January 7, 1986

For Foucault, there are only ever practices. Still, the two practices, that of knowledge and that of power, are irreducible. So that power cannot be known. And yet there is a reciprocal presupposition, or at least, power will be known indirectly. It will be known indirectly, that is, it will be known in relations of knowledge. It is knowledge that provides us with a knowledge of power.

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

Edited

 

Part II of the seminar focuses on power, the second axis of Foucault’s thought, starting with Deleuze’s review as well as with key terms: historical formations or strata of statements and visibilities; the intertwining of the latter, also as language and light, yet remaining heterogenous, constantly battling, occurring via a “non-relation”, but a relation between the forms occurring due to the dimension of power. While knowledge is formal and stratified, power is informal and unstratified, with a boundary like a curve regularizing relations between singular points, with knowledge relations linked to power relations, the latter thus understood as relations of forces. Deleuze notes three aspects of these relations and their heterogeneity and joins Foucault to Kant through heterogeneity and reciprocal presupposition, but while quickly observing that Foucault needed to add a third axis (part III of the seminar), he remains focused on the first two axes. With no systematic account of the principles of power in Foucault, Deleuze pursues these principles by understanding power as practice, which is how, Deleuze says, Foucault approached all historical problems, linking his published works to his interviews. Situating Discipline and Punish within ongoing European leftist debates of the 1960s-70s, and to Foucault’s activism in the GIP (Prisons Information Group), Deleuze considers three practical problems: the new types of struggles (concerning power), the role of the intellectual (concerning knowledge), and the meaning of “to be subject”, with Deleuze selecting different examples of ruptures (e.g., in Yugoslavia, in Italy, and France). Besides the transversal type of engagement, Foucault also proposed the “specific intellectual”, influenced by American types of “new subjectivities”, the subject defined as a small incarnation of singularities. Hence, the three practices (struggles, intellectuals, subjectivities) form the linkage between historical formations and the here-now. Deleuze then seeks principles, first being: do not seek large aggregates, but rather microphysics, corpuscles in contrast to the wave, distinctions that Foucault links to Durkheim versus Gabriel Tarde (Foucault tending toward the latter). After outlining Foucault’s denunciation of six postulates of classical theories of power, Deleuze leaves the last one for the next session, the linkage made between the State and law.

 

Deleuze at Place Vendôme in Paris on Monday, 17 January 1972, during a demonstration for prisoner’s rights, with Michel Foucault and Jean-Paul Sartre in the background.
Photograph by Élie Kagen.

 

Gilles Deleuze

Seminar on Foucault, 1985-1986

Part II: Power

Lecture 09, 07 January 1986

Transcribed by Annabelle Dufourcq; time stamp and revisions, Charles J. Stivale

Translated by Christopher Penfield; subsequent revisions, Charles J. Stivale

 

Part 1

… Many of you followed his courses at one time or another and were affected by him and loved him. I believe the tribute to offer him, the kind we can offer, is to reread his books in order to evaluate their importance, starting from The Birth of History, because I believe he truly created a major body of work.[1]

Thus, moving forward, we begin the second part of this study on Foucault. The second part is the second axis of his thought, and this second axis concerns power. It was necessitated by the first axis, which concerned knowledge. All of last quarter, we were led in effect to see how the domain of knowledge demanded, under very precise conditions, a response that had to come from elsewhere. And we rightly sensed that this response, which had to come from elsewhere, could doubtless only come from an analytic of power. Only from an analysis of power relations. I will try to summarize once again what we learned from the previous analysis of knowledge.

The first point is that historical formations appear as strata, as stratified formations. We will see, perhaps, that in relation to the problem of power, this notion of stratum or stratification, such as it appeared very rapidly at the beginning of the Archaeology of Knowledge, takes on a new importance in relation to the problem of power. We must feel right away that the question will be: is power itself stratified? But we are not there yet, after all. How do these stratified formations present themselves? They appear as veritable sedimented layers. Layers of what? Layers of seeing and speaking. Spoken words accumulate, visibilities accumulate: layers of seeing and speaking.

Second point: these layers refer to two forms, seeing and speaking, but, more precisely, the form of the visible and the form of the statable.[2] And each stratified formation is composed of the intertwining of these two forms: the visible and the statable, or their formal conditions, light and language.

Third point: although these two forms intertwine to constitute stratified formations, there is heterogeneity between them. These are two irreducible forms, without common measure. The visible is not the statable. Speaking is not seeing. So, the intertwining of these two forms is a real battle, and can only be conceived as grappling [étreinte], hand-to-hand combat, battle. And finally, in a practical sense, isn’t this what interests Foucault and explains much of his style? That is, everything transpires as though for him, it was something like a matter of hearing cries beneath the visible and, inversely, of extracting visible scenes from words. Flashes of lightning beneath words, cries beneath the visible, a perpetual grappling of the two. This is what we saw in [Raymond] Roussel: lightning flashes that escape words. In [Jean-Pierre] Brisset as well – another unusual author, whom Foucault commented upon more briefly than he did Roussel –, he will seek cries beneath the words.

I’ll open here a very brief parenthesis because I have said little about Brisset and Foucault’s text on him, but Brisset’s book on language contains very curious operations, which it would be wrong to take as exercises of fanciful etymology.[3] There is a beautiful passage by Brisset on the word “saloperie” [“muck”]. And how does Brisset proceed? He says: well, what is “saloperie”? It is sale [dirty], unclean, eau [water], because water is the universal origin out of which frogs come, and we are all frogs – that is Brisset’s great idea – “sale – eau – pris”, être pris [to be taken], meaning that the captives, in war, are put in a kind of wet field, thrown into dirty water. The captives are put in dirty water: sale – eau – pris; they are taken, pris, in dirty water, l’eau sale. There you have it.

You see the procedure? The word, if this were an etymology, would be a pitiful trick, but it is better than that: from the word, he will extract a visible scene, the captive who bathes there in a kind of pit of dirty water. Good. Thereupon, from this visible scene, he extracts a cry. The vanquishers, around this pit, normally cry out: “sale eau – pris” [“dirty water – taken”]. You see, they insult the captives. From this, a new return to a visible scene, sal – eau – pris. La salle [the room] (like the one we are in), aux [at a] prix [price]. In effect, the vanquishers are not content to just insult the captives by saying “sale – eau – pris”, that is, taken in the dirty water; further, they buy them to make them slaves. They buy them in what is thereupon a salle [room] aux prix [at a price]. I will stop myself here because, in Brisset, it never ends.

But in what way isn’t this an etymological exercise? You see, perpetually, he sets out from the word, extracts a visual scene, and adds sound effects to the visual scene. The sound effects, the first sound effects of the visual scene induce another visual scene. And he will evoke the sound effects for the second visual scene. It is a very interesting poetic procedure that produces Brisset’s most beautiful passages, and perpetually there is this kind of animated story jumping from a cry to a visual scene, a visibility, and from a visibility to a cry… This is, of course, why Foucault could not pass on Brisset. Thus, heterogeneity of two forms, which are perpetually in a relation of capture, of hand-to-hand grappling with each other.

However, fourth point: even if one says these are relations of battle, how is the hand-to-hand combat possible? How is the grappling possible, since the two forms are irreducible? And we have seen the reply at the level of the fourth point, which can ultimately be nothing other than this, namely: there must be a relation between these two forms without relation, the visible and the statable, light and language; there must be a relation between these two forms without relation, therefore the relation can only come from another dimension. It is another dimension that will bring out the relation in the non-relation of the two forms.

You see, — if I’m insisting here because this is going to important for us, even before we understand what this is — I have no choice, this other dimension must be informal and non-stratified, otherwise it would not respond to the problem. It would not be a response. This other dimension must differ from that of knowledge, and it must be distinguished from knowledge, among other things, by the following: it will not be stratified, and it will not be formal.

In other words, there will be no forms of power. I would almost say that we must understand abstractly before seeing concretely. And we saw in the last point, in fact, why and how knowledge was itself surpassed toward [se dépassait lui-même vers] another dimension. And that was the subject of our last meeting: how knowledge is surpassed by another dimension. And the response to which… and this was the analysis with which we ended, that of azert, A Z E R T. The analysis of this unusual example of azert, which was like Foucault’s own contribution alongside the exercises of Roussel and Brisset – these very curious pages of Foucault where he played around with azert, azert, saying: well, you ask for an example of a statement, I give you one, azert. Take that! Well, we saw where it led him. Namely, it led him to the idea that the boundary, the distinction to make, does not pass between the statement and what it designates, nor even between the statement and what it means/signifies.

What, then, does the boundary pass between? The boundary passes between the statement and what it embodies or actualizes. And what does it actualize? What is this boundary between the statement and what it actualizes? A statement is defined by its regularity, which is to say, it is the analogue of a curve. [Pause] But what does a curve do? It regularizes relations between singular points, between singularities. Azert regularizes relations between singular points, that is, the relations between the letters in the French language and the fingers. Between the frequency of letters, the proximity between the letters and the relations of fingers. Here, then, are relations between singularities. Azert is a statement as the curve passing through the neighborhood of these singularities. In other words, to what does the statement azert refer? It actualizes relations of forces. These force relations are the relations of forces between the letters and fingers, in the French language.

I would say precisely: this is how knowledge is surpassed by power. Why? Because power is a relation, and the power relation is strictly the same as a relation of forces or the relation of forces. And in Foucault, the power relation, in the singular, and relations of forces, always in the plural, are strictly synonymous. If knowledge is surpassed by power, it is to the extent that the relation between the two forms, the form of the visible and the form of the statable…, is surpassed by the relations of forces that it (the relation between the two forms) embodies. So, we have the abstract formula for the knowledge-power relation before concretely understanding what power is. And you see, then, the importance, for what lies ahead, of Foucault’s remark on the statement as an element of knowledge: the statement is always in relation to something else, although this other thing is distinguished infinitely or little from it. This amounts precisely to saying: relations of knowledge are fundamentally in relation with something else, which are relations of power, although the two – power relations and knowledge relations – are distinguished infinitely or little. The other thing that is nearly the same, but this is a “nearly”.

Hence our problem becomes: what is power? We already know Foucault’s fundamental answer: power is a relation. Just as knowledge is a relation of forms, power is a relation of forces. We will see. Foucault thinks that if we understood what force relations were, the concept of power would have to be radically altered. Now, you may say, rightly so: but fortunately, he is not the only one to have defined power as relations of forces. If there is originality to Foucault at this level, it must be sought in his concept of relations of forces.

There we are: I have tried to consolidate our gains. So, I will make a strong appeal: are there any problems with what happened in the first term, or no? All is well? Ah, there is another question: whether, in your reading of Foucault, you agree with the way the problem of knowledge has been presented, but we can hold onto that until the end of the year … I don’t know. Well, it is now that we would have to speak, if… if you would like to. You don’t wish to? Fine. Ah!? Yes?

A student: [Inaudible comment]

Deleuze: Speak up, I cannot hear anything.

The student: The term “discursive formations” has never had a place in your analyses. You speak of historical formations… [Inaudible comments]

Deleuze: It’s not false, it’s not false, yes, yes, yes. “Never” is perhaps a bit severe… Um. I think I mentioned it. I prefer …. I used instead the term “statement regime”. Yes. I will tell you why I believe…, why, in effect, you are completely right… It’s … but well, one always makes a choice with terminology. On the other hand, he very, very rarely uses the word “strata”, and I have accorded essential significance to “strata”. Uh… Clearly, these are the little choices that one makes.

But I have another reason for why I have, in effect, spoken very little of “discursive formation”. It’s that I fear ambiguity in this regard. By contrast, I suspect that Foucault much enjoyed the ambiguity in this regard. For “discursive” has a specific meaning in French and in philosophical terminology. It is a certain regime … of deduction. It is a certain deductive regime that defines a discourse. And he himself will take up the word “discourse,” for example, in the title L’Ordre du discours.[4] It goes without saying that he comes up with a completely new conception of discourse. And what constitutes the novelty of Foucault’s conception of the discursive? It’s that a discursive formation is a family of statements. So … It can be said at once that he makes the statement into a concept that is very, very paradoxical and completely new. Discursive, then, in Foucault is a certain way of getting to his conception of the enunciative.

