June 1, 1982
The movement of thought, you see, is going to be: the implied time involved … or the signified of power; and the signifieds of effect, that is, such as language gives them to us, … These are going to be positions taken on this signified of power which is process, procedure, aspect. As a result, everything is fine. … I would say that optical images, if I return to my theme of cinema, are only positions in relation to these temporal implications, to these thought processes. And I can discover the idea that it is always in the nascent state, that is inevitably in the nascent state, because if I take them through the nascent state or the ending state, whatever, I would then fall again into the domain of explicit language; I would then fall back into the domain of the movement-image as such, etc. In any case, I think that this would allow us to maintain that cinema has nothing to do with language, or that, at least, it’s the status of language’s relation with pre-linguistics, and that this is what constitutes the essence of what is by no means a code language, but only if, in fact, if we insist on saying that it’s a language, that it’s an analog language.
Seminar Introduction
In the first year of Deleuze’s consideration of cinema and philosophy, he develops an alternative to the psychoanalytic and semiological approaches to film studies by drawing from Bergson’s theses on perception and C.S. Peirce’s classification of images and signs. While he devotes this first year predominantly to what he considers to be the primary characteristic of cinema of the first half of the 20th century, the movement-image, he finishes the year by emphasizing the importance of the post-World War II shift toward the domination of the time-image in cinema.
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

In this truncated final session (the opening segment is missing), Deleuze provides a succinct review of the different signs and images proposed in the Seminar, in some ways, recapping the contrast he has made between the movement-image and the time-image. At the start, Deleuze seems to continue from the previous session’s review of “modes”, particularly the three he had not discussed fully (didactic mode, critical mode, temporal mode), the last of which seems to have allowed him to introduce the linguistics proposed by Gustave Guillaume. He then moves on to a complete review of eight levels as the course overview, from the opening movement-image, through fifteen kinds of signs (some not having been proposed earlier, in session 12), finally coming to the time-image that overlaps and deepens the opening movement-image. To a student’s query at the end about the 1982-83 seminar topic, Deleuze initially answers that one semester will be devoted to the history of philosophy, and then asked the question that he failed to address during this first year on cinema: how is cinematographic perception a different perception from our so-called natural perception, and what are the relations between them? As a result of this “lapse”, for want of a better term, the next seminar will be, in fact, a second year on the material already considered during year 1, but from an alternate angle of approach, as he explains in the opening session in fall 1982.
Gilles Deleuze
On the Movement-Image, Bergsonian Lessons on the Cinema
Lecture 21, 1 June 1982
Transcription: Part 2, Noè Schur; Part 1, no recording; supplementary revisions to the transcription and time stamps, Charles J. Stivale.
Translation, Christian Kerslake
Part 1[1]
Part 2
… there, as an internal element of the set [ensemble]. Let me call that the ‘first position’.
‘Second position’. I say: ‘I see a man’. But something in the tone, in the intonation of my phrase, indicates I understand that if I see a man, it means that there are others not far away. He is the first in a series I have not seen yet. That is to say, seeing a man, I have the certainty that he is not alone. Because, for example, if I say – assuming I know that wolves go in groups – ‘Oh, I see a wolf’, that means: I see a wolf, but I know very well that it is the first in a series. Fine. This is no longer an element taken from the set, it is an element taken from the border of the set. I do not see the set: the set is announced to me by this border element. This is another position of ‘one’ [‘un’]. Second position of ‘one’.
Third position of ‘one’: I take a fly. I have a little box full of dead flies. I take a fly, with pliers, and I show it to you, saying: this is a fly, a specimen of a fly; I have taken it out of its set. Third position of ‘one’.
I would say these are three “signifieds of effect”.[2] Good. Let us refer them to such a movement. I will be able, across my positions, to fill the intervals and to trace a movement of thought which will be the “signified of power” of the indefinite article ‘one’. And [Gustave] Guillaume is going to come up with a theory – here I’m only giving the result which will appear rather disappointing to you, but it still seems like a very interesting method to me – he’s going to say, the indefinite article is the movement of individuation, [Pause] while the definite article, the, is the movement of generalization.[3] He is going to come up with graphs where each time, these movements, these movements of thought are going to be taken with positions on them that mark signifieds of effect. The movement of thought, you see, will be: implicated time [temps impliqué][4] (if I summarize), or the signified of power; and the signifieds of effect, that is to say, such as the language system [la langue] presents us with them, such as a language system presents us with them: they will be the positions taken on this signified of power which is procedure [procès], process [processus], aspect.
As a result, all goes well. I mean, at this moment in these implicated times, in these processes, [Pause] I would say that optical images, if I come back to my theme of the cinema, are only positions in relation to these temporal implications, to these processes of thought.[5] And I can rediscover the idea that it is always in the nascent state, it is necessarily in the nascent state, because if I take them indiscriminately in the nascent state or the finished state, I would then fall back into the domain of explicit language, I would then fall back into the domain of the movement-image such that …, etc. In any case, I think that would allow us to maintain that the cinema has nothing to do with language, or that, at least, it raises the status of the relation of language with the pre-linguistic, which is what constitutes the essential in what is not in any case a language of code. But only if, indeed, we insist on saying that it is a language, that it is an analogical language. Only today, it seems to me that people no longer even fully understand what an analogical language means, because they think it’s a language of mimes, because they think it’s a language where one imitates, that it’s an imitative language, whereas analogical language has never been that.
