory. There you have it, that is the first characteristic of the diagram.
And we saw last time: the second characteristic by which power will also be distinguished from knowledge. It is that knowledge, as we saw previously throughout the first term, always proceeds from within a form and, even better, from one form to another. It proceeds following the form of the visible, light, following the form of the statable, language; and better, it proceeds from one form to another, that is, it interweaves the visible and the statable. Whereas power, for its part, goes from one point to another. Power is a set of relations that are punctual and not formal. The point is what… there is a text where Foucault uses the slightly strange expression ‘state of power,’ a ‘state of power.’ I believe that what must be called a power, according to Foucault, is the point, once it has been said that power ceaselessly establishes relations between points, between points in the plural.
What, then, is a point or a state of power? A point of power? Well, here, I believe this must be taken very seriously because it will be important for us. You recall what I told you: the diagram is the exposition of relations of forces. Now, the relation of forces does not add to the force, there is not the force plus its relation with other forces. Force is fundamentally plural, there is force only in the plural. That is, force is, in its essence, a relation with another force. It is with force that the word ‘one’ loses all meaning. There is no force that is not a relation with other forces. It is on this basis that Nietzsche founded his pluralism of forces. Forces are not unifiable. It was a very simple idea, yet so well confirmed by physics, by the physics of forces, which is very important. Good.
“On this basis” means: a force is fundamentally related to other forces, that is, a force has no essence [Pause], but is defined by the fact that it affects other forces and is affected by other forces. That a force always affects others… You see, it has no essence, a force has nothing but affects… That a force always affects others, this is what we will call its spontaneity. Even if it is determined so as to affect other forces, the aspect through which it affects others is its spontaneity; the aspect through which it is always also affected by other forces is what we will call its receptivity. In other words, a force has active affects and reactive affects. [Pause] Its active affects express the manner in which it affects other forces, and its reactive affects express the manner in which it is affected by other forces. I could refer to its active affects as points of spontaneity, like this, I dream… I am trying to build a vocabulary… The other aspect, the reactive affects, I could refer to them as points of receptivity. Good.
So, we sense… okay, what is a point? Well, we have seen – I am returning to my whatsit [truc] – it is a diagrammatic trait expressing a relation of forces, that is, a force that acts on other forces. To impose some task on a limited multiplicity, to impose a force that imposes a task on a small multiplicity of other forces. And I told you that we can diversify according to the aspects retained of a space-time, and this will yield: to arrange [ranger] – a force which arranges others – to arrange, to be arranged; to arrange, on the side of the force that affects, that arranges other forces; to be arranged, on the side of the forces affected. And I told you that there is more than just arranging, there is classifying, a hierarchical series, first, second, third, fourth, fifth, which is also a diagrammatic trait, that is, a category of power, a force that classifies other forces. Here I will again have two affects: to classify – to be classified. To be arranged is a reactive affect, and to arrange is an active affect. To classify and to be classified, etc., I have given all kinds of examples…
You see that we must not believe that reactive affects are simply symmetrical, for there is a specificity of the affect “to be arranged” in relation to the affect “to arrange”. Reactive affects are not the simple… the simple repetition of the active affect. To be arranged responds to arranging, yes, it is a relation of forces. To classify, then, to arrange is, for example, at school… it is the schoolmaster who says, who… all the students wait at the door without smoking, … the schoolmaster arrives and he places them two-by-two. The force of the schoolmaster is to arrange. The force of the students is to be arranged, for there must be a force to support force. To be arranged is a reactive affect of force, like arranging is an active affect of force. An affect of receptivity, to be arranged. But it is not the same thing as the results of the arrangement. The result of the arrangement is a classification. First, second,… thirtieth … all that. Good, you can go in both directions, since you have two kinds of teachers. Teachers are distinguished by the following: those who give the results of the arrangement by beginning at the end, and those who give the results of the arrangement by beginning at the beginning. They do not follow the series going the same direction: these are not the same affects that are distributed in the class, the young passive forces that receive the instruction of the schoolmaster do not at all feel the same affects.
I remember very well in my childhood — I am opening a parenthesis — better or for worse, the teachers who began at the end were considered to be properly sadistic. For there is no doubt that it was original. There are fewer teachers who begin at the end than at the beginning; any teacher who distinguishes themselves from the average does not appear clear to the child, and from the moment when it does not appear clear to the child, he has good reasons for thinking that there is some kind of dirty trick going on. And, in effect, with the teachers who begin at the end, it is not at all the same type of emotion, that is, the affects are going to be … fine. First of all, it is completely despicable, eh — and I’m emphasizing it quickly — because in the last place, it’s a hammer blow, immediately, it is a heavy blow. The guy begins: thirty-first, ah… it is a hammer blow.
In the other case, let us consider, when the last place comes at the end, it’s not joyous obviously, the anxiety grows, but to me it still seems less inhuman than the massive… um… and, after all, inversely, the anguish grows … when one begins at the end, the anxiety grows for the one in first place, who says: fifth place… ah, am I going to be first place? It’s not good, it seems to me, it is not a good thing. It is better for the first place student to be immediately situated in first place, and humaneness must grow as the order descends…; whereas in this case, on the contrary, by beginning at the end, one makes the first place student more and more prideful. So… the little last place student was stunned… In any case, it is not at all the same distribution of affects in the classroom. It is only the middle that doesn’t give a damn. [Laughter] The middle doesn’t give a damn either way, then. They don’t give a damn because, whether it goes in one direction or the other, they are still there in the middle and are the only ones who don’t move.
But, thus, … the affects … understand that, based on a small variation like that, the distribution of affects in the social field, that is, in the class, in the space considered, changes completely. So, I would say that I can define power as every relation from one point to another, if I understand what “point” means. And here, I am simply suggesting that “point” must be understood as the strict equivalent to “affect”; a point of force is not the origin of a force but the affect, that is, its relation with another force that affects it or that it affects. That is what defines a point. I would say, then, that power, the power relation, goes from one point to another. Which amounts to… it is the development of the simple idea: all force is plural, there are only pluralities of forces.
So, I am here returning to a theme … that we have sketched a little bit, saying: among the diagrams… Foucault, at the end of his life, accorded a great deal of importance to pastoral power, to this ecclesiastical power. Why? Because he saw there the first power that proposed to individualize its subjects. But how to define pastoral power, this power of the church? Foucault defines it in his manner… But, for me, this evokes something because I tried to tell you: this whole conception of Foucault is very Nietzschean. It is very Nietzschean, and it is so Nietzschean that, strictly speaking, he had no reason to say so. Now, it is all the more important for our purposes that Nietzsche is surely the first to have posed the question, in cold and cynical terms… no, not cynical, in cold terms: what does the power of the priest consist in? And this is perhaps one of the most novel things in his philosophy: what is the power of the priest? What does it consist in? I am not at all saying “cynical,” I am saying: he demands a positive theory of the power of the priest. Where does this strange power come from? Hence Nietzsche also can exclaim at the end of his life: “I am the first to have done a psychology of the priest!” By psychology, he means something very elevated. “I am the first to have done a psychology of the priest.”
