December 21, 1982

In fact, once again, the sign will be the image which will represent either the composition or the genesis of the type of image to which it refers. So, this definition has nothing to do with Peirce’s, and yet we will find common signs, it’s perfect. If I go on at length, on the contrary, I would say, there are cases where I can conceive of five signs per type of image. Indeed, the sign of composition will be divided into three, depending on whether the two poles are supported or one rather than the other. So, in this case, I will have three signs of composition. And then, the sign of genesis will divide in two, eventually, the sign of genesis and sign of extension. So, I will have at least two signs per type of image: zeroness, oneness, twoness, a maximum of five signs per type of image. Fine.

Seminar Introduction

In the second year of Deleuze’s consideration of cinema and philosophy, he commences the year by explaining that whereas he usually changes topics from one year to the next, he feels compelled to continue with the current topic and, in fact, to undertake a process of “philosophy in the manner of cows, rumination… I want entirely and truly to repeat myself, to start over by repeating myself.” Hence, the 82-83 Seminar consists in once again taking up Bergson’s theses on perception, but now with greater emphasis on the aspects of classification of images and signs drawn from C.S. Peirce. This allows Deleuze to continue the shift from considering the movement-image, that dominated early 20th century cinema, toward a greater understanding of the post-World War II emphasis on the time-image.

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

 

Antonioni’s “The Passenger”

With the previous outline of Peirce’s categories of Firstness, Secondness, and Thirdness fully linked to types of images, Deleuze now reflects on how an image becomes a sign. Following Peirce’s terminology of the sign – “represantamen”, object, interpretant – linked to the three categories, Deleuze specifies nine classes of signs. Deleuze then enhances Peirce’s categories, particularly regarding Thirdness and adds a “zero degree” (Zeroness) or the perception-image, with Firstness then linked to the affection-image, Secondness to the action-image, and Thirdness to be determined. After deriving successive definitions of the sign, Deleuze offers several examples, first in light of the quality or power within the affection-image as regards faciality, and he undertakes a parallel list of signs as a sign of composition of the affection-image. In terms of the face’s two poles, Deleuze multiplies this list into facializing contour and the aggregate of discontinuous features (hence “contour icons” and “feature icons”), and as signs of genesis, he identifies the any-space-whatever (espace quelconque), with two types as well (“qualisign of deconnection”, “qualisign of vacuity”). The session (and year) ends with the new graph’s outline, to be taken up following Christmas break.

Gilles Deleuze

Seminar on Cinema: Classification of Signs and Time

Lecture 6, 21 December 1982 (Cinema Course 27)

Transcription : La voix de Deleuze, Franck Baron (Part 1), Una Sablajakovic (Part 2), et Franck Baron (Part 3); additional revisions to the transcription and time stamp, Charles J. Stivale

Translation, Charles J. Stivale

 

Part 1

… For the return to classes, right? And since you’re not going to remember, that’s going to cause problems, but you really have to remember, right?

So, there you have it: we are in this story of Peirce, and what we saw is, you remember, the distinction of three types of images in his work. And these three types of images for him, it is Firstness that corresponds, roughly, as we know in advance, to our affection-image; it is Secondness that corresponds, roughly, to our action-image; and it is Thirdness that corresponds to what? At the moment, we don’t know; from our previous session, we are just retaining that this Thirdness has for characteristic the domain of the mental according to Peirce, what he himself calls the “mental”, that is, “something thought” [le pensé], the domain of “something thought”, all that is “thought”, and that Peirce expresses it under the example sometimes of the law, a law, sometimes of signification, sometimes of the relation. And we had seen that the best way perhaps to define this mental domain of Thirdness was the relation taken in its two senses — and this is what I had tried to develop the last time – namely, in the sense of concrete or natural relation — first sense — or in the sense of abstract relation — second sense.

And here I insist on this summary so that you have that truly aggregate in your mind; this distinction seems to me very important of two meanings of the word “relation” that many other philosophers than Peirce — which still does not appear in Peirce, but that does not matter, it’s of no importance – that many philosophers have sensed and indicated, and I was saying, it would seem to me to correspond to something quite concrete when we speak of relations. Once again, all relations are external to their terms, no problem. But there are relations which consist of this: how a given image leads naturally — understand, effortlessly — to another image which is not given, which was not given. An example once again: a photo reminds me of, the photo of Pierre, reminds me of Pierre, and as Hume said, “smooth transition”, that’s what we would call a natural or concrete relation. The resemblance here acts, the resemblance between the photo and the model acts like a concrete relation.

And we noticed that there is all the more reason to call it natural or concrete since these effortless transitions do not go far. They scatter very quickly, or else they are interrupted. A natural relation works between two things that resemble each other greatly. But as we have seen a thousand times, anything looks like anything else; it’s enough to find the point of view, and there you don’t have a natural relation. The photo of Pierre reminds me of Pierre, Pierre reminds me of Paul who looks like him. But from resemblance to resemblance, the natural relation will quickly deteriorate. And there, it will no longer work. There will be no more transition; there is indeed a “smooth transition” from A to B, from B to C, from C to D, and therefore at the extreme, from A to D, but beyond: that no longer happens; natural relations wear out quickly.

Whereas you remember the abstract relation, it is not a “smooth transition” leading us from a given image to the image of something that was not given. The abstract relation is a circumstance which determines a comparison of the mind between two images which are not relatively related. If I say, for example, between A and X, what resemblance is there? Since anything looks like anything, there will always be a circumstance whereby any two things whatsoever will be alike, ultimately, because they are things. Good. And the circumstance or the reason why I compare two ideas, two images, which are not in natural relation to each other, that is an abstract relation. See, there are therefore two meanings of the word “relation”.

So, I have that, we’ve reached this point: Firstness, Secondness, Thirdness according to Peirce. These are the three types of images. On that point, the problem in my opinion, it seems to me, is a very, very important one: how does an image become a sign? And this problem, as Peirce creates this new discipline — semiotics or semiology — this problem must be fundamental for Peirce. For after all, every sign is an image, but every image is not a sign. We will call “sign” certain special images.

So, see that “sign” launches us into a whole new element to discover, into a whole new world. I have my three types of images, fine, one, two, three, but then the “signs”, and why is there the need for “signs”? Well, I say to my amazement, or in general amazement anyway, Peirce doesn’t care that much about this problem. You really have to look through the pages, and yet there are pages where one can discern his conception: “why do images give rise to signs?” and “what is the difference between an image and a sign?” And it seems to me that for him, there are two answers: he always tells us, well, see, a sign is always a Thirdness. Ah well, that will complicate things because if a sign is a Thirdness, it is because, in a sign, there are “three”. In any sign, there are “three”. But it is out of the question to confuse Thirdness and sign, in the sense that any sign is Thirdness, but any Thirdness is not a sign. Thirdness is a type of image. It is not by itself a sign.

So, if it is true that every sign is a Thirdness, the question becomes: What type of Thirdness will require, demand to be called sign? And Peirce’s first answer is that Thirdness is the domain of the mental, of something thought, but something thought does not therefore constitute knowledge. All that is thought does not constitute, does not form knowledge, and that the sign is a Thirdness constituting knowledge. [Pause] In other words, it is a Thirdness since the domain of Thirdness is relation, well, he tells us, the sign is a Thirdness which constitutes a cognitive relation. [Pause]

Fine, we tell ourselves fine, [Pause] but at that point, this is a program to be fulfilled. He gets a little more specific when he tells us this: well, a sign is a Thirdness that makes relations efficient, but this doesn’t explain much. Here, I’m already more interested. Notice that the two are equal, because maybe making relationships efficient is an excellent definition, perhaps, of the act of knowing. It would be an original and perhaps very satisfying definition, which would answer the question: “what is knowledge?”. Knowledge is an operation that makes relations efficient. Good. What does that mean?