Good, so you see that these would be our four modes: imaginary mode, critical mode, didactic mode, temporal mode (of implicated time); and what would then have to be shown (which would be very easy) is the extent to which all that is completely … how everything overlaps. So in that case, the moment has come for me to make a recapitulation, and I’d rather not just do a summary. I mean, let’s try … – earlier on I devoted a session to Peirce, and his peculiar classification of signs,[6] making use of it for our own purposes.
And now I’d just like to say: let’s forget everything about Peirce apart from one thing, which seemed excellent to us and of which we have made much use: the distinction between the three modes of existence[7] that cut across signs [qui recoupent les signes]: Firstness, what only has to do with itself; Secondness, where there are two terms, action/reaction; Thirdness, where there is a real relation between two terms, that is to say, where there is a third. And we saw that firstness was for us the affection-image. And indeed, the affect has nothing to do except with itself, even when it has a cause … but that’s not what’s in question here, I do not want to come back to all these points; secondly, secondness was the action-image; and thirdly, thirdness, that was an image we were still looking for.
And I only retain the words here. I say this clearly, so there will be no confusion: I will only retain Peirce’s words, but for my part, I will never keep the meaning that he for his part gives to these words. I will need the words, but – this is justified, I think – on the proviso that I will be giving them a different sense, which I will specify. So, summing up everything we have done, I will just give you a classification of the signs that we would have obtained. And it will be an odd affair in which there are sometimes quite bizarre words. … But, look, it’s 11.30, I have to go to the secretary … I’ll be as quick as possible. [Interruption of the recording] [7:25]
[Fragmentary discussions among students with Deleuze near to the microphone] Ah, afterwards, you’ll go take a nap … the whole day, followed by a little cinema in the evening … [Pause; unintelligible comments] He’ll fill up his head with this, I don’t really know…
Right, so you see … here, although it’s not absolutely, not quite on point, is the classification of signs that I would like to propose to you so that you can learn it by heart during the vacation [Laughter]. But I’ll leave it up to you, change them at your pleasure … Maybe you’ll put this or that one in a different order; whatever you like.
I’m saying: first – and you will see that this recapitulates everything that we have done this year – first of all, in our Bergsonian starting point (and I made a lot of this), image = movement = matter. This was the movement-image. And we started from that: the image is movement, movement is image. Obviously, that doesn’t mean much for those who haven’t followed everything, but no matter. So, my first level is the movement-image.
Second level. Question: under what conditions and when does the image become a sign? A completely Bergsonian answer: it is when the movement-image, or movement-images, are related to a center of indetermination, to a center of indetermination defined simply as this: a gap between an action and a reaction, that is to say, a gap between two movements. But then if movement-images are related to centers of indetermination, they become signs, and I call sign – and here, once again, this does not coincide with Peirce at all – what I call sign is the movement-image in so far as it is related to a center of indetermination.
Third level … no, I’ll just add something: we’ve seen starting from this that movement-images related to a center of indetermination gave us three great types of images: perception-image, action-image, affection-image. From that moment on, we must expect that signs correspond to these three types of images. This is normal. So everything is fine so far.
Third level: the perception-image. I will call it strictly, in order to fix … no, it sounds like I’m joking, but I’m not – I will call it embryonic firstness, because it makes me happy – although it’s only me that it makes happy, but hey … but not that much, eh – embryonic firstness because, in order to say that it is from there that firstness will be born, although it is not yet firstness. And why is it not yet firstness, or secondness, or thirdness? For a simple reason: it is because, at this third level of the perception-image, we will be witness to the formation, to the determination of the constitutive elements of the sign. It is the sign related to its constitutive elements. So there is not yet either firstness, secondness, nor thirdness, which refer to the constituted sign. It is the sign related to its constitutive elements. Therefore I call this ‘pre-firstness’, embryonic firstness.
And I’m saying, at this third level, that the perception-image had two poles, and we stayed with this for quite a while: objective pole, subjective pole. But then strangely enough at this third level, we are indeed given the two poles of perception, but not yet the nature of the signs of the perception-image. Why? Because the perception-image properly speaking never stops oscillating from one pole to the other, from the objective pole to the subjective pole and from the subjective pole to the objective pole. These are relations between perception and the thing.
As a result, the first sign corresponding to the perception-image is what we called the ‘semi-subjective’ [mi-subjective] image, [Pause], such that it seemed to us it had a status, a very consistent status, namely as an intermediary between objective and subjective. And it is there that there will be a sign. And that appeared to us to receive a very consistent status thanks to Pasolini when he elaborates his notion of the ‘free indirect image’, the free indirect image, which will account for the equilibrium, and for the nature of this equilibrium, between the form and the content of the image; ‘free indirect image’ being a transposition of what, in grammar, is called ‘free indirect discourse.’[8]
And in what way is the image semi-subjective in the cinema? Well, in effect, it’s when you have both, one intertwined in the other, a character who sees and who acts – subjective image – and a camera-consciousness that will operate the framing [cadrage], etc. And in the analyses that Pasolini offered us of Antonioni, of Bertolucci, and even of Godard in certain respects[9], there was a kind of formation of these free indirect images which will be very, very important, and which will establish the equilibrium of the form and of the content of the image.