As a result, for me, I had no reason to take back up the term “discursive” in Foucault, because it is not a term of malice. It is a kind of relatively neutral word, where one could insert many things and by means of which he comes to his own in particular, namely, statements, a family of statements, the statement regime. But each time that I have spoken of a statement regime, you could say “discursive formation”. And what I have tried to show is how a family of statements in Foucault – which was not at all defined by a resemblance of the statements with one another, but nearly by the possibility of prolonging the series of completely different singularities – was an even more original concept. So, in effect, my “discursive formations” are “a family of statements”. Yes, your remark is completely fair.

As in many other cases, it seems to me, I am led to make – well, to each their own reading, eh? — I am led to privilege certain terms, to barely cite other terms… I don’t know, I don’t know. This will surely again be the case for power. There is a passage in Foucault where he uses a word once that appeared to me so important and so illuminating for the whole of his theory, the word “diagram,” that I will insist enormously on the diagram, … even though Foucault uses the word only once, yet in an essential passage. Thus, everything also depends on … you know, …each of us are like that. If we are led to privilege certain terms in relation to others when we read, it is also to the extent that we accord to such and such a passage a decisive significance. A book is never homogeneous, eh? A book is made of strong moments [temps forts, ‘highlights’] and weak moments. The weak moments are sometimes brilliant. … I am speaking rhythmically, “weak” moments in the rhythmic sense.

It is clear, then, that two people who read a book with passion – it suffices that there be passion – I am not sure that the distribution of strong beats and weak beats will be the same in the two readings. So that the differences between readings, as you know, come well before problems of interpretation. When a book is rich and beautiful, the differences between readings occur already at the level of the rhythm of the reading. And though you may well read quietly to yourself, there is no reading that is not rhythmic. That is, before you have even understood what it is about, there are signals that reach you. And these signals are like little lamps that light up, and where you say to yourself: ‘ah! There is something important.’

And it is true that reading is entirely a breathing exercise, a rhythmic exercise before being an intellectual exercise. There you make … the criteria for choosing when to say: ‘ah! That is what’s essential.’ And it is not because the author italicized phrases … well, sometimes. Because when he uses italics, that means he is giving his own reading, that means he is in the process of reading himself and addresses himself to his reader, saying: this is what is important! Then you must listen, because the author still has a privileged point of view, it is he who knows, right? But you are often compelled to go beyond the italics. So, this is a question of the theme of rhythm, of the distribution of strong and weak beats, such that, again, interpretation follows from this rhythm. This is even how the reader participates in the creation of the author. It’s a bit like when…, you know, when you read philosophy or, even more so, when you read literature, it is very similar to when you listen to music. Strictly speaking, you do not hear music if you do not grasp the rhythm, or even sometimes something else. It is often said, and this seems obviously correct: one literally does not hear Mozart if one does not appreciate the distribution of accents. If you do not distinguish the accents, Mozart, ultimately, is a mediocre musician, a musician of the axis… Okay. So, you see, it’s not so bad that you can fail to literally perceive it. In literature and philosophy, it is also like this.

I am thinking of an author like Leibniz. Take a passage from Leibniz. But even before … in reading it, you cannot fail to ask yourself: but at what height is it situated? Like music. At what height? At what level? A thought always has several levels, always discloses itself on several levels. To read is to assign such and such a passage to this or that level.

So, in fact, I come across … I am returning to Foucault … “discursive formation,” if I come across “discursive formation,” I would place it on the lowest level. Not at all that it is a bad notion, but it is a tricky notion. I could very well conceive how someone else, on the contrary, would make it central. That would be a completely different distribution of rhythms and strong and weak beats. It is for this reason that what I truly aim to do is not at all to impose on you a reading, but only to propose one to you, so that you can then come up with your own. And understand that when that moment comes, …it will not do – you have not at all said this to me, you had such kindness and it showed that your question was perfect – it will not do to tell me, ‘Therefore you’re wrong.’ If you distribute the accents differently in your reading…. Obviously, there are untenable readings. There are always untenable readings. These are the readings that banalize, that transform what is new into cliché …. Just look at what imbeciles are saying about Foucault today …. So, at this point it must be said that these are not untenable readings, these are non-readings. They have never read, they do not know how to read … just as there are people who do not know how to hear music. I say it all the more cheerfully because I am one of them. It is a sense that you lack. What is irritating is to do a book on Foucault when you lack any reading whatsoever; that is unfortunate. But otherwise, all readings that are actually readings are good ones.

Let’s get going, then. Especially because in this story … no, I will still conclude these general remarks by saying the following: all that we found, in a certain manner – and here again, it is a question of rhythm – all that we found at the level of the axis of knowledge will be as though displaced at the level of the other axis, the axis of power, and will take other resonances. Namely, at the level of the knowledge axis, what did we find? We found three points that concerned the two forms of knowledge. To wit, firstly, there is a difference in kind or heterogeneity between the two forms. Secondly, this does not prevent there being reciprocal presupposition and mutual grappling. The visible presupposes the statable, and the statable presupposes the visible, the two perpetually hand in hand. And thirdly, this does not prevent, however, there being a primacy of one over the other, namely, a primacy of the statement over the visibility.

We must maintain these three tonic accents: heterogeneity, reciprocal presupposition, primacy. Primacy of one over the other. Good. This time around, we will rediscover these three themes, these three accents, at the level of power-knowledge relations. And it must be the case that, between power and knowledge, there is heterogeneity, difference in kind, and in a certain manner, non-relation. And it must also be the case that, at the same time, there is reciprocal presupposition – there is no knowledge without power and no power without knowledge. And it must be the case, again, that there is primacy of one over the other, [Pause] namely, that power is determining.

And, if there is heterogeneity, you see right away what that means, namely: power in itself is not known, is not an object of knowledge. However, in the first term, I drew a connection between Foucault and Kant precisely at the level of the irreducibility of the heterogeneity of the two forms, which in Kant were not the visible and the statable, but rather intuition and understanding. Here, I could equally make a connection with Kant, for Kant is doubtless the first to have posited a difference in kind or radical heterogeneity between two functions of reason. And he called these two functions of reason ‘practical reason’ and ‘theoretical reason,’ and the two were heterogenous, yet practical reason had primacy over theoretical reason and was determining. And the primacy of practical reason was bound to be a fundamental theme for Kant. Yet what did the heterogeneity of the two functions of reason, the practical function and the theoretical function, entail? That practical reason is not known and does not give rise to knowledge. Practical reason was determined by the moral law according to Kant, but the moral law was neither a matter of knowing nor an object of knowledge. There was nothing to know in the domain of practical reason. [Pause]

It is different in Foucault because the two, knowledge and power, both refer to practices. For Foucault, there are only ever practices. Still, the two practices, that of knowledge and that of power, are irreducible. So that power cannot be known. And yet there is a reciprocal presupposition, or at least, power will be known indirectly. It will be known indirectly, that is, it will be known in relations of knowledge. It is knowledge that provides us with a knowledge of power. Thus, all the themes we discovered in the relation between the two forms of knowledge are going to be surpassed following the other axis. As a result, if I were called on – you indeed see! [Deleuze goes to the board] – if I were called on to present Foucault’s thought, for the moment, I would do so as two axes, the knowledge axis and the power axis, with the displacement of a type of problem from one axis to the other.

What is going on there now? A third axis gets added. It will have been necessary for something in the distribution of the two axes to appear insufficient to Foucault – for a more or less urgent problem to become, for him, increasingly so – in order for him to add this third axis and, in the end, to reorganize his thought. But we are not yet there. Although … we are going to touch on it already… And what I would like is, for all this time that we are going be focused on the problem of power, that you be aware of the numerous occasions that suggest, even in a very indistinct manner, the necessity of a third axis. But for now, we will contend with the two axes, that is, this kind of excrescence on knowledge, the power axis, which thereby comes to intersect the axis of knowledge. This will be an increasingly three-dimensional thought as soon as he will have found the third axis. Good. Good, good, good.

So then, what is power? Today, I would mostly like to stay focused on what Foucault did not provide, namely, an account of the principles of power. And why did he not provide an account of principles? In a way, it is obvious. Because in a way, all his thought consists in saying that power does not have a principle, … and further, he wanted, and selected for himself in the writing of his books, the point of view of immanence, namely: power is taken up in relations of knowledge, thus it must be grasped immanently in relation to knowledge. But, as we have seen from the beginning, immanence does not prevent there being a difference in kind between power and knowledge. So that, for me, I would assert the other possibility and emphasize the difference in kind between power and knowledge. And at that point, I am entitled to try, on the basis of Foucault’s texts, to draw out some principles of power. Only, the sole task, for me, will be at every instant to not forget that this is all well and good, but it does not change the fact that power does not exist except in its relations of immanence with knowledge.

You see, I believe that Foucault, in his texts, privileges the relations of immanence with knowledge, but maintains nonetheless that there is a difference in kind between power and knowledge. As for me, in order to clarify Foucault’s thought and only for this purpose, I would like to do the opposite, to emphasize the difference in kind without forgetting in any case that there is immanence. Well, when I read, therefore, “principle of power,” what does that amount to saying? It is already…. Is “What is power?” a legitimate question? By this I mean, is power answerable to a question of the form “what is?”, once it has been said that power, like knowledge, is a practice? In other words, we must take seriously that power is something that is practiced. But you will tell me that this is also true of knowledge. Yes, okay, as we have seen. Knowledge is something that is practiced, it is to see and to speak, and nothing preexists seeing and speaking. Well, power, too, is something that is practiced. Only, these are two practices that differ in nature. It does not suffice to say that the question “What is power?” refers to a practice; the inspiration for the question must itself be practical.

What does it mean for the question’s inspiration itself to be practical? It means: what difference does today make? And here, we touch upon Foucault’s method: in a certain sense, Foucault never posed anything other than historical problems. Yet Foucault never posed a historical problem without centering his thought on the concerns of today, here and now. Why does he admire Kant? No doubt he admires Kant … for the whole of his philosophy, but he particularly admires Kant because Kant, according to Foucault, was doubtless one of the first philosophers to pose the question of the subject in the coordinates here-now, and because the transcendental or universal subject in Kant is inseparable from a subject here and now, which is to say: what difference is made by Kant’s present-day, by the age of the Enlightenment? And Foucault opposes Kant to Descartes in saying that Descartes remained at the level of a universal self, for the subject of the Cogito is any subject, whereas the Kantian subject is always a subject who says “me, in the age of Enlightenment.”

And how, for Foucault, is the historical problem fundamentally tied to the question “what difference does today make”? Precisely by the notion of practice. It is the notion of practice, it is practice itself that is the sole continuity of history up until now, up until the present. It is the chain of practices that is the sole historical continuity. Taking into account ruptures, mutations in practices, etc., it is the practical element that goes from the ancient past to now… [Interruption of the recording] [42:40]

Part 2

…and inversely. This allows me to respond to a question which greatly interests me, namely, in what sense does that which is not explicitly part of an oeuvre — this is the kind of question that Foucault posed very well, for example, and [Pierre] Klossowski too, apropos of Nietzsche — in what sense, for example, are Nietzsche’s letters part of his oeuvre? Are they part of it, or are they not? The question I would pose is: in what sense do Foucault’s interviews and discussions form part of his oeuvre? Say, in effect, that a problem was posed of a publication of interviews. What meaning would they have? It seems clear to me… The interviews always develop the “here-now” that corresponds to a book, the book bearing upon a historical period. As a result, at every moment, there is a strict correlation between the interviews and the books. And this is where I would like to begin, before even really beginning my analysis. This is what I would like to develop, the importance of these two questions: “how did things happen in a given period?”, “what difference does today make?”.

There will be continuity, but a kind of subterranean continuity. And only practice enables us to grasp this continuity, which is not a reflective continuity. This is why he will not speak of it in his books. But the topicality of his books arises quite clearly from the urgency of the here-now. So that the coherence of Foucault from the perspective of his life, of his oeuvre, appears to me very…, very clear. How so? Let’s take Discipline and Punish. Discipline and Punish, as we have seen, treats a short duration, from the mid-eighteenth to the early nineteenth century. What is it about? It’s about criminal law and the prison. Good. In parallel, what is Foucault concerned with in the interviews he gives at this time? The prison today. Practically speaking, Discipline and Punish can be considered as the book to which corresponds what practice?