This implies that relations, you remember, they belong to the image of Thirdness, they belong to Thirdness, but by themselves, they are not efficient. Why? Abstract relations! Concrete relations that naturally shift from one image to another image, are they efficient? Not exactly. In Peirce’s vocabulary, we should say that they are put into action, they are put into action, they act, that is not what efficiency is. And abstract relations, they are inefficient, in fact, as long as I grasp them as a pure image of Thirdness. I see fit, a circumstance by which I see fit to compare two ideas, two images arbitrarily united. The very definition of the abstract relation, they are not efficient. Concrete relationships act, they cause the mind’s movement from one image to another, but they are not efficient: they act. They are active, or if you will, they are acting [agissantes]. And abstract relations, they’re inefficient. I am comparing two ideas arbitrarily united in the imagination.

What does it mean, to make relations efficient? I guess, although he doesn’t explain much about it, I guess that means making them operative. And that’s why I attach a lot of importance to what I suggested the last time, that it is impossible to think the relation; this is the only contribution here that I would like to offer you. For the moment, there, in this perspective, impossible to think … if philosophy has failed in, or has encountered so much difficulty in an effort to think the relation, it is because we have not seen enough or not said enough that the relation strictly implicated, enveloped change, because we did not see that the relation was in the domain of becoming. I am stating a platitude, but it’s true, this is how it is. If we see that the relation is inseparable from becoming and change, there I think it becomes much easier to think it.

But that does mean that the relation itself is, how to say, dependent on a system of relationships between relations such that relations can be transformed into each other. I would say that at that point, relations become operative. What does science do? What does math do? They make relations operative. Not just any situation… and not just any relation maybe; relation of situation, one could even distinguish branches of mathematics, relations of situations are made operative; quantitative relations are made operative; linear relations are made operative, whatever you want. We could define mathematics as the first system which makes relations operative or efficient. All right, so let’s accept all that for now.

And we have a definition of the sign which is: the sign is an image of Thirdness defined by this, that it makes relations — relations referring to Thirdness — that it makes relations efficient. Oh, but here we have a definition like that. I mean, this is a purely nominal definition. So concretely, how will a sign be composed? One must discover some Thirdness in it. Well, yes, Peirce tells us. What will be the real definition of the sign? When I say: the sign is a Thirdness which consists in making relations operative, I only have a nominal definition. I am not saying what the sign consists of.

If I look for a real definition which states what the sign consists of, very well, what Thirdness will the sign consist of? Peirce tells us, well, a sign is an image: one, two, which is valid for another image; [Pause] three: through other images that relate the first to the second. This all sounds annoying, but it seems to me, that’s what a big, big part of modern philosophy has been built on. That’s why it’s important, this kind, to me, this kind of return to Peirce. Let me explain: you see that in this real definition of the sign, what is important is that, in fact, it is clearly a Thirdness. It is an image which is valid for another image by the intermediary of other images which refer the first image to the second.

Let’s explain ourselves better with the words Peirce uses. The sign is an image; in itself, it is an image; this is its aspect, I would say: this is the Firstness of the sign. This is the sign in itself. He calls it: “representamen”. [Pause; Deleuze writes on the board] It is the sign in itself, the Firstness of the sign, the sign considered as an image. This is valid for another image. This other image for which it claims to be valid, [Deleuze writes on the board] there he calls it — curious, this alternation of bizarre words and quite ordinary words, he plays on it a lot; we will have to ask why as well — he calls it: “the object”.

Understand that the object of a sign does not need to be real. A sign is an image, and its object is another image. Sometimes the object is a real object, sometimes the object is not a real object. The drawing of a Centaur is a sign that applies to an object, an object which is not real. [Pause] Fine. That, I would say then, the object, that is, the image for which the sign claims to be valid, is the Secondness of the sign.

Third, through the intermediary of other images, which relate the first, the “representamen”, to the second, the object, [Deleuze writes on the board] and that, he will call it, and that’s what poses such a problem when, [Deleuze writes on the board] when we started to know Peirce, his texts, he calls that “the interpretant”. This is the interpretant of the sign. The images that relate the first image to the second… What?

A student: In general, people who read Peirce never understand it like that; they always manage to understand something other than that.

Deleuze: Yeah, yeah. You see what a misunderstanding it would be to believe that the interpretant is an interpreter. Peirce has always, and some of his followers in America, rushed in that direction immediately. For what purpose? To create a psychology of the sign. But Peirce is a logician who, as such, does not like psychology at all and claims to reject all psychology. The interpretant has nothing to do with an interpreter. The interpretant is the set of third images that relate a first image to a second image for which the first image claims to be valid, but such that it could not have this claim without the interpretants. It is this system that makes relations efficient. So obviously, phew, an example; let’s select an example. See already the importance of this triadic system because if you compare it, roughly, with a contemporary of Peirce who is well known to him in France, [Ferdinand de] Saussure — he is not French, but he is well known in France. [Laughter] — Well, it is well known that the whole theory of the sign from Peirce is based on a kind of dual distinction: signifier-signified…

A student: From Saussure!

Deleuze: From Saussure, sorry… I said … Yes! Notice that it is already relatively important for us that Peirce absolutely cannot fit into this, into such a lineage. Hence the importance of his idea of Thirdness. There is no place for the signified, the signifier, all of that. It’s, it’s a whole different atmosphere, okay. But phew, some examples, really, I mean, we ask for examples. The French presenter of the book, about whom I told you how excellent he was, gives one [example], but one which is obviously worrisome since he will select a linguistic example. And I too, I remind you, I am very attached to the idea that for Peirce there is no difference in nature between signs which would be natural and signs which would be conventional.

So, we’ll need at least two examples; it’s up to us to find another. Following the example of that commentator, of [Gérard] Delédalle, he chooses a word, and he says: let’s take the word “grenade”.[1] There it is: this is a sign. [Deleuze writes on the board] In what sense? The “representamen” is the word considered in itself, what linguists would call: “the verbal image”, you know. The word claims to be valid for something, this something being another image. Ah yes, but what? What? I can tell the object, well, it’s a city. [Deleuze writes on the board, no doubt “Grenada”] In other cases, I can say, the object is a fruit. [Deleuze writes on the board] In yet other cases, I can say the object is a weapon.

What makes me choose? What makes me select are the interpretants. Namely, for example, an entire verbal context for which it is a question of Spain, Spanish cities, what there is in Spanish cities. I situate myself with this series of interpretants, I situate myself and I say, when I hear the word “grenade”, the object is the city! — [Deleuze speaks in a very low voice] I hope it’s in Spain [Laughter], I’m wary, I’m wary of everything now, fine — If I hear, however, if I have a series of interpretants of the type: demonstration, police, etc., I tell myself: this is the weapon, this is the weapon, so if I hear “a Grenada grenade”, I have a short circuit of two sets of interpretants, fine, etc. See what Peirce means when he says the “one” of the sign only relates to “two” through the intermediary of interpretants, only through the intermediary of three. One can say a very simple thing. In fact, if I remain at one, two, the subject will necessarily be incomplete to say the least. Or else, it will remain indeterminate, that is, disjunctive. It is either a city, or a fruit, or a weapon, or it will remain incomplete. It is therefore only the three, that is, the interpretants which completely relate one to two and, thereby, make the relation efficient, operative. So, he is in the process of keeping his bet.