Now, these free indirect images give me the first sign of perception, which I call, borrowing the word from Peirce, dicisign [Deleuze spells it out], a dicisign.[10] You can see why the word is useful to me, since it concerns, in effect, something free and indirect, [Pause] as in free indirect discourse. Therefore, the word dicisign is well-founded although, once again, Peirce uses this term, creates this term and uses it, in a completely different sense. And so I take this word because it seems to me to be useful for us. I would say the first sign of the domain of the perception-image is the dicisign, that is to say, the free indirect image or the semi-subjective image. This sign is related to its two elements: form and content. Or rather, the image is related to its two elements: form and content.
But, equally, the perception-image gives me another sign which I would call this time figure. And we have seen – all this is just a recap, I really want to make a recapitulation almost just of vocabulary – we have seen that the ‘figure’, in its difference from the cinematographic image, in its relation and difference to the movement-image, is the ‘vibration-photogramme’. And I would call this type of sign a ‘figure’, the emergence of the photogramme and its vibration. So, the perception-image gives me two first signs which participate in embryonic firstness, and which I call dicisign and figure. Whew!
Fourth level: the affection-image. With it, as we have seen, begin the signs of firstness. And there are two signs corresponding to the two poles of the affection-image. Two signs of firstness.
In the first place, affect is expressed by a face. It is the close-up, perhaps not only, but it is in an exemplary sense the close-up. And I call that icon, also borrowing the word from Peirce, but giving it a different meaning, since for me an icon is the presentation of an affect on a face, on a face in close-up. And the other sign of the affection-image, the other sign of firstness, I again borrow the word from Peirce, is the qualisign [Deleuze spells it out]. And the qualisign is this time the affect, not as expressed by a face in close-up, but the affect in so far as it is exhibited in what we have called an any-place-whatever or an any-space-whatever, in correspondence here with the work of Pascal Auger: when the affect is presented, exhibited, no longer on a face, but in an any-place-whatever: fear as it emerges in an any-place-whatever. And we have seen the consistency of this notion of an ‘any-place whatever’ at the level of the cinematographic image. These are the two signs of firstness.
So, I already have my two first signs: dicisign and figure for embryonic firstness; icon and qualisign for firstness. This is a table of categories of the sign, is it not? I do love tables of categories – when there are lots of them, you understand.
Having said this, what we’ll do, all the same, we’ll not forget our principle of embryonism. We are not going to jump right away to secondness, which is the action-image. We will make a passage, a passage from firstness to secondness, or if you like – fifth level – it’s the passage from firstness to secondness or embryonic secondness. And it will correspond to two new signs, which do not appear in Peirce’s classification. So I take parts of it, leave other parts, and so on – we have the right to do this, no? As long as it’s necessary. It’s necessary to me. These are fetishes … no: symptoms and fetishes.
And we have seen what types of movement-image this corresponded to in the cinema. It was truly the passage of the affection-image to the action-image, but in a particularly rough and violent form, namely the ‘symptoms’: these were signs through which a determinate milieu, no longer an any-place-whatever, no longer an any-space-whatever, but where a determined and qualified milieu refers back to an originary world. It was the world of the impulses [pulsions], and the impulses were signs, that is to say, symptoms of the originary world. It’s a bit like a novel. And the fetishes were the pieces, the pieces which the impulses tore from the milieu and on which it feeds [dont il fait sa pâture]. And cinematographically, that appeared to us to found one of the dimensions of what should be called the great naturalist cinema, both in Stroheim, the non-realist, but naturalist cinema, and in Buñuel, where the signs or the image-signs are made of symptoms and fetishes.
In this light, we have the sixth level: signs of secondness, because symptoms and fetishes were signs of embryonic secondness. Signs of secondness as such now: sixth level. And we have them, it was the two great signs of the action-image: [Pause] the ‘synsign’, expression of Peirce, for when affects and qualities and powers are considered as actualized, actualized directly in a well-determined milieu, in a geographically, historically, socially determined environment. Those were synsigns. And the other type of sign of the action-image, of secondness, was the ‘indices’ [indices] [Pause]. From the point of view of the synsign, we went from the situation to the action, this is what we called ‘the large form’; from the point of view of the index, we went from the action to the situation, this was ‘the small form’. And we saw that completely different types of space corresponded to it.
In light of this, there is one thing that I had to skip because we wouldn’t have had the time, I wouldn’t have been able to finish today if I’d gone into it. It is, well, we’ve still got to do thirdness. But then I would say, it’s not just that we’ve got thirdness to do, we still have to do the passage to thirdness, from six to seven, the seventh level: the passage to thirdness.
Thirdness is the mental, it is the third party [le tiers], it is relation. It is the relation between two things, and it is the relation that can only be thought. It cannot be seen. It can only be thought. So thirdness, following Peirce’s expression, thirdness is the mental, okay. But we will see what that yields for our perspective.
Well then, at the seventh level, I need an embryonic thirdness, which is still as it were contained, retained by the movement-image, by the action-image, by the narrative-illustrative cinema and which is nevertheless already akin to the shadow borne by true thirdness. So if I had had the time I would have said, well, there are two sorts of signs there, of embryonic thirdness. And some of these, I call them – I would prefer another word, but I did not find it – I call them marks, and the others I call symbols [Pause]. ‘Symbol’ does not pose any difficulties; ‘symbol’, for me, that has strictly nothing symbolic about it. I use the word ‘symbol’ in an operational sense, namely, a symbol is the sign of a relation to be established or an operation to be carried out. So I give to ‘symbol’ a strictly positive meaning – this whole classification claims to be positivist – strictly a sign of an operation to be done or of a relation to be established. You can see clearly enough that here there is a kind of thirdness, which if you have in movement-images, in a film of movement-images, in a … [Deleuze does not finish the sentence].