During that period, around 1970, Foucault forms a so-called leftist group, which, I believe, had a great deal of importance because it was ultimately the only post-’68 formation. It seems to have functioned, or to have proposed functioning, in a very particular way; here, too, it was a question of practice, and this group was the “Prison Information Group” or GIP. I would say: it is the only leftist group that functioned, though perhaps I am exaggerating. In any case, it spread, since a psychiatry information group and emigrant information groups constituted themselves on the model of the GIP.  And at that time, there were a great many proliferations.

What is unfortunate, among those who speak today of May ’68, is that they turn it into a kind of intellectual phenomenon, forgetting that it was a global phenomenon and global practice, that May ’68 was the expression in France of something that happened or would happen in Italy, Japan, America, etc., and that we cannot even think this period if we fail to globalize it. Now, I would say that after ’68, what happens? Well, there was a reconstitution of relatively centralized groups. I remember: the Gauche prolétarienne [Proletarian Left] was very centralized; there were leaders there [Laughter]. leaders out for blood. What did Foucault do during this period? I speak of practice, but I believe that … it’s not a way of making a parenthesis, it’s that it is very difficult to understand, to live … to live today if you do not sense a bit what always remained for him a practical problem.

I have the feeling that, … there was first an evaluation for Foucault, namely, a practical evaluation: something is going to take place in the prisons. Good, it is a political evaluation. In my view, it is very difficult to understand anything in politics without making these evaluations: what is going to take place…, an impression that something is going to play out, take place in the prisons. You will tell me that it isn’t difficult, that there were movements, and yes, the movements had begun. But I will reprise the same point made about reading, that what is difficult is to say: well, yes, that is important, that, that is not going to fail. There is a great evaluation by Foucault, in saying: there is something there. As if, in the torpor following ’68, a hotbed was reignited, but now in the prisons. This had been preceded by the great prison movement in America. Foucault was working on the prisons, working theoretically on the prison, and while he appeared to be working theoretically on the prison from the eighteenth to the nineteenth century, he was acutely aware of the prison movement in America, … acutely aware of the case of George Jackson who, at that time, was of great importance, … and he sensed that something was about to happen in France.

And what is his idea for the formation of the GIP? Well, it is to form a non-centralized group. In this way, it is a descendent of ’68. How did ’68 present itself? As a non-centralized movement that wanted to be non-centralized. To what did it immediately give rise? A new kind of struggle. [Félix] Guattari, before ’68, had taken up – in what was already an interpretation of a new form of struggle that was emerging – the idea of transversal struggles, by opposition to centralized struggles of the classical kind, which is to say, centralized around the union, around the party. You see immediately that it was clearly a question of the Communist party and the CGT [General Confederation of Labor]. Transversal struggles: it seems that May ’68 was the bursting forth of a transversal network where struggles ceased being centralized. Thus, a new kind of struggle. We will see what importance this has when we return to theory. And how did the GIP take form? What is a transversal struggle, by opposition to centralized struggles? It is a struggle where there is no representative. No one is represented. No one can say “I represent this”.

You understand [Deleuze laughs], if you believe in philosophy, you will never think it is an intellectual affair when one speaks of a critique of representation today. One cannot critique representation if one is not attentive to the practice entailed by such a critique. And the practice entailed by such a critique is very simple; it means: “I will never speak for others,” “I will never take myself to be the representative for someone else”. Now, this was strangely new. Foucault, I believe …. There are people who do not have the right to critique representation because, when they critique representation, it is really lip-service, and they critique representation while claiming to represent something or someone. I would say that this is the academic critique of representation. There are only ever practical critiques. If I critique representation as a philosopher, I commit myself to something as well: not being part of any committee, it’s as simple as that. That is, never represent anything. Otherwise, it won’t go well. How do you wish, on the one hand, to say that “representation isn’t okay,” or to do philosophy that is not a philosophy of representation, and then unconcernedly continue representing? None of that can be taken seriously. I mean, if we pose the question of life and the oeuvre, it’s not that one must be coherent at all costs, but that one cannot have very good ideas, right?, if one doesn’t realize that one is still within representation in saying “down with representation”. Fine.

What did Foucault grasp? Well, he grasped something rather curious in favor of the prison movement. He grasped that people never ceased speaking for the prisoners. But that the prisoners themselves never spoke. You must remember…, though it isn’t ancient, things have actually changed quite a bit in not so many years. Because now it is no longer true, there are the small announcements in Libération, for example; there, the prisoners speak on their own behalf, good… But, you know, even after ’68, it was astounding, on T.V., there were always shows on the prisons, since it had already become a fashionable topic. And literally everyone did the talking, lawyers from the left, lawyers from the right, prison visitors (they were marvelous, the prison visitors, often these remarkable women), all that, judges, people in the street, the janitor and all that, anyone except for the prisoners. What’s more: except for former prisoners. So, for Foucault, that made him crack up. There was only one kind of person who did not have the right to speak about the prison, those who had been there or those who were there. Thus, the critique of representation would mean: prisoners do not need representatives in order to say what they have to say. Clearly, a group had to be organized.

How is it made? A transversal group, a non-centralized group. One starts with a working hypothesis: Foucault and others had devised a questionnaire. It wasn’t a joke, huh! — I mean … a practice, a practice… I am trying to say what a practice was for Foucault. — It wasn’t a joke because, as for the questionnaire, where were they going to place it? Where was he going to place his questionnaire? First, he makes a questionnaire, a series of hypotheses, but what mattered was that it be the prisoners who respond and who rework the questionnaire. Yet Foucault was not a prison visitor, he could not go to the prisons. So fine, that happened all on its own. First, how is a network constituted? A network isn’t secret, you know. One had to stand in line. One had to go stand in the lines at the prison gates, and then pass out the questionnaire to wives, to wives and to parents who were there, standing in line to see the prisoners. It’s not that easy because the people who were standing in line asked each other what this thing was, and then of course, the guards quickly understood… And as soon as they had a copy of the questionnaire, it started to go wrong. Good: a practice was constituted, and Foucault constituted his practice.

And he started, truly, in the lines outside the prison, in the procession of people who stood in line and who often, well, like anyone, they did not want a scene … one passed them the paper, they did not understand why … it is bizarre what was being asked of them, since one asked them ultimately to pass the paper to the prisoners or to respond themselves afterwards… etc. And little by little it worked. It required teams, you see, and it happened. It’s not at all centralized, it wasn’t at all pyramidal, it was one group there, another group there… transversal relation, it was the people who stood in line at the prisons.

Good, and then, the groups expanded. To wit, a third group: Foucault and the others joined the former inmates. Good, and that was a great addition, with various tensions and relations of power. All that wasn’t simple, each time it was necessary… There was, as I was just saying, the reaction of panic from certain people who stood in line at the prison gates because they did not want a scene. But, when the former prisoners were brought in, there were tensions among them, there were relations of power, which provided support already to Foucault’s claim: there are relations of power everywhere. There was one prisoner who arrived, he arrived first, good, thus he had prestige…. Foucault had his prisoner. It was… at the same time, it was a great parade, all that… There was the prisoner, exactly like the philosophy department had its worker. [Laughter] The prisoner, there was Foucault’s prisoner. And when a second prisoner arrived, the first was jealous. So, there were conversations between them that were very, very curious; they looked each other up and down, and then it didn’t go too well because one said “how many years did you do?”. The other said “I did eight years” – “where at?” – “At Melun”. And the other distrusted him because the first had done fifteen years and in harder jails. There was a question of the severity of the housing, the prisons, all that… So, it made for a turf battle [ça faisait des petits foyers].

Foucault knew very well how to negotiate all that, he arranged everything, good, but it was very tiring, it was… New transversal relation with a third group, the former inmates. And then… seek out lawyers, seek out magistrates… then, in the union of the judicial authorities, all that… The network … moved around, each time with relations of forces, relations of power, all that. Relations… good. And then, well, it had to extend outward [déborder]. At a certain point there were movements in neighboring countries… And then to go beyond each province. It was necessary to not reconstitute … a pyramidal organization, but how to do this? How does one ensure, for example, that the GIP of Lyon remains absolutely in charge of its initiatives? Um… Collect money? This made up a fourth group: who was going to give money? And pamphlets, making the pamphlets, all that… Right, and then especially, that it be the prisoners who speak: not easy to go looking for prisoners in prisons so that the prisoners speak. It’s not easy at all.

What is happening in this type of struggle? There was, in all that, a Foucauldian inspiration, which next led to the development of groups that investigated psychiatric asylums, psychiatric hospitals… And I believe, I would hypothesize, that one of the reasons for the silence and kind of dejected hopelessness that Foucault experienced much later on was … what might be called the failure of this movement. It was the failure of this movement, where “failure of this movement” refers to the fact that around 1971-1972, many people had the impression that something was going to change under so much pressure. That something was truly going to change in the regime of incarceration. And then, afterward, it was part of the post-’68 situation, when the lid was put back on, it was, it was… There were changes, one cannot say there were no changes, I believe there were many more than Foucault thought; but Foucault, who had wanted for there to be even more, was rather beaten down, because he considered it – wrongly, in my view – to have been a failure.

Still, for a moment — okay … I would add … you will see where I would like to take this …. For a moment, there was a keen interest in making a connection with the American movements. Good, it was necessary that this GIP network develop in this direction, pushing literally a kind of transversal, always a kind of relation of transversality, pushing a transversal toward America. Good. What happens? This is where I might use something like Spontaneist metaphors, it happened almost by itself: that’s when Jean Genet appears. Jean Genet, who was closely tied to the Black Panthers …, who himself was not an intellectual reflecting on the prisons, but who had had a notorious and substantive experience as a prisoner…. He comes just then. Very good. And he plays a pivotal role as a kind of hinge. And Jean Genet does not act within the GIP in his capacity as a prestigious author or intellectual. No doubt his prestige is useful. Clearly, but he is active, and insofar as he is active, it is not in his capacity as intellectual but as intermediary with the Black Panthers. What does all of this mean?

Well, I want to get to this point: there are three practical problems of the here-now. And this long parenthesis on the GIP…, I could not draw out the three problems unless… It seems to me that there are three practical problems for every here-now. It’s: what new type of struggle, if there is any, what new type of resistance to power? You will tell me: ‘but you speak of resistance to power before having spoken of power.’ Well, yes, it doesn’t matter: there are sites of power, there are sites of resistance to power, and the sites of resistance to power are sites of power. That’s clear. Good. When I say “resistance to power,” I am also saying “power”. Thus: what new type of struggle, today, here and now, whatever the period? It may very well be that I reply to myself: the period is not promising, so there’s no new type of struggle on the horizon.

Second question: is there a particular role, today, here and now, which would be that of the intellectual? [Pause] Is there yet another question? Ah perhaps, but what would it be? The other question. Perhaps there are other questions. Here and now, what does it mean “to be subject”? Perhaps… Perhaps not… Why do I say that, are these the three questions? Let me go slowly.

First, if you take a good look at these three questions, I would say: the one clearly concerns power. What new type of struggle today? The second clearly concerns knowledge. What is the role of the intellectual today? If there is a particular role; perhaps there is no particular role… this is the practical question of knowledge. The third question, if there is a third question, what is it? Good, all of that is not just in Foucault’s head. No more than May ’68 was in the head of intellectuals. What is it, then?

We must go back a long time, this is part of our history. And what is part of our history, I believe – I am giving a very brief history –, is that it all began to change starting with and based upon 1950. As a function of what? As a function of the Yugoslavian experience and rupture. It is the great date of what seems to me the first time, in a communist country, that centralism was called into question. What presents itself at that time as the experience of Yugoslavian self-determination, with a theoretician-practitioner [Milovan] Djilas who at that time is a comrade of [Josips Broz] Tito and whom Tito would later imprison, is the Yugoslavian rupture, which was the great detonator. And starting with the Yugoslavian rupture, the question, which no doubt had been subjacent before, bursts forward in the form of a new, non-centralized politics. And the chance, in a capitalist country, of establishing a new kind of struggle. The theme of transversality begins to take form after 1950.