I am saying, I need an example of natural signs because I care greatly about, because there is no difference between natural signs and conventional signs. It’s fine if you understand that; that already removes all kinds of linguistic problems. There are differences, but they arise from entirely others, at very low levels, really… very… There, there I am walking, I am walking in the forest, and I see a track; a track is a sign. I am saying, the track in itself, the track in itself, is not a “representamen” in Peirce’s language. It is a sign, the object. I say: this track was left by an animal; an animal passed that way, good. The animal image is the object of the “representamen,” the track. Alright, yeah. And then an animal is the object, but the incomplete object.

What will I need? Well, for example, it will take a string of interpretants, not mammals: birds. [Deleuze writes on the board] I am excluding, see, a whole series that would lead me to a mammal. No, it’s a bird [Pause] that I link with other interpretants. The track sends me back to …, [Deleuze writes on the blackboard] if I am very, very clever, it was made at night; this is not a daylight track. So, I already have a string of interpretants: bird, nocturnal. Fine, that can go on: raptor, man-eater, [Laughter] etc., etc., etc. See that here too, my object remains incomplete if there is not the interpretant and the chain of interpretants which brings one to two. So that’s what a sign is for Peirce.

So, I am indeed saying, for Peirce, a sign — we have nonetheless made a lot of progress here — it is a special image in the sense that nominally, it makes… It is still a Thirdness. Nominally, it makes the relations which belong to Thirdness operative or efficient; they really consist of a particular organization. It is a Thirdness which consists in this and only in this, and it is only in this case that we will speak of a sign: an image which is valid for another image through the intermediary of other images. OK? All this is understood because from this will emerge, will obviously emerge — [Deleuze is distracted by something and speaks in a low voice] no, it doesn’t work? — [His attention returns to the subject] So fine. What? … Do I want another example? [Pause] If it’s very short, sorry, because I would really like to move on to… [He speaks to a student] Yeah, give us one if you want … Yeah?

A woman student: The French flag, as you know, is blue, white, red…

Deleuze: No, I didn’t know that. [Laughter]

The student: Blue, blue represents both freedom and nobility or the bourgeoisie become noble, well, blue, or today the right. Red represents both equality and the people. And white represents the king. This is how the flag was made up during the French Revolution. That is, the flag was constituted in such a way that the king or now the State permanently maintains the balance between the rich and the poor, or else the aristocrats plus the bourgeois on the one hand and the people; they manage everything they can so that things aren’t disturbed, and that the State is a stabilizer of society. And if the flag was really: liberty – equality – fraternity, as we say in France, it should be blue – yellow – red, and at that point, we realize that the three colors are composed in white and that the whole people become king. [Laughter]

Deleuze: Why not? But in my opinion, this is a Thirdness, that is not a sign! The flag is indeed a sign, but in your story, from a completely different point of view, but in your story, blue – white – red, you cannot situate a “representamen”, an object and interpretants…

The student: What…, but yes, yes, I can, it’s obvious: the “representamen” is the flag, right? The object is the political balance that is chosen to be maintained, well, within society, and the interpretant is how all this is constituted, as we know, for example, that the colors in Freemasonry were blue and yellow, that is: freedom plus fraternity …

Deleuze: Yeh, well yes, your example complicates things. Well, OK, if… OK, OK. But it’s complicated.

A student: But ultimately, this example is still important because there are examples where one proceeds through concepts, so this is practical. But when we are, when we are faced with an analogous code, an analogous expression, it’s always fragmentary, really, or else the expression has to be very formalized.

Deleuze: Yeh, yeh, yeh.

The student: … otherwise the series, the interpretant series…

Deleuze: Yes, but then even if it meant formalizing it, it would have to be done symbolically; we couldn’t do it anymore, for example … in fact, that’s what Peirce does.

So suddenly, you understand, his list of signs, you will see in which … Understand immediately in which thing he is located… He is facing… I will write here… [Deleuze writes on the board] This is why it is very important here. I ask those who are really following this to note down this graph because we will really need it when I reach that point… of what I would like to do here and that … [Deleuze writes on the board] Here’s where we are. On one hand, I have Firstness, Secondness, Thirdness. I can say, these are the three cases of the movement-image. But finally, since we are within Peirce, there is in his vocabulary, he calls it, I stated it from the start, not “phenomenon”, he feels the need to select another word so as not to be confused with phenomenology, he calls this: “phaneron”, the “phaneron”. Fine, as we have seen, the word “phaneron” interested me since, finally, it seemed to me to be translatable by “light-image”, and then we had seen that light-images and movement-images, for us, were the same. So anyway, it’s basically the world of phenomena or images, the image: it’s the phenomenon, it’s the “phaneron”, it’s whatever you want. Well, so the image, right? According to Peirce, three types: Firstness, Secondness, Thirdness.[2]

If I now create this very special type of image in the other direction, the sign, then it is no longer images, it is signs. I see that they have a special Firstness: the “representamen”; a special Secondness: the object; a special Thirdness: the interpretant. The sign is itself a Thirdness since, as we have seen, its Firstness and its Secondness presuppose the Thirdness, which is proper to all Thirdness. Fine, to have the graph of signs, well, the simplest graph — because this will get complicated, but we have to… us, we will not be concerned with it anymore, because… and still… it is enough … — And we can predict how many large classes of signs we will have. [A very long pause, Deleuze writes on the board] In other words, I am entitled to expect nine classes of signs… So, let’s start with… What am I going to say?[3]

My question… So, I am considering the sign in itself. In other words, I ask the question, for this whole heading, this whole slice, I ask the question: in the sign, what “yields a sign” [fait signe]? What “yields a sign”? [Pause] My answer is:

First case: what yields a sign can be a pure quality, a pure quality. If this is the case, I’ll call it a “qualisign”. When a sign in itself is such that what yields a sign is a quality, we find ourselves in the presence of a qualisign. You will tell me: there is no such thing in advance as a pure qualisign. Maybe, maybe not. If there is no pure quality, it doesn’t matter. We will call qualisign those in which this aspect dominates. It is even probable that these are pure signs which are never realized. What is realized are combinations of signs. And what is very important is that some combinations cannot be realized. Good. What yields a sign is a quality; in this case, I will speak of a qualisigne.

Let us suppose that what yields a sign is a state of things, not a pure quality, but a state of things which has several qualities, actual, actualized qualities. We’ll choose right away, we’ll choose an example in a little while, but I’m forced to start with the abstract. I would say, according to Peirce, that here, when what yields a sign is a quality, but a realized, actualized quality in a state of things, actualized in a thing, it is a — necessarily an individual thing — an individual state of things or an individual thing, this is a “sinsign”, S-I-N, sign. Why the “sin” prefix here? — Peirce explains it – it is the same as for singular, an individuated or individual state of things.