And ‘marks’, what are they for me? ‘Marks’ are images which form an ordinary series [Pause], referring to a habit, a custom. I specify that all the theories of signs among the English were affected by the need to make a connection with the notion of a ‘habit’ or ‘custom’, which is fundamental for them. So, we could therefore enrich that with more properly logical, philosophical considerations. Therefore, marks are signs, they are the state of sign that a movement-image takes on when it is inserted into a chain, into an ordinary chain which completes it. You can see, for example, that this is not the same thing as an index. An index: I have a footprint. And I say: that is the index of a man who passed by there.
What I call a mark is not that. It is not the same thing as the index. It can be the same example: but it’s a footprint in the sense that if I see a footprint, I expect there to be another footprint, in the same direction; it forms a chain, guaranteed by what I would call a habit. Suppose that the traces disappear, and the ground is all wet. No more tracks! I look everywhere, I ask myself, has he retraced his steps? But … nothing, it’s all vanished. I would say that in this case I come across – this is not a new type of sign – an adventure of marks, namely, I find myself before a de-marked image. I would call ‘demarked image’ precisely an image which has lost its value of sign-trace, that is to say, which breaks the customary chain with other images, with images that all together constitute a habit or a custom. The object is de-marked.
See: here is my cinema of thirdness. So, if we inevitably raise the question ‘who?’, who does this? Who in the narrative-illustrative cinema, or in so-called ‘classical’ cinema, if you like, who does that? I would have liked, if I had attempted, if I had had the time to attempt to do an analysis of this embryonic thirdness, I would have said, it’s all those types who, even in the cinema of action and narrative, have given an immense importance to the problem of relations. To the point that at the limit, their action cinema looks very, very classical and is even quite brilliant as action cinema, but in fact, there is something completely different going on in there. In fact, this is a cinema, I would say, of symbols, in the sense of operations to be carried out. It is not actions anymore. It is operations which are being carried out, equations which are being established, relations which are being constructed. And oddly enough, it’s always the same ones who handle, it’s always the same authors who handle [relations] and symbols. At the end of the day, I would say with a symbol, it’s quite simple: it is a bearer, it is an object that bears relations. You therefore have here at the level of these two types of sign: demarked objects and the objects bearing relations.
So, I say to myself, that’s like an extreme pole of action cinema, of action-image cinema. And again, I haven’t said a word about this, which is why I’m expanding a little bit. But then when this kind of cinema (for example, the Actors Studio) pushed it to the limit (the cinema of the action-image), they pushed it so far that they ended up opening onto something else; but they were no longer able to master it, they couldn’t get on top of it using only their method from action cinema. But they succeeded anyway. I’m thinking about certain texts by Kazan which interest me. These are the texts where he says, you know, it’s true what they say about the actor, about the Actors Studio type of actor: he doesn’t stop moving; even when he is still, he cannot be completely immobile; he is moving in every respect, he doesn’t stop. However, it must be added, says Kazan, that there is not just that; it must be added that he always has an object to play with, there is always an object. He does not move in a vacuum. And Kazan says: “For my part, I love objects.” He lists the objects that come up in his films. He says, it’s very important, the objects in a film.[11]
That poses problems obviously connected to the framing, linked to all that; it would allow us to review all sorts of very important questions because objects can be exchanged. Objects are exchangeable. I ask for nothing more in order to define the symbol: the object in so far as it is an object of exchange. That can happen. But it goes further: [the object] is thrown, it is kissed, it goes into the other’s mouth. The object is a support of relations. It is a support of, and it is a sign of operation. It is an operational sign. It is here, it seems to me, that the cinema-image profoundly attains symbols, those objects which are the supports of multiple relations or ambiguous operations. Ambiguous operations, I mean, there were signs in the old mathematics, and there still are, to designate that the operation can either be an addition or a subtraction. You see, this is what one called signs, and it’s what Leibniz called ambiguous signs. The object that bears … the object that is an ambiguous sign. Is the object an object of exchange? Is it a gift? All this concerns mental relations. To give, to exchange, concerns mental relations. Good.
Well, at the end of the day, who is the great author who pushed this, who made a cinema of the operational symbol? It’s not Kazan, it’s not the Actors Studio, although they came close to it. In fact it’s almost the anti-Kazan, it’s the one who can’t bear the Actors Studio, it’s Hitchcock, that’s Hitchcock. Hitchcock seems to me to be the pure author, then, of what I am so crudely calling embryonic thirdness, the pushing of the action-image right up into the apprehension of the mental. And in fact, if I take the classical book – although we no longer have time – if I take the classical book by [Éric] Rohmer and [Claude] Chabrol on Hitchcock[12], what did they basically demonstrate? The role of relations.
He’s an Englishman, not without reason. What Hitchcock does is a logic of relations. It’s a logic of relations. With his great theme: a logic of relations seen again, obviously, seen again, or even not seen again, just seen – by Christianity. For after all, if the problem of the logic of relations was first posed in the Middle Ages, it was Christianity with the three Persons, with the Holy Trinity, which in effect founded logically – and here I’m not just babbling – founded in the logic of the Middle Ages, in Mediaeval logic, founded a whole logic of relations. Well, Hitchcock, it is well known, he is a Catholic, very much a practising Catholic, and his theme is what? It is indeed the one that would be identified by Chabrol and a whole gaggle of Jesuit students. It’s not the only one; it’s not mine, I’m not inclined that way myself. I am a product of secular [laïque] education, me. [Laughter]. But that’s what they keep talking about: the exchange of guilt [l’échange de culpabilités].[13] The operation arrived at does not belong to action, you understand. The action-image is … The exchange of guilt as Christian mystery, as Christian mystery analogous to the Trinity, this is what seems to me to explain this very curious cinema which pushes the action-image into an absolutely different cinema, which is one of the relation-image.
And is it by chance then that the same Hitchcock precisely makes a cinema of the demarked object, and that it’s on this that he will base himself, on the two complementary aspects, the symbol and the demarked object? The demarked object in Hitchcock is constantly the object that slips out of its ordinary chain, out of its usual chain. It’s the object that jumps out of its marks. The most famous example is in North by Northwest[14]: the plane. The guy finds himself stuck at the crossroads, he’s stuck there, and what’s that over there? There’s a plane spraying something to fertilize [sulfater] the crops. Yes, but it’s strange, there are no crops. So what the hell is the plane spraying? You remember this shot: the crossroad, the usual routes leading off, the plane in the distance, everything’s normal, we don’t even see or notice the crop-spraying plane, it’s all so normal; then all of a sudden, it comes out of its marks. Ah. Where are the crops to fertilize? There aren’t any! The demarked object will then become charged with suspense. What is it going to do? It approaches the man, it approaches the man [Deleuze repeats], and the man understands just in time that it’s going to sulphate something else entirely [Laughter]. Good. This is quite characteristic, and if we had the time, we could dwell on it further.
So, I would say that this is really the passage to thirdness, where the actions are only there in order to highlight the operations and relations, and where the indices are only there in order to highlight the marks and demarkings. And I would have added, if I had been able, that there is another case which seems extraordinary to me; and it’s not by chance that thirdness is as we have seen it to be.
Firstness, it’s one; action is one, two; thirdness, it’s one, two, three. It’s not surprising that embryonic thirdness in the burlesque would be the Marx brothers. And what would have interested me in an analysis of the Marx brothers, would have been to ask: why there are three of them? And how do they function? Because it really is American positivism. I mean, the Marx brothers are functionalist comedy. Why?
One could say that yes, there are comedians of firstness, one; there are the comedians of secondness, two: Laurel and Hardy. And if the Marx brothers are three, it is not at all by chance. It is absolutely not by chance. For when it comes down to it we would have to agree that Harpo is the man of firstness, and not only that, there is also a curious functionalist movement at play. He’s the man of firstness, he’s the man of affects. He has carried affects up to the level of the burlesque. He is the man of affects, but he is already the man of impulses. You see he is already in a firstness that prepares something else. He is the man of wild impulses, sexual impulses, alimentary impulses, ultimately. He is the man, he carries with him at the same time, he expresses from the moment he plays the harp, that he is pure affect, and as soon as he sees a woman, he is unleashed impulse. Good. Okay.
If I pass to the third, to Groucho: Groucho is the man of the logic of relations in the pure state, that is what he is. That will obviously be translated through the nonsense that is proper to Groucho Marx, which will in fact be a kind of nonsense that belongs to a logic of relations. Nonsense, a logic of non-sense has always been the indispensable complement of a logic of relations; it’s been like that since Lewis Carroll. Well, that’s what they did at the level of burlesque, in other words, there will be a veritable burlesque of relations, and always guaranteed in my opinion by Harpo … by Groucho, sorry.
Then between these two, oh what’s he called, he’s the one … [A listener: Chico]. Chico! Chico, what does he do? Well, it is very curious. He never stops, it seems to me … he has his own specificity: he never stops preparing marks and ratifying demarks, playing an intermediary role. That is to say, the ordinary figure of marks is: Harpo talks to Chico, they both form a group, and it is Chico who translates and who will translate for Groucho’s nonsense.
In the same manner, Harpo is a reserve of fetish objects. In effect, he is the impulse; he is not only the affect of the harp, he is the impulse unchained, like any impulse. He carries his object-pieces with him, and in his immense pockets, and from there he takes out any piece you can think of. And the task of Chico, it often seems to me, but anyway I don’t have the time to … As when he translates Harpo’s gestures and whistling, it will be to arrange [agencer] the objects provided or to select these objects, in order to constitute chains, in order to constitute chains that are apparently habitual, but which are in fact quite original, with which Groucho will then precisely be able to link up in order to assert his reversals of relations. And at the level of Chico, there will be a kind of demarking of the objects that the other one, Harpo, provides in series, etc. So that we would have here a fairly complete figure of this ‘enveloped thirdness’.
Finally, we are reaching the goal. You see, good. I would say that this is truly the moment where the movement-image finds its limit. It finds its limit. And we have seen from that moment – eighth level – what are we going to have? Well, we go back to the very first level. The movement-image was only the slice [coupe] of a more profound image, it was only the perspective of a more profound image. This deeper image will have its own signs, and it will be this that will constitute the domain of thirdness.