So, Foucault would be the second. What does this mean? But consider the great psychology of the priest that Nietzsche provides in the second treatise of the Genealogy of Morals, for he does not only provide a psychology of the priest, but a psychology of the mutation of the priest. And this second treatise of the Genealogy of Morals begins by providing the psychology of the Jewish priest, so as to pass on to the psychology of the Christian priest. Now, how – I do not want to go back over all of this, but it is just so you better understand what is at stake in this matter of the diagram and relations of forces – how does Nietzsche define the Christian priest, that is, the pastor? It is the pastor, and Nietzsche often uses the form “the pastor,” the expression “the pastor”. How does he define the pastor? He gives a very strange definition, but it is so beautiful: it is one who returns force against itself… [Interruption of the recording] [1:00:32]
Part 3
… Hence the possibility of defining power without ever speaking of man. For all the histories, like the death of man, etc., we will see later on, we will see very soon that if these ideas of Foucault were so misunderstood, it is because, from the beginning, nothing was understood of what he was saying. A diagram of power makes no appeal to the “man” form, even if it presupposes it. It must be entirely articulated in terms of affects and relations between the affects and relations of forces.
This is why I can say: if the first determination of the diagram is the swirling together of non-formed matters and non-formalized functions, the second definition of the diagram, equal to the first, incidentally, is an emission of affects or singularities, an emission of affects or singularities corresponding to a social field, corresponding to a collective field.
And grant me that there is a strict implication from one definition to the other. Ultimately, they are two ways of saying the same thing. This wouldn’t be difficult to show, but it doesn’t much matter. The two say absolutely the same thing. And the two definitions give us two differences between power and knowledge. Knowledge passes through forms and operates within forms. Power establishes relations from a singular point to another singular point, from one point to another. Knowledge regards formed matters and formalized functions; power regards only non-formed matters and non-formalized functions.
Hence you may understand my question better. Foucault uses this word “diagram” one time, and he uses it concerning disciplinary societies, which is to say, what he considers to be our societies, and I told you last time: good, very well, but there remains a problem for us, we readers of Foucault. Does this mean that the diagram of power exists only for our societies? And that, before these societies, other previous societies were ruled [régies] by something other than a diagram of forces? For example: did the sovereign replace the power of the diagram? There was a power of the sovereign and not a power of the abstract diagram. I would say: definitely not. Quite evidently not. For the sovereign, in turn, is only a formalized and finalized function that presupposes a diagram. It is thus for every society, for every social field, no matter what it is, that we must say there is a diagram. Which brings us finally to an important problem, which we saw last time, and in this regard, I would like to add something that will prepare us going forward.
I was saying that just before our disciplinary societies, you have, according to Foucault, the societies of sovereignty, and this is another diagram, their diagram is not disciplinary. “Disciplinary” designates a certain set of relations of force with force, but it is not the only relation. Once again, in a diagram of sovereignty, it is no longer a question of composing forces, as in discipline…; it is a question of levying forces. The sovereign is a force that levies other forces. It is a power of levies. And levying is an active affect, just as “being levied” or, rather, “being the force on which one levies something” is a reactive affect. It will thus be another distribution of singularities in societies of sovereignty, other relations of forces, fine, but there will be a diagram. It must be said that the diagram is not reserved for disciplinary societies, but there is a disciplinary diagram just like there is a diagram of sovereignty.
And so, in this spirit, I was saying, in the limit case, regarding matters with which Foucault was never the least bit concerned, in so-called primitive societies, there is a diagram, the diagram of alliances, which is a very particular diagram, some aspects of which will remain in the diagram of feudalism, for example, with a network of alliances but conceived in a new way. So with diagrams, you can multiply them, you can make as many as you like, especially since there are inter-diagrams, I was saying.
And, finally, we must ask ourselves… we will see what kind of problem arises from this, we will be compelled to see… It will be necessary in the end to say that diagrams are always intermediary, much more than they belong to a social field. Diagrams are always intermediary between a social field in the process of disappearing and a social field in the process of emerging. Why? You will see where I want to take this. It is something that will concern a fundamental instability of the diagram. The relations between forces in a social field are fundamentally unstable. What is stable? You will see right away what is stable, namely, the strata, yes, the very formation, but the diagram that is like the motor of the formation is itself fundamentally unstable.
But, at any rate, we will see this… This is why there are, according to the text of Foucault, there will be…, we will be able to speak of a Napoleonic diagram, as I told you. The Napoleonic diagram is Napoleon precisely at the hinge point of the conversion from the society of sovereignty into disciplinary society; he presents himself as the resurrection of the ancient sovereign, of the great Emperor, but in the service of a completely new society that will be a society…, a diagram of the organization of detail, a kind of secular pastor, the surveillance of cattle. And the Napoleonic diagram will be the conversion of the society of sovereignty into the society of discipline.
But, to better emphasize the diagrams, which are everywhere and vary greatly from each other, I say to myself, okay then, for the first-year students…, I could give them an assignment. It would be: construct… There would be problems, like in geometry, you know. Construct the diagram of feudalism. So, if they speak to me of the knight, I will cross it out and mark zero: [Laughter] they have not understood what a diagram is, they must only speak of affects of singularities and relations of forces. They will therefore be forbidden the word “knight”, and each affect will be defined by a force relation, but there will surely be an “x” (small x) that will only be able to be performed by horse. Very interesting. Do you want me to give you an assignment…? Even for everyone, eh? Or a written quiz, that would be marvelous… And then I will give you the grades, and after, the comments, eh? I will begin with last place, obviously… Okay. Alright, then.
And then another topic would be: in your view, …are we still within the disciplinary diagram, or are we already within a society that refers to a new diagram? I was telling you last time, the partisans of the idea of a postmodern era, in which we would have entered, means above all that, according to them, the diagram has already… how do you say it? “Molted” or “mutated” [mué ou muté]? You say “mutated”? You are sure? Mutated. The diagram has mutated. The diagram has molted… Yes, you say “mutated,” good. They are not the same, but yes, good. Alright, then.
Or I could ask: and the Greek city-state, does it have a diagram? I mean, this interests me greatly because Foucault, at the end of his life, converts all his reflection into a reflection upon… in part, not all his reflection, but a reflection upon the Greek city-state assumes a particular importance. So that, if I jump into his last book, one of his last books about which nothing at all has been said, The Use of Pleasure, I say to myself: is there enough there to find a diagram? And here we would see the originality of Foucault’s method because we would have to see how Foucault defines the Greek city-state, how ancient historians defined the Greek city-state, how contemporary historians define the Greek city-state. This would then be a privileged terrain where we could directly evaluate Foucault’s originality in relation to historians. Now, if you read The Use of Pleasure, it seems to me that what he retains as an essential characteristic of the Greek city-state is very interesting, because, if I take as criteria the best current historians of ancient Greece, [Paul] Vernant or [Marcel] Detienne, for example, I would not say that it is in opposition, but that it does not coincide. It is not opposed to, but does not coincide with, the principle traits retained by Vernant or Detienne.