A third possible answer: what yields a sign is a law, it is a law. [Deleuze writes on the board] So, sorry, what a mistake I was making; this is a “legisign”, it’s a “legisign”. Let’s select an example. I can say green is a sign of hope, green is a sign of hope; whether it’s true or false, it doesn’t matter, right? “Green is a sign of hope” is a proposition in which green is a qualisign. I do, indeed, consider green as a sign of hope, and I regard green as a pure quality, whether it is realized, whether it isn’t realized. [Pause] When I say — another proposition – “the cow recognizes what is good to eat by the green”, that is, it eats grass, I consider this time here the green as realized in a state of things: meadow, the green of the meadow. The cow does not confuse the green of the meadow with, insofar as it is realized, with the color green, pure quality. So, what yields a sign to the cow when it’s let go into the meadow and bows down with its mighty head, in that way obeying, and showing that it has grasped a sign, and starts to chew… [Interruption of the recording] [42: 50]

… what yields a sign is a sinsign. [Long pause] I go through the green light. This is a legisign. The green works there like a legisign. There is in fact a law, in this case, conventional law according to which red signifies stopping the car and green signifies the car’s passage. Good. This is indeed a conventional law since, for the pedestrian, it is the other way around. He moves on red; he is the red that functions as legisign of passage. Good. Very well.

Let’s take this example of the green light. To show you, you have to sense that there are a lot of problems because all the concrete signs that I could analyze, don’t they have several aspects? There is one aspect through which the green light is a “qualisign”. This is not surprising, since ultimately any sign is a Thirdness. And also in the green light, the green does not just have value for its quality; the green has value insofar as realized in a state of things, the hole in a pole with another light which will be red. Finally, this is a legisign. With a given sign, you may always ask yourself, what is it in the first place? But also, in what other types of signs do they participate? There you are.

So, we have our first series. I would say that the qualisign is the sign in itself of the Firstness; the sinsign is the sign taken in itself of Secondness; the legisign is the sign taken in itself of Thirdness. OK? Within this, I notice that the three terms are created by Peirce, and here, he didn’t tire himself greatly to create because he did it, he turned it into a prefix, because I always think of that as regards this side problem: in which case and how is it necessary for a philosopher to create new terms? In what case can he be satisfied with terms already known?

Let’s move on to the second aspect of the sign. This time, I no longer consider the sign in itself, nor the question of what yields a sign, but I consider it in relation to its object, according to the question: of what type is the relation of the sign to its supposed object? From the point of view of Firstness, I would say that there are certain signs which refer to their objects, to their object, through qualities which are intrinsic to them, [Pause] through qualities which are intrinsic to them. You fully understand that, in this case, the relation of the sign to its object is grasped under the species of Firstness. It is through intrinsic qualities that the sign refers to its object. An example, yes, I would call such a sign according to Peirce an “icon”. According to him, an “icon” is a sign that refers to its object — well that, you have to … in any case, those who, you have to remember all this at all costs because, like us, we’ll then give a very different meaning to all these notions, it is necessary that… well yes, I mean, this is at the level of a kind of axiomatic, we’re justified, we’re justified… — And an icon, you see, it is, in fact, it seems to stick, his definition of the icon, because… well, in fact, an example of a sign which refers to its object by qualities which are intrinsic to it, to it, the sign, well, a photo, a photo, it is through its intrinsic qualities that it refers to the model.

But I would say more: a geometric figure, isn’t that an icon too? A geometric figure in what form? One must be careful; the triangle that I draw on the board, a triangle that I draw on the board is a sign that refers to its object. The image of the triangle, through qualities that are intrinsic to it. It must come as no surprise that — here I do not want to develop Peirce’s theories at all at this point — that Peirce says, for example, that something like diagrams are cases of icons. In any case, you see, the icon, well, from the point of view of the relation of the sign to its object, is indeed the sign of Firstness, no longer from the point of view of the question of what yields a sign, but from the point of view of the other question, of the second question: what relation is there between the sign and the object?

From the point of view of Secondness, the relation of the sign to the object can be such that the sign would not exist, would not exist as a sign without its object, that is, without the existence of its object. So, the sign, this time, relates to its object by virtue of the supposed existence of the object itself. [Pause] I am saying, for example, smoke is a sign of fire. [Pause] Such a sign relates to its object by and in the supposition of the existence of the object; I call it, or rather Peirce calls it, an “index”, and he tells us, smoke is the index of fire. In other words, the sign is there a sign of something posited as existing. You can see that it is Secondness since Secondness, its natural state, was, you remember, the duel. The index is a case of a duel or a dyad relation, or a dyad. The sign refers to, etc.

Third, the relationship of the sign to the object is regulated by a law, natural or conventional law. This is what Peirce will call a “symbol”. [Pause] And from the point of view of interpretants, [Pause] well, interpretants are thus, you remember, the ones through the intermediary of which … it’s the Thirdness of the sign, those through the intermediary of which, the third party by which the “representamen” makes operative, makes efficient the relation which it has with the object. And in fact, there are three interpretants. Not only is the interpretant the Thirdness of the sign, but there are three interpretants. There is an interpretant — and here, then, as I would still like to go quickly — I am indicating that Peirce himself in his successive texts says — ah, there it is terrible, it is very difficult, this is not developed – he never stopped varying, it never stopped varying because … and then, it interested him a lot, and then he ended up finding lots of interpretants, types of interpretants then, about which I am telling myself that I’m sticking to some very, very elementary things. But this is already, this already varied enough like that.

There is an interpretant that one could call the “immediate interpretant of the sign.” He calls it: the immediate interpretant of the sign. [Pause] And the immediate interpretant of the sign is the signification. [Pause] What does signification mean? Why is that signification? Because it’s hypothetical; this is neither true nor false. This is the domain of what is neither true nor false. The word “grenade”, its immediate interpretation, is city, weapon or fruit, question mark. You see this is the set of significations of the word; this is hypothetical since I don’t know, I have no reason to choose, and it’s neither true nor false.

In language, one would say that the immediate interpretant is the word. A word is neither true nor false, it indicates a possibility. This is, as might be said, a propositional function. This is hypothetical, it’s all you want. Well, the immediate interpretant, he calls it very mysteriously, he calls it: a “rheme.” Why does he call it a “rheme”? To my knowledge, because I have looked at the English texts, the American texts, I do not see any justification for the word, because the word, roughly speaking, derives from the Greek, and implies a becoming, a flow. And he uses “rheme” for, as the equivalent of the word or the immediate interpretant, of the signification insofar as it is neither true nor false. I can’t see very well. I do not know. What…?

A woman student: [Inaudible comments]

Deleuze: He doesn’t know! The object in becoming, he doesn’t know, he doesn’t know the object in becoming.

The student: [Inaudible comments]

Deleuze: Ah, okay, yes, no, no, but our discussion dealt with what he had in mind. He only knows one thing: it’s possible. In fact, since Firstness, you remember, yes, that was it, it’s in the domain of the possible. Secondness is in the domain of the real or what exists. Thirdness is the domain of the necessary. [Deleuze writes on the board] He only knows what is possible. So, the “rheme” is indeed a figure of the possible. So why did he choose this word? Well, it escapes me completely, and then it’s not terribly important.

For the rest, things are better because, what is there other than the immediate interpretant? There is what he calls the dynamic interpretant. And then, the dynamic interpretant, we grasp this more because I ask myself — here I have a doubt — am I not forcing, am I being quite faithful? But really to finish this, for me, it seems to me that it would correspond entirely to what we called: a natural relation. This is the natural relation. The dynamic interpretant is the set of images with which the first image naturally associates.

A student: [Inaudible comments]

Deleuze: No, they aren’t given with it, but the natural relation results in the mind passing from the image which is given to these images which were not given. And how is he going to do it? Well, by forming a proposition. This is the proposition, in fact. In the immediate interpretant, we had a one-word equivalent; in the dynamic interpretant, we have the equivalent of a proposition. It is the proposition that will determine the set of images through the intermediary of which image 1 refers to its object. “Granada is a Spanish city”, proposition; you see that “city”, “Spain”, are the dynamic interpretants by which the word: “grenade/a” refers to the object, the city located in Spain. So, I have a series of dynamic interpretants, that is, natural relations: “grenade” makes me think of “city”, makes me think of “Spain”, etc. I have a series of dynamic interpretants that is referred.