And if the starting place, at least generally, was the Yugoslavian experience, by what path did it spread? It did not spread from there as though from top to bottom; it spread through networks, by way of broken lines. It passed by way of Italy and the reinterpretation of Marxism by the Italians, notably by one Italian – we will see, because all this is tied fairly closely to Foucault – by one Italian whom I believe to be very important named [Mario] Tronti, T-R-O-N-T-I, who was a member of the communist party and who attempted a reinterpretation of Marxism on the basis of Italian conditions. What were the Italian conditions? The Italian economic conditions were very different than the Yugoslavian conditions, but the existence of a kind of double market in the Italian economy, a black market with temporary labor, off-the-books work, etc., which had taken hold in Italy very early on in a kind of institutionalized form, had been fundamental for the formation of a theme of autonomy. And autonomy, the theme of autonomy, began with Tronti, and was also permeated by the idea of new forms of non-centralized struggle.[5]

Now, from the beginning, this theme of transversal struggles, of non-centralized struggles, inspired first by Yugoslavian self-determination, then by Italian autonomy, had been mingled with a question that was more unclear, more difficult, which was what? Something like: toward a new subjectivity. Are we subjects in the same way as forty or fifty years ago? What does it mean to be subject? Can we attempt to disengage ourselves from centralization without being subject in a new way? Without there being a new style of subjectivity? Tronti went very far in reintroducing a new subjectivity into Marxism, and for him, Marxism was the promotion of a new subjectivity. Which clearly ties Tronti, in a certain way, to the Frankfurt School. All that is quite meticulous as a network.

Yet at the same time as in Italy, and each reciprocally influencing the other, there formed around Sartre in France the theme of: toward a new working class, notably associated with a close friend of Sartre, one of whose pseudonyms was [André] Gorz. And, with Gorz, there emerged: toward a new working class, in the double sense of new struggles, a new form, new forms of struggle, new forms of resistance to power and a new subjectivity. And even in France, other groups before ’68 were developing this question. This triple question: new forms of struggle, a new role for the intellectual, new subjectivity. It was all three. It seems to me that these were the three great poles of what turned around ’68, of what exploded in ’68, and again, these are only theoretical questions if you see that they do not await theoretical answers; they are undertaken practically, materializing practically in history.

And in France, these will also be the three questions, doled out differently, in a very subtle way, as the groups already hated each other, but… There was thus: around Sartre, but there was also Socialism and Barbarism – they turned entirely around these three questions. There was also Situationism. Finally, there were the dissidents of the Communist Party, namely, the Communist Voice, where at that time [Félix] Guattari, who launched his theme of transversality and who had already launched his theme of a micropolitics of desire; and there will clearly be an echo in Foucault, when Foucault launches his own theme of a microphysics of power. You see precisely what the transversal, non-centralized struggles isolate as a kind of element that must be analyzed: whether microphysics or micropolitics, it’s the transversal relations between restricted groups… Good.

In this light, I want to get to the following point: the three questions respond more or less to the three famous Kantian questions. The three Kantian questions are: what can I know? What should I do? “What can I know?”, Critique of Pure Reason; “what should I do?”, Critique of Practical Reason. For what can I hope? Here, we must admire Kant’s perspicacity when he thought that these were the three fundamental questions since, so many years later, we see… “What should I do?”, that is: what are the new types of struggles today? What are the new pockets of resistance to power? “What can I know?”, that is: what is the role of the intellectual? “For what can I hope?”, that is: is there a new subjectivity? Perhaps you sense that my third axis is planted there, is planting itself in there, is in the process of growing on this terrain. So, what will Foucault do?

This is not at all to diminish, I believe, the depth of his originality, to say: Foucault is, in this regard, the last, I believe, to have taken up this set of questions in order to propel them to a point where they had never been. But these questions have their origin beginning with 1950, and they also permeate Italian Marxism, Yugoslavian Marxism, the Sartrian circles, the pre-’68 milieu… Foucault will give them an echo after ’68, and perhaps on the first question (the new types of struggles), he  will not provide – he’ll create a… he’ll express this theme very strongly – but perhaps he will not provide not it with a radical innovation, and his innovation will have been much more to have brought it into practice, that is, to have constituted the GIP in the mode of a new type of struggle. But the new type of struggle, transversality, transversal struggles, etc., to which Foucault, at the end of his life, will still refer, about which he will still sum up the principal features, I believe that this is not what is most innovative in his work.

On the other hand, the two other questions – that we’ll see [indistinct words] – the two other questions, specifically, “Is there a new role for the intellectual that will be renewed?”, by proposing the following outline which has, it seems to me, a great historical value: “we are done with the era in which the intellectual was the guardian of values”. In fact, we are not yet done with it, since the ancient figure of the intellectual as guardian of values was to be reawakened, resuscitated in the form of the intellectual as guardian of the rights of man. But Foucault was able to think, at a certain moment, that this figure of the intellectual had died. And he opposed to this ancient figure of the intellectual, the so-called universal intellectual, what he called the “specific intellectual”, or what we could just as well call the “singular intellectual”, if you have held onto the theme of singularity in Foucault. It was no longer in the name of his universality that the intellectual acted, but in the name of his specificity or singularity.

What does this mean? Well, in fact, if you ask what is … it’s not at all despicable at this or that period. When does the intellectual begin to be a bearer of the universal? There, one would have to provide a very good and very precise history … I don’t know … Can we already say this of the Renaissance? That the intellectual in the Renaissance is the bearer of a kind of universality? I do not know, because universality is firstly Catholicism. So, is the catholic intellectual, the catholic cleric already a figure of the universal intellectual? Perhaps, I do not know, it’s too complicated.

In the eighteenth century, if you take the far-reaching interventions of Voltaire in politics, there is something there like an intellectual as bearer of universality. In the name of what does he interfere in juridical affairs? In the name of the Enlightenment, in the name of justice. The Enlightenment! If I make a giant leap in history, I jump to [Émile] Zola. In his famous intervention in the Dreyfus Affair, Zola explicitly presents himself — in what may be the first time the very term “intellectual” appeared as such — The “Manifesto of the Intellectuals” in support of Dreyfus is archetypically a manifesto in which the intellectual presents himself as guardian of universal values, namely: no national reason can justify a false judgment, a tampering with evidence, etc. It is in the name of the universal that the military is denounced.

It would be necessary to follow — if I am still taking up those who played a great intellectual role — André Gide, when André Gide denounces the conditions of a jury in a famous trial at the time, the conditions of justice, or when he denounces the conditions of colonial labor, these are the interventions of a great intellectual who has a great impact, for which Gide will never be forgiven… His homosexuality was something that could be readily forgiven at the time, but the history of the Congo was too difficult for him to be forgiven. Alright.

And then, more recently, Sartre. It’s not wrong to say, as has been said, nor is there anything … hostile in saying that in Sartre, there was in fact something relatively close to Voltaire or to Zola, that the courage of this man … perhaps when it comes to Sartre there were both aspects: the birth of a new role for the intellectual and the maintenance and completion of the ancient figure of the intellectual, the intellectual as guardian of universal values. But according to Foucault’s analysis, there was a moment when this breaks off. He situates this break with the atomic bomb, when physicists intervened against the atomic bomb. For you must realize that when physicists intervened against the atomic bomb, they did something that was in fact very curious, namely: it wasn’t in the name of universal values, but in the name of their situation that was specific to them as physicists. They said: we who know what we are talking about, we are telling you that… The intellectuals did not make use of values like justice etc., they made use of their specific situation, namely: we who have made this bomb, we know that… and that’s what has been hidden from you. It is on the basis of their singular situation… The distinctive case, although relatively late, is [Robert] Oppenheimer, the atomic physicist, it’s Oppenheimer who spoke from the depths of the laboratory where he was, in the name of the laboratory where he was. It was a completely new figure of the intellectual that Foucault identified very, very well.

And if Genet was going to speak about the prisons, it was not in the name of rights and eternal values, but rather in the name of the singular experience that he, Jean Genet, had undergone, which could be linked to the singular experience of the Black Panthers in America, which could be linked to the singular experience of the French in prison at the moment when Genet had been put in prison, etc. It was in the name of singularity, of one’s proper singularity as intellectual, that the intellectual was going to be able to speak. In a certain manner, it would never again be in the name of rights, even if they were the rights of man, but rather in the name of life and a singular life. [Pause] Well, then, I really believe that in Foucault, the three questions are reunited. They are reunited as though for the last time, for now, not for the last time forever, but the period is very bad today … and thus the questions are again disunited in a kind of night of the non-question. … But they are reunited for the provisional last time, in the most original and strongest way in Foucault… And once again it is not a matter of abstract reflections.

When Foucault asks: “are there new types of struggles today?”, “is there a new role for the intellectual?”, “are there chances for a new subjectivity?”, what does that show?  On the horizon of the question about “new types of struggles”, there is the experience of the GIP. On the horizon of the question “are there chances for a new subjectivity?”, there is surely the attraction he felt toward American communities, the interest he had in forms at once … sometimes solitary forms, sometimes communal forms, where he clearly emphasized that for him, in effect, the problem of the formation of a new subjectivity was concretely a mode of evading identification. Modes of evading identification are equally modes of community; there are group subjectivities, notably the American groups… and what the American groups are today developing — You know, things that develop do so in mediocrity and inanity just as much as in grandeur, no? It’s always, as that other guy said [no doubt, Brisset],[6] it’s always also [your] muck [c’est toujours, comme disait l’autre, c’est toujours aussi (ta) saloperie] – so are there in American communities today, in German communities today, is there a new form of subjectivity being developed? Quite clever someone who would respond [Bien malin celui qui répondrait]. In any case… There again, these aren’t theoretical questions, one must go and see. With the Americans, it’s not easy. Fine.

Thus, it may very well be that, in his own life, Foucault passed from one question to the other, that he discovered quite late the practical questions: new subjectivities — it was necessary that he understand them in his own manner, and the question that I believe he understood that the most profoundly – [is] the necessity of a new situation for the intellectual – whence the very ambiguous relations between Foucault and Sartre. Very ambiguous relations … I would say: these relations were very good, but they were relations where, theoretically, Foucault surely had a great admiration for Sartre but had no affinity with Sartre’s thought. I believe he had a great affinity with Heidegger, as we will see later, but not with Sartre. But that does not prevent him from having a great amount of respect for Sartre, and then … things were very good between them… But the ambiguity stems from the fact that Foucault considered Sartre to be the last figure of the bygone intellectual and Foucault, personally, had to live like – to the extent that he considered himself an intellectual, or that he agreed to be taken for an intellectual – Foucault had to live under another image, not better but something other… [Interruption of the recording] [1:29:28]

Part 3

… I imagine he does not mean “as a function of the self,” as a function of the self as individual, of the self as person. No. I can think that I embody singularities, but with regard to the singularities, I can only ever embody them. …So, I embody a small lot of singularities, we are all like that, no? What does it mean, a subject? No doubt a subject is an incarnation of singularities. It is here already that I would like you to feel the third axis, which will be revealed to us much later, but which is already there. This axis of subjectivation…, it is in the process of… and yet we are quite far from it. Hence, now at last, we are ready, and subsequently you will see that what I am saying amounts to this: the only historical continuity going from past times to now is what? Practices, and in what sense? The practice of struggle, the practice of knowledge, the practice of subjectivity. This is what establishes the correlation between historical formations and the here-now.