What, in fact, strikes Foucault about the Greek city-state? He does not think that the Greeks were some miracle, eh, he does not make the Greeks into an astounding idea. In this regard, he is full of implicit irony vis-à-vis Heidegger, for he does not see in the Greeks the world-historical gesture or gest. He says: oh no, these are interesting people, but in the end, it is nothing earth-shattering; no, they are not at all the shepherds of Being. They are not the shepherds of Being, but in a sense, as always, Foucault says, this is his form of humor: they did much less and much more, depending on your perspective. So, what is greater than “being the shepherds of Being”? If you look at the text, you can read The Use of Pleasure by searching for it: what diagram of the Greek city-state does this book imply? He has a very curious idea, which is that the Greeks invented something very strange in the relations between forces: they think that force is only exercised legitimately over other forces when the forces are free. You will tell me: great, but what are free forces? Well, no doubt it refers to preexisting diagrams. And this obviously refers to the Greek citizens… You will tell me: yes, but here I have lapsed into the error I was denouncing, and I cannot help myself to the Greek citizen, the free man. I cannot help myself to the free man since the free man is not diagrammatic. It’s not that this does not exist, but that it is not diagrammatic.
So, I must choose another word. It’s not just a semantic distinction: I will speak of free agents, free agents, that is, those who are presumed to not be already determined by a pre-existing power relation; this is an abstraction, which I can very well make. “Free agent,” by contrast to agents who are already determined by pre-existing power relations, for example, slaves. And the great idea of the Greeks is a rivalry between free men. I believe that is what Foucault is saying: I believe it, but is is up to you to say it and to see if… His idea of the Greeks appears to me to be very strong… We will pick all this back up, so it’s not pressing, it’s just to anticipate where we’ll be going. His idea of the Greeks is that what they invented politically is something that has not ceased inspiring democracy afterward. If they are democrats, it is in this sense: for them, free men are in a state of free rivalry. Why is this diagrammatic? Because I do not say in what domain. In all the domains. In any domain. So, we must not be surprised that the Greek city-states did not stop warring with each other, since they lived fundamentally as rivals. It is not the barbarians: the barbarians did not compete with the Greeks, from the Greeks’ perspective. However, the Greeks compete with Greeks. Greekness is the rivalry of free men.
So let us remove “man,” since I can no longer say it. It is the rivalry of free agents, with the corresponding affects. Take Plato; it is striking, in Plato, the extent to which he registered [recueilli] this. It’s not that he was in agreement: Plato said that this was what the Greek city-state was going to die from. But I am saying “in all the domains,” and in politics, it is a matter of competing: this is precisely what defines free men, what defines citizens, to be agents who are able, who are capable of competing with one another. Hence the Greek theme, namely, to be the best. What is the best? So, Socrates exults, saying: ah yes, but in which domain? In which domain? You talk about racing, so, if talking about racing, the best is the fastest, he says. For the footrace, it is the fastest. Here, there is a simple question and a simple response. But can you say this for politics or for medicine? Is the best doctor the one who cures the fastest? Then the interlocutor says: “no, Socrates, it is not the one who cures the fastest, it must be the one who cures well.” “Ah, yes, but isn’t running fast the same as running well?” Then the interlocutor says: “Ah, yes Socrates, but already I no longer understand…” So, Socrates says: “Okay, come on, we’ll start again. So, the best doctor is not the one who goes fastest, thus the good is not the fast?” The interlocutor, he… [Deleuze does not finish the sentence] So okay, it’s not the fast. What will rivalry be, then?
There is a dialogue that I find to be Plato’s most beautiful, namely, The Statesman. And now, Plato comes to say: well, yes, the politician – this is how this text by Plato inspires all of pastoral power – the politician is the pastor of men, and with respect to that, one sees plenty of … – it is great theater – plenty of people who arrive and say: it is I! It is I! You see: it is I. Then everything goes. The pastor of men is the one who takes care of man as a flock. So: the butcher arrives and says: who takes better care of man than I do, by providing him with meat? The weaver arrives and says: but no, it is I! Yes, it is I who clothe him, giving him his own wool. Without me, he would be like a completely shorn sheep. It is I, the pastor of men, etc., and everyone battles: that is the Greek city-state.
You pose a question and he replies to you: it is I, it is I, it is I! Go say that where the emperor of China resides; there will be no rivalry among free agents. There will be free men where the emperor of China resides, certainly, but they will not enter into a rivalry, into a pure and direct rivalry. They will not compete. They will perhaps compete in secret, but in order for rivalry there to be precisely the relation by which forces pass… And then, it may not be the footrace… Rivalry in the Olympic games – that is why the Greeks invent the games, with their theme of competition among free people, among free agents – so they invent the Olympics, absolutely, but they also invent democracy… Democracy is not liberty, nor is it equality, but rather the power to compete [rivaliser]. The power to compete is an affect. You understand, I would say, that rivalry is the diagram of the Greek city-state, just as I was saying that alliance is the primitive diagram. And is it surprising that they invent eloquence? Rivalry, I would say, implies the form of the good adaptable to each domain. The form of the good, insofar as it is adaptable to each domain, is what defines rivalry.
What is the form of the good in the domain of speaking? The form of the good in the domain of speaking is eloquence. The Greeks will thus have institutions of eloquence. The form of the good in the domain of physical exercise is, as we put it, speed, at least in certain physical exercises it is speed; consequently, the Greeks will have the Olympics. And just like that. I could say that the diagram of the Greek city-state…– just as there is a disciplinary diagram, a diagram of sovereignty and a thousand other diagrams – I would say that the diagram of the Greek city-state is rivalry among free agents. And, once again, in a despotic society, in an imperial society, rivalry among free agents cannot take place. Good.
So then, you see our problem: we now find ourselves not, as we first believed, before a sole diagram, but before an open multiplicity of diagrams. This suits us, but remember that each diagram is itself a multiplicity, multiplicity of relations of forces, multiplicity of singularities, multiplicity of … affects. Well, yes. Each diagram is a multiplicity, and there is a multiplicity of diagrams. But then, where do the diagrams come from? This is an importance question. What traverses them? [Qu’est-ce qui les parcourt?] … What is this diagrammatic materiality? What is that kind of thing? Well, no doubt there is neither beginning nor end. There is neither beginning nor end, neither an origin of forces nor an end of forces. There are only reincarnations [avatars] of forces. There are only metamorphoses of forces. The question of origin and the question of destination are disqualified from the point of view of forces. What does this mean? It means that not only are there mutations from one diagram to another, but every diagram is a place of mutation, and that this is what defines the diagram: it is a place of mutation.