So, I can say that for me, these are natural relations, the dynamic interpretant, and it is quite normal [Deleuze writes on the board] that they become operative in the proposition which, unlike the true, unlike the word, the proposition is the possibility of something true or false. What will be true or false is the judgment which affirms or denies the proposition. [Pause] — But no, I’m completely … Proposition, yes, that’s the “dicisign”… Yes, yes, because I haven’t given the name of the sign yet. I am saying, it’s the equivalent, etc. — This sign corresponding to the interpretant, to — I am starting to get confused — therefore corresponding to the natural relations present in a proposition, he calls it: dicisign, D-I-C-I-sign, it is the dicisign. [Pause]

And finally, there is what he calls the final interpretant. See: immediate interpretant, dynamic interpretant, final interpretant. The final interpretant, well, that’s what is obviously on another plane, but which must in a certain way end the series, because otherwise, the series of dynamic interpretant will go to infinity. And as Peirce says, every thought refers to another thought which refers to another thought, etc., to infinity. But on another level, there is a closure.

I would like to say, because it would be too beautiful for us, but — there is no reason that it be good for us — that this, for me, I cannot understand. I can’t understand what Peirce means, [Deleuze writes on the board] it’s not finished… Because I’m saying, at the point we’ve reached, there is only one way, there, it can only be in the domain of abstract relations. And in fact, he seems to say, since the sign is going to be named “argument”, “argument”, and since, just as the “rheme” vaguely corresponded to the “word” in linguistics, and the “dicisigne” vaguely corresponded to “proposition” in linguistics, now the “argument” corresponds, in fact, to reasoning, to the syllogism. So, there I can say, and feeling stronger about saying, well yes, this is the domain of what we analyzed the last time under the name of abstract relations.

Unfortunately, he maintains in all kinds of texts, but reading it is so difficult, that the final interpretant is, he says, habit, habit, the habitus, habit. The texts become very difficult because … It seemed obvious to me that habit is precisely what makes me pass from one term to another in the series of dynamic interpretants. For him: no. What is his conception of habit? Here it becomes excruciating. Here we find, in fact, a whole, very important aspect of English and American philosophy, in fact, — I don’t care, I don’t care; I can’t manage that, I can’t sort it out anymore[4] — it is enough for me that, at least in one aspect, it is perhaps habit, perhaps habit, but then a radically new habit, but, it was also the fabric of the abstract relationships presented in the argument. Well, you see, you see, this is the graph we must select… We have seven, eight, nine! You see if we continued, but here I am going to stop because we have to move on to something else.

Here I have my pure signs or what I might call great signs. Any sign is a combination. Already you can think of how many combinations are possible. But among the possible combinations, there are some that cannot be realized. Why? Because, for example, the whole domain of Firstness cannot be combined with just anything. [Pause] I’ll choose an example. He’ll tell us, a rhematic iconic qualisign is very possible, right? From that point, we are quite delighted here because… So, Peirce — you understand everything I’m telling you, this is only the start for Peirce. – So, fine, here, he does not provide a name. Sometimes he provides a name, right? And for the rhematic iconic qualisign, this is one of the most beautiful, in any case, it is the first that is possible! But there are completely impossible cases. Fine. And then at the end of his life, the more he moved forward, the less this satisfied him, all that. So, he multiplied the headings there because the three interpretants, he noticed that the sign had, no matter why, two objects and three interpretants, which we put here, but he took it upon himself to put them there! [Deleuze points to the table] So he had two objects and three interpretants. [Interruption of the recording] [1:06:48]

… Suddenly, he obtained many more signs. It ended up fantastic; obviously his dream was to create a classification of four thousand, five thousand signs. So fine, that’s what we call a lifetime’s work. But at the same time, he was doing formal logic. He put it all in, he formalized it all, it’s a very extraordinary work, fine. Curtain, because then here we are faced with the problem, well, there, we, I mean, it is not that… We cannot compete with a philosopher like that, it is not a question, it’s not at all a question of that.

But for us, and these are not objections. When I try to tell you all the time, when you read something, anything, for example, there are things that you find useful and then you say, this is for me, and then there are things that are not for me, okay. When you listen, when you listen to a course, there are things for you and things not for you. It’s part of the mysteries. Why is that for us, why is it not for us? No one will know. Good. Well yes, maybe, but we would have to look carefully. This is, this is really the problem with the affinity of concepts.

Well then, so I do not allow myself any objection, really; it’s learning to be quiet in order to go faster. I just ask myself, what’s, what works for me, what doesn’t work for me? I said that the last time, it’s, I’m already just clarifying, what it is that might work for me, what isn’t going to work for me. I take note and I tell myself: and my perception-image, and my perception-image, there is no place for it in all this.

What strikes me as obvious is that to me, once again, if you accept this not to be an objection, he says too much, and he doesn’t say enough. On one hand, he says too much because for the moment, I would develop nothing with Thirdness. Obviously, it is disturbing because he tells himself, those who do not understand Thirdness, they do not understand anything. But I would have to develop something else regarding Thirdness. But, on the other hand, I would badly need of a, of something. Because, in fact, this is still frightening.

There I can say, Peirce is exaggerating because he allows himself these three images, up there, Firstness, Secondness, Thirdness. He tells us, the “phaneron”, that is, the Appearing, is organized like this, and that’s just how it is! But that’s not good at all! Where do they come from? What will account for this triple root of the “phaneron”, this differentiation of the “phaneron”? He doesn’t tell us. That’s how it is because that’s how it is. But I don’t like this! I would like that to be clearly stated, and you can see why. There will be no analysis by him of the “phaneron”, but we – so he’s justified — but we are required to do so because we did not start from the “phaneron” like that. We started with a long description of the movement-image or the light-image. That is, we have made a very long description of our own “phaneron”.

So, our movement-image, our own light-image, it already had a perfect consistency. Moreover, it spanned a whole plane of immanence, if you remember. Hence the need for us to account for the differentiation of images of Firstness, Secondness and even Thirdness. In other words, we need something that accounts for this differentiation, in other words, a root of the differentiation, one, two, three. We need something which is already not the movement-image anymore, and which is not yet either Primacy, Secondness, or Thirdness. What other word do you want to give it other than “Zeroness”? So there needs to be a Zeroness which itself will have signs. But then, if we do that… Moreover, then if for other reasons, we are led to reject Thirdness temporarily, what then will a sign be for us? It will obviously have to be something else entirely.

I mean this, and this is my first point: [Pause] it seems to me that he doesn’t say enough. If I sum things up, it is because there is no generation of varieties of images for him. [Pause] We need a degree zero which explains how and why the movement-image gives rise to Firstness and Secondness. [Pause] In other words, we need a zone that accounts for the birth of special images at the heart and within the universe of the movement-image or the light-image. And we have seen, if you refer to what we did at the very beginning of the year, this is what we saw. What is it that ensures the passage from the movement-image to special images, and which therefore is both a special image already and still participates in the plane of the movement-image? It was precisely the perception-image.