And we saw how, for the first time – and here I’ll just refer to the sequence [ordre], I no longer need to proceed two by two – in the sequence, for the first time, I would say, the action-image encounters an obstacle. It is what? It is what I call for convenience the ‘opsign’, the ‘opsign’, that is to say, the pure optical image, or ‘sonsign’, the pure sonorous image; that is to say, the sensory image cut from its motricity. This is a new type of sign. And this new type of sign comes in relation to what? We have seen it, with what I called the ‘modes’, and there will be as many signs as there are modes. That is to say, the opsign will be in relation with four sorts of signs that one could call the ‘noetic’ signs of the cinema, that is to say, the signs of ‘cinema-thought’.
And these four signs, I call them scenes in the case of the imaginary mode. I’ll call them that because I was looking for a word at any cost. Eras [ères], in the sense of ‘geological era’ or ‘historical era’, eras [Deleuze spells it out] for the didactic mode. [Pause]. I’m missing a word, but … When I get there, I might utilize the word trial [procès] for the critical mode that belongs to Godard, calling ‘trial’ any mode of extraction of the image from clichés, a trial brought against clichés, trials. And finally, aspect for the time-image, since the word ‘aspect’ would be guaranteed by the grammatical use that [Gustave] Guillaume makes of it in his theory of implicated time.
As a result, we would have two, dicisign and figure; icon and qualisign, four; symptom and fetish, six; synsign and index, eight; mark and symbol, ten; opsign, eleven; sce …, eleven, [scene] twelve, [era] thirteen, [trial] fourteen, [aspect] fifteen; lovely. I think Peirce had even more. [Laughter] [15] It’s not serious because we could make sub-divisions. These are signs after all; we’d have to divide them again, then we’d have to ask ourselves, what’s the best, the ideal of that, what’s its purpose?
For me, you understand, a table of categories is not as Kant conceives it. I dream of a very simple table of categories that would be like Mendeleev’s table in chemistry, that is to say, where there are empty boxes, and where we could say to ourselves, okay, that’s fine, these are images which do not yet exist. We could already give them a name, but they wouldn’t exist, you see? So, we could proceed like that, it would be necessary to invent other modes; we would say to ourselves, ah well yes that … no, that does not exist. And no matter, perhaps it will exist; or else it will never exist. A table of categories has to bear on the real, the possible, but even also on the impossible, … and indeed the … [Interruption of the recording] [46:12}
… For my part, what I wanted from you, throughout this year, was for it to awaken for you avenues of research that would be your own. I mean, it’s not about … it’s that it serves you in something which is for you … [unfinished sentence]. In this sense, it could be at any level; it could be at the level of you wanting to treat some type of image in a different way. I assume, if you accept … or it could be that you would have made another classification of images or signs in cinema, etc., etc. This is what I would have liked. Well, does anyone want to say anything …?
A student: I have a question, it’s not an important question, but next year, will you be talking about the Critique of Judgment?
Deleuze: Next year, next year, you know ….
The student: At the beginning of the year, you said …, you made a great topos; you told us that there will be a part on Kant’s Critique of Judgment.
Deleuze: Yes, that has completely disappeared … [A student close to Deleuze says: So much the better, which inspires some reactions to which Deleuze does not respond]. But next year, I’m sure, yes, I’m sure … I’ll be doing a semester for sure, in any case, I’ll be doing a semester on the history of philosophy, on Kant or on … Kant, I did him anyway, but not on the … Yes, I will perhaps do the Critique of Judgment, yes, yes…
The student: And what did you have in mind when at the beginning, you said we’ll talk about cinema, Bergson and all that, and then we’ll talk about the Critique of Judgment …?
Deleuze: What I had in mind was quite simple. It’s the admirable texts of Kant that consist in saying this: the beautiful is a pure question of form. Well, that might not appear to be much, it’s just …, okay. But, he says, that means that one cannot now carry on talking about the beauty of nature. There is no beauty of nature. But he says, there is all the same an essential problem here, and well, it is this that interests me. It is not that the beautiful is a question of form, that is nothing. It is that in nature, there is an aptitude to produce things whose form will appear to us beautiful.
In other words, the beautiful is a question of form. But there is a question, a very curious question of the material and natural production of pure forms. So it’s this activity of nature I thought to put in relation with the image which is not the movement-image. Because obviously what I completely suppressed in the sessions we have just done was the question I wanted to come to, amongst other things, which was: in what way is cinematographic perception another perception than our so-called natural perception? And what are the relations between the two? Last time I just suggested that we should call ‘diagrammatic perception’ this perception that grasps cinema-images. But what is this diagrammatic perception, and what relations does it have with nature itself? So, this is the case, it’s in this ensemble that Kant would have served me and no longer Bergson.
But next year, yes, I plan to go back to a very severe history of philosophy, especially since we don’t have a room, so there’s all that to deal with … all the same there is no question of us staying holed up in this vault; we’d all go mad, one after the other. So, I would like to do a very precise history of philosophy for the use of philosophers. But after all, you’re all philosophers, so that doesn’t help matters.
But listen, there it is, I’ve had a happy year anyway because, well, I said to myself, I took risks by talking about cinema, because I started low. But you have taken risks too; I mean, you have, we all took risks, and it was [A student is in the process of holding a sort of parallel discussion with someone] … potentially … yes ….
A student: … Imaginary, he called that the imaginary signifier.
Deleuze: Oh. You’re talking about something else … yes, yes, okay. And well anyway, thank you very much indeed.
A student: Happy holidays …
Deleuze: Of course, happy holidays [End of the recording] [50:34]
Table of categories of signs [please note: this revised table is our reconstitution based on Deleuze’s comments]
Perception-image (embryonic firstness) with objective pole and subjective pole: Dicisign (semi-subjective image, free indirect images) – Figure (emergence of the photogramme and its vibration).
Affection-image (firstness): Icon (close-up, affect on a face) – Qualisign (affect exhibited in an any-space-whatever).
Passage from the affection-image to the action-image (embryonic secondness): Symptom (a determinate milieu refers to an originary world) – Fetishes (pieces torn from the milieu by the impulses).
Action-image (secondness): Synsign (affects, qualities, powers actualized in a milieu; from the situation to the action, or the great form) – Indices (from the action to the situation, or the small form).
Passage from the action-image to embryonic thirdness: Marks (image forming an ordinary series referring to a habit; demarked objects) – Symbols (relation to be established or operation to be carried out; objects as bearers of relations)
Thought-image (thirdness, return to the movement-image as slice of a more profound image): opsign or sonsign (arrest of the action-image: pure optical image or pure sonorous image; the sensory image cut from its motricity) entering into relation with the modes, hence four sorts of ‘noetic’ signs, of ‘cinema-thought’: Scenes for the imaginary world; Eras for the didactic mode; Trials for the critical mode; Aspects for the time-image.
Notes
[1] Despite the fact that the first cassette of this session is missing, one can see that, in the light of the four great modes of images sketched in the preceding session (where Deleuze developed the sub-modes of the first, imaginary mode), Deleuze seems to have begun this session by continuing the development of the three other modes (didactic mode, critical mode, and temporal mode), and the discussion of Gustave Guillaume’s linguistics, with these linguistics, corresponding to the fourth mode. Drawing on Guillaume’s distinction between “signifieds of power” and “signifieds of effect” (in his 1953 essay “Psycho-systématique et psycho-sémiologie du langage”, reprinted in the posthumous 1964 collection Langage et science du langage [Language and the Science of Language]), Deleuze identifies various “signifieds of effect” of the indefinite article “one” or “a” [un(e)]. The logical basis of the distinction between the indefinite and definite article is also a recurring topic in the essays collected in Langage et science du langage.
[2] Cf. Gustave Guillaume, ‘Psycho-systématique et psycho-sémiologie du langage’ [‘Psycho-Systematics and Psycho-Semiology of Language’], in Langage et science du langage (Paris: A.G. Nizet et Québec, Presses de l’Université Laval, 1964), p. 246-247: “A no less important distinction […] is that between the signified of power [signifié de puissance] permanently attached in the language system [langue] to the sign (which becomes a signifier of it) and the signified of effect [signifié d’effet] with which the sign is momentarily charged, through the use which is made of it in discourse. The sign in language is the mediator between the signified of power and the signified of effect. […] The signified of power is the upstream of the phenomenon; the signified of effect is downstream.” Deleuze will return to Guillaume’s linguistics three years later, during a long discussion in ‘Cinema and Thought’, the fourth seminar on Cinema (1984-1985), in particular sessions 4.16 and 4.17. There are only two brief references to Guillaume in the Cinema books (both in the second volume, The Time-Image): first a reference to the “signified of power” as “a visual material which is the utterable of language” (p. 241), and then in the Conclusion (p. 262ff.), where Guillaume’s name is mentioned, along with some concepts, such as ‘psycho-mechanics’, and ‘chronogenesis’, that appear to be derived from him. The Cinema courses suggest a much greater dependency on Guillaume’s linguistic theory than is apparent from the books (see notes 3 and 4 below).
[3] Guillaume uses the term ‘particularisation’, rather than ‘individuation’ (as does Deleuze himself in his later reference to this theory in ‘He Stuttered’; cf. Essays Critical and Clinical, translated by Daniel W. Smith and Michael A. Greco, Minneapolis: University of Minnesota Press, 1997, p. 109). For Guillaume’s theory of the indefinite and definite article (and the underlying linguistic movement of universalisation and particularisation), see ‘Observation et explication dans la science du langage I’ [‘Observation and Explanation in the Science of Language, I’] (p. 34 ff.), ‘Particularisation et généralisation dans le système des articles français’ [‘Particularization and Generalization in the System of French Articles’], and further essays on the indefinite and definite article in Langage et science du langage; cf. also the extracts from Guillaume’s lectures in Foundations for a Science of Language, translated by Walter Hirtle and John Hewson (Philadelphia: Benjamins, 1984), pp. 52-54.
[4] On temps impliqué, cf. G. Guillaume, ‘Immanence et transcendence dans la catégorie du verbe: Esquisse d’une théorie psychologique de l’aspect’ [‘Immanence and Transcendence in the Category of the Verb: Sketch of a Psychological Theory of Aspect’] (originally published in Journal de Psychologie, January-April 1933, reprinted in Langage et science du language, pp. 46-58). It is not obvious how to translate temps impliqué, as temps can also be translated as ‘tense’ in the context of linguistics, while impliqué could also be translated as ‘implied’. The translation ‘implicated time’ can be justified by referring to Guillaume’s discussion of the term at the start of the article in question: “The verb is a semanteme which implicates and explicates time. Implicated time is what the verb carries with itself, what is inherent to it, making up an integral part of its substance and the notion of which is indissolubly linked to that of the verb. It suffices to pronounce the name of a verb like marcher [to walk] in order to awaken in the mind, along with the idea of a process, that of a time destined to carry it to realization. Explicated time is something else. It is not time that the verb retains in itself by definition, but time divisible into the distinct moments – past, present, future and their interpretations – that discourse attributes to it. This distinction between implicated time and explicated time coincides exactly with the distinction between aspect and tense [temps]” (‘Immanence et transcendence’, pp. 47-48).
[5] The connection Deleuze draws in this lecture between (grammatical) aspect and the idea of the time-image may indeed be rooted in Guillaume’s first major work, Temps et verbe: Theorie des aspects, des modes et des temps [Tense and Verb: Theory of Aspects, Modes and Tenses] (Paris: Champion, 1929), the first chapter of which seeks to determine ‘Les instants caractéristiques de la formation de l’image-temps’ [‘The Characteristic Instants of the Formation of the Time-Image’] (p. 7): the ability to use tense is said to rest on a fundamental ‘chronogenesis’ [chronogenèse], involving a basic temporal schematic ‘image’ of past, present and future (p. 8).
[6] See ‘Cinema: The Movement-Image’, Lecture 12, 9 March 1982.
[7] Peirce uses the phrase ‘modes of being’ in ‘The Principles of Phenomenology’ (in Philosophical Writings of Peirce, ed. Justus Buchler, New York: Dover, 1955, p. 75); Deleuze also relies on Peirce’s well-known essay ‘Logic as Semiotic: The Theory of Signs’ (also in Philosophical Writings).
[8] See Pier Paolo Pasolini, Heretical Empiricism (translated by Ben Lawton and Louise K. Barnett, Washington, DC: New Academia, 2005), in particular ‘Comments on Free Indirect Discourse’, pp.79-102. The phrase ‘free indirect image’ does not occur in the English translation of the text.
[9] Pasolini, ‘The “Cinema of Poetry”’, Heretical Empiricism, pp. 178-181.
[10] For Peirce’s discussion of the dicisign, see ‘Logic as Semiotic: The Theory of Signs’, p. 103. See Deleuze’s discussion of the dicisign in Cinema 1: The Movement-Image, p. 76.
[11] In Cinema 1: The Movement-Image (p. 238, n.21), Deleuze refers to Michel Ciment’s book of interviews, Kazan par Kazan (Paris: Stock, 1973; English version, Kazan on Kazan, London: Secker and Warburg, 1973). The references here to the mobility of the actors and to the use of objects probably refer to the following passages: “There’s a basic element in the Stanislavsky system that has always helped me a lot in directing actors in the movies. The key word, if I had to pick one, is ‘to want’. We used to say in the theater: “What are you on stage for? What do you walk on stage to get? What do you want?” I always asked that of actors; what they’re in the scene to obtain, to achieve. The asset of that is that all my actors come on strong, they’re all alive, they’re all dynamic – now matter how quiet. The danger of the thing may be a frenzied feeling to my work, which is unrelieved and monotonous” (Kazan on Kazan, p. 41). Shortly afterwards, Ciment goes on to ask Kazan, “How did you use objects to convey feelings?”, to which Kazan responds: “In Wild River, for instance, Clift is coming back to Lee Remick’s house. It’s raining, and Lee Remick has a towel in her hand. The inside of the house is warm. Whatever shyness she had, it’d be natural for her to welcome him into the house. But she’s shy about touching him except for the towel. The towel is an excuse to touch him. If it were a nice day and he’d just come in, the scene would have been impossible. One of the basic things in the technique of the Method is to use objects a lot. All objects are symbols of one thing or another. It’s something you can see move from one hand to another, you can see it break, or you can see it captured, you can see it sold, you can see it bought, you can see it transferred, you can see it embraced, you can see it thrown away. That’s like making an act out of a feeling, through the object. Of course, it helps actors who are self-conscious, because if they concentrate on the object they won’t be concentrating on themselves” (p. 45). Kazan then discusses the device of the swing in Baby Doll and an improvised scene in On the Waterfront in which Marlon Brando picks up a glove accidentally dropped by Eva Marie Saint and holds onto it, in such a way that “the glove was his way of holding her” (p. 46).
[12] Éric Rohmer/Claude Chabrol, Hitchcock (Paris: Éditions universitaires, 1957); translated by Stanley Hochman as Hitchcock: The First Forty-Four Films (New York: Frederick Ungar, 1979).
[13] Rohmer and Chabrol write: “The idea of the ‘exchange,’ which we find everywhere in his work, may be given either a moral expression (the transfer of guilt), a psychological expression (suspicion), a dramatic expression (blackmail-or even pure ‘suspense’), or a concrete expression (a to-and-fro movement) (p. ix). Cf. p. 108, on the “exchange of crimes” in Strangers on a Train; p. 114 on “community in sin” in I Confess; p. 139 on “the exchange of secrets” in The Man who Knew Too Much. Chabrol and Rohmer conclude: “We have called this movement ‘exchange’: let us recognize that it here finds its most noble expression in the idea of the interchangeable guilt of all mankind” (p. 149).
[14] Deleuze uses the French title, La mort aux trousses.
[15] There are a number of additions that we have added to this table which differs from the glossary Deleuze adds to chapter 12 of Cinema 1: The Movement-Image, pp. 291-293.
For archival purposes, the augmented and new time stamped version of the transcription was completed in February 2021. The translation was completed in May 2022, with additional revisions added in February 2024.[NB: The transcript time stamp is in synch with the WedDeleuze recording link provided here.]