And then, how to define a mutation? A short text by Foucault will say it, in The Order of Things, page 217: “mutations that suddenly decide…” — here, then, this will interest us because it once again prepares our future — “mutations that suddenly decide that things are no longer perceived” — I am putting in parentheses the following: “(described)” — I close my parenthesis: “expressed…” — There you have it: listen carefully — “…these mutations that suddenly decide that things are no longer perceived, (described), expressed, characterized, classified, and known in the same way…” Suddenly I no longer speak, I no longer perceive — remove the “I” — suddenly one no longer perceives in the same way. Suddenly one sees something that one did not see before. Suddenly things are said that were not said before.
You will tell me: we are not in the domain of the visible and the statable. Of course, of course, we are in the domain of the visible and the statable, but there was a mutation. For there to have been a mutation, the diagram must have passed through it. What assures the mutation is the diagram. The diagram is the mutation itself. Relations of forces are in perpetual mutation. There is no longer, I can no longer even speak of current society. So-called current society is only the conjunction of what is still in the process of disappearing and what is still in the process of emerging. Parenthesis: [Antonio] Gramsci said extremely profound things about this, about this intermediary characteristic of every society. [Pause] So, good.
But finally, this brings us to a… a very delicate point, which is what? I would almost put it: where does the diagram come from? Where does the diagram come from? Since we saw that it did not presuppose any forms. I can invoke no form. It is primary in relation to forms. Forms follow from it. We do not yet know how, but it is primary in relation to forms. It is primary in relation to knowledge. It precedes all knowledge. No doubt nothing could be known if there were no diagram, but the diagram does not belong to the domain of knowledge.
So then, where does the diagram come from? We will see Foucault’s reply. For the moment, let us content ourselves with what we can. It must be said, there are two replies… The diagram … it will be necessary to consider one after the other…. The diagram always comes from outside. For me, I’d like this: the diagram comes from outside, but what does that mean? Either it means nothing or else it requires an analysis of this notion of outside. We can be sure that it requires an analysis of this notion of outside and that it will be necessary to take account of this notion of the outside in Foucault, which is all the more important given that, in an article written in homage to [Maurice] Blanchot, he titled his text “The Thought of the Outside”.[2] So, I would say: the diagram comes from outside, but we do not know what Foucault might have in mind by “outside”.
Second reply: the diagram always comes from another diagram. Oh yeah: why? Because every diagram is mutation. Every diagram is a mutation of a preceding diagram, which was itself already mutation. The diagram is fundamentally mutant. It even expresses, in a society, the mutations that are possible. Good. How to join the two together? If you have followed everything I have said, I would say: if you want a definition of the diagram, it is very simple, namely, the casting of a dice throw [l’émission d’un coup de dés].[3] What is it to think? To think is to cast a dice throw. We are Mallarmeans. What are the points on the die? The points on the die are affects or singularities. These are singular points. To think is to cast a dice throw. To think is to cast a dice throw. As soon as you think something, you have cast a dice throw. Your molecular brain cast a dice throw, and you say: I think. What allows Descartes to say “I think” is that his brain cast a dice throw. Each time that you think, you cast a dice throw or, at least, your brain casts a dice throw for you. I cast a dice throw, good, which is to say: I produce a distribution of singularities. To cast a dice throw means: to produce a distribution of singularities. Okay? You have released some such combination. That is what a diagram is, a distribution of singularities, a roll of the dice.
Only, here, understand, we’re going to… …we are going to understand everything. For I would say: every diagram comes from a preceding diagram. Well, yes, there are drawings [tirages]. There are successive drawings. There are successive drawings. So, you give yourself – because we cannot go back infinitely – you give yourself a first roll of the dice, you call it “diagram 1”, and then you give yourself a second roll of the dice, which yields “diagram 2”. In what relation is diagram 2 to diagram 1? You could conceive several cases. Either the drawings are completely independent of each other, which is the independence of the cases claimed by pure statistics. Each roll of the dice is presumed to be radically independent of the other. That is the extreme case.
Second case: the preceding dice throw fixes the conditions under which the following dice throws are cast. This does not remove chance. But it makes a mixture that we will call a mixture of chance and dependency. We discussed this, if… those who were here last year, we discussed this with regard to cinema, it was what we thought should be called “re-linkages” [réenchaînements] rather than linkages. When there is a succession of dice throws such that the second dice throw partially depends on the conditions produced by the first, the conditions determined by the first. Those who know a little bit in this domain know that what I am in the process of describing are called mixtures of the aleatory and dependency, or “Markov chains,” named after the scholar [Andreï Markov] who studied this kind of linkage or re-linkage.
For me, I believe that the relation between a diagram and another diagram, between a diagram and the diagram following, is typically a Markov chain; that is, each diagram is a dice throw, but the second diagram re-links together with the preceding one because the preceding one fixed the conditions under which the second throw, the second casting took place. Typical Markov chain. We dwelled on this for longer. Now I don’t have the time, this year, and then it would be necessary to start over, but those whom this interests would do well to refer to statistics manuals, to probabilities, to a chapter on Markov chains that currently plays a very important role in mathematics and physics.
But then, perhaps you understand certain themes, after all…To think is to cast a roll of the dice: of course, this is from Mallarmé, but it was also, independently of Mallarmé, from Nietzsche. And [Thus Spake] Zarathustra is full of this conception of thought: to think is to cast a roll of the dice. When you will have thrown the dice on the table of the earth. When I will have thrown the dice on the table of the earth. It is even a question, at that moment, of the table which splits – this is more than we should take on – which splits according to a principal disjunction. Good.
And with regard to this conception of the dice throw thought, it will be important to confront Mallarmé and Nietzsche, for it is doubtful that they meant exactly the same thing, yet in the end, it is nevertheless the same family. This is better explained, in light of what we were just saying, by Nietzsche’s formula: the iron hand of necessity – I am citing this formula because Foucault loved it and cited it, notably in his article on Nietzsche – the iron hand of necessity that shakes the cup of chance. The iron hand of necessity that shakes the cup of chance. This is the fundamental formula of Zarathustra. The iron hand of necessity that shakes… I repeat it because you must know it by heart. The iron hand of necessity that shakes the cup of chance.
For me, I don’t know…, I would like especially to not flatten, to not make this subject to a scientific formula, but I would say: it is clear that this implies – I am not saying that it is reducible to – this implies a conception that is already a Markov chain. What is the iron hand of necessity? It is that every preceding dice throw determines conditions. That is the necessity, the cup. Every dice throw… What is the cup, when you play dice? The cup, it appears to be a cup, but not at all. The cup is the preceding throw. That is the cup. The preceding throws determine certain conditions on the basis of which, on the basis of which a new casting takes place. The new casting is random, yes, but within a mixture of chance and necessity, within a mixture of chance and dependency. It depends on conditions fixed by the preceding throws, so that I have a Markov chain. Ah.[4]
I would say: from one diagram to another, there is a re-drawing and redistribution following the second drawing. So let us say that from primitive societies to imperial societies, from imperial societies to ancient or archaic societies, from archaic societies to the Greek city-state, from the Greek city-state to the Roman world, from Rome to feudalism, from feudalism to the societies of sovereignty, from the societies of sovereignty to disciplinary societies, etc., what must be conceived is a multiplicity of re-drawings where the state of forces, the singular points, the affects, are redistributed at each level… at each diagrammatic state. You understand? This would be obviously a way of restoring universal history, which would horrify Foucault, thus it must be said at the same time: no, this is not universal history since you make constant appeal to mutations and re-drawings. Everything is re-drawn [retiré]. Constantly, everything is re-drawn. It is only in the dreary periods, in the periods when we are beaten down, that we no longer believe that everything is re-drawn. Yet there always come the dawns, when one perceives that a new roll of the dice is possible. Which is to say: a new roll of the dice is possible, that is, thought, thought again becomes possible.