And why did the perception-image have this privilege? Here it must be remembered, otherwise it makes no sense, the perception-image had this privilege because it participated in a dual system. It was inseparable from a double system. It participated first in the system of movements-images which all vary with respect to each other on all their facets and in all their parts, and in this sense, all the images, whatever they were, were perceptions in themselves. And the atom was a perception in the sense that it was the perception of all the actions it underwent and all the actions it performed on this plane of immanence of the movement-images.

So, I am saying that image-perception, through one of its poles, referred to the world of movement-images and their universal variations. And on the other hand, at the same time, the movement-image… the perception-image, the perception, expressed through its other pole the movement… no, the process through which the system of movement-images was going to be organized around a center of indetermination, that is, where all the images were going to vary for one and in relation to one alone. And in relation to this second system, I no longer said all things are perceptions; I said: someone perceives something. Someone, a center of indetermination, perceives something. The images ceased to be in themselves, the images became, the “phaneron” ceased to be in itself. It became for someone, and in relation to someone, in relation to someone who perceived him, a perceiving subject.

But the perception-image did not cease oscillating from one of its poles to the other. An obvious example that we analyzed considerably last year regarding cinema: it suffices that the perceiving subject — so I have my two systems; first system: all the images vary for one another; second system: the images vary in relation to a privileged image. These are the two poles of the perception-image, okay? — But if the privileged subject, if the privileged center, if the privileged image begins itself to move, the more it is set in motion, the more it will tend to reconstitute universal variation. Right? Very well. [No interruption in the recording] [1:17:13]

Part 2

So, the perception-image does not cease to oscillate. This is why I say that it is presupposed by Firstness and Secondness, that it is something other than the movement-image and yet it is not yet Firstness. It is going to be what makes Firstness and Secondness possible, and in fact, quality is quality, but it will have to be perceived in one way or another. Action is action, but it will have to be, in a way, not only acted, but perceived. So, the perception-image will accompany the special images; it will be a special image itself, it will have a very odd role. On one hand, it plunges into the pure movement-image, on the other hand, it is already a special image, on the third hand, it accompanies other special images. This is why for me, this imposes a Zeroness that refers to the perception-image. So, at the top, I will have — we will see it later — Zeroness, perception-image; Firstness, affection-image; Secondness, action-image. For the moment, I leave aside the problem of Thirdness.

But what did I just say? I just affirmed that starting from perception-image, there was at the same time one characteristic as what? A genetic characteristic. We must clearly mark the generation, the genesis of each special image. So, there is a genetic aspect. On the other hand, I noticed that there was a bipolar aspect of the perception-image. It participates in two systems: a system of universal variation, a system of variation relative to privileged cen… to a privileged image. Okay… To me, that’s what a sign is. So, we’ll have to say… Good. A sign, I would say, well, you understand… A sign is a special image, a particular image, rather, a particular image that represents a type of image. There we are. A sign is a particular image that represents a type of image.

So, for the moment, I have three types of images; I am still not taking Thirdness into account. I have perception-images, affection-images, action-images. I am saying, a sign is a particular image insofar as it is valid for a type of image. In fact, I would say at that point, ah this is affection-image, it’s affection-image, ah it’s action-image, ah it’s perception-image. So, it’s a particular image – this is essential for me — that represents a type of image. From what point of view? I have not yet introduced the story of interpretants. I leave it aside. I can’t continue anymore, I can’t continue.

On the other hand, I have two points that interest me, and perhaps many more. How can a particular image, from what point of view can a particular image represent a type of image? Apparently, from two points of view or maybe four, it will depend on your choice. It depends; if we sum up, if we extend, from two points of view, [Pause] from the point of view of the bipolar composition of each type of image.

So, in this case, a sign will be a particular image that represents the bipolar composition of the type of images to which it refers. Obviously, the examples will come, the examples. [Pause] You understand, it gets complicated, because already by looking for them, well, I can say, I will have at least three signs: I will have a bipolar sign, a sign that emphasizes the A pole, and a sign that emphasizes the B pole. That will make three slots for me. I mean in the numbers race, this is very important.

The other aspect… [Interruption of the recording] [1 :23 :10]

… [The next sentence is missing from the recording but is furnished in the Paris 8 and WebDeleuze recordings] I would say that the sign is also a special image that refers to a type of image, no longer from the point of view of the bipolar composition of this type, but this time, from the point of view of the genesis [return to the recording] of this type. And there, I might have two signs of it, because why only the genesis? Also, the disappearance. It could be either the sign’s disappearance, the type of image… Sorry, not the sign’s disappearance; the sign will be a sign of disappearance or a sign of genesis of the type of image to which it refers. Sometimes, understand that you can’t decide anything in advance; it doesn’t seem like it, but it’s very, very amusing. Sometimes I would realize that the sign of extinction is necessarily confused with the sign of genesis. Sometimes to my amazement, the sign of extinction will not be the same as the sign of genesis. So, I would say that through the type of image, I will have, if I take things as accurately as possible, I will have two signs: a sign of composition and therefore bipolar, [Pause] and a sign of genesis, a genetic sign. In fact, once again, the sign will be the image which will represent either the composition or the genesis of the type of image to which it refers.

So, this definition has nothing to do with Peirce’s, and yet we are going to find some common signs, that’s perfect. If I go to the longest, on the contrary, I would say there are cases in which I can conceive of five signs per type of image. In fact, the sign of composition will split into three, depending on whether the two poles are supported or if it is one rather than the other. So, in this case, I will have three signs of composition. And then the sign of genesis will split into two, eventually, the sign of genesis and the sign of extinction. [Pause] So I will have at minimum two signs per image type: Zeroness, Firstness, Secondness, at maximum five signs per image type. Good.

And then, why, why am I leaving Thirdness aside for the moment which will nevertheless be essential? Good. So, it’s tragic, because I have to go to the main office. Are you still hanging on or not? Because if you can’t continue, there’s no point in doing so.

I feel like I should give a little example; maybe that would fix everything but… You understand, yes, I’m offering giving an example so that… Well… I’m taking Second… Firstness. For those who were here last year, they will understand, and you will understand why I need Peirce’s words, why I am justified in a way to give them another sense.

For me, Firstness is the affection-image, and in fact, the affection-image is the quality or potency [puissance] considered in itself, independently of a state of things, independently of a state of things in which it would be actualized. Okay, that’s clear. When you take a quality or a potency, when you consider a quality or a potency, regardless of a present, existing state of things in which that quality is actualized, at that point, you will find yourself facing an affection-image.

I’m selecting an example, fear. Fear always has a cause. I’m facing a precipice, I’m scared, well, fine. Fear always has a cause; that does not prevent it from existing as a fear independent of its cause. And you can consider it regardless of any causes. In other words, you can look at it independently of any current state of things. You will tell me, there is always a current state of things; at first, I’m not sure, it’s not so obvious, but you see it as pure potency, pure potentiality. We saw that with Firstness in Peirce, he conserves all that. I would say, here we have an image of Firstness. Okay… So… Fine, okay… There you go.

The analysis I tried to do last year on affection-image gave me this, but, a potency-quality considered in itself, what does that mean? Notice, here as well, these are not objections at all. Peirce is happy with that since it refers to a pure immediate consciousness, which is never given. So, he doesn’t need anything else. I told myself, I need something else, to define it, I don’t know, more ontologically, less logically.