For, after all, when we posed the question of the here-now, we can say to ourselves: but hasn’t history perpetually been made by means of transversal struggles? Hasn’t history perpetually been a fabric, a network of transversal struggles, before these struggles are centralized? Doesn’t everything happen this way? Doesn’t centralism each time come to cover over with its pyramid and stifle all that had been rich and had been created through a form of transversality, through a transversal form? I do not wish to convince you, I am simply saying that …

I have no sympathy for the Surrealist movement, and I have no sympathy for Surrealism because I see there a dirty, centralized organization, with tribunals, excommunications, etc. What had there been before? Can we clearly see the succession? There had been Dada, there had been Dadaism, and Dadaism is a network. Dadaism is a transversal network that affects all countries…; from Eastern Europe to America, Dadaism traverses the entire world. Precisely because it is not centralized. And what does André Breton do? Those who like André Breton, you correct this yourselves, eh! What does André Breton do? He reestablishes order. He reestablishes order, and he makes something national, something quite French. Surrealism feels French. And he establishes his tribunals, and he launches his excommunications, and he will subject everyone to forced labor, namely, the pages of automatic writing and idiotic little games … and all that. Okay. He gives the order. And he makes it into a French centralism. Okay: Dada will not recover from this, and if there is a politics in art or in literature, you can take this example because it really was a political struggle. It was a political struggle, namely: Dada was absolutely consumed, devoured, was no longer possible because of the reordering imposed by Surrealism.[7]

Well, if this is so, one can say that after all, transversal struggles do not date back only to 1968. So, let’s take a giant leap back in history. This is what Foucault, by the way, foresaw very clearly: at the time of the Reformation, but before Luther had made his own kind of surrealism, that is, his own recentralization, the struggles of the Reformation – at the level of the peasantry just as much as, indeed before the reforms supported by the Lords – had been composed of all manner of movements of transversal struggle, engaging the people of the cities, the people of the countryside, etc.; and the period of the Reformation was a very great moment for transversal struggles. So, finally, perhaps we will discover that the three questions – the new types of struggle, the situation of the intellectual, the new modes of subjectivity – have not ceased being posed throughout history, and that this is the true continuity of history, though only practices provide the continuity of history.

So, in light of all this, which allows us to take the necessary precautions in posing the question, we can ask: what are the principles of power in general? We no longer risk, in effect, succumbing to responses by way of centralization. I have tried to … what is the word for chasing away the devil?

Lucien Gouty: To exorcize

Deleuze: To exorcize! What I have tried to exorcize here is a response to the question “what is power?” that would be a central response. To the question “what is power?”, the only suitable response is a transversal one that splinters power into a multiplicity of sites. So, what will this be? Are there principles for this multiplicity? Power… [Interruption of the recording] [1:35:52]

…on Rivière, on Pierre Rivière who had slaughtered his family… How did he distribute what goes into the notebook – written or sound image – and the crime itself – visual image? That there would be, there, a relation between the statement and the visibility, and to know if this is the place to find an original solution, given that very original solutions were found in cinema for taking account of these relations between the visible and the statement. Especially since, in the case of the murder of Pierre Rivière, or rather the murders committed by Pierre Rivière, …the notebook where he recounts his whole story has a very mobile role, sometimes conceived as coming after the crime and sometimes as coming before it etc., which raises all sorts of questions. Thus, for the problem of seeing-speaking relations, I myself would be very interested to rewatch the film on the basis of how [René] Allio cinematically treated these relations between seeing and knowing. — But there must not be too many of you because we will run out of room.

A student: It’s like that [Inaudible comment]

Deleuze: It’s like that, so there can be many of you. — So, let’s continue, but I think that everything which came before was absolutely indispensable as practice, already, and as practice of power.

Now, then, it is as if we returned to the question: what are the principles… What is power? And well, if there are principles, they will clearly be practical principles. But only if there are principles. What is the principle that Foucault always brings out? Well, it is, strangely, that in the texts of Foucault, there is no long study at the level of principles, as I told you. If you want to locate that level, there are a few pages, two or three that go very quickly, at the beginning of Discipline and Punish, pages 22-24 – it doesn’t go very far at the level of principles, I am saying –, and in History of Sexuality Volume 1, a little bit longer this time, pages 82-85.[8] This is to say that…, and well, we will do the opposite; we are going to go a little off-center, though only with the idea that it is really in order to be clearer, and we instead are going to focus at some length on this question of the principles according to which we can think power.

Now, the greatest of these principles in Foucault, it seems to me, is constantly called forth by his concrete analyses, namely: one must not start from large aggregates. One must not start from large aggregates. What are these large aggregates? Well, in the modern world, for example, they are the large institutions. You see right away this means that in order to evaluate power, one must not start from large institutions that are already made, like the State, the law, and classes. In the end, Foucault is the only one to have made a leftist theory of power, I believe. He is not the only one to have proposed it, but to have done it… Good, must not start from large aggregates, why? No doubt because the large aggregates are given as already made, the genesis and exercise of which must both be shown. The law, the State, classes are already entities, how to put it, that are much too big. This immediately explains his appeal, from the beginning of Discipline and Punish, to a microphysics. Once it has been said that microphysics consists in saying, in the domain of the physical sciences, that, beyond the great aggregates – if I give a very rough summary of microphysics – beyond the great aggregates that are statistical aggregates, a science of molecular corpuscles must be attained and constituted.

What, then, are corpuscles? Power must be grasped at the level of molecules and corpuscles rather than at the level of great institutions. This is what “microphysics of power” means before anything else. It means something very simple in appearance, but what is this molecular power? The State, classes, the law, these are powers that we will call molar. And, well, no, in order to grasp power, one must go all the way to molecules. It is a metaphor. Good. History of Sexuality Volume 1, page 92: “The analysis, made in terms of power, must not assume that the sovereignty of the state, the form of the law, or the over-all unity of a domination are given at the outset”. In other words, the theory of power must be local rather than global, molecular rather than statistical. “… [T]hese [the sovereignty of the state, the form of the law, or the over-all unity of a domination] are only the terminal forms power takes”. It’s not that there is no State or no law, but that these are statistical expressions of an agitation of another nature. Power is a molecular agitation before being a statistical organization. Hence, once again, all I am doing is justifying the term “microphysics of power”. But, if the word “microphysics” must be taken seriously, it’s because we will find there the great microphysical unity, namely: the corpuscular wave. Power is not a matter of large aggregates, but rather of waves and corpuscles. There are political waves and political corpuscles. So that should remind us of something or someone about which Foucault, however, never speaks.

I will tell a story today, which will be…, which will conduce toward … a reprise, that is, a series of stories. I am telling a story. At the end of the nineteenth and beginning of the twentieth centuries, two sociologists opposed each other in an inexpiable war. What the hell! [Diable!] If I say this, it is because it is true to the letter. It must be believed that there was not only the question of knowledge, but also the question of power. The one will be crushed by the other, but, as always, the one who was crushed is the best. It’s about [Émile] Durkheim, the great Durkheim, and Gabriel Tarde. Gabriel Tarde is the founder of microsociology, which will disappear in France and go on to America, whence it will return to France, by the American route, without us knowing that America was only returning to us what it had taken. O what sorrow! [Laughter]

Okay, what happened between Durkheim and Tarde? Here again, I think that this parenthesis is necessary. When I say that Durkheim crushed Tarde, it was a veritable war and a conquest that Durkheim won, a conquest of primary education, of secular morality, and of the university. The deans were Durkheimians. I mean, statistically, the number of deans of faculty who were Durkheimians, you’d think that… they were high-ranking civil servants. Tarde was not an academic by training, he was a justice of the peace… Fortunately, he had been received by the Collège de France – there was quite a strong opposition at that time between the Sorbonne and the Collège de France – he taught his courses, which must have been marvelous courses, but… well, his influence…, his non-Durkheimian disciples were massacred, to the point where they fell back on the law and, very oddly – I abound in anecdotes today – very oddly, the disciples of Tarde were very important in the formation of European movements and also in legal disciplines; for example, a great jurist named [Maurice] Hauriou, not the more recent [André] Hauriou, but his father, the old Hauriou… who was of great importance at the time of the Matignon Agreements [1936] and had a political influence, was a direct disciple of Tarde. And Tarde went in two directions, on the side of the Matignon Agreements and on the side of Europe. Very bizarre. He was a federalist who announced federal Europe… in the end, very, very bizarre, very curious. No matter![9]

But all of this is to say: what is it that Durkheim did? If there has been a sociology of large aggregates, it was this one. And how did Durkheim found sociology as a science? He said: it’s quite simple, there are two sorts of representations. There is individual representation, which is the concern of psychology, which is worth nothing, by the way, … and then there are collective representations. The two differ in kind. Collective representations are not of the same nature as individual representations, and sociology is the science of collective representations. Okay. I am simplifying a great deal. But the notion of collective representation is fundamental… I can say that it is typical of molar sociology. It is the sociology of the large aggregate. And it was wonderful, Durkheim’s sociology is quite beautiful, at least for those who like that sort of thing.

What did Tarde say? Tarde said: well, yes, what Durkheim says is very good and interesting, but in the end, these collective representations, these large aggregates do not emerge readymade; where does something like that come from? So then, Durkheim went into a rage, the question “where does it come from?” was not to be asked, and perhaps he was right. But Tarde renewed the charge, saying: no, no, you know, not everything that is socially composed presupposes the large aggregates, there is always… What is there always? And well, …in large groups, there are currents of imitation and then there are moments of invention. And then he said: you see, what interests me is the way in which an employee of the ministry …writes his signature. There is a moment, for example, or a period where the type of signature of the ministry changes, and Tarde says: that is what’s good, something small, a small social invention. Society is formed by currents of imitation and movements of invention, and they are always small imitations and small inventions. You see, there, I am not even giving the theory, I am saying what interested Tarde. You will tell me: fine, but what is all that about?

Thereupon falls the inexorable judgment of Durkheim and the Durkheimians, which is “Death to Tarde. He wants to reduce sociology to psychology.” In fact, he wants to explain collective representations, that is, social phenomena, by imitations and inventions. In other words, he wants to explain the social by the individual; an individual imitates another, an individual invents. You see? Tarde never recovered from this in France. And yet… and yet Tarde said something that had absolutely nothing to do with what Durkheim would have had him say, for in Tarde, social imitation never goes from an individual to another individual, and social invention does not depend upon an individual. So then what does he mean in referring to imitation and invention? This is where it becomes quite beautiful. But you understand, it must… Well, he had said it clearly, repeatedly, but nothing would do; he was always taken to be saying that the social is explained by the psychological. There are misunderstandings that one can never take back. It is sad, no? It is sad. Tarde was very sad. So, what did he mean? He meant the following: imitation is a wave or a current of propagation. It is a current of propagation.

Okay, let us accept this, we can understand vaguely, and we will see what comes of it. It is a wave of propagation. What is invention? Invention is an encounter between two different imitative currents. Good. Invention always comes from the crossing of two orders. Thus, imitation is some wave of propagation, and invention is the point of encounter between two currents, two waves of propagation. So obviously, the question… This is still nothing. The question that becomes serious is: okay, but between what and what does imitation happen? From where does the current of propagation depart, to where does it go? And that is where Durkheim awaited him, saying: “well, you see clearly that it goes from one individual to another individual, thus you are doing psychology. However much you say that imitation is a current of propagation, this current of propagation can only go from one individual to another individual, since you reject collective representations.”

Now, Tarde in turn said that a current of imitation or of propagation does not go from an individual to another individual. It goes from where to where? It goes from a state of belief to a state of belief, or from a state of desire to a state of desire. What is propagated is belief or desire. You see where I want to take this. It is not a matter of doing a sociology of representation, but rather of doing a sociology of what there is beneath representations and what representations presuppose; and what representation presupposes is belief or desire. And a representation is not propagated, which is why Durkheim had to posit it ready made. But beliefs and desires are inseparable from waves of propagation and constitute true sociology, that is, microsociology; to do microsociology is to conduct the study of waves of propagation of beliefs or desires that traverse a social field. This becomes a great idea that has nothing to do with psychology, but that has to do with microsociology. Beliefs and desires are social corpuscles.

You see the force of the critique against Durkheim. Durkheim remains at the level of representations, he does not see what there is beneath representation. What is beneath representation… Representation is a large aggregate, a molar instance. Beneath representations, there are corpuscles of belief and desire, and corpuscles of belief and desire are inseparable from waves of propagation, and the wave of propagation of belief or desire is imitation. So it is only secondarily that one can say the believer imitates someone; that is conversion. Someone converts to Christianity at the first moments of the Church; that means that Christianity as belief, that is, as molecule, as social particle, spreads following a wave of propagation. That is what imitation is. And invention will be the formation of a new desire at the meeting point of two currents of propagation of belief or desire, or the formation of a new belief.

I will not say more about it because that would be … I dream of one year teaching a course on Tarde, but… Tarde is very beautiful. It is the true microsociology, which is to say, the Americans never attained this point of non-psychology. Yeah, it is only Tarde who knew how to maintain a microsociology psychology [une microsociologie psychologie] without the least … um … with its history there, … notably his great theme that representations are never quantifiable, whereas there is a social quantification of beliefs and desires, and he made a great theory of social quantification that involves an entire logic, … and he redid everything, then, on the basis of his microsociology, since he will reproach logic for being a logic of representation, instead of being, as it ought to have been, a logic of belief and desire, but only, if logic becomes a logic of belief and desire, at that point new quantifiers must be introduced into logic. And he sketches an astonishing logic of belief and desire. So, this is to indicate the importance…

Now, I believe – I don’t mean at all that there was an influence – that there is often in Foucault a veritable Tardian tone. I would say to the letter that, you see, imitation and invention in Tarde correspond completely – you will only be able to understand what I mean a little bit later – correspond completely to what Foucault calls relations of force, because, I will say right away, to avoid any misunderstanding, for Foucault, and he is formal in this regard, relations of force have nothing to do with violence, in other words are irreducible to violence. They have an entirely different nature. Why call them “relations of force” since they cannot be explained in terms of violence? We will see, we will see why, precisely relations of force are something that exceed violence from all sides [rapports de forces quelque chose qui excède la violence de toutes parts], but you understand that if relations of force exceed violence, I can very well say: imitation is a relation of forces, invention is a relation of forces, it’s obvious. It’s obvious. Imitation implies a relation of forces between what imitates and what is imitated. Invention implies a force relation between the currents whose encounter gives rise to invention. Good.

And I am saying that very often, Foucault’s tone strangely recalls Tarde. For example, Foucault has an extraordinary taste for what he himself calls…, one time…, there is a passage in Discipline and Punish where he speaks of … small social inventions, and he says: the prison car, one never speaks of it, he says, but it’s an odd little invention. Obviously, one speaks of the blast furnace, one speaks of big technical inventions, and one does not speak of small social inventions. I have the feeling that … I don’t know if Foucault knew Tarde, I think he must have known him, but even if he didn’t know him, it is an astonishing encounter. Because there, it is literally almost a text that would be signed by … The prison car as small social invention, by contrast to the big technical inventions, well it’s … it’s pure Tarde. So, all of this is only a starting point. You see, the relations of force are molecular relations, micro-relations between elements that function like corpora. Good, it’s with this that we will need to sort things out.

So, in trying to arrange Foucault’s texts well, I would say that from this, we could abstract six principles from his microphysics of power. Six. That’s a lot. And these six principles are equally denunciations of what he took to be postulates in classical theories of power. Thus, we can extract six postulates denounced by Foucault. Let’s do it. – Ah, I thought that was a watch… it’s a… you see, I thought that was a bracelet; thank you very much! —

First: we will call this the postulate of property. Foucault tells us: political theories proceed as though power were the property of something or someone. Well, power is the property of no one. Property of what? Well for example, even the Marxists, at least at first glance, make of power the property of the dominant class. Discipline and Punish, page 26: “Now, the study of this micro-physics presupposes that the power exercised on the body is conceived not as a property, but as a strategy”. It’s the first time we encounter this word that will take an ever-greater importance in Foucault. Power is not a property, but a strategy.

In other words, there is an absolute functionalism of power. Power is not possessed, is not a property, but rather is exercised. It is exercised. It is never possessed, it is not a property. If you have already understood everything we have said, this shouldn’t surprise you, because… at the risk of getting ahead of ourselves, if it is true that power is fundamentally relational, it cannot be a property. A relation is not a property. If power is a relation, and if we take seriously the expression “relation of forces,” and if we say that “power is relationality itself,” it cannot be a property, even of a supposedly dominant class. It is exercised, that is, power is a strategy. A strategy… Good, good, good. … And now, what should this new word make us think of? But we will have to pay close attention, and in particular, we must not confuse strategy and stratum. Because knowledge refers to strata and stratification, but power, on the other hand, refers to strategies.

Moreover, strategy appears there where there is no stratification, there where there is no property. The sedimentary layers can be possessed, but power is exercised rather than possessed. And how will Foucault define strategy? “Innumerable points of confrontation, innumerable points of confrontation, sites of instability.” It is forced, but he does not say “sites of instability” for the pleasure of being clever; rather, it is that in the micrological domain, in the microphysical domain, there is by nature no equilibrium, it is a domain that repudiates equilibrium. This is no corpuscular equilibrium. There is no wave-like equilibrium. There is no microphysical equilibrium. There is no stability. There is only statistical stability, which is to say, the stable is the large aggregate. Large aggregates are possessed, it is true, but you can guess that power is fundamentally diffuse, fluid, unstable at the level of microphysics.

This does not mean that at the level of large aggregates… so understand, at the level of large aggregates, of course it is possessed, of course it is stable, of course it is in equilibrium. But that is not the source of power. It’s not that, and we must attain to this microphysical layer which does not admit of sedimentation, which is pure strategy and not stratum, innumerable points of confrontation, sites of instability; and what are these innumerable points of confrontation, these sites of instability? It is what we saw before, singularities rather than psychological individuals. Tarde would say they are quantities of belief and desire, corpuscles of belief and desire that are fundamentally unstable, fluid, swept along by waves, etc. Innumerable points of confrontation, sites of instability: relations of forces are relations between singularities. These are waves of singularities, just as microphysics speaks to us of corpuscular waves, that is, waves which guide a corpuscle; and, well, relations of forces guide singularities.

But you … have not yet done anything if you stick to the large aggregates that this microphysics will come to form. It is not a matter of saying that the large aggregates do not exist; the large aggregates are effectively constituted by this microphysics, which is to say, microphysics constitutes a macrophysics. The large aggregates are nothing other than states of affairs [faits] composed of these corpuscles and these waves, but at the same time, the states of affairs stratify the corpuscles and waves, [Pause] which is to say, they make large aggregates out of them, the global effects of which are all anyone considers. The large aggregates are the global effect of microphysics.

Okay, everything alright? Then I will again open a parenthesis, just as I will keep doing later. There is one thing that Foucault never speaks about – we will do a test, eh, we will do a little confirmation test by taking an example outside those given by Foucault. I say to myself: Foucault never spoke about primitive societies. Why did he never speak about so-called “primitive” societies? He never spoke of them for a very simple reason, I believe, and it is not at all that he did not like ethnography and ethnology; it is because he wanted to operate on short and well-determined series. We have seen it, right… He had such horror of universal history that he feared, evidently, that if he did not take up a well-assigned, well-determined series, then he would seem to lapse into a kind of universal vision of history. But if we do not have the same fears, then we can always try, because, if you followed everything a moment ago, I would like to find in modern ethnology the same story as that between Durkheim and Tarde, for after all, the same stories can be found.

In modern ethnology, it is well known that primitive societies are said to be Stateless. Foucault could say: yes, perhaps they are Stateless, but they are not without power. Good, but after all, if the so-called “primitive” societies do not present us with a large aggregate that exceeds them, they present us all the same with large aggregates, which are what? They are the great lines of filiation. Like the Marxists say, what stands in for the State in primitive societies is kinship. Okay, the great lines of kinship, the great lines of filiation that we refer to fairly as great lineages. I would say that the great lineages in primitive societies are the collective representations of the base or, if you prefer, the large aggregates. In modern societies, the large aggregate is instead the State and the law. Well, we are assuming, it’s easy to say.

Here I notice that an oeuvre as admirable as that of [Claude] Levi-Strauss… At least, I am not speaking of the Mythologiques, because when it comes to the Mythologiques, perhaps everything changes, the Mythologiques are a curious story… but I am speaking of the Elementary Structures of Kinship, an already antique book. At the level and moment of the Elementary Structures of Kinship, there is no doubt that Levi-Strauss develops a structuralism based on large aggregates, and, of course, structures are only ever molar. Here, notice that we are reaching the point where we will be able perhaps to confirm our analyses on the difference between Foucault and structuralism. Structures are fundamentally in equilibrium. A structural disequilibrium, well, it happens, it happens, but a disequilibrium is still the structure in question. The structure qua structure designates a state of equilibrium. In other words, Levi-Strauss says this explicitly when he affirms that malfunctions and disequlibria are consequences, the consequences of something else. Good.

In this way, then, with regard to primitive societies, Levi-Strauss develops a macrosociology based on the great lineages of kinship, the great lines of filiation. And what does he tell us? He tells us that there are relations between great lines of filiation. These relations are alliances – among which one privileged case is marriage – alliances, marriages, which are forms of exchange between one filiative line, one lineage of filiation, and another. And he tells us finally that if one considers the totality of exchanges within a sufficient social space, one will find that there is a generalized exchange – this is what he called generalized exchange, by contrast to restricted exchange, where one limits oneself to two lineages, to the relation of two lines – and that generalized exchange forms a closed cycle. Closed cycle = structure. There you have it.

A number of ethnologists reacted, saying that while Levi-Strauss’s oeuvre, with respect to the Elementary Structures of Kinship, was very admirable in its systematization of ethnological data, in their view, no primitive society had ever functioned like that. They said, in a sense, that that had never worked like that. It works like that in the mind of Levi-Strauss, fair enough, but never has a primitive society been able to work like that. You see that those who objected to Levi-Strauss invoked a functionalism, a practice: what are the effective social practices? In other words, it was the English and the Americans. And notably, one of the greatest ethnologists, along with Levi-Strauss, was an Englishman named [Edmund] Leach, a great ethnologist. And just as there had been a Tarde-Durkheim polemic, here was born another polemic, Leach-Levi-Strauss, where each showed a kind of genius in the art of not understanding what the other was saying… [Interruption of the recording] [2:15:47]

Part 4

… because in all the societies I know of, well, in all the primitive societies, their great concern is that what happens not seem like an exchange. That is their first theme: it must not seem like an exchange, and when one speaks to them of exchange, they are not at all happy about it. For example, for them, a woman is not exchanged, not ever. A woman is either given or stolen or sometimes both at once. It must be either a theft or a gift. But exchange, that won’t do… It is a gift or a counter-gift, a gift that obliges etc., the whole well-known theme that Leach takes up in his own way. And you will see what he will draw from it. This is his first point, which Levi-Strauss, in a certain manner, had recognized in saying: exchange is unconscious. And the strength of Leach is to have said: but why is exchange unconscious? Why do they care so much… If it was a matter of exchange, why were they so anxious to hide it? It is bizarre, there is no shame in exchanging.

So, Levi-Strauss was forced to say that it is because, if they recognized that it was a matter of exchange, then they would understand that the cycle is closed, and they do not want the cycle to be closed. Leach says: but if they do not want the cycle to be closed, perhaps that is because it is not closed, that is, because there is no structure. Good, it is getting interesting. And why would the cycle not be closed? It is that Levi-Strauss’s whole hypothesis presupposes that relations of alliance are deduced from a line of filiation and stem from lines of filiation. There are first the lines of filiation, and then they exchange something. They exchange women, they exchange products, they exchange titles, emblems. Exchange takes place between two filiative lines. Good.

Is alliance an exchange? Yes, alliance is an exchange if it passes between filiative lines, between lineages. And never has a primitive society functioned on the basis of lineage. No one believes in lineages. I am simplifying a bit, there. Never, ever have they functioned like that, says Leach. They function entirely otherwise. They function, in fact, through alliances. But the alliances are not at all exchanges because alliances do not presuppose the great lines of filiation but are constituted otherwise and elsewhere. Alliances do not admit of being derived from lines of filiation. Alliances are autonomous. And in fact, they are literally wangled, manipulated by groups that Leach calls local groups, Leach’s local groups in opposition to groups of filiation. And the local groups do not cover the groups of filiation, and it is the local groups that organize the marriages. It is the local groups that decide the gifts and counter-gifts; in other words, what must be understood? What many ethnologists had said, but it is Leach who brings it all together. The alliances are a matter of practice. There is an ethnologist who will say and who will employ a very curious word, saying that it is not a structure, but a process [procédé]. It is a matter of practice, of a process. Let us say hardly a strategy. A society strategizes before structuring itself.

And Leach poses his great question: does the position of an individual in the social field owe to their belonging to the lineages of their father or the lineages of their mother, or to the two lineages, or even – which is quite different – does their position come from the fact that their father and mother are allied? It is a very good question, because if you read with care the Elementary Structures of Kinship, you will see that Levi-Strauss is for the first response, and if you read Leach with care, you will see that Leach is for the second response. The position of the individual in the social field comes from the fact that their parents are allied and not at all from their participating in the lineages of the father, the mother, or both.

In other words, what is Leach’s perpetual theme? It is that the great lineages form a vertical structure. It is incontestable. On that point, Levi-Strauss is unbeatable, and he definitively established the status of this vertical structure. But the network of alliances is not derived from this vertical structure. The network of alliances, and here I cite Leach by heart, the network of alliances is a lateral network. In other words: transversal. Perpendicular, says Leach, to the structure, to the filiative structure, irreducible to this structure, organized by small local groups and constituting not at all an exchange in a closed cycle, but in perpetual disequilibrium, which is to say, constituting a physical microsystem in perpetual instability. [Pause] And in effect, the alliances are constantly broken in favor of other alliances etc. It is at the level of alliances that you find the microphysics of power.

And I would say, in primitive societies, what is the relation of forces? The relation of forces passes by the network of alliances that is irreducible to the structure, irreducible to the filiative structure. What’s more, there is such an opposition between the two, it is so clear the network of alliances is not derivative and always calls into question the filiative structures, that the filiative structures will be reorganized according to the network of alliances at this or that moment. There will be an effect and, finally, the lines of filiation will translate only the state of the network of alliances. There is a microphysics of alliance at work beneath the macrophysics of filiation. In other words, there are series beneath the structure. See, then, how Foucault, who had nothing to do with this example, will be able to oppose strategy and structure. Saying: but a social field is not defined by a structure, it is defined by the set of its strategies. In a primitive society, the networks of alliance are really, yes really, the relations of forces.

And you see that this can involve violence, but that this does not entail it necessarily. A relation of forces is something entirely different. If I give you something, it is a relation of forces. We know it well, even in our society. If I give you something, it is a whole system of gift/counter-gift, but it would be idiotic to compare a relation of forces to violence. If I give you a present, eh… you see. I give you a present, you say: wow, what is he going to ask of me next? You say: no no no, I do not want it, you are too kind, it’s too much. You say: yes, yes, yes, I’ll take it. What a relation of forces! I press it in your hand, I put it in your pocket: yes! Keep it, keep it! – No, I don’t want it! It is a fantastic relation of forces, there is no violence. Well, there is no apparent violence. That’s the relation of forces. If force relations were a punch in the mouth, but that would be… the world would be so clear! [Laughter] But it’s not at all like that.

You understand, a man and a woman, if the relations of force were simply the moment when they say difficult things, when they sleep with each other… but the world would be a delight! But the force relations are disgusting precisely because it’s… it’s of this nature… I am doing this, ah, you ask me to do it? Well, yes, well yes. I am doing it, I am doing it yes yes! So, the other is infected. It doesn’t bother you too much? Oh no, no no… that is a relation of force… It’s filthy, true muck. That’s it, to do something all while showing how unhappy one is to do it, that one does it for the other. And there, you say “oh that’s nothing!”. And one starts coughing, “oh I’ve taken cold!” … “oh no it’s not your fault at all!”. All that which composes our everyday life, all these extortions, these disgusting extortions, eh… of which one is sometimes a bit ashamed afterward, but one says to oneself… those are relations of forces. So long live the moment when it breaks out into punching! We say: perhaps it will release something, eh… if not… well, that’s what force relations are like. So… societies work like that, strategizing, one has a strategy. A domestic quarrel is an incredible strategy, it’s a…. [Deleuze does not finish the sentence]

So, then, long live Tarde! Long live the microsociology of Tarde, you understand? Because to do sociology is precisely this, to grasp relations of forces, and not at all to do psychology. So all I am saying is that you see what Foucault might say – and I have taken this example of primitive societies, once again, to which nothing corresponds in his work – because he will be able to tell us that in modern societies, it is no different. There, too, the network of alliances infinitely overflows the large bodies like the State etc. The large bod…, if you take what will be called an important family, if you take the network of its alliances, you see how it overflows the institutions, the relations of forces overflow…

Take… you understand, we must make use of recent events; so, what is it that is most surprising and joyous in the Boutboul affair? [Laughter] It is a marvel, the Boutboul affair. It is a marvel, what do you see? Here you can make your transversal network: woman jockey [Darie Boutboul], good, that’s how I say it, woman jockey. A lawyer husband [Jacques Perrot], lawyer but mixed up in the races, so I make my transversal line there. Good. Husband murdered. Well, on that point, a stepmother, [Laughter] a strange lawyer, blacklisted [radiée], a blacklisted lawyer [Marie-Elisabeth Cons-Boutboul]. Ahh, she has relations with the… – and so the network branches out to infinity – with the Jesuit missions. [Laughter] These are the Jesuits, what are they doing in there? [Laughter] It is they who blacklisted the… So, the lawyer is dead, then who did it? Is it the stepmother who killed the lawyer, or is it the Jesuit seen by the stepmother, who says that the Jesuits are terrible assassins… Good: a cluster. Social assemblages are clusters. And it strategizes on all sides. The declarations of the stepmother, of Madame Boutboul, are declarations of high strategy, it’s clear. …Everyone strategizes, and then they show it to us on TV. It is truly a variety show in its pure state, [Laughter] it’s…. Oh! I forgot the dead father off in his corner! [Laughter] So how did he die? Forming his strategy, he says: “Oh, it is because I was separated from my wife, so it was not necessary for the girl to feel too much grief!” [Laughter] It’s very strange.

Claire Parnet: He made an even more bizarre declaration, he said: at 45 years of age, one is not the master of his actions like at 73 years of age, so we don’t know at what age it begins!

Deleuze: Ah yes, that’s good too. It’s in order to spare the girl pain that he passed himself off as dead rather than pass himself off as separated from his wife, that’s very interesting! That’s high strategy. Good, this is how a social field, a society, works. That’s right. All of this to conclude: a society is strategic. Understand that what this entails is really very important: power is not the property of anyone, but rather is the exercise of everyone. That is what “strategy” means: the Strategic Principle. And we saw and we begin to understand how strategy is opposed to structure, but also, more finely and profoundly, how it is opposed to strata.

Second principle. Localization. Which is to say, this is the second denounced postulate, namely, the tendency to localize power in an apparatus. And here, Foucault clearly has in mind one of the theses, from the same time period, of [Louis] Althusser, who, in a very beautiful text, showed or tried to show that even private powers had only the appearance of dispersion and actually functioned like special State apparatuses. Foucault is literally the opposite of this. I do not pose the question here of what brings about this difference between Althusser and Foucault because it would be necessary to go back to how Althusser himself posed the problem. But I want to just retain Foucault’s point of view, namely, that even public powers have an origin, have procedures, have exercises, that is, everything essential to power, which the State no doubt supervises or covers, but which it does not constitute. You see in what sense this is the anti-Althusser, Althusser saying: even the private powers are special State apparatuses. Foucault says: even the State apparatus only administers procedures of power that come from elsewhere. And, in effect, you have guessed why: if the State apparatus is a constituent element of the large aggregates, if it forms large aggregates, then by definition, the State apparatus presupposes relations of power; it does not explain power relations. It presupposes relations of forces that come from elsewhere.

Where Foucault is very strong, of course, it is because he has very concrete examples. He says: take a cog of the State apparatus like the police… It would be all too easy if he took more ambiguous institutions. But, if there is something that belongs, seems to belong to the State and to be a part of the State apparatus, it is the police. This is the understanding of all those who define the State as the monopoly on force. Well, Foucault says: well no. If we concretely consider policing techniques, we see that the procedures of police power are by nature, not by exception but constantly and always and even originally, covered by the State, reutilized by the State, but that the State is absolutely not the origin of them. The police has its own techniques of power, its procedures of power. You will say: sure, but the police, what does that mean? Well, there are sources of policing that reveal themselves already at the level of communes without a State apparatus.

You will say: yes, but there is the commune, there is always another power. No, not necessarily. Not necessarily. And in any case, even when there is another power, the police is measured by its independence in relation to other sites of power. The police force is itself an autonomous site of power, and it always has been; this is how it purports to function. So, the State, in fact, can globalize police power or can appropriate for itself police powers. Police powers do not come from the State. In the same way, he will show why. Because police powers are disciplinary powers, and the disciplines always preceded the moment when the State appropriated them. The State appropriates the disciplines, it is not at the origin of the disciplines. The disciplines of the school, the disciplines of the army, the private disciplines, the disciplines of the Church etc. always preceded the State. Disciplinary techniques are recuperated by the State. They do not find in the State their origin.

And in Discipline and Punish, Foucault will show the same thing, especially with regard to the prison. As we have seen, the prison was not part of the horizon of penal law; it is irreducible to juridical power, relatively independent of juridical power, having a power proper to itself, the power of the prison or power in the prison. Power in the prison, where the judge and right have nothing to do with what Foucault will call a “disciplinary supplement,” in that disciplinary technique, and where the State serves the prison more than the prison being a cog in the State apparatus. [Pause] In other words, you see, there, power does not admit of being localized in an apparatus. We must pay attention to the word “local” because Foucault will use it in two senses. There is no contradiction between the following two kinds of phrases in Foucault: power does not admit of being localized, that is, it is diffuse, this is what that means, power does not admit of being localized, it is diffuse, that is, it swarms throughout the whole social field; and the other kind of Foucault phrase, that power always consists in local sites, the affirmation of the local character of power, which signifies this time that power is never global. And the two are perfectly coherent. It does not admit of being localized because it is diffuse; however, it is always local because the global is what the large aggregates are, and relations of power operate below the large aggregates.

Third postulate denounced by Foucault: postulate of subordination. This is the postulate according to which power would be subordinate to a mode of production as infrastructure. [Pause] This is to say: there would be relations of production that would define an infrastructure, and power, which would be only political, would express this infrastructure, whatever the complexity of that expression. Foucault’s idea is very simple: you cannot speak of relations of production without already placing there, and without interlacing them with, relations of power. Thus, the relations of power obviously do not stem from relations of production, because there are no relations of production definable independently of their interlacing with power relations.

And in fact, let’s return to an example in our primitive societies. The relations of production are strictly inseparable from relations of alliance, from the transversal network which we spoke of earlier, since it is this transversal network of alliances that will determine who works for whom, who the members of such-and-such a filiation are that work with what members of such-and-such other filiation. It’s simple. Thus: refusal of every subordination of power relations relative to so-called economic relations or to relations whatever they might be.

Fourth postulate. Postulate of the essence or attribute. Power would have an essence and would be an attribute, and, qua attribute, it would qualify those who possess it as dominant, distinguishing them from those over whom it is exercised: the dominated. That is the postulate of the essence or attribute. Foucault’s response – here, too, we are ready and waiting, now I can go quickly – is that power has no essence but is functional, operational. It has neither essence nor interiority. And why? It is not an attribute, but rather a relation, and it is because it is relational that it is not an attribute. In other words, the power relation is the set of force relations in a social field, and therefore power passes by way of dominated forces no less than by dominant ones.

Power passes no less… here we must reflect just a little bit on what he means. Power passes by way of dominated forces no less than by dominant forces. You see: it is not an attribute that distinguishes the dominant and the dominated, but a relation that relates the dominant to the dominated and the dominated to the dominant. Hence the text from The History of Sexuality Volume One or from Discipline and Punish, I no longer know which: “power invests the dominated, it passes by and through them, it finds support in them, just like they themselves, in their struggle against it, find support in turn in the holds that it exercises over them” [Discipline and Punish]. And perhaps here is one of the most precise examples, a moment ago I invoked the prison for the preceding postulate, now one of the most precise examples analyzed by Foucault is the example of the lettres de cachet. About which I have already said a few words to you. And the lettres de cachet – which are an institution truly proper to the French monarchy, I believe, following Foucault, there is no equivalent during that period – seem typical insofar as they indicate a transcendence of power. And they are often presented as being the expression of the pure arbitrariness of the king, where, independently of any procedure of inquiry, the king decides on imprisonment, the internment of someone, and no doubt that was true in several cases, notably, in the case of the great lords. But the technique of the lettre de cachet is completely different, I said.

And what does the technique of the lettre de cachet demonstrate? That the true process of the lettre de cachet is this: a member of the family, a neighbor, a colleague, an officemate, whatever, sends a request and says: Mr. so-and-so is completely nuts, completely crazy, and must be locked up …, put him in prison. My lord, put him in prison.” After which there is an investigation. In other words, it’s not the arbitrariness of the king, but the way in which the dominated participate in the arbitrariness of the king. Power passes through the dominated as much as through the dominant. The lettre de cachet is fundamentally demanded by the families, the neighbors for every creature who causes a little trouble, that is, against whom one cannot unleash a judicial procedure.

It must be a small trouble, in the manner of Tarde, a micro-trouble. If it was a large trouble, if he had committed a crime, the normal procedure would be used. The “lettre de cachet” procedure is made for the microphysics of offense, for the infractions, the small infractions that are not punishable from the legal point of view, and Foucault gives an example of a lettre de cachet, very moving incidentally, the letter of…, the petition sent by…, what was her name, ah… well I made a mistake.

Ah no, I did not make a mistake. The lettre de cachet, I lost it … — ah! The wife of Nicolas Bienfait. Madame Bienfait, who, in the seventeenth century… Here it is: “I take the liberty to very humbly point out to your Royal Highness that the said Nicolas Bienfait, livery coachman, is a most wanton man who is beating her to death and others and who is selling everything, having already caused the death of his two wives, the first of which had her child die inside her body, the second, after his having sold and devoured her by his ill-treatment” – I do not understand that, no matter – “did die in languor even going so far as to want to strangle her on the verge of her death.” There is nothing condemnable by the law in all that; the lady cannot do anything. “The heart of the third, he wants to eat on a spit, not counting all the other murders he has committed. Your Royal Highness, I throw myself at the feet of Your Majesty in order to implore Your Mercy. I hope you will render me justice from your goodness, for my life being at peril all the time, I shall not stop praying to God for the preservation of your health.” [10]

Now, as I was telling you, what happens today when one undertakes a so-called “voluntary commitment”? A woman is massacred by her husband, what can she do? Well, she requests a lettre de cachet, that is, she requests that the guy be locked away if he is… She requests that the guy be locked away, there is an investigation, the police come to see the concierge, the lettre de cachet in its pure state, and then he is locked away if the police investigation… There is a psychiatrist who comes, all that. Good. Power passes. Power is in a force relation, passes through the dominant, through the dominated, no less than through the dominant. This is the postulate, then, the denunciation of the postulate of the attribute: power is not an attribute, but a relation.

Fifth postulate. Postulate of modality. In many classical theories, power is treated as though it had two modalities: either this, or that. Either it proceeds by violence and is repression, or it proceeds by ideology. Repressive or ideological. [Pause] Repression or ideology, one gets the sense that this is a very impoverished alternative, because what Foucault ceaselessly shows is that power can act on souls and on bodies, but even when it acts on souls, it acts otherwise than by ideology. And even when it acts on bodies, it acts otherwise than by violence and repression. If we had to wait for power to be repressive, you know… for quite some time there would no longer have been power, it proceeds entirely differently. It is repressive in the last instance, yes, when it cannot do differently, but otherwise it can very well do without being repressive [mais sinon il se passe très bien d’être répressif]. It has more subtle means that are neither ideology nor repression. Why? Here is perhaps the most essential point, but we will have to return to it; for now, I will indicate it, introducing it very quickly.

The force relation is not violence. Why isn’t it violence? Because the relation of forces is the relation of force with force. This is very important, because if I say that the force is essential in the relation, this means that there is a reason, in the concept of force, for why force is never by itself, for why force always forms part of a multiplicity; it is a property of force to be set in relation to another force. Therefore, every force is a relation of forces. There is not one force, there are relations of forces. What is violence? Is it a relation of forces? No. Violence is not a relation of forces. That is, violence is not the relation of a force with another force. What is violence? Violence is the relation of a force with a being or an object.

And this is fundamental, obviously, to the analysis of Foucault, who does not even insist on it because, for him, it is so self-evident; but still, we must make it clear. It is… and yes, when I suffer violence, it is not my force that suffers violence, it is my body, which is perhaps the seat of a force, my own, but what is destroyed by the force? It’s not another force, another force is not destroyed by force; what would it mean for a force to be destroyed by another? A force is quite incapable of destroying another force. On the other hand, a force can very well destroy a body, yes, a being or a thing. The force of a bomb destroys a city as a body or a living being as a body. Violence expresses the relation of a force with a thing, an object, or a being.

What do we call a relation of forces? It is the relation of a force with a force. What is the relation of a force with a force? Foucault will say in an interview – though he does not develop this point in his books – that the relation of a force with a force is an action on an action. It is not an action on a body, but an action on an action, the latter action being real or possible. Force is an action on a real or possible action. It’s quite precious, we’ll have to return to this. An action on an action is not violence, never does violence act on an action; violence is exercised on the support of an action, on the subject of an action. Violence does not act on an action.

What is an action on an action? He gives some examples in an interview that is republished in the Gallimard edition of [Hubert L.] Dreyfus and [Paul] Rabinow’s book on Foucault.[11] He gives examples that are very unusual to us, but this will be precisely the opportunity for us to take a look. He says: relations of forces are not at all of the type “do violence” or “repress”, but rather: incite, provoke, combine. This may seem like nothing, but we will have to take a closer look. You see what he means at first sight: a force does not do violence to another force, but a force can incite another force. A force can combine other forces. Those are indeed relations of forces. Thus, the relations of forces are not at all of the type “do violence,” but rather: incite, provoke, combine. We will leave this for now, it’s too obscure for us when we don’t yet have the means to give a commentary on it. We will have them soon. So, we are leaving it to the side.

And I say to myself: in Discipline and Punish, what are the relations of forces that he studies? If you take the whole of Discipline and Punish, I will try to make a list. In Discipline and Punish, I would say: what do force relations do? Organize space, namely, like … “lining up” in school. Line up in school, that is, straighten up, enclose, divide up, serialize. To make a series is not the same thing as lining up. To serialize, for example, is to make a list for a composition: first, second, third. There you have it. In other words, there you have a first aspect: distribute space. The functions of which Foucault speaks in Discipline and Punish, the title of the first great function: distribute space, enclose, divide up, straighten up, serialize.

Second great type of function: put time in order. This is to subdivide time itself, hour, half-hour, minute, second, program the act, program an act, decompose a gesture. These are prescriptive functions in time, no longer distributive functions in space. You will tell me: where does this happen? Well, the place it happens is quite evidently the workshops, the first factories. The decomposition of the gesture, mechanized labor. It is a prescriptive function in time. And well, it’s … there is perhaps an intense underlying violence, there are no lashes, the workers were not formed by lashing the whip, in a sense it was worse.

Third great function: to compose in space-time. And this time, what is it to compose in space-time? It is to produce a useful effect that is greater than the sum of elementary forces. To produce a so-called “useful” effect that is greater than the sum of composite forces. There it is. Those are relations of forces, not repression and ideology.

And this links up again with History of Sexuality Volume One; you remember our whole theme, what was there in History of Sexuality Volume One? If we remain at the level of words and phrases, we can always believe that sexuality was repressed. But if we know how to read, if we know how to extract the statements of the period, we will perceive that on the contrary, sexuality was perpetually solicited, invited to express itself and to speak. On what condition? Precisely on the condition of letting itself be distributed in space, ordered in time, and composed in space-time. This is what Foucault will translate by saying in a nutshell: modern societies proceed not by ideology or repression, but by normalization. What is it to normalize? To normalize is the relation of forces par excellence, namely, to distribute in space, order in time, compose in space-time. Hence the denunciation of the fifth postulate.

Finally, there will be a sixth postulate denounced by Foucault, the postulate of legality. The link that most theories make between the State and the law. And since that one is more complicated than those preceding it, we will save it for the next time. That’s the last one. [End of the recording] [2:59:44]

 

Notes

[1] Deleuze is no doubt referring to his friend and colleague François Châtelet who died during the Christmas break, 26 December 1985.

[2] Of the different translations in use for l’énonçable, notable, the enunciable or the articulable, we opt for the statable.

[3] Brisset, La Grammaire logique (1878), re-edited at Éditions Tchou (1978).

[4] While this title translates easily as “the order of discourse”, that is not the title under which this important work by Foucault (Paris: Gallimard, 1971) is known in English: see “The Discourse on Language” published as an appendix (pp. 215-237) to The Archeology of Knowledge (New York: Harper & Row, 1972), translation of the former by Rupert Sawyer, of the latter by A.M. Sheridan Smith.

[5] Deleuze will develop these political perspectives briefly in Foucault, pp. 89 and 115, and p. 144 note 28 and p. 150 note 45, and at great length later in this seminar, notably in session 19, April 15, 1986, session 23 (in which Félix Guattari participates and intervenes), May 13, 1986, and session 24 (in which Éric Alliez intervenes on Italian Marxism), May 20, 1986.

[6] See the reference to Brisset earlier in this session.

[7] Deleuze will return to this distinction regarding Dada and Surrealism in “P as in Professor,” in L’Abécédaire de Gilles Deleuze.

[8] The references in the text are to the translations; in Foucault, Deleuze provides different page numbers for Discipline and Punish: “Only a few pages, since he adopts a method that is completely different from the ‘thesis’, He is content to suggest abandoning a certain number of postulates which have traditionally marked the position of the left” (pp. 24-25), with footnote reference DP 26-29, and SP 31-33. He concludes this comment by saying, “And we have to wait for The History of Sexuality for a more detailed exposition” (p. 25). On the six postulates that Deleuze will outline further on, see Foucault, pp. 24-30.

[9] What complicates this anecdote are several factual details: the historical event of the Matignon Agreements took place in 1936; Gabriel Tarde died in 1904, but certainly could have known and influenced Maurice Hauriou, a jurist and law professor who died in 1929. The younger Hauriou, however, a leftist aged 39 in 1936, could well have been part of the Agreements. As Deleuze concludes here, “no matter!”

[10] “The Life of Infamous Men”, 1977, transl. Paul Foss and Meaghan Morris, in Michel Foucault: Power, Knowledge, Truth (Sydney: Feral Publications, 1979), p. 88.

[11] Michel Foucault, Un parcours philosophique (Paris: Gallimard, 1984; Michel Foucault, Beyond Structuralism and Hermeneutics (University of Chicago Press, 1982, 1983); see “On the Genealogy of Ethics” (interview with Foucault conducted in English), in Foucault (ed. Paul Rabinow), Ethics: Subjectivity and Truth, The Essential Works 1 (New York: The New Press, 1997).

Notes

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 revisions added in May 2022 and also February 2024.

Lectures in this Seminar

square
Reading Date: October 22, 1985
right_ol
square
Reading Date: October 29, 1985
right_ol
square
Reading Date: November 5, 1985
right_ol
square
Reading Date: November 12, 1985
right_ol
square
Reading Date: November 19, 1985
right_ol
square
Reading Date: November 26, 1985
right_ol
square
Reading Date: December 10, 1985
right_ol
square
Reading Date: December 17, 1985
right_ol
square
Reading Date: January 7, 1986
right_ol
square
Reading Date: January 14, 1986
right_ol
square
Reading Date: January 21, 1986
right_ol
square
Reading Date: January 28, 1986
right_ol
square
Reading Date: February 25, 1986
right_ol
square
Reading Date: March 4, 1986
right_ol
square
Reading Date: March 11, 1986
right_ol
square
Reading Date: March 18, 1986
right_ol
square
Reading Date: March 25, 1986
right_ol
square
Reading Date: April 8, 1986
right_ol
square
Reading Date: April 15, 1986
right_ol
square
Reading Date: April 22, 1986
right_ol
square
Reading Date: April 29, 1986
right_ol
square
Reading Date: May 6, 1986
right_ol
square
Reading Date: May 13, 1986
right_ol
square
Reading Date: May 20, 1986
right_ol
square
Reading Date: May 27, 1986
right_ol
square
Reading Date: June 3, 1986
right_ol