Meanwhile, what must be done, between two moments when it is quite difficult to think? There is but one thing: bow one’s back. Bow one’s back beneath the iron hand of necessity and wait for the iron hand of necessity to shake the cup of chance; good, that’s a matter of practical work. In any case, each time that you think, you rattle a kind of cup, and then what is it…, what is it that makes it so that sometimes it works and sometimes it doesn’t? Surely the force [puissance] of the re-linkages, with what it is re-linked together… this is complicated, but it is no longer our problem [ce n’est plus notre problème]. [Interruption of the recording] [1:38:41]
… This is a… he’s very important… in research on new pedagogical methods, and he faced the problem of schizophrenic children, of autistic children, and he conducted some extraordinary studies, which are not in fact studies but a kind of practice where he followed these kinds of children along their trajectories.[5] He followed them not at all by hiding…. And in order to understand what this space was… and it is very difficult to understand… Whether it be an animal, a man or a patient, it is impossible, I believe, to understand whatever it is if one does not follow it, and this is the limitation of psychoanalysis, namely, that from the couch it is difficult to understand, whatever the case may be… But the trajectories of a schizophrenic child, their stoppages, the lines traced in space, the customary paths – by “customary paths” I mean the paths that the child ordinarily takes – all this, he treated in an extraordinary manner. And here, too, just as I was saying that Pierre Rivière calls for cinema, even more so, the trajectories of a schizophrenic child call for a cinematographic transcription, which is indispensable for understanding these trajectories, these markers in space… Now, this is not a theoretical study because it must be supported by the child’s spatiotemporal markers in order to enter into the least degree of communication with the child. And with regard to this, the studies of [Fernand] Deligny, in my opinion, have nothing anywhere that would be their equal. It is very beautiful. Good.
So, I have arrived at the conclusions of this second part. You see: my first part on power was a very general discussion of postulates. My second part on power, where we have arrived at the end, was: what are the differences between power and knowledge? And these differences, at the heart of this second part, you see that we have gone over them using several terms. Power is strategy, by contrast to strata; it is the diagram, by contrast to the archive; it is microphysics, by contrast to macrophysics or molar physics. Good, next we could add…. I would nearly add the forms of knowledge… Forms are always the forms of knowledge, the forms of the archive. Everything that is formed is already archived, a matter of strata… Well, the forms are subject to history. Forces are instead subject to a mutation, subject to the diagram; I would say, for the sake of convenience, that forces are taken up in a becoming. And just as we must distinguish power and knowledge, we must distinguish the becoming of forces and the history of forms. History is morphological. Good.
This amounts to saying what? Well, once again, regarding this whole domain of power, if I multiply its characteristics, then I would say: it is informal, so that whereas knowledge is always formed, the diagram is informal, going from one point to another, a many-pointed system and not a formal one; it is non-stratified, the non-stratified matter invoked by [Herman] Melville[6] — “and we go from stratum to stratum in the hope of finding the non-stratified element” –, non-stratified and unstable; it is diffuse; it is in perpetual mutation; it is abstract and yet is not general — you see that this is the only way I can mark abstraction and, at the same time, variability– it is abstract without being general, it varies, in fact, and is not general because it varies with the coordinates of space-time, as we saw.
And if I wanted to try to summarize everything, I would say: in a certain manner, but one which we will have to try to understand, it is virtual, it is virtual. But just as I said that “it is abstract without being general,” I would say: it is virtual without being irreal. It is virtual without being fictional. Why is it virtual? Well, because it is made up of small emergences, micro, small emergences and vanishings. At every moment it reworks itself. That which materialized in it vanishes in favor of something else, in favor of another relation of forces. Abstract without being general, virtual without being fictional, virtual without being irreal. Indeed, we will make a great deal of the distinction which consists in recalling this: the virtual is opposed not to the real, but to the actual. There is a reality to the virtual. The virtual is real; it is un-actual, but it is real. What is opposed to the real is not the virtual but the possible. Possible is opposed to real, yes. But virtual is not opposed to the real; it is opposed to the actual.[7]
So, I can very well say of the diagram that it is, once again, informal, non-stratified, unstable, diffuse, many-pointed, abstract without being general, virtual without being fictional or irreal. Last characteristic: it is… it all revolves around the same thing, it never ceases to be made the object of re-linked drawings [tirages réenchaînés], Markov chains… [Interruption of the recording] [1:46:48]
Part 4
… azert, when he said: “when I say ‘azert,’ I state something [j’énonce]; it is the statement of the series of the succession of letters on the keyboard of a French typewriter, A Z E R T.”[8] If I say “azert,” I am stating the succession of letters on the keyboard of a French typewriter, that is a statement. He said: but these same letters are not a statement. Why? It is something other, he said, than a statement, and yet, he added, something that is nearly the same. It is nearly a statement, and it is not one. You remember.
Now this should become illuminated, this text by Foucault, for we ought to distinguish, in effect, two levels. First level: a roll of the dice, which is not simply random, but it is normal, since we saw that, in a Markov chain, there is no pure chance. It is not pure chance, it is a casting of singularities, and each letter being a singularity, a singular point, azert is a casting of singularities, following what? We have seen, not following pure chance…. I could throw out a casting of letters following pure chance, if I said all of a sudden, “b, e, k, p.” In this case, I have just cast something randomly. So, one might ask me to do this 20 times. However, it would have to be difference persons than me because my own first casting will have already predetermined me. So, someone else makes a different casting, good, all that… I would make a list of random castings of letters. A Z E R T is not a random casting, it is a casting that already follows the statistical laws of frequency in the French language, but not purely: the laws of frequency paired with laws concerning the relations among fingers, digital relations.
Thus: a mix of relations of literal frequency and relations of digital dynamics. It is complex. Here, I could say, this is typical of a mixture of necessity and chance. Azert. Good, you have thrown a casting of singularities, but this is not yet a statement. However, you say “azert,” you state something. In that case, it is a statement, why? Foucault tells us: it is like a curve…, you have traced…, it is as if you had traced the curve passing through the neighborhood of singular points.
And I told you: in the theory of equations – it does not matter whether one is familiar with it or not, that is not the question, so those who know nothing will understand … should understand what I am saying. In the theory of differential equations in mathematics, you have the distinction between two levels of mathematical existence. What are the two levels of mathematical existence? Distribution of singular points. The singular points, in mathematics, are, for examples, vertices, nodes, nodes of a curve, focal points of a curve, points of tangency [points de rebroussement]. Singularity is itself a mathematical notion. So, you have a first domain of existence. The distribution of singularities in a vector field. It is amazing because, believe me, you do not need to know a single mathematical word in order to understand this. You would have to know if you wanted to do mathematics, but to understand it with regard to concepts, you do not need to. Thus: the distribution of singularities, there, I cast singularities within a vector field. I throw my points. That is one domain of existence in mathematics.
Second level of existence: you trace the curve that passes through these singularities, or rather, that passes through the neighborhood of these singularities; now, this is what you call … what is this? It is no longer at all…, this no longer refers simply to the vector field, but presupposes a… what we will call an integration. The curves are integral curves. You thus have: (1) the domain of singularities distributed in the vector field; and (2) the trace of integral curves that pass through the neighborhood of the singularities. And of course, in mathematics, the two are indissociable. But the fact that they are indissociable does not prevent them from differing in kind. These are two distinct operations. Good.
And, well, understand that this is exactly the power-knowledge problem. Power is the casting of singularities, the distribution of singularities, as we have seen. The strata, knowledge, are the trace of curves that pass through the neighborhood. Good, you grant me, therefore, that the following goes without saying: the two are inseparable, and yet they are distinct. The two are inseparable because the singular points remain indeterminate, which is to say: we do not know if they are either points of tangency or inflection points, or nodes or focal points. You know their existence, but you ignore their nature insofar as you do not have the integral curves that pass through the neighborhood, which alone will say what their nature is. Thus, the casting of singularities is not separable from the integral curves. Inversely, the integral curves are inseparable from the pure distribution of singularities in the vector field. And well, it is the same thing. The relations of forces are actualized in the relations of forms that constitute knowledge. You understand? It is a question of actualization. Therefore, if I say, very roughly, that power is actualized in knowledge, strategics is actualized in the strata, then the question that remains for you, in this third part, is: what does it mean “to be actualized”?
And well, to be actualized means two things. I will say it right away… I will say it right away so that you will follow better. To be actualized means, on the one hand, to be integrated, and on the other hand, it means to be differentiated. What is actualized? What is actualized is a virtuality insofar as it is real but not actual. What does a virtuality do when it is actualized? It does two things, it is integrated, and it is differentiated, both at once. If one understands that, then one will no longer have problems regarding the power-knowledge relation. To be actualized is to be integrated and differentiated.
Hence the first point. When relations of forces are actualized, those which are vanishing, unstable, etc., what does this mean? Well, it means that they have a fluid materiality, non-formed matter, and that they have diffuse functions, non-formalized functions. To be actualized is to fix matter, therefore, to form it, and this is to finalize the function, therefore to formalize it. All I am saying is that in order to be actualized, and actualized well, the relations of forces must be integrated. So, this idea is at once very complicated and very simple. So that you can choose: either the simpler or the more complicated. Either the simpler: it is very good, very clear.
Foucault tells us: you know, we must never begin with institutions in order to understand the social field, because institutions are integrations. These are integral curves, curves of integration. These are the curves of integration of relations of forces. Relations of forces, which are micro-powers, are integrated in large institutions. The institutions are molar. The molar institutions, that is, institutions as a whole are the integrations of the relations of forces of this entire microphysics of power.
And that is why it was necessary to not begin with the State, nor with any other institution, in order to understand power. It is because the State and plenty of other things are processes of integration. Integration of what? Well, of relations of forces that will be differential relations. It is simple, that is what is meant by “microphysics of power”. As he says: the State is a terminal state, a terminal form. Institutions are forms, they are forms that integrate the micro-relations of power. So that if it was necessary to designate by a name the relations of power, the relations of forces, then the name “State” would be not at all suitable. What would be the suitable name? He says it in an interview published by [Hubert L.] Dreyfus and [Paul] Rabinow in their book on Foucault.[9]
There, in this interview, Foucault says very well: we must resuscitate the old word “government” by taking it in the most general sense it has ever had in order to designate, by the letter, all the relations of forces, whatever they be. The shepherd governs the flock. The nurse governs children, on calls this a governor. Government is force, and all force will be said to be governing that imposes a task on some multiplicity. That is, government is a matter of micro-power. Every relation of forces, such as a force imposing a task on a multiplicity of other forces distributed in a confined space, is a government. One can be a governor of children, a governor of citizens… all that amounts to the same. Government precedes the State. Since the government adequately refers to relations of forces, whereas the State is only an incarnation. You see: government expresses molecular relations that constitute power. Whereas the State and other institutions are the integral curves that actualize these relations of forces.
As a result, Foucault will insist strongly in saying: but the State, you know, there is no essence of the State; why would people ask themselves what the essence of the State is? The State is a process, there is no State, but a process of state-ification [étatisation]. And here again, this remark by Foucault must be generalized, for the same thing will be said about the family. The micro-physics of power is not an operation that would consist in searching for the secret of large ensembles in small ensembles. One is not doing microphysics if one explains the State by the family. Micro-physics makes appeal to a difference in kind between what is instituted, the domain of what is instituted, and the relations of another nature that are presupposed by all that is instituted.
As a result, the family is also a molar institution. It must be said: there is no family, there is a process of familialization that varies and that is defined how? By that which it integrates, some sort of relations of power. It integrates such-and-such affects, such-and-such singularities in a social field. The State will integrate others and, according to the social fields, it depends: things that were in another field, integrated by the family, may very well be able to be integrated by the State. It may well be that one institution outweighs others; and this varies in all periods, hence: the strata are perpetually reworked according to the relations of forces that they incarnate in themselves. If you take a given period. Good.
For example, in the sixteenth century, you will have to ask yourselves… an institution like the family, what kind of relations of forces does it integrate? And the State, what kind? And the church? And today you will not have the same responses. Things that were, in fact, as one always says, things that were a family affair have passed over to the State, let us suppose, and that simply means that the processes of familialization, of ecclesia-…, of ecclesia-…, anyway, you see what it is, of pasteurization [Laughter], the processes of pastoralization, the processes of state-ification vary according to the strata under consideration. But whatever the stratum under consideration, you do not have families, nor even a structure… – you see the extent to which he was never structuralist – you have processes of familialization, you have processes of state-ification, these are exactly integrations, … not exactly, it is the equivalent, if you like, of what mathematicians call an integration.
But then, what will be very interesting in Foucault, in our reading of Foucault, is that, I would say: institutions are exactly molar agencies [instances] that actualize, that integrate molecular relations of forces. So, what is interesting is to follow history… I was saying: there is a becoming of forces, but there is a history of forms, therefore there is a history of institutions. It would be necessary to conduct a history of each large molar agency that integrates, in a manner varying greatly according to the period, relations of forces, and which relations of forces. You understand, this opens up an entire enormous field to history. What are these molar agencies? I would say, take the case of a molar agency, the family, say: the father. The integration as institutional process takes place according to a giant molar agency: the father for the family, right? [Pause]
For politics, it might be the sovereign, the king: giant molar agency. In societies of sovereignty the relations of force are politically integrated around the person of the sovereign. Juridical integration takes place around the molar agency: the law. Therefore, it should surprise you even less. You have seen, in our first part on power, we saw the critique of… the great critique of law in Foucault. When he told us: but, you know, law is only ever a result of illegalisms, and he proposed to us this curious notion of illegalism. Illegalism taken at the microphysical level. You see what he meant: law is a form of integration. It is a form of actualization. Why yes, it is a process of actualization. I could also say that there is a giant molar economic agency, since economy, politics, familial, etc. are distinguished at the molar level, at the level of forms. I could say: there is an economic agency, there is a giant economic agency, which is what? Well, let us say it is money. Good, I can get all that from the molar agencies which themselves have a history. History of money through the social fields. History of the father through the social fields. Good.[10]
And for sexuality, is there a molar agency? Well, yes, there is a molar agency. The molar agency around which all the micro-sexual, micro-sexed relations are actualized and integrated is what one calls “sex” [le sexe].[11] Sex. Alright then, whence one of the fundamental theses of Foucault’s History of Sexuality, Volume 1, when he begins his history of sexuality: it is to align himself… It is the entire end of the book that is splendid, a sexuality without sex. You will understand nothing about sexuality if you do not extricate it from this giant molar agency that only actualizes it or actualizes its effects, namely: sex. But a micro-physics of sexuality should ignore this molar agency. But what is a sexuality without sex? Now would be a good moment to re-establish both the difference and the complementarity of the molar and the molecular. Of micro- and macro-physics. So, we must not let this opportunity slip away because what is… what is it that he can call sexuality without sex? One senses that it is important to him; that, for him, it is a very bad way of conceiving sexuality to introduce into it… to start from the notion of sex. It doesn’t work like that. Yes, yes, yes, sexuality is integrated in sex, but it is not at all a good idea to conflate it with this agency … What a settling of scores with psychoanalysis this is! This is the end of History of Sexuality, Volume 1.
So, we will try, since he left us a bit… here we must take some risks… happily, they are not large risks that we are taking. What could Foucault have had in mind by this history of molecular sexuality or of sexuality without sex? For this expression is not one that I am making up… the appeal to a sexuality without sex is the object of the entire end of History of Sexuality, Volume 1. Well, I tell myself, he was thinking of something very specific. Well, he was thinking of something quite precise… what I mean is that it is a nod to a great author who drew this out wonderfully, namely, Proust. I don’t mean that he was inspired by Proust; I mean he encountered him.
For I am going to recount for you… Because I have the impression that this is obviously not well understood. What is the very curious conception of sexuality that Proust developed? And you will see in what sense this completely opens up our understanding of this molecular sexuality–molar sex problem. When I say a girl–a boy, I designate…, I speak in molar terms; I nearly said that I speak in statistical terms. There are, statistically speaking, girls and boys, but a girl is a statistical creature, and a guy is a statistical creature. Yes. Well, I could stop there, by the way.
But Proust in Sodom and Gomorrah… – it’s not by accident that the book takes this title – he is going to explain that there are three levels. Only, he explains it in a way that is so beautiful, so beautiful that we cannot manage to follow it. It is too beautiful. There are three levels. There is a level of large molar ensembles. And this might be a re-reading of all of In Search of Lost Time. There are the large molar ensembles. What are these? In a social field, I would say that the large molar ensemble is the set of heterosexual loves distributed within this field. That is Proust’s idea, eh. It is a statistical set. And if you read all of In Search of Lost Time and all of the loves… you will see that the heterosexual loves are presented in this form. A kind of first layer, a set that traverses the entire social field.
But this set contains so many accidents that one says to oneself… It is so difficult for all that to work normally, this set of heterosexual loves, that one says to oneself: this is fishy, there is something else underneath. And what is there beneath this horizontal ensemble? This horizontal ensemble, in music one would say a melodic ensemble… What is there beneath this horizontal ensemble of heterosexual loves that are distributed in a society, according to Proust? There is a discovery that makes one quake in terror: two vertical homosexual series. And beneath the fabric of relationships where the man relates to the woman and the woman to the man, there are two fragmented [en pointillés] series, two vertical series, one according to which the man relates only to the man and the other according to which the woman relates only to the woman; these are the two homosexual series. The one is called Sodom, for the man, and the other is called Gomorrah, for the woman. And these two homosexual series, independent of each other, face to face with one another, fall under the abominable prophecy: each of the sexes will die individually. And this is the domain of shame and guilt, or in any case the domain of the secret. All that is what Proust says.
And oftentimes, we readers stop ourselves there. …Because we always love for our great authors to be authors of guilt… So then, one says to oneself: ah good, so yes, there is nothing else beyond this. But in the end, for Proust, that’s not the case, because although guilt exists, he has his affair with a certain ease. And in fact, there is a third domain. And how… it would have to be called… it is neither horizontal nor vertical… We must name it: it is transversal. It is transversal. So, today, I would like for us to end on this point, so that you are able to reflect.
What is this new domain? This third domain? I am being schematic because… [Deleuze goes to the board] this third domain… Here you go: I will have to draw the following table, in order to understand. [He draws on the board] Here you have a global man, you see, global or molar. I will call him M1. And Proust tells us: you know, he is globally a man, by the predominance of one pole, but he has the two sexes. He has the two sexes… He is predominantly m, but he has m and w. [Pause] Are you following me?
Quite simply, that is the misery of our condition, that we have the two sexes, but that they are divided. This is a fundamental idea in Proust: everything is always divided. Which is to say, this is never communicated [ça ne communique pas]. Never in Proust. In Proust, things are always boxed away. And this is never communicated. You must open the box and pull out what is there, and it is always like some kinds of demons, from a Japanese garden that are.. that are pulled from a box. There you have it.
So, I have my man, M1. He has the two divided series. In other words, he is not going to be able to settle on his own. There is an animal like that. We are snails. The snail is hermaphroditic. But it cannot fertilize itself all alone. It must go seek out another snail who is also hermaphroditic. It is marvelous. There is no reason to be jealous of snails. [Laughter] You see, there is already the virtual here, but in order to be actualized, it must pass by way of the other. The snail is a hermaphrodite, yes, but a virtual one. However, it is real, but it’s not actual. It is real: it has two sexes. But it is not actual. It can only be actualized if it enters into relations with another snail. That is what we will call “the knowledge of snails”. Well. … I can mark down… this is “woman 1,” and she, too, has two sexes. With the predominance of the feminine pole. She is woman-man, but she is globally woman. Molarly. You see?
But you sense well enough that, in my…, this does not suffice. If I want to make a true combination of all the possible situations, I cannot do it for two [à deux]. I need four terms at a minimum. In fact, why is a man necessarily united with a women; and a woman, necessarily with a man? It is possible, it may be considered reprehensible, but it is possible that a man be united with a man and a woman with a woman. I need to have another man and another woman. I need an M2, who himself is both man and woman, and a W2, who herself is both woman and man. You understand? Good. I have the minimum molecular basis.
Let us try the combinations. The molecular combinatorics. Sense that, insofar as I remain at M1, M2, W1, W2, I am in the domain of sex [je suis dans le sexe]. Already – perhaps it is only the beginning – when I am in m, w, with the possible combinatorics, I have passed into the molecular, which is to say: it differs in kind. Let us try the combinatorics. I start from M1, and I say: M1m, you see? M1m – that is, man considered with respect to his man pole, if you are following me – can enter into relation with what? He can enter into relation with… I will start there: with M2m or with M2w. Ah. Globally, I would say: this is homosexuality. Molecularly, it’s not the same. Proust attaches enormous importance to this. Good. That is all for that relation there.
A student: W1
Deleuze: What?
The student: W1
Deleuze: W1? W there?
The student: Not there, W1.
Deleuze: Ah, okay! So, he can re-enter into relation with W1w; he can re-enter into relation with W1m. I have not taken account of W2, since it will not present a new combinatorics, a new term. I have four of them. On the other hand, I can consider, now,… I am never going to find it. I didn’t make my graph sufficiently [wide] …You can consider: M1w. So, this can re-enter into relation… But it is already necessary for me to beware, you understand; if I hesitate, it is because I already have cases that are taken. So: M1w can enter into relation with: M2m. You see, do I have it? Yes. I have it there. And can this enter into relation with M2w, no, it is not a relation. It can enter into relation with W1w, that is fine, I don’t have that.
A student: Yes, yes you do.
Deleuze: I cannot have it: I have W1w with M1m. Eh? Not W1w — Ah, I sense that I can’t do this, I’m going to snap [Laughter] — And then he can enter into relation with… W1m. W1m, I do not have it either. W1m [Pause] … And then, after 2, 4, 6… afterward I should have 7 combinations. So, good, that works… Oh yes! Oh yes! And then I have [Pause]… A seventh combination, this should be: W1… so now, and two w… W1… oh la la. We are too tired. W1w and W2w. I don’t have that? And then W1w and W2m. Do I not have it? So: 4, 6, 8…
A student: There is one too many.
Deleuze: There is one too many… it seems to me… Wait, ok? [A student makes a suggestion] Listen, I will research it, I cannot go on any longer. It’s the one up high that is wrong? W1w, W2w…
Comments by several students at the same time: [Inaudible]
Deleuze: Ah no, that works. What?
A student: Seventh. It is W1m [Inaudible comment]
Deleuze: Can you repeat?
The student: You did W two times, not M1w with W1m.
Deleuze: Ah, I did M1w two times with W1m.
The student: On the board that should be the sixth.
Deleuze: 2, 4, 5, 6. [Pause] No.
The student: … seventh
Deleuze: M1w… Ah listen, I no longer understand. [Pause] I find myself in the grotesque situation of the mathematics professor who messes up during his demonstration. [Laughter] So, I will come back to this next week, I will have the time… but you see the principle.
I am asking: why is this so important in Proust? Because… only… this is the only way to explain the way he multiplies the kinds of homosexuality, in order to show that…, and his entire theme is that there are combinations such that the homosexual can expect from a woman…, the male homosexual, for example, can expect from a woman what the man would give to him, and inversely. There are homosexual women who can expect from a man what they would expect from a woman. So that the two homosexual series crumble completely. It is in this sense that I say: he completely pulverizes guilt, it is there on a second level, but at the third level, it bursts apart on all sides. And it is obviously the third level that is interesting, which is to say, that is his great idea, his great theme of a sexuality that is vegetal and thus innocent, a vegetal and innocent sexuality that will go beyond the series of guilt; this is why it strikes me as so annoying when one stops at the two great series of Sodom and Gomorrah, which are still statistical series.
And then, at this level, I was saying – it is only for this reason that I was developing this theme – at this level, understand, it is a question of a sexuality without sex. A sexuality without sex, then, presents itself how? With only, only the following elements [données], the following variables: the body and its pleasures. And the expression “the body and pleasures” occurs constantly in Proust, and when Foucault will speak of a sexuality without sex, we must not be surprised to commonly find, under his pen, the same terms: “the body and pleasures.” And if Foucault, I believe, refuses the notion of desire, it is precisely because he cares about this idea of the body and its pleasures as being the sole direct expression of a sexuality without sex. You see that, here, we come to a molecular sexuality, whereas the two other levels were at the statistical or molar or global level. Moreover, Proust will go as far as to say: local pleasures; local is opposed to global in every sense of the term…There you have it.
Listen, remember where we are because, I believe, we can no longer keep going, or in any case, I cannot. [End of the recording] [2:28:26]
Notes
[1] On these filmmakers and different aspects of their work, see sessions 21 and 24 of Cinema seminar IV, May 7 and 28, 1985.
[2] See Critique 229 (June 1966), Dits et Écrits I, pp. 518-539, and “Maurice Blanchot: The Thought from Outside”, trans. Brian Massumi, in Foucault/Blanchot (New York: Zone Books, 1987).
[3] Reference to the poem by Stéphane Mallarmé, “Un coup de dés n’abolira jamais le hasard” (A dice toss will never eliminate chance) (1897).
[4] On the relinkings and dice throws, see Foucault, pp. 116-118.
[5] The reference throughout this paragraph is to the researcher named near the paragraph’s end, Fernand Deligny.
[6] This is reference to Melville’s Pierre, or The Ambiguities; see the previous session, 10, January 14, 1986.
[7] On these distinctions, see above all Difference and Repetition, pp. 208-209.
[8] Deleuze develops this example in earlier sessions of this seminar, notably session 5, November 19, 1985, at the start of session 8, December 17, 1985, and will return to it in sessions 12 and 13.
[9] Michel Foucault, Beyond Structuralism and Hermeneutics (Chicago: University of Chicago Press, 1982, 1983).
[10] Much of this development, and what follows on sexuality, is succinctly indicated in Foucault, pp. 74-76.
[11] In this translation, the English “sex” is retained to translate the French le sexe so as to preserve consistency with the English translation of Foucault’s History of Sexuality, Volume 1. However, the semantic range of the French word should be kept in mind: le sexe also, and quite relevantly, refers to genitalia, including specifically the penis, as well as to “gender” more generally.
For archival purposes, the transcription of this seminar by Annabelle Dufourcq took place starting in 2011 with the support of a Purdue University College of Liberal Arts grant. The translation was completed in 2018, and subsequent revision of the transcript and translation and time stamp were completed in January 2022, with additional transcript and translation revisions in May 2022 and also in February 2024.