I would say, it’s not difficult, it’s a quality. It is a quality or a potency, such as it can be expressed by a face. It is not actualized in a state of things. It is a quality-potency insofar as being expressed. Good. It’s even saying too much, by a face. It is an expressed quality-potency and not actualized. Good. But what is the variable of expression? You don’t have to go very far in the analysis to say, well then, the variable of expression is a face or the equivalent of a face, something that functions as a face. A foot can very well function as a face, this is what we will call a faceification. Alright, very good. Or even a proposition, a proposition can function as a face. A proposition is a mental face. The proposition is a face insofar as it “expresses”. Expressing is just one of its dimensions. It is only in this dimension that a proposition is an equivalent of the face.

So, I would say, there is a first sign of the affection-image, and the first sign of the affection-image, it is quality-potency as expressed, three little dots, insofar as expressed by a face or face equivalent or proposition. I call it an “icon”. So, I need the same word as Peirce; however, the sense is very different. For me, an icon will be a quality, or a potency insofar as expressed. By implication, if it is expressed, it can only be so by a face, an equivalent of a face or a proposition acting as a face. There we have the sign of composition of the affection-image. Ah, good! This is the sign of composition, and in fact, it is not genetic at all. This is what the affection-image is made of. It is composed by expression.

Hence a suspicion: but the expression or its variable expressing it, that is, the face, the equivalent of a face or the proposition, let’s try to search a little, aren’t there two poles? Yes, there are two poles. We saw it, here I am recapping something for… even I hope that those who weren’t here last year grasp this, nonetheless. Yes, there are two poles. And in fact, there are two poles of the face. The face can be considered as a contour which I will call at that point, a faceifying contour; it is essentially expressive because it is a contour. It is the contour that it is expressive. This is the pole of reflection. The face is reflective and pensive.[5] [1: 33: 21]

Part 3

And then he has a whole other pole, not a faceifying contour, but a set of features, a set of discontinuous features: an eye that gleams, a nose that sniffs, a mouth that opens, and we jump from one to the other. I would say that the face is taken as a dynamic aggregate, this time of features of faciality, and not of facifying contours.

And in fact, I had tried to show last year how painting entirely had these two poles and that the art of the portrait presented to us sometimes a pole of contour, a prevalence of the contour, of the faceifying contour, and sometimes a prevalence of features taken from the mass, with jumps from the eye to the nose, etc., and not at all the contour line, the great contour line that you find in the 16th century, for example. In the 17th century, there the portrait undergoes a kind of, we pass from one pole, everything happens as if they were passing from one pole to the other of the face. I’m simplifying a lot.

This is great for me; I could already say “the icons”, which is for me the first sign of the affection-image — and therefore everything will be shifted already, that will not be the icon at all — that will be the first sign of the affection-image, and I would say, this is the sign of composition. But as such, there are two kinds of icons: some — I could look for some very complicated terms — but not to, not to exaggerate, some we will call: “icons of contour”, the others, we will call them “icons of features”, [Pause] good, dependent for us, in the true art of icons, would there not be equivalences? Wouldn’t we find icons of contour and icons of features — the history of cathedrals – from this point of view, shouldn’t we consider sculptural things, etc.?

And then, [96: 00] and then, I’ll quickly finish this example, and then we’ll leave it there — so you can go on vacation faster — and then, that’s not all nonetheless, it is not necessarily all, because I say to myself, well, the icon is a sign of composition, but in fact, it does not tell me anything about the genesis of the affection-image. It tells me so little about the genesis of affection-image that the face, whether as an icon of contour or an icon of features, this is a huge unit, it’s a huge unit. It generally works, right? That’s about…

There’s a silly story in the movies, you know, that’s the famous effect: a stream of shock, right? There is an actor there who has the same expression, and then there’s a montage, and this expression is supposed to correspond, in the first instance, to a woman, so it becomes “desire”; at a second instance, to a meal, to food, then it becomes “hunger”; in a third instance… I don’t remember after. This is silly because, this is silly, it has absolutely nothing to do with the problem of associations with a context. There was no need to have this experience. It has been a long time since all psychologists and people with good sense have noticed that the face is a very, very little differentiated instrument in the expression of affections, that it has very few expressions. Long ago… I remember in Sartre’s novels, there is always a metaphor, not a metaphor, a note that returns, that was entirely typical of Sartre: nothing looks more alike than sexual desire and the need to sleep — not always true, but often, [Deleuze laughs] often — sexual desire and the need to sleep — Ah yes, I reversed, hey … well — but in fact, in the end, as crude materialists would always say, in all emotion, there is what? An adrenaline rush sounds the same.

I can say at the most that if I don’t consider the state of things where it is, this is required, this is a crude expression, it’s composition. In the end, the face expresses only composed affects. To arrive at expressions of pure affects, some very special efforts are necessary, otherwise you are always expressing composed affects. There, we can see very clearly that the affection-image, insofar as signified by an icon, that is, as expressed by a face or a facial equivalent, it only finds a sign of composition. The icon is only a relatively crude sign of composition. At most, it can accentuate one pole, accentuate the other pole.

But what would the genesis of affects be? There is something much finer than the genesis of affects. I therefore claim a finer instance of the face. There is something much finer in the face. A face can always lie; it can always, all that, that can always, there is something much more differential, something which would perhaps be the “genesis of the affections”. What is the genetic element, the differential element? It is, for the moment, it is certain places or certain types of spaces in which, there, affects are much finer to the extent that these spaces are even repertoires, are even conservatories of still unknown affects. We have the impression that we are penetrated by affects that come from another world. Good.

Cinema, and I was talking about it last year because cinema has greatly played a role. Be careful, these are not determined space-times, since determined space-times, they are part of Secondness, they are part of the action-image. No, I don’t have the right to contradict myself to that extent. So, they are not actualized space-times. What will we call it? Last year, it was one of you [Pascal Auger] who proposed this term, which we would call “any spaces whatsoever” [espaces quelconques], which does not mean universal spaces at all times, in any place,[6] which means spaces that have no other functions than to exhibit affects, exhibiting pure qualities or potencies. For example: the space of shadows in German Expressionism, there we have any space whatsoever. [Pause] Another example of any space whatsoever: spaces in which one does not orientate oneself. That is, it’s not that one does not orientate oneself; these are spaces in which the parts of that space are not oriented with respect to each other. They are spaces without connections or, on the contrary, it comes down to the same thing, where each part is united to the other by a multiplicity of possible relations, by all kinds of possible connections, where the connections are not determined univocally. They are therefore literally disconnected spaces. Connecting one part to another can be done in a thousand ways. In mathematics, in geometry, we know that well; this is even what appears in Riemannian geometry where neighborhoods have no univocal coordination and the connection of two neighborhoods can be done in an infinite number of ways.

Well, I am saying, quality-potency, a first mode in itself, regardless of its actuality, regardless of its actualization, a pure quality-potency has two modes, no existence, two modes of Appearing. It can appear insofar as expressed by a face or a facial equivalent, as we have seen; and it can appear insofar as exhibited by or in any space whatsoever. [Pause] I would say, the any space whatsoever, much finer, is the real genesis, this is the real genetic element of affects. It is in any spaces whatsoever that affects are born, are formed. Only then will they be gathered by a face. You drop a face into some any space whatsoever, you have a horror film. Good.

So, well, what does that give me? I am saying: affection-image, Firstness, that is, quality-potency in its purest form. First sign: of composition, I call it “icon”. [Pause] Second sign: any spaces whatsoever that exhibit, [Pause] any spaces whatsoever that exhibit pure quality or potency. That’s what I call “qualisign”. I would say the affection-image has two signs: the icon and the qualisign, at its shortest. At its longest, it has four signs: the icon of contour, the icon of features which corresponds to the two major of poles of the sign of composition. [Pause] And on the side of spaces, qualisigns. Once again, I call “qualisign” any space whatsoever insofar as it does nothing other than exhibit pure quality or potency. From this point of view, there are two types of any spaces whatsoever; there are disoriented or disconnected spaces which are defined as follows: no connection, or at least no univocal relation from one part to another. But there is also another type of space, empty or emptied space, another type of any space whatsoever, empty or emptied space. I would say the first is a sign of birth, the second a sign of extinction.

Okay, I’m taking this movie stuff to be very clear, right? I come back to it very quickly for my four signs. A close-up of the face, we saw it last year, has its two poles. There are close-ups of faces that are centered on the contour, often, not exclusively, often, for example, in [D.W.] Griffith; there is the close-up of the face, which is almost exclusively based on the feature, it is much more the [Sergei] Eisenstein close-up. There I find my two signs, my two icons, the close-up being the cinematographic icon.[7]

For the any spaces whatsoever, this is also very common, and that goes beyond the horror film. I was quoting the shadow space of Expressionism when shadow drowns contours, drowns lots of things, etc. This is obvious, but in a more modern way, research on spaces among great filmmakers has been fantastic for a long time. I’m selecting examples: disconnected spaces in which really, precisely that is what we call, for example, the problem of false continuities [faux raccords] or no connection at all.[8]

Rendering, minimizing the connections between a place and a neighboring place, what does this belong to? One of the greatest, but one of the first, is [Robert] Bresson, disconnected spaces precisely because he wants to obtain in this way spaces with tactile value, which become visually arbitrary spaces, completely disconnected spaces. For example, “Pickpocket” [1959] is a masterpiece, but already “The Trial of Joan of Arc” [1962] where you find the cell which is never taken in wide shot, but which is filmed, always there, in succession of closed angles without connection or with equivocal connections. So, I would say, Bresson is a great case.[9]

But, but in all contemporary cinema, it was a very, very exciting research on these disconnected spaces that you find, for example, completely in Italian neo-realism; think of the fortress of “Paisan” [1946], for example, of [Roberto] Rossellini. It is, well, that you find completely in the school of New York in cinema, for me, one of the greatest of the New York school, it is obviously [John] Cassavetes. Now Cassavetes has multiplied the any spaces whatsoever,[10] then the disconnected spaces in which we absolutely do not understand how a guy passes from the site to the same neighboring place, in which there is a type of false continuity, he is a genius of the false continuity. Fine. The French New Wave, I’m not discussing it, Godard’s disconnected spaces, so it’s, it’s famous, all that, well. There would be all kinds of research that we had outlined last year.

But you also find the empty spaces, they’re famous, they’re famous, there are not so many, two, there are three geniuses — there are so many geniuses from everywhere — but three names immediately come to mind: [Jean-Marie] Straub’s empty spaces; the empty spaces, for me, this is the one that would interest me the most of all, the empty spaces of [Michelangelo] Antonioni, the empty planes of Antonioni, you know when there is — after things have happened, after, after the actualization, after the event, he throws an empty space at us. For example, think of the end of “The Passenger” [1975], the camera stays for a while in the space which is completely empty. Splendid, splendid, it is a beautiful case of emptied space, and then, not at all depotentialized, but which, on the contrary, only exists through its affective potentials. And the, the one that … I should have started with him, because historically he must have been the first, obviously [Yasujiro] Ozu, Ozu, so the empty fields of Ozu are also a value, a very intense, affective, intense value, and where there, it’s so meticulous, and his empty fields unlike those of Antonioni, are so deliberately insignificant that the genetic element of affection is then really created there.

Let me summarize this: you see, and all this is what I would like to systematize at the start of next year. Not only would we have a completely different — it was to give an example; for me this example, we will have… There, I would have, for example, in a graph of a completely different nature, we will see — I will have: affection-image Firstness; and then I’ll have up there: sign of composition, which would give me “icon” here; and then there: genetic sign, which would give me: “qualisign”. And then I will have my two signs of composition, “icon of feature” and “icon of contour”, and then I will have my two “qualisigns”: “qualisign of disconnection” and “qualisign of emptiness”. This would also be pretty, that would be nice, see. There I would have four in that case. There would be cases in which I would only have two because that especially must not work with a law. That has to change every time, right? So that, I absolutely need that for the start of the year, those who have not been completely disgusted really remember that, and especially to remember the graph by Peirce. Have a very good vacation. [A minute of sounds of voices, from Deleuze and some students] [End of the recording] [1: 54: 00]

Classification of Signs (initial version)

Firstness: Sign in itself, qualisign; Sign in relation to the object: icon; Sign in relation to the interpretant: rheme

Secondness: Sign in itself, sinsign; Sign in relation to the object: index; Sign in relation to the interpretant: dicisign

Thirdness: Sign in itself, legisign; Sign in relation to the object: symbol; Sign in relation to the interpretant: argument

Affection-Image (Firstness): Signe de composition: Icon (icon of contour/icon of feature); genetic sign: qualisign (qualisign of disconnection/qualisign of emptiness)

 

Notes

[1] Reference to Gérard Delédalle who edited Peirce, Écrits sur le signe (Paris : Seuil, 1978); Deleuze referred to this text the previous spring, in session 12 of the seminar Cinema I, March 9 1982, as well as several times in the current seminar. The example taken from Delédalle, “grenade”, plays on the different senses of the word in French (the city Grenada, pomegranate, and the weapon).

[2] For the term “phanerons”, see session 4 of this seminar, December 7, 1982.

[3] The graph that Deleuze develops subsequently takes up the distribution introduced during the previous seminar, session 12, March 9, 1982; see the breakdown reproduced directly below the translation.

[4] It is possible here that Deleuze refers to the error he made about Peirce in session 4 (being English rather than American), and that he indicates his annoyance at trying to keep the distinction straight.

[5] At minute 93, approximately, in the transcriptions presented at Paris 8 (in particular, in parts 2 and 3) and at WebDeleuze, a text segment, corresponding to some 55 minutes of the session, is placed into the transcription. We have omitted this long repetition, and the Part 3 marker is only an approximation since Part 3 in Paris 8 begins in full repetition. Unfortunately, in Les Cours enregistrés de Gilles Deleuze, 1979-1987 (Paris: Vrin; Mons, Belgium: Sils Maria, 2006), Frédéric Astier relied entirely on these transcriptions (rather than listening to the actual recording), and his summary of this session unnecessarily reflects this repetition of the text (pp. 54-55).

[6] On this term, see especially session 11 of the Cinema seminar I, March 2, 1982.

[7] On the face and close-up in Griffith and Eisenstein, see Cinema seminar I, session 8 (January 26, 1982), 9 (February 2, 1982) and 10 (February 23, 1982), as well as The Movement-Image, chapters 5 and 6).

[8] On the any space whatsoever and Expressionism, see Cinema seminar I, session 11 (March 2, 1982), as well as The Movement-Image, chapter 7.

[9] On these spaces, see The Movement-Image, pp. 168-171.

[10] On spaces in Cassavetes and Godard, see The Movement-Image, pp. 250-252.

Notes

For archival purposes, the augmented and new time stamped version of the transcription was completed in March 2021; an updated version of the transcript was completed in November 2021. The translation was completed in November 2021. Additional revisions occurred in February 2024. [NB: The transcript time stamp is in synch with the WebDeleuze recording link provided here.]

Lectures in this Seminar

square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol