January 26, 1982

We have to finish the perception-image today since we have so much to do. … We are now – thanks to image-perception – trying to identify something like a genetic element. A genetic element of what? A genetic element which would be the genetic element of the movement-image or which would be “a” genetic element of the movement-image … And this element therefore, thanks to the perception-image, thanks to our analysis of the perception-image, distinguished a genetic element from the movement-image which would be at the same time another perception. …, that is, another way of perceiving or what I was trying to call, quite simply … a non-human perception or a non-human eye.

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

Edited
Bergman, Persona
Bibi Andersson and Liv Ullmann in Ingmar Bergman’s Persona, Sweden, 1966.

 

After reviewing the key points of the previous session, Deleuze recalls references to Vertov’s cinema experiments, the double operation in his work causing transformation within the movement-image, and he links this to more recent analyses of American “structural cinema” and, notably, the experimental films of Michael Snow and George Landow. These film techniques, including blinking and looping, reveal molecular perception, especially evident in the development of video imagery in American cinema. But, Deleuze links these effects beyond the arts to mind developments with drugs, with Zen Buddhism, and alternate perceptions revealed in the writings of Carlos Castaneda: “stopping the world,” “grasping things within their web” (at a molecular level), creating “lines of power” and light. While these are translated cinematographically, Deleuze argues, in experimental films, they also tend to be reinjected into the cinema movement-image (three examples named are from Antonioni, Bergman, and Godard). After Deleuze’s summary of the perception-image analysis with a nine-point outline, Richard Pinhas gives a brief presentation on looping and recording innovations, focusing on the experimental musician, Robert Fripp. Deleuze then moves forward discussing of the affection-image for which he immediately summarizes with the expression, “the affection-image is the close-up, and the close-up is the face.” Of course, Deleuze must distinguish characteristics of each kind of cinema shot, and he links the particular importance of this shot and the fact to the horror genre, notably with the two poles of the close-up, an intensive pole or series through micro-movements, and a pole of reflected or reflecting qualitative unity.  Moreover, he links the former to traits of faciality and the latter to surface facialization, both of which he examines at length, notably through a brief history of portraiture. Then, with different cinema references (Pabst, Griffith, Clouzot, Albert Lewin), he creates a polar comparison between Griffith and Eisenstein. Hence, after devoting this session’s first part to chapter 5, he shifts to the first section of chapter 6 in The Movement-Image. Following this presentation, the session ends with an exchange between Deleuze and Georges Comtesse (heretofore absent from the French transcript).

Gilles Deleuze

Seminar on Cinema: The Movement-Image

Lecture 08, 26 January 1982

Transcription: Hamida Benane (part 1, duration = 1:10:31) and Nicolas Lehnebach (part 2, duration = 1:01:44); augmentation of transcriptions and time stamps, Charles J. Stivale

Translation, Christian Kerslake

 

Part 1

We have to finish the perception-image today since we have so much to do. Now happily we have made progess in the analysis of the perception-image, because we no longer distinguish, or oppose or play against each other, two poles of perception of which one could be called the objective pole and the other could be called the subjective pole. Where have we got to then? We are now – thanks to the perception-image – trying to isolate something like a genetic element. A genetic element of what? A genetic element which would be the genetic element of the movement-image, or which would be a genetic element of the movement-image – one must always look out for nuances. And this element therefore, distinguished thanks to the perception-image, thanks to our analysis of the perception-image, distinguished a genetic element of the movement-image which would at the same time be an other perception. An other perception, that is to say, another way of perceiving, or what I was trying to call, in a summary fashion – but also feel free yourselves to introduce all the nuances you like – a non-human perception or a non-human eye. And in fact, it should be expected that these things are the same: genetic element of perception, genetic element of the perception of movement, and non-human perception or non-human eye.

Now in this regard, the development or the progress made by [Dziga] Vertov, which I had attempted to show at the end of the last session, is obviously something very important for the history of the cinema. For Vertov’s whole attempt began, as we saw, in news and documentaries, where Vertov started out from a free treatment, an original treatment, of the movement-image. But his treatment of the movement-image was so free and so original that we have seen in what sense his later work, Man with a Movie Camera, exceeded the limits of this first attempt. And it exceeded its limits in what sense? Well, in the sense of a double operation; and it is this double operation which it seems to me is analyzed very well by Annette Michelson in the text I invoked, namely the collective work Cinéma, théorie, lectures, published by Klincksieck, in which there is an article by Annette Michelson on Vertov.[1]

This double operation, if I try to summarize it, by referring you to this article, would consist in this: on the one hand, extract from the mediate movement-image one or several photogrammes.[2] And by that to ascend to a couple of the type photogramme/interval, or already, photogramme/blinking [clignotement]; and it is evidently this couple photogramme/interval, photogramme/interval of movement, or photogramme/blinking, which you perhaps sense, in effect, that we could consider as the genetic element of the mediate image of the movement-image. And at the same time, as giving us an other perception, as constituting the camera eye, that is to say, a non-human eye. And at the same time, the inverse of the operation: it is only valid if one reinjects the couple photogramme/interval, or photogramme/blinking, into the movement-image, whether it be in order to change the nature of the movement, that is to say, to reverse it, to accelerate it, etc., etc…, or whether it be in order to obtain an alternation between the two, an alternation that can be more and more precipitated.

A concrete example from Man with a Movie Camera: the presentation of a series of photogrammes and it is only afterwards – for example, a photogramme of a peasant or a photogramme of children, of the heads of children – and it is only afterwards that one recognizes them in movement-images of a traditional type. It does not seem like much, but we will see that there is perhaps something very interesting that occurs in the cinema at this moment. Or another procedure: presentation of a normal image, of a mediate movement-image, the path of a cyclist, and then a refilming presenting the same movement-image, but refilmed under the conditions of the movement unfolding on the screen. A procedure of refilming and then comparison, an alternation between the two situations that is more and more precipitated. The cyclist’s path presented under the ordinary conditions of the movement-image; the same cyclist’s path presented under the conditions of refilming, and the passage from one to the other precipitated more and more, with increasing rapidity. [Pause]

When I talk of the importance – you can see this tendency: in fact it indeed concerns the isolation of a genetic element within cinema, a genetic element from the point of view of cinema, a genetic element of the movement-image, and by the same token, an invitation to us to an other perception. One understands that there is a kind of operation here in which, in effect, the new couple photogramme/interval is perpetually extracted from the mediate movement-image, but on condition of also being perpetually reinjected into the movement-image, even if it means literally changing the ‘look’ of the mediate movement-image. If this is a very curious kind of set [of images], perhaps the importance of these attempts by Vertov could only appear later, when one is more sensitive to the importance of this type of endeavour. Why? Because one has been made sensitive to it by everything that has happened afterward, namely by the American cinema known as independent or experimental. And in the same text published by Klincksieck, Cinéma, théories, lectures, there is another article that is just as good, by an American critic called [P.A.] Sitney, on what he calls ‘structural cinema’[3] in America. And I will just identify – I refer you to this article – three procedures, or three directions, belonging to this so-called ‘structural’ cinema.

First direction: extract the photogramme or a series of photogrammes. Thus to substitute for the mediate movement-image of the 24 images-per-second type, extract a photogramme or a series of photogrammes and, now prolonging the photogramme, now repeating it with an interval, [one arrives at] what one will call – and here the term is important, we will see why soon – a ‘loop procedure’; and in fact if you take a loop constituted by the repetition of a series of photogrammes with intervals, the possibilities of playing with the delay [décalage] allows the procedure in this first direction to become extremely complicated; that is to say, you can even obtain at the limit ‘superimpositions [surimpressions]’, superpositions of the series, of your series of photogrammes at different moments. For example, a superimposition of the end of the series onto another moment. So with the loop procedure, you allow for not only repetitions and echo phenomena, but phenomena of superimpositions.

Second direction: no longer to substitute for a mediate image, the series of photogrammes taken in a loop operation, but to substitute blinking for the movement-image, as a type of vibration of matter. Now, I am not too familiar with the nuances between two American words here; two words seemed to be employed: blink and flick[er]. One of the first blinking films was in fact that of [Norman] McLaren – I can’t remember the date [1955], but McLaren made a very beautiful experimental film with blinking procedures, called Blinkity Blank. Blank, I believe, is the empty, it is a void but a special kind of void, no? [Someone in the audience confirms]. It is a type of Zen void, eh? Literally, the blinking void, Blinkity Blank. And then someone else, I believe, by the name of Tony Conrad, made a film called The Eye of Count Flickerstein [1967]; that is to say, the eye of the blinking Count [du comte clignotant]. That would be the second direction.

Third direction: to substitute for the space of perception a flat space, grainy [granuleux] and without depth. A granular space, obtained how? Obtained by a very simple procedure, through a series of refilmings, through re-recording, a re-recording of images projected on the screen.

Of course there are all kinds of procedures, but I identify these three as particularly important in experimental cinema or in ‘structural film’ in Sitney’s sense. These three procedures – which animate so many experimental American films – perhaps permit us to better understand that of which Vertov was simultaneously the herald with Man with a Movie Camera, and which was not yet singled out as a procedure, for it concerned what? I will borrow from the pages of Sitney a type of description of a film that I like a lot. (It’s being shown currently, you should go: the classics of American structural cinema are being shown or will be shown at Beaubourg … but there is a strike at Beaubourg; anyway they were on the schedule … but I saw that there are other places they are being shown at the moment, so if you feel like it … In particular, there is one of the greats of structural cinema, we will talk about him a little later on, called Michael Snow, they are showing Michael Snow. [A student comments on showings of these films]. Ah good, good, those who have not seen it, do go, everything will become clear.)

I have extracted [Sitney’s] synopsis of the film Bardo Follies [1967] by [George] Landow, another of the great men of the structural cinema. This is what Sitney tells us: it is something like a kind of summary of everything I have just said rather badly, but a concrete summary: “The film begins with a loop-printed image” – understand that the image is a photogramme – “the film begins with a loop-printed image of a water flotilla carrying a woman who waves to us at every turn of the loop.”[4]

This is the first time. I would say – I talked to you about it very quickly last time – I would say that it is not by chance that it all begins in a blink of an eye with an aquatic image, for one will witness in the film, it seems to me, the singular passage from an aquatic image to a typically gaseous image. “The film begins with a loop-printed image of a water flotilla carrying a woman who waves to us at every turn of the loop. After about ten minutes …” – obviously, enough is enough – in parentheses, he says that a shorter version exists, go and see the shorter version [Laughter] – “the same loop appears doubled” – observe that this is not a a phenomenon of superimposition, it is a phenomenon of juxtaposition – “the same loop appears doubled into a set of circles against the black screen. Then there are three circles for an instant. The film image in the circles begins to burn”. That is to say, the burning of the photogramme is recorded. “The film image in the circles begins to burn” – this is a fundamental stage, provoking the expansion of “a moldy, wavering, orange-dominated mass.” And that is truly the passage from the liquid state of the image to the gaseous state. I have not invented that, this is exactly why Landow is one of the greatest, no? Good, where was I?

“The film image in the circles begins to burn, creating a moldy, wavering, orange-dominated mass. Eventually, the entire screen fills with one burning frame, which disintegrates in slow motion in an extremely grainy soft focus.” Obviously the refilming already guarantees the granular space – all that is in the film, Bardo Follies. “Another photogramme burns”, then more, and the whole thing will start to burn. “Another photogramme burns; the whole screen throbs with melting celluloid.” Ah, that is beautiful. “Probably, this was created by many generations of refilming from the screen – its effect is to make the screen itself seem to throb and consume itself. The tension of the desynchronized loop is maintained throughout this section, in which the film itself seems to die.”[5] “After a long while”, we reach the end. You see, after a long while, what has happened? “It becomes a split screen of air bubbles in water filmed through a microscope with colored filters, a different color on each side of the screen. Through changes of focus the bubbles lose shape and dissolve into one another and the four filters switch. Finally, some forty minutes after the first loop, the screen goes white. The film ends.”

Good, so that’s that then; you understand. If you grant me what we saw last time at the end and what I have said just now, what does it involve? It involves truly constructing with the cinema what you could call what? A molecular perception: the genetic element of the movement-image will itself be grasped in a molecular perception as a kind of equivalent of a micro-perception, or of what I called, in opposition to liquid images, a gaseous perception. Why gaseous perception or molecular perception? I must reiterate what I have said before: what is involved here is not a metaphor, it is evidently the same thing once again; a gaseous state is the state in which molecules can move freely, in an average free path, as opposed to with the liquid state and the solid state. [Pause]

Now such a procedure, the attainment of a molecular perception, all that, with the procedures – the loop, granular space, you see, the photogramme which would truly be treated as a cinematographic molecule … the looping procedure, the granular space, all that – is to be compared with things that happen in other art forms. I mean in both painting and music. In painting itself, it is necessary perhaps to go a little further back, but not very far back, in order to find a famous “granular space” which certain Americans reclaim, namely the pointillist space of Seurat; and one sees types of structural cinema which are very familiar with Seurat, and which through their procedure of refilming think of themselves as obtaining a species of granular space. Or take what also happens in music in the same period, namely a musical procedure which is that of loops with the possibility of making use of intervals and playing with the intervals in such a way that one is able to obtain superimpositions, superpositions of two different moments of the series. And in a moment when I have finished with that, if Richard Pinhas would like to say a few words on loop procedures in a musican like [Robert] Fripp today, you will see that there is a kind of analogy between sonic loop-procedures and cinematographic loop-procedures.

Now, I have said that it is thus this molecular perception that simultaneously gives us the genetic element of the movement-image and what we called the non-human eye, non-human perception, and we have just seen that one could also say that it is thanks to all the attempts of American structural cinema that one is more sensitive to what is extraordinary in Vertov’s venture, Man with a Movie Camera. Must we not say something more? Namely, that what currently makes us sensitive to these experiments in American structural cinema is the advent of a new kind of image, the video-image. And that for a simple reason, which I do not want to go into now, but perhaps Richard Pinhas will give us some indications on that. If one very crudely defines the video image as being no longer an analogical image but a coded image, as a kind of digital and no longer analogical image, the first thing that defines, that characterizes, the video image is that it plays on an infinitely greater number of parameters than the [analogical?] image … [Interruption of the recording] [24:30]

… in this sense, the video image procedures would be a confirmation of the same line – but because we’ll come back to the video-image later, I won’t develop that here – for all that after all also has a relation, I would point out, with things that happen on granular spaces in painting, because it is not Seurat who … I mean, Seurat in this regard is like a kind of initiator, if you like, in the same way as Vertov, bringing about a prodigious beginning, but the current preference of some (not all) [artists] for reconstituting granular spaces is quite extraordinary, we talked a little about it last year, notably among those who paint on the reverse side [of the canvas], or those who paint on special materials like tarlatan: here there is a very interesting constitution of granular space.[6]

Good. What I have just said not only has a relation with what happens in other arts, it is obviously in relation with what? With the movements of the mind, if one can put it that way. It is very difficult to dissociate in principle – I do not mean in fact, it does not matter if they are on drugs or are not on drugs, that is not that interesting – but it is very difficult to dissociate in principle certain experiences connected to drugs from the most beautiful experiences connected with Buddhism, with Zen Buddhism.

Why? It’s not difficult: here one would have to take up again from this point of view the film by Michaux[7], but also the so-called ‘American structural cinema’, for drug experiences were active in this type of cinema, and why? Why is this? It is because with drugs or Zen – Zen is better, but philosophy already sufficed – there is truly the accession to a molecular perception, in what sense?

Here I take the classic book – which will involve at the same time molecular perception and the genetic element of perception – the popular book, a book I find very interesting, and which was a great success for a while: Castaneda’s book, or the books by Castaneda, on his initiation through hallucinogens by the Indian sorcerer, the good sorcerer. Well, what do I get from that? I retain what interests me, and you will see that it is exactly the same thing.

The lesson, the lesson of the Great Sorcerer, is what? First great proposition: you will have done nothing if you have not arrived at stopping the world.[8] Ah: stopping the world! It is necessary that I stop the world: good. But what does that mean, stopping the world? You can sense it: to extract the photogramme from the immediate movement-image. But that is not quite what is meant, there is something else involved in ‘stopping the world’ according to the Great Indian sorcerer, it means acceding to a ‘not-doing’, to accede to ‘not-doing’.[9] One must break ‘doing’, one must to prevent any doing. In some cases, there is no trouble in our breaking ‘doing’, let us shatter ‘doing’, let us accede to not-doing. Let us stop, let us stop the world. Good. ‘Not doing’: what is what? The doing is the subjective image, recall what we saw with Bergson: the action-reaction relation, subjective perception, is the perception that consists in grasping the virtual action of the thing on me and my possible action on the thing.[10] Subjective perception is ‘doing’. To stop doing, what is that? It is to accede to another type of perception, to stop the world: good, there, that’s our first theme.

Second theme: if you have arrived in your perception at stopping the world a little, it is indeed a kind of effort to transcend the movement-image. But why do it? Or rather, why not-do it? Well, the first phenomenon that is given to you already as a splendid compensation is magnification, the insane magnification of things. Things literally become close-ups, things enlarge themselves. You see a face, not under the experience of drugs, which is always miserable, but in a Zen illumination, and apprehend it in colossal dimensions. Why is that interesting? [30:00] If that doesn’t happen to you, it’s because you haven’t taken enough, by why is it interesting?

It’s prodigiously interesting because at this moment the thing becomes holed [trouée].[11] The larger it is, the more holed it is. You will not accede to molecular perception unless you attain to the ‘holes’ in each thing. Take the video image, which is easy to make holes in. All that makes for a certain kind of whole [ensemble]. It is necessary that you grasp the thing as Castaneda says: “to grasp things in their pattern [trame]”[12], each thing’s pattern; the thing is holed. When you have stopped the world, the thing becomes enlarged, reveals its holes. Castaneda’s analyses are beautiful even when taken in a literary sense: it is the perception of water, the molecular perception [31:00] of water, the molecular perception of air, the molecular perception of movement, and each time molecular perception means: to have stopped the world, to obtain this magnification of the image and to grasp the holes in the image.

Water does not have holes in the way air does, for you do not know how each thing has holes: a face does not have holes like another face, so that you never know in advance in what way a thing has holes. What is that, then? I would say that this is precisely the theme of the intervals of movement, to grasp the intervals in a movement, in a movement that appears to you as continuous with ordinary perception; but it is not at all! To grasp the intervals; it is not nothing to grasp the intervals in the gallop of a horse. What wisdom! It is only if you have known how to stop the world that you will grasp the holes in the thing; and once again each thing has its way of having holes, there are no two things that are holed in the same manner. This, then, is the interval, and it is through it that you will typically obtain a blinking world, which is everywhere blinking at different rhythms: this is the vibration of matter.

And, thirdly, through these holes, you make pass … and this is the most mysterious operation.  At a pinch, you can understand the two first points, but for the third, a little magic is necessary! It is necessary to have arrived at the Zen stage, and then to enter a state that everyone dreads. Final stage: through these holes, you will make pass lines of forces, lines of forces, which are sometimes lines of light, and which trace themselves in this stopped universe; and on these lines of force, accelerated movements will produce themselves: this is the famous hyper-rapid montage of the structural cinema.[13]

Everything connects here; I mean, each of these three aspects. And it is on these lines of force that pass through the holes of things that the initiate in Castaneda sees the sorcerer dance, that is to say, leap, leap at a speed that transcends all conceivable speed, that is to say, leap from the height of a mountain onto a tree, and then leap from the tree to the mountain, etc., a sort of prodigious … [The barking of a dog is heard. Deleuze laughs]. Ah, I know that one, I recognize him. Who excited the dog? Who hurt the dog?

So you see, these three aspects, one can present them in three stages of the Zen experience or the hallucinogenic experience, and also as three stages of the photogramme image, of the photogramme/interval couple, which constitutes the genetic element of perception and gives us at the same time an other perception, a so-called ‘non-human’ perception.

I have almost finished with this story of the perception-image, and I would just like to say: does that mean – I recall my warning, which I really would like you to take seriously – that structural cinema is ahead [en avance] of the other forms we have come across in this long analysis of the perception-image? Once again, no: I would even say, if one calls structural cinema a form of the avant-garde, I would say and I have already said ten times, that what is proper to the avant-garde is without doubt that it has to exist, it must be made, it must be made, and there have to be people who dedicate themselves to it, this is evident. But who is most creative? I mean, what is proper to the avant-garde is to be without issue [sans issue][14], it is to be without issue, that is to say, to receive its issue from something else. That means that all that is not worth as much than the conquest of a molecular perception and of a genetic element. It seems to me that this only has worth in so far as it is reinjected either back into a cinema with stories and narrations – even if the boundaries are so fluid that there is really no place for attaching any importance to these categories – or in any case back into the movement-image; and if you do not reinject it, if you have do not have the genius to effect this reinjection, experimental cinema can only sterilize itself, to the point that it would be necessary to ask: who is the strongest, the most inspired? The one who sets out on the experimental path, or the one who reinjects the experimental data into the movement-image?

And here I will take two, three examples. [First,] Antonioni. What happens in this cinema in which it is well-known that it is much more the strong affective moments, rather than the movements, that count for what Antonioni himself presents as his problem: the intervals between movements? Second example: What happens when – this is an example that someone gave me last time, because I have not seen this film – Bergman feels the need to burn a photogramme and refilm it, to burn a photogramme of a face, in Persona [1966]? Third example, what is happening in the famous cycle promenade in Godard’s Sauve qui peut [1980]?

Good, I give three examples: who is the creative one? The experimentalists, who are if need be great cineastes, or the other great cineastes who reinject? I mean, where is the maximum of invention? There is no way to measure that, but it must not simply be said that the one comes after the other, that the latter just utilize what the former have discovered. That is not the case: there is a kind of … there are directions, it is as if there are differentiated lines in creation and invention, on which I would say that one of Antonioni’s intervals is obviously just as important as one of Landow’s experimental intervals, just as the burning photogramme in Bergman is just as important as the one in Bardo Follies. Good, so it’s up to each of us, there you go.

So: taking account of that, and because I am loyal, I am concerned that you get a sense of exactly where we are now. I will simply summarize what I hold to be our acquisitions – after all, we are not taking too many acquisitions for granted – corresponding to this new part which has just been concluded, namely, the perception-image as one of the cases of the movement-image. You will recall that in effect the movement-image had three cases, according to our analysis: Perception-image, action-image, affection-image. I have just concluded the analysis of the first case, the perception-image. I would like to summarize what we have acquired in the form of nine remarks.

The perception-image is therefore the first type of movement-image, under the conditions we saw previously. First remark: what has allowed us to organize an analysis – it was not necessary, but we found it to be so – of the perception-image is the distinction between two poles of perception. The one we called, by convention, objective, but we had reasons for calling it ‘objective’: this was the regime of universal variation, the universal interaction of images, that is to say: all images vary at the same time for themselves and in relation to each other. From that moment, for me at least, one has suppressed the ridiculous false problem of ‘an objective perception should be without montage’! Not at all, not at all, if one understands what objective means, that is to say: the place of universal variation, universal interaction; then it is obvious. On the other hand, we called the subjective pole the variation of all the images in relation to a privileged image, whether that of my body, or that of the body of a character. This was our point of departure, and this is our first remark.

Second remark: these two poles being thus defined, the objective and the subjective, one does not cease passing back and forth between one pole and the other. We have seen that in a sense we intersect here – and I’m really just summarizing what we have acquired – in a sense at this level, the perpetual passage from one pole to another, we intersect with the problem: shots, reverse-shots.

Third remark: one could posit the following principle: that the more the subjective center of reference – since the subjective image is the image related to a center of reference, it is the images that vary in relation to a center – is itself mobile, the more one will pass from the subjective pole to the objective pole. Example: the marvellous images of Dupont’s Variety[15] [1925], in which the subjective center of reference is an acrobat in movement, and in which the vision of the whole of the circus, from the point of view of such a dynamic center of reference of movement, already passes from a regime of universal variation, that is to say an objective pole.

Fourth remark: in these perpetual passages from objective to subjective and from subjective to objective, it is as if a specific form of the perception-image is born in cinema. This specific form of the image in cinema is the ‘semi-subjective’ [‘mi-subjective’], as Jean Mitry baptised it; the half-, the semi-subjective, the semi-subjective image, as named by Mitry. And in fact, the status of the semi-subjective invites us to isolate a kind of ‘nature of the camera’, defined as ‘being-with’. The ‘being-with’ of the camera, this being-with, consists in what? What is effectuated, for example, in the travelling[16] of a closed circuit when the camera no longer contents itself with being-with, with following a character, but displaces itself among the characters.

Fifth remark: if one tries to give a true status, a conceptual status, to this semi-subjective, one must seize the opportunity provided by what is simultaneously, it seems to me, one of the exceptional theoretical attempts in the effort to think cinema, namely Pasolini’s venture. An exceptional theoretical venture, to the extent that it culminates with a concept that I attempted to analyze as closely as I could this time, ‘the free indirect subjective’. The free indirect subjective image: which refers to precise technical processes, which are no longer simply the camera displacing itself among the characters, but the technical procedures Pasolini defined like the zoom (we have not yet spoken about that, because I want to keep it for later, but it doesn’t matter), the doubling of perception and the immobile shot or obsessive framing. And which, according to Pasolini, indeed defines one direction in cinema, for example in Italian cinema, not only, but Italian cinema after neorealism and which would have its privileged examples in Antonioni, Bertolucci and Pasolini himself.

Sixth remark: at the stage of analysis to which Pasolini has carried us, if you like, with this new notion, something decisive already emerges, namely, that the mediate movement-image, in which we have been installed since the beginning, tends to differentiate itself along two directions: first direction: subjective perception to characters in movement, subjective perception of characters in movement who leave and enter the immobile frame. Second direction: objective self-consciousness of cinema through itself. In the form of the obsessive frame. The danger, if there is a theoretical danger, is that this ‘self-consciousness’ of cinema presents itself once more, if you like – from a theoretical point of view, I am not talking of a practical progression – as an idealistic consciousness, or as a pure aesthetic consciousness.

Seventh remark: the fixed consciousness of cinema by itself must be that of the objective pole, that is to say, that of universal variation, or of universal interaction. That is to say, if you like, it is quite simple in relation to the preceding remark, it should not itself be simply a component of the perception – it must be itself an object of perception. It is in this sense that we have found, in a certain direction of cinema, the coexistence of two objects of perception, if one can say that: the liquid object as simultaneously objective and veridical, the solid object as subjective and partial. And in this coexistence of two regimes of perception, liquid perception and solid perception, there already began to be born what preoccupied us, what began to preoccupy us, that is to say, the possibility of a molecular perception: at that moment, the cinema consciousness, in a strictly ideal sense, was flowing water; and it seemed to us that that defined a whole French school between the two wars, at which historical point we pause.

Eighth point to note: one more step. It was necessary that the two poles from which we started no longer be simply two polar objects of perception like the liquid object and the solid object; it was necessary that there should be two forms of perception, and what is more, two forms of which the one would play the role of genetic element in relation to the other, that is to say, that the one truly plays the role of a ‘micro-perception’, of a molecular perception; and this, it seemed to us, was the direction that was sketched out by Vertov, which was taken up again by structural cinema, and where the movement-image is now found to be transcended towards the photogramme/interval couple.

Ninth and final remark: here, once again when we summarize our acquisitions, we cannot prevent ourselves from acting as if there were a kind of progression here. It goes without saying that here there is still no progression, and that the great creative arts make advances as needed, if you take my eight levels, when a more developed level is reinjected into the preceding level. So that what counts is the whole of the schema, without one direction being worth more than the other, and so that the whole of the schema consists in saying what to us? That the perception-image is run through by a kind of history – not by History [as such] – by a kind of history which pushes it to put in question the notion of the movement-image.

The perception-image starts by being a type, from the point of view of our analysis, of movement-image. But it only develops itself and only develops its poles, in tending to transcend the movement-image towards something else, towards another type of image, and that in two ways. This is our last remark: First way: it is no longer the movement which is an interval between positions in space, it is now, on the contrary, the interval between movements that will raise us to a reality. Second acquisition: it is no longer the cinematographic movement which is more or less illusory in relation to the real movement, but it is the real movement and its cinematographic transcription which are illusory in relation to a real cinema. What an acquisition! What gains!

What remains to us? You can see what is left, it’s very simple, we will have to do the same thing, provided that nothing falls out of place! The ideal would be to succeed in doing the same thing for the two other types of images, the affection-image and the action-image, and if we succeed in doing that, we will have exhausted the movement-image. Moreover, we will not have exhausted it, because it will have led us to another type of image, for the other types of image than the movement-image, there are ten, a hundred of them, there are so many, so that they will keep on coming until the end of our days. Perfect: of mine perhaps. Great, perfect!

Okay, so we will soon begin a new type of image, after a very short break for you, but I would like it if Richard Pinhas feels himself … if you are ready to read what I … but you don’t seem to be ready … oh well. Ah, you are getting up! If you wouldn’t mind, because we can’t hear you that well. Great, you’re coming over. [54:35]

Pinhas: [Only fragments of Richard Pinhas’s intervention are audible] … I’ll take a very simple example … a sound, for example, is defined by four parameters … very, very schematically … We’re going to be able to analyze and connect the components at the level of a single element, for example, at the level of the timbre, one will no longer … We’re going to see ten thousand little parameters which are analyzed starting from a real instrument …. and it’s from the moment when we will be able to vary in an unreal [irréelle] manner, that is to say, not modeled on real variations  … we will manage to be able to reproduce … In the milieu of video, because of the coding, it seems to me that this gives access … to much finer parameters than the usual parameters …. The actors, the light, the depth …. and from the moment we create or recreate an image from calculations or syntheses, we will be able to reach micro-elements that will be similar to those of the film, and we will manage to make images … with micro-movements [55:00-58:32]

Deleuze: Although in music the first one to come to mind to have invoked a type of gaseous state, or at least a chemical state, is Varèse, it is Varèse who is really at the origin of … Yes?

Pinhas: What is fantastic in the case of Varèse, is that [inaudible sentences] …

Deleuze: He absolutely lacks the technical instruments that are necessary for his music [Pinhas talks, inaudibly] … Yes, that’s it! Aside from that, all these linkages are grotesque, I can’t help thinking that it’s grotesque, but it’s because it seems to me to be somewhat the same situation as that of Vertov in cinema. It is to have the idea of … and yet one cannot say that Varèse’s work is lacking anything! … And in fact it proves, in a certain way, that he makes something which could only be heard when instruments of the synthesiser type become available, which do not yet exist in the period in which he is making it. And in fact I think there are a few cases, there are many cases like that; I think that there are many cases like that! I think that one can love and admire Varèse during his lifetime, but could one hear him fully? Of course! There were people who fully heard him, and who understood him very, very quickly! But for us, it appears obvious to me that for the average listener, one can only understand, or hear, truly hear, Varèse, once the synthesiser exists. It is very, very curious, that!

If you don’t mind, would it trouble you to try to say a little – if it can be said clearly enough – about Fripp’s procedure of loops and superimpositions, because that would maybe help everyone!

Pinhas: [Inaudible sentences; commentaries on techniques of delay in Fripp][17]

Deleuze: I heard it! Did anybody here hear Fripp when he came to Paris? It was very, very … In my opinion, it was beautiful, very beautiful!

Pinhas: [Inaudible sentences, still on Fripp’s techniques]

Deleuze: Formed of patterns [trames], it’s a good thought, we have our theme, to form a pattern, and finally the loop procedure is the formation of a pattern, to form adequate patterns, which vary according to each theme or each thing.

Pinhas: [Inaudible sentences, still on Fripp’s techniques].

Deleuze: Yes, this is the ‘fixed consciousness’ aspect! That’s the ‘stop the world’ aspect!

Pinhas: [Inaudible sentences].

Deleuze: Yes, all of that was enlightening. And so now are you all rested? [Laughter]. Yes, then we will continue.

So now we start a new part … You don’t want a break, then? And I might as well say it now: you’re smoking too much, you have to stop smoking! One can hardly see, the eyes are stinging [Sounds of movement, people coming and going]. We’ll catch a cold, that’ll be good, won’t it? [Pause]

Well, now and next time we are going to be occupied – don’t you want to close the door? Because open doors give me anxiety – we are going to be occupied with the second type of image: the affection-image.

And here I want to come straight out with a kind of formula which could serve as a guide, although one would literally never anticipate where it will lead us. What I want to say is – the fact that I want to say this is a sign of something, of the fact that with the perception-image, it took us a long time to find a formula which sketched things out – so I want a kind of formula straightaway – well, we’ve got one straightaway, it’s very simple, it’s totally simple, we have the impression that the secret is to be found here, well, I have the impression that the secret is found here: namely, the affection-image is the close-up [gros plan] and the close-up is the face. So, one point there, and salvation. The affection-image is the close-up and the close-up is the face. I repeat it.

I repeat it and it is obvious that there are all sorts of problems here straightaway. The formula appears fully satisfying to me, I almost want to say nothing else. And then one feels that there are all sorts of things, namely, that there are possible objections, which are so obvious, which is fine, but nevertheless, [despite] all these possible objections, I have the impression for my part that the formula holds even with regard to all the objections. But that would make it a mystery and then suddenly this formula which appears so simple, one perceives that it is more complicated.

So first of all, there are all sorts of objections. Immediate objection: well, what does all that mean? Already there are all sorts of close-ups that are not of the face. There are all kinds of close-up. But what the close-up also means, is the face. That implies that the formula would not give me the same contentment if I said: a close-up of the face. It is not that the close-up is a close-up of the face, the close-up is the face. It might be that the formula is false; but it isn’t, although I do want to say that its simplicity is false, that it risks drawing us into paths which do not work, which work less than one might think.

Fine, for our problem is what? What transpired with the perception-image, in our preceding analysis of the perception-image, was that we arrived at a criterion to govern the analysis. We had the criterion that governed the analysis from the moment we were able to distinguish two poles of the perception-image. Even if our analysis makes us take these poles in progressive directions, which vary according to a progression. Obviously, understand the conditions of the problem: it is not a question of saying: “well, in the affection-image there are two poles, an objective pole and a subjective pole”, since if that holds for the perception-image, it does not hold for the other. We need a completely different guiding line of analysis. Now, this guiding line, I come back to it, I’m happy with it, I repeat it (it is not that it is a good formula, but I sense that it is not untrue, that a truth resides in it): The affection-image is the close-up, and the close-up is the face.

Now, a little text by Eisenstein, which has been translated, a very short text which has been translated in the Cahiers du cinema, says something which is very intriguing, very interesting, it seems to me.[18] He says: take the three great types of shot [plan]: long shot [plan d’ensemble], medium shot, [plan moyen] and close-up [gros plan]. One must see that it is not simply three sorts of images in a film, but it is three ways in which it is necessary to consider any film, three coexisting ways in which it is necessary to consider any film.

And he says that with the long shot, there is a way in which, whatever the film you are viewing, Eisenstein proposes, as a way of seeing the film, it is necessary that you see it as if it had been made uniquely on the plane of the long shot; and then at the same time it is necessary that you see it as if it has been made uniquely on the plane of the medium shot; and then it is necessary that you see it as if it has been made uniquely of close-ups. And he said it is necessary because, the long shot is what refers back to the whole of the film, and when you see a film you become sensible to the Whole. And then the medium shot is what refers to something as the action or intrigue or story, and when you see a film, it is necessary that you are sensible to the action. And the close-up is the detail, and when you see a film, it is necessary that you are sensible to the detail.

Fine. For our part, we would say something different, but which comes to the same thing. We would say: well of course! The long shot is the perception-image, the medium shot is the action-image, and the close-up is the affection-image. And when you see a film it is necessary that you see it as simultaneously made exclusively … [Interruption of the recording] [1:10:45]

 

Part 2

… [77:34][19] And then, what succeeded him was a tone of critique of cinema that was a little more guarded in tone: “That doesn’t happen to us. Anyway, that doesn’t happen to us, don’t exaggerate.” As if one felt shame at this lyricism of the first men of cinema, whereas now there is a more critical vision of the close-up, sometimes inspired by psychoanalysis, which assumes or suggests that perhaps the close-up is not unrelated to castration.

And then we find ourselves before a liberation of close-ups. We don’t even know who to think of here, as every great man of the cinema has them, and has signed them as his own. And furthermore, although before we did not feel any need to talk about actors, it is more difficult here, and we will sometimes have to envisage and talk about the actor, as if a close-up were co-signed by the one who lends their face to the author of the film. So it is with the couple, Marlene Dietrich – Sternberg, for example. And with the great couples, it is at the level of close-ups that the great director – actor/actress couples come about. Indeed it’s our only – you understand – it’s our only liferaft [bouée], or at least, it’s my only liferaft,[20] but we will have to escape from that. My problem here is to start from this formula, however disputable it is: the affection-image is the close-up and the close-up is the face. Are we going to be able to extract a method of analysis from the affection-image?

So, first point. It is necessary that I do not assume anything, it is necessary that you don’t reproach me for slipping anything into the formula, so I start from an example of the close-up which is precisely not a close-up of a face. A close-up that crops up all the time in the history of the cinema, a close-up of a clock [pendule]. How is this a close-up? Well, surely because it is seen up close, because it is seen from close by, as one says, because it is seen from right next to it. But that is not it. What makes it a close-up? For example: take the close-up of a clock you have seen many times. These close-ups of clocks exist very early on in the cinema. I will cite two types of close-up of a clock: the Griffith close-up, there are many examples here of close-ups of clocks; and the Lang close-up, there are many close-ups of clocks in Lang; and elsewhere too.

What is a close-up of a clock? I do not say “What is a clock?”, “What is a clock which can be a close-up?”. I say, “What is a clock as a close-up?” Well, it seems to me that it is two things, so perhaps we are saved. It is two things. On the one hand, there are the hands, by which we measure seconds if need be. Sometimes we measure to the hour, sometimes to the second. I would say that these hands are only valid in close-up as susceptible to movement. They have a virtual movement, since the close-up can without doubt show them moving, but it can also show them fixed; even when they are fixed they have a virtual movement. Indeed, they can have a minuscule movement and then skip a minute.

The horror film – about which we have much to say actually, because it articulates something in relation to the close-up – has played a lot on the infinitesimal instant before the fatal hour, before midnight. I would say that in the close-up of the clock, the hands are inseparable from a virtual movement, or a possible micro-movement. And even when a close-up is first shown, ‘11 o’clock at night’, followed by the close-up ‘midnight’, the emotion we associate with the close-up comes from the fact that, with each reading, we animate the hand with a movement – I should just say, this is just to [illustrate?], I’m not saying I found this schema [in a specific film?] – with a movement that we could call, at the limit, a virtual movement or a molecular movement.

I just note that this molecular movement or this virtual movement would be nothing in the close-up if it did not enter into an intensive series which is at least possible, even if it is not shown to us, a series in which intensity increases. I would say therefore that they are micro-movements in so far as they enter into a virtual intensive series, one which is at least virtual. That is the first aspect of the close-up of the clock.

Second aspect of the close-up of the clock, co-existing with the first. This is an immobile receptive surface, it is a receptive plate [plaque]. Immobile receptive surface, receptive plate, or if you prefer, it is a reflecting and reflected unity. Reflecting and reflected unity represented by the dial and the glass. And how are they complementary? Well, obviously it is the unity of the micro-movements. In other words, it is the qualitative unity of the intensive series, represented from the other side. There you have it. Fantastic, that’s it – I mean, we’ve got our poles. And what reassures me is that we have not at all just modeled them on the two poles of the perception-image. Here we have newly found two poles of the close-up, which are not at all modeled on the two preceding poles of the perception-image, and which hold for themselves. The two poles of the close-up are: micro-movements taken in an intensive series on the one hand; on the other hand: reflecting and reflected qualitative unity. [Pause] [86:00]

What is a face? A face is the complementarity of a reflecting and reflected unity and of micro-movements, and of micro-movements which determine an intensity. That is a face. Good, so that’s a face? Well, yes, obviously that’s what a face is! We will call the reflecting and reflected unity ‘surface of faceification [visagéification]’; we will call the micro-movements which enter into intensive series ‘traits of faciality [visagéité]’. And we will say that the face is the product of an operation of faceification, through which the reflecting and reflected unity subsumes, grasps the traits which are from that moment traits of faciality – intensive traits which then become traits of faciality. And that is a face: qualitative unity/ intensive series. Good![21]

You will say to me that that is a face, but it is also something else and that is a close-up. And a close-up is what? A close-up operates. A close-up is not necessarily a close-up of a face, but a close-up is necessarily a face. It is not necessarily a facial close-up of course, but it is a face: a close-up operates the faceification of what it presents. The close-up of the clock operates the faceification of the clock. Which means what? Which means a very very simple thing: it extracts from the clock the two correlative and complementary aspects of the intensive series of micro-movements and of the reflecting qualitative unity. And what that amounts to, the two aspects of the face, is obvious, for what makes something a face?

What makes something a face is two things, but it can only be that way: [1] a face feels, [and [2] a face ‘thinks of’ [pense à]. A face feels, that means what? It means: it desires, or – which comes to the same thing – it loves and hates. It loves or it hates or indeed both at the same time, as we often say. That is to say: it passes through an intensive series which decreases and increases. And it thinks of, it thinks of something. [90:00] And that is no longer the pole of desire, it is what one could call the pole of admiration. He admires … Why do I say that? Perhaps one will understand later, but the English have a word that suits us here, the French, alas, do not. I shall say it with my best accent: I wonder [in English]. ‘I wonder’ is ‘J’admire’, but it is also ‘I think of’. On the side of the face as reflecting and reflected qualitative unity, I would say it is just as much ‘I admire’ as ‘I think of’. Fine. On the other side, it is ‘I desire’, ‘I love’ or ‘I hate’. I traverse an intensive series; I increase and decrease.

The face therefore has two components, which are those of the close-up. There is, on the one hand, a component which one could call the content … no, no that would not be good! There is a first that one will call: trait of faciality. Traits of faciality are the movements on the spot, virtual movements which traverse a face in constituting an intensive series. On the other hand, it has a contour. This is a true pendule[22] in both senses. There is a contour under which it is the reflecting and reflected unity. I come back home in the evening exhausted, or indeed a woman comes home … No, the husband comes back in the evening exhausted after a long day’s work. Then, his wife says to him … No, no … He looks at his wife, he opens the door … no, let’s try to be cinematic – he wipes his feet, he opens the door, his wife looks at him, and he says to her moodily: “What are you thinking?”, to which she responds: “What’s up with you?” – you see, it is from one pole to the other. What are you thinking? This is the communicating and reflecting face, that is to say, it is the quality he has on the face. What quality emanates from your face? And the other answers: What’s up with you? What’s the matter with you? What is this strange intensive series that you traverse in rising and descending? From such small beginnings, the domestic scene is engaged with close-ups, it’s all done in close-up.

That already existed in painting, these two aspects of the face. In painting these two aspects of the face existed in such a marked fashion that I read or re-read (for completely different reasons, we needed it last year) a text by Wölfflin on the development of the portrait in the 16th and 17th century.[23] The portrait of the 16th century and the portrait of the 17th century. And here again, we don’t make any progress. Let us choose a portrait of the Dürer type, or again Holbein. He establishes – this is the first type of portrait – he establishes its form by appealing to a very sure and categorical line. The contour of the face progresses – you see, this is the contour-face – the contour of the face progresses from the temples to the chin in a continuous and rhythmed movement by means of a line that is regularly accentuated. The nose, the mouth, the edge of the eyelids are drawn with a single enveloping trait, from one continuous enveloping line. The various kinds of hat belong to this same system of pure silhouette. The artist knew how to find a homogeneous expression even for the beard, sometimes modelling them out of smears, he had absolute resort to the principle of palpable form. This is therefore the continuous line that makes a contour and which in this sense refers to the tactile no less than to the eye. It is the contour-face or the qualitative face, reflecting and reflected unity.

Another type of portrait: in perfect contrast with this figure, consider a head by Lievens, a contemporary of Rembrandt. Here, all the expression that is refused to the contour has its seat in the interior of the form. Two dark eyes with a lively gaze, a light quivering of the lips, here and there a line that sparkles, in order to disappear again. One would seek in vain for elongated traits in the drawing of the lines. A few fragments of line indicate the form of the mouth, a small number of dispersed traits, those of the eyes and eyebrows. Here, it is the face led back to the side of its other pole: the traits of faciality. The portrait will be constituted by discontinuous traits of faciality, and no longer by an enveloping line which makes a contour. A few fragments of line indicate the form of the mouth, a small number of dispersed traits, those of eyes and eyebrows; often the drawing is completely interrupted. The shadows that figure the model no longer have objective value. In the treatment of the contour of the jaw and the chin, it seems that everything is being done to stop the form becoming a silhouette, that is to say, it is capable of being deciphered [déchiffrée] through its lines.

Therefore painting already confirmed to us the two correlative aspects of the face: it is the traits of faciality dispersed in such a manner to constitute an intensive scale on the one hand; on the other hand, the contour line that makes a qualitative unity of the face. Every time a thing is reduced to these two poles in such a way that the two poles co-exist and refer to each other, you can say “there has been faceification of the thing”, and you will see with stupor and astonishment that you find yourself before a close-up. Hence you could already say: the close-up is the face, and there is no close-up of the face, but there is no other face than the close-up. And when the close-up focuses on something which is not the face, it is in order to faceify the thing.

This is what Eisenstein certainly understood in an obscure way when he started a celebrated and admirable text[24] by saying: the true inventor of the close-up is Dickens. When Dickens starts a famous text with the sentence “The kettle began it …”,[25] there is in effect a close-up at work and Eisenstein has extracted it. Everyone can see it’s a close up, but what makes it that? It is a Griffith close-up. It is a Griffith close-up, but we can also ask, what has made it a close-up and by virtue of what is it a close-up? It is a faceification of the kettle and it is by virtue of that that it is a close-up.

Indeed, let us take a sequence of close-ups, or rather a sequence that includes close-ups … All that is … what’s the word …. we haven’t really begun yet, we’re just amassing confirmations, we’re amassing data, materials. Always on these two poles of the face. You see I start at … this is why I thoroughly insist that there is no mere modeling from the two preceding poles of the perception-image. We are in the process of finding the two poles proper to the face, that is to say, to what will reveal itself – but I have not yet explained this – to be the affection-image. And all that is complicated, it is necessary to go very carefully – well, it’s not necessary, if you are tired, give me a sign and I will stop straightaway…

I am thinking of a long sequence – because sometimes I’m forced to cite films I haven’t seen, but I believe it’s best when … well, sometimes it’s films I have seen, not that that changes anything – it’s [in] Pabst’s Lulu.[26] The splendid conclusion where Lulu encounters Jack the Ripper and is killed by him. The script of Pabst’s Lulu has been published in English. I read it with the help of a dictionary, and it gives the end there; I will just introduce some divisions since the cutting [découpage] is not indicated.

This is the encounter of Lulu … they have encountered each other down below, but I’ll take up the film at the moment that Jack the Ripper is in the bedroom, the sordid chamber of poor Lulu. There is a surprising scene of relaxation, with facial close-ups. They are relaxed [Ils sont détendus], they admire each other, they surprise each other, they play … and the English word ‘wonder’ appears here. And they play, she plays, Lulu plays with his pockets. She asks him for money but he doesn’t have any, and it does not matter; this is a great moment of sweetness, as if Lulu had recovered all her youth, all her freshness. She rifles through the pockets of Jack the Ripper and pulls out a little sprig of mistletoe just given to him by a woman from the Salvation Army; she clasps the mistletoe, she puts it on the table, and the faces are happy and relaxed. This is the first time. I would say this is the first pole of the the face. Lulu is even quite mischievous, she thinks of something, she clasps the mistletoe. She expresses the first ‘wonder’: to marvel; and the second ‘wonder’: to have a thought. A charming scene – my God, why can’t it stay like that? For everything ends badly.

Second moment: she has lit his candle on the table, and he sees the knife which gleams, he sees the gleaming bread-knife. Good, let’s stop with this second movement, for in the history of cinema we know a number of facial close-ups [gros plans de visage]; I would rather now say “close-up-face” [gros plan (trait d’union) visage]. We know a number of close-up-faces which are famous because they were particularly audacious. This is the succession of the facial close-up and the subsequent image, what the face is thinking about. And at the beginning, it literally made the audience shout out – as the histories of cinema recount – because it was visually difficult as an association. A facial close-up, and then, after the connection has been made with the subsequent images, it became simply that in the subsequent images one sees what the face was thinking about.

And that was one of Griffith’s great tricks. In a famous example, Griffith makes a facial close-up of a young woman who is thinking about something – you see, it is always the pole ‘thinking about something’ – and then the subsequent image is her husband, in a completely different place. Translation: she is thinking about her husband. Good, easy! Fritz Lang also used this procedure: facial close-up, connected to what the face was thinking about. Okay, put that on one side. Why am I mentioning it now? Well, presently we have the same procedure, but much more logical. One begins by making a close-up of what the character thinks about: the close-up of the knife, the face of Jack the Ripper which already manifests a terror. You see, one begins by showing the object of his thought and then one shows the close-up-face afterwards. That happens very often in the cinema.

There is a very beautiful case that I like very much in the history of cinema, and it is – I’m not saying I like the film, I’m saying that I very much like this moment – Clouzot’s The Murderer Lives at Number 21.[27] There is a singer, she sings. And while she sings, there is a sudden close-up: three roses, three flowers. And then – I’m not quite sure, I no longer recall – three candles. That whole succession, if I remember rightly, is in close-up – it would be necessary to see the film again, … but in any case, there are some close-ups – three flowers. There is a group of three people who are listening to her. But a succession of close-ups. And then when one returns to her, to her face: she has understood everything. Here, in The Murderer Lives at Number 21, one goes from the thing she has grasped, and thinks about, to her face in so far as she thinks, and what she is thinking there. She understands everything, namely, she understands that the murderer who is being looked for is not a single person but three persons, three persons at once. Thus, it’s the same procedure as with the knife.

You see that you can therefore do both procedures with the face-thought [le visage-pensée]; face, then linked with: ‘that about which it thinks’; or, on the contrary, the object grasped as ‘that about which it thinks’, and then the face comes afterwards. But then, at the threshold: the bread-knife, which imposes itself upon Jack the Ripper. He grasps the bread-knife, this is his thought, and he tips over into the other pole of the face. He tips over into the other pole of the facial close-up, because, third moment, starting from this initial terror – when he grasps the bread-knife – it is clearly said in the script that his face will run through all the degrees of terror to the point of paroxysm. End of the tranquillity, the serenity of the face-thought, of the face-amusement, of the face-admiration, of the face-surprise earlier on. He has entered into the terrible intensive series of traits of face-unity.

In effect, what one sees here is a trait of face-unity. This is all the traits of the face which escape from the beautiful qualitative organization of the thinking face. The mouth goes to hell, the mouth stretches, the eyes become exorbitant, as if the traits which earlier on composed the calm face take on an autonomy. But at what price do they take on an autonomy? It is to enter into an intensive series which will burst open the face, which will burst open the face in a panic, a panic of mad terror. All that will serve us well for the cinema of horror. Good: but until what moment? The script puts it clearly: up to the moment where he resigns himself.

The final moment, the fourth moment of this sequence (from the point of view which occupies us, that is to say, of close-ups): he resigns himself, he knows that he can no longer fight it. He accepts, he accepts what? He accepts the thought. He comes to the thought in the form of resignation. Well, it is decided: I will kill her. And at that moment, he experiences a sense of relaxation. He experiences a sense of relaxation. He has returned to the organization of the face, this time no longer in the form of the innocent face that admires, but in the form of the face that receives its destiny, that is to say, he has changed, the face has changed quality, but it has returned to the qualitative stage of the face. And so he grasps the knife – and I won’t say how it ends because it’s too sad.

So, we have given ourselves this material in order to justify what? In order to justify the face, what the face is, the sense in which the face is the close-up. Here, we have advanced a little. But we have not at all justified the beginning of the formula: the affection-image is the close-up and is therefore the face. I propose to you that this should be our last effort for today, because we have done a lot. It’s already true – I add this at any rate in order to maintain my acquisitions – the great actors, notably the great actresses of the close-up, and the great directors of the close-up, what do they know how to do which is not nothing? It is the only moment, I think it’s the only moment where there is a true collaboration between the actor  and the director. If not, the actor does nothing; but the close-up is not nothing; what I’m saying, of course, only holds for the cinema. What does a great director of the close-up know how to do? And it is not easy. He takes a close-up and he will show, sometimes in one order, sometimes in another order, the face which first of all thinks about something, and which then feels something; or the other way around.

For those of you who have recently seen that very beautiful film Pandora [and the Flying Dutchman] [1951], there is a close-up – it is true that one cannot talk of the close-up by just talking about a Sternberg close-up (or whichever [director] you like), without also adding the name of the actress. In the very beautiful Pandora, which is by [Albert] Lewin, there is an Ava Gardner close-up; she is a great specialist in close-ups. And when you see this close-up, it is fantastic, because she starts by putting itself into the arms of the man she loves, the Flying Dutchman. And at that moment, she tilts her face towards him, and there is a facial close-up of Ava Gardner, and all you would have to do is see this face for you to say: this is the reflecting unity that expresses a single quality: love. And the contour of the face [is] of a purity, of a beauty; it’s very beautiful, very beautiful. Then, all of a sudden, an air like … sometimes it is truly nothing, a little play of the lips and eyes. And you cannot help saying to yourself: look, she’s thinking about something. It passes from one plane to the other. It is not the same type of face; it could be in the same close-up, but it is absolutely not the same type of face, there has been a reorganization. A tiny edge of the lip that has split off, that has split off beyond the qualitative organization, leading one [to think]: what is she thinking about? She thinks about what ceases to be a problem for her: that she will give her life for the redemption of the soul of the Flying Dutchman. There, that is a thought. Good, so you have close-ups that are mainly: the face is thinking of something. You have close-ups that are mainly: the face is traversing an intensive series.

Moreover, in the famous history written by Eisenstein,[28] again, splendid pages, but where everything is tangled together, could one not say that at the beginning of classical cinema, there were those two great poles: the Griffith pole and the Eisenstein pole? If one asks oneself what the contribution of Griffith is when he imposes the close-up on cinema, the answer is simple: he is the one who does the most beautiful face–contours, which are face-contours to such an extent that they are often surrounded by a masking [cache].[29] In the Griffith close-up, there is a masking, in the middle of which there is a face, all the rest is in black, a circular masking. One cannot better indicate the face, or underline the face, as contour, except that in the images that follow one discovers what this face perceives, or what this face is thinking about. I would say that with Griffith the close-up-face is formed as a reflecting and reflected qualitative unity.

And what can reflect a face more beautifully – even if it is tragic … I’m going to say that what reflects is the white, or ice. I have already said too much, but it doesn’t matter. The white or ice. A close-up of Lillian Gish, a Griffith-Lillian Gish close-up with frosted eyelashes.[30] It is Eisenstein who says: Griffith has understood that the icy aspect of a face could just as well refer back to the physical quality of a world as to the moral quality of an atmosphere, that the face can express ice for better and for worse, that it expresses the white – the white of ice, or the white of love. Okay, but this could be the Griffith pole.

What is Eisenstein’s fantastic contribution? It is not him, one cannot say that, any more than … but what did he truly invent? We will try to say next time in what sense he invented it. But the intensive series of faces, each one only valid through a trait of faciality, will tend towards an insane paroxysm. In monsters, in the cinema of horror … okay, no matter. Is there not a Griffith pole and an Eisenstein pole at the beginning of cinema, where the face is understood by Eisenstein in the form of traits of faciality which enter into an intensive series; by Griffith …, etc. But of course, this leads into vast realm, and we will have to do it next time.

But you see, what preoccupies me, the last thing that now preoccupies me, I find very interesting. Affection: what is affection? I’ve assumed it, I’ve presupposed it from the beginning. Why is that also affection? What I have just explained is roughly what I’ve amassed from the material in order to justify the identity face and close-up. But my identity was not double, it was triple. Why is it also affection? It would be necessary to arrive at [an account] with the same certainty.

Well, what is an affection? What is it? Well, we will seek it finally on the face. For philosophers, this is not their problem, even if it can be their problem, but in the end it’s not their principal problem; but in turn, the-close up, that’s also not at all their problem; however, when we come to affection, that is their problem. Perhaps then they have something to say to us, the philosophers, on affection … they are the specialists. And what is affection? If you have not completely forgotten Bergson, and if you liked – perhaps you recall – the definition Bergson proposed of affection (and this is my first attempt to amass the material from this side).

He says to us: an affection, it is not difficult, and he comes out with his splendid formula: it is a motor tendency on a nervous surface, it is a motor tendency on a nervous surface [Deleuze repeats] … No, what am I saying? I’ve garbled it; it’s a motor tendency on a nerve, on a sensible nerve – which comes to the same thing, but I don’t want to over-complicate it. You’ll see straightaway where I want to get to. The right way is simpler: it is a motor tendency on a sensible nerve.[31]

And I have already given a preview of this; what more beautiful and what better definition of face is there than this? If you want a true definition of the face, it is not what I have just attempted to give, because what I’ve just attempted to give, I believe it is true, but it is not beautiful. On the other hand, Bergson says the same thing, it’s the same thing after all. We’ll see that it’s the same thing. A motor tendency on the sensible nerve, that, as one can see, is a face. What does he mean? He says that this is what an affection is. This is what an affection is: fine.

He means that in these very particular images[32] – we have seen, I make a very quick tour back – in these very particular images, where there are not reactions and actions that follow immediately, but there is a phenomenon of delay – you recall? – well, these very particular images or these very particular bodies are fabricated out of sense organs. That is to say, instead of reacting with all their organism, instead of grasping the excitations with all their organisms and reacting with all their organism, they have delegated certain parts of their organism to reception. [Pause] This is a great advantage, they are made of sense organs, instead of reacting in bulk. They are made of eyes, a nose, a mouth, across a long history which is that of evolution. That had a great advantage, specializing certain parts of the bodies for the reception of excitations.

And yet – what an inconvenience! The inconvenient thing was that from then on, they would immobilize certain regions if they were to delegate the reception of excitations. They would immobilize … This is a terrible thing for a living being – at least if it is not a plant – to immobilize its organic regions. And what is affection? The advantage was that that permitted them to perceive at a distance, without waiting for contact, thanks to these sense organs. [Pause in the recording] [2:04:20]

If we now split Bergson’s definition, we rediscover our two aspects of the face. The traits of faciality which refer to an intensive series on the spot, or to micro-movements, this is the motor tendency and its virtual movements. The nervous surface immobilized, the receptive plate, this is the communicated face, reflecting and reflected. [125:00] So that Bergson’s definition is indissolubly a definition of affection, we can now add that he did not realize that he had just defined the face better than anyone had ever defined it. Good, for after all – another twist – it’s in any case not by chance that our sense-organs, except for our hands, which are so faceifiable [visagéifiables] – in parentheses, the latter were quite often the object of the close-up – it is not by chance that our sense organs are localized on what one calls the face.

Good, the sense organs are localized on the face … and what has that to do with affection? [126:00] It is because our face, whether deceptive or not, expresses the affections that we have, or which we feign to have. And, once again, it is the most convenient means for expressing the affections, whether it be involuntarily, or voluntarily. And why is it that the face expresses affections? And why is it a function of the face to be the expression of affections? Here we leave Bergson, for there are so many philosophers just waiting for us to read them, and we jump to a great text of philosophy: the Treatise on the Passions by Descartes.[33] For in the Treatise on the Passions, there are certain articles – it is divided into articles – which explain to us why there is a link between the passions and the face [127:00].

Right, I’m afraid that you can do no more. Is that okay, or not? Shall we stop? Do you have any remarks to make, any short remarks to make? [Pause] No remarks? Ah! [127: 30] [The end of the session is omitted from the transcription at Paris 8].

Georges Comtesse: [Remarks barely audible; he comments on ‘the operation of faceification’, and on a book by Raymonde Carrasco on Eisenstein, Hors-cadre Eisenstein (1979) with regard to two kinds of close-up; Carrasco will participate in the seminar and will intervene several times during the fourth year, 1984-1985] [128:00-129:28]

Deleuze: [He interrupts Comtesse] Can I point out in any case that it’s me who wrote all that.

Comtesse: What? [Laughter]

Deleuze: I don’t usually quote myself, sorry, but at the end, that bothers me a little because almost all those last sentences have to be attributed to me. They are literally written in Mille plateaux, the sentences you read.

Comtesse: But I was reading from the text of Raymonde Carrasco, and there is evidently a plagiarism! [Laughter].

Deleuze: I have done lectures on the face … [130:00] It doesn’t matter, it’s not a question of priority, but I mean, what you’ve said suits me fine, as even if I haven’t developed that here, these are things I’ve already said!

Comtesse: [Barely audible; here Comtesse seems to use what Carrasco says in order to make an objection against what Deleuze has just said with regard to faciality] [130:13-131:00]

Deleuze: You do know that this often happens to us, Comtesse, between you and me. Me, what I call face, you call non-face. [Laughter] Then, if I say, this is finally how to put it: there is no face other than the close-up – then you, you say, no, the close up is non-face. At that point, I say, okay, what I call face, you will call the non-face, but I do not know if there will be a real difference. And everything you say, I would have already said, I think I have already said it, once again. That does not bother me too much; it’s simply that you call it non-face; I call it face; adding that we have only one thing in common, that the face is a horror, yes, it is horror, it is terror. The face is terror as much on the side of thought as on the side of intensity. You say that the non-face is a horror, it is a terror; then we are in agreement. Okay? [132:00] I see between us only one difference – and this often strikes me when you speak, it is that when you speak (and I do want you to talk), but every time, I say to myself, this is simply a question of words. I could have done my whole session in the form, ‘I am going to speak of the non-face’. I have reasons for calling it a face. You, you have reasons for calling it a non-face. But this is a minimal difference. In my opinion, I feel, once more, our agreement is more profound than … [Interruption, voice of Claire Parnet]. How life is beautiful! [Laughter]. Well, listen, we’ll see next time. [End of the session] [2:12:38]

 

Notes

[1] Cinéma, théories, lectures, ed. Dominique Noguez (Paris: Klincksieck, 1973). Annette Michelson’s article, ‘L’Homme à la camera. De la Magie à l’Epistémologie’, was originally published in English as ‘The Man with the Movie Camera: From Magician to Epistemologist’, Artforum, March 1972.

[2] Cf. Cinema 1: The Movement-Image, pp. 84-85.

[3] P.A. Sitney’s ‘Structural Cinema’ (anthologised in Sitney (ed.) Film Culture Reader (New York: Praeger, 1970)) was translated into French as ‘Le Film structurel’ (in Dominique Noguez, (ed.) Cinéma, théories, lectures, op. cit.). Deleuze cites this text in Cinema 1: The Movement-Image, 230 n.22.

[4] Here Deleuze cites and comments on P.A. Sitney, ‘Structural Cinema’ (in Sitney ed. Film Culture Reader (New York: Praeger, 1970), p. 340. Deleuze uses the French translation, ‘Le Film structurel’, also included in Dominique Noguez, (ed.) Cinéma, théories, lectures, op. cit. The paragraph from Sitney is cited without commentary in Cinema 1: The Movement-Image, p. 86.

[5] Sitney’s English text has been modified here to reflect the French translation used by Deleuze. Sitney’s original reads: “The tension of the silly loop is maintained throughout this section, in which the film stock itself seems to die” (‘Structural Cinema’, p. 340).

[6] Cf. Deleuze’s sixth seminar on painting of 19 May 1981.

[7] Henri Michaux, Images du monde visionnaire (1964).

[8] Carlos Castaneda, Journey to Ixtlan (Penguin, 1974), pp. 9-13, 120, 151.

[9] Castaneda, Journey to Ixtlan, p. 196ff., 202-214.

[10] Cf. Henri Bergson, Matter and Memory (translated by Nancy Margaret Paul and W. Scott Palmer, New York: Zone Books, 1991), pp. 21, 56.

[11] Castaneda, A Separate Reality (Penguin, 1972), pp. 223ff.; Journey to Ixtlan, pp. 204, 212.

[12] Castaneda, A Separate Reality, p. 223. The word Deleuze uses, trame, comes from the French translation of Castaneda’s original word ‘pattern’. The French translation of A Separate Reality, by Marcel Kahn, Voir: Les enseignements d’un sorcier Yaqui (Paris: Gallimard, 1973) has “[Alors les sons reprirent et] immédiatement je saisis leur trame” (288) for Castaneda’s sentence “I picked up their pattern immediately”, and ‘pattern’ is consistently translated as trame in subsequent passages. In this passage of A Separate Reality (which is the only one of Castaneda’s books to be cited in Cinema 1: The Movement-Image [p. 230, n.13]), Castaneda is mostly concerned with ‘holes’ between sounds.

[13] On luminous lines and the ‘lines of the world’, see Castaneda, Journey to Ixtlan, pp. 207, 266-67. At the beginning of his initiation, Castaneda also experiences the transformation of a dog into an “iridescent being” after taking peyote for the first time (The Teachings of Don Juan [New York: Penguin, 1970], 42-43); described again in Journey to Ixtlan, 266.

[14] Sans issue could also be translated as ‘no way out’, or ‘deadlocked.’

[15] Also known as Vaudeville. Cf. the discussion in the 6th session of this course, 12 January 1982, and in Cinema 1: The Movement-Image, p. 77.

[16] In English in the original.

[17] Richard Pinhas discusses Robert Fripp and ‘frippertronics’ within the broader musical context in the chapter ‘Le rhythme et la modulation synthétique’ in his Les Larmes de Nietzsche. Deleuze et la musique (Paris: Flammarion, 2001), 155-207.

[18] Eisenstein, ‘A Close-Up View’, Film Essays and a Lecture, ed. Jay Leyda (Princeton: Princeton University Press, 1970), pp. 150-154.

[19] The tape recording here starts again, with the repetition of approximately the previous seven minutes (i.e. starting from Deleuze’s words at 64:00): hence the apparent gap in the time stamps.

[20] The word in the recording, bouée, alone means “bouy”, but as bouée de sauvetage, it would mean “liferaft”, which we’ve selected. In any event, Deleuze seems to mean a resting spot from which the discussion needs to move forward.

[21] On the face, cf. Cinema 1: The Movement-Image, pp. 87-90.

[22] Pendule can mean ‘clock’ or ‘pendulum’.

[23] Cf. the 6th session of the seminar on painting, 19 May 1981; Deleuze also refers to Wölfflin’s Principles of Art History [1915] (translated by M.D. Hottinger, New York: Dover, 1932), in Cinema 1: The Movement-Image, p. 230n.1.

[24] Sergei Eisenstein, ‘Dickens, Griffith and the Film Today’, in Film Form: Essays in Film Theory, edited and translated by Jay Leyda (New York: Harvest/HBJ: 1977), pp. 195-255. Cf. Cinema 1: The Movement-Image, pp. 91-92.

[25] This opening sentence of Dickens’ 1845 novella The Cricket on the Hearth is also the first sentence of Eisenstein’s essay.

[26] Pandora’s Box [1929]; original title Die Büchse der Pandora; directed by Georg Wilhelm Pabst, starring Louise Brooks.

[27] L’Assassin habite au 21 (1942), directed by Henri-Georges Clouzot.

[28] Eisenstein, ‘Dickens, Griffith and the Film Today’, op. cit.

[29] In Cinema 1: The Movement-Image (cf. p. 16) Deleuze makes use of André Bazin’s distinction between the cadre (frame) in painting and the cache (masking) in film: “The outer edges of the screen are not, as the technical jargon would seem to imply, the frame of the film image. They are the edges of a piece of masking that shows only a portion of reality” (‘Painting and Cinema’, What is Cinema, Vol. I, translated by Hugh Gray (Berkeley: University of California Press, 1967), p. 166).

[30] Deleuze returns to the example of Gish in the following session.

[31] Bergson, Matter and Memory, p. 55 (translation modified).

[32] Deleuze reverts to the terminology of the opening pages of Matter and Memory, in which one’s own body is one “image” among many others (Matter and Memory, p. 17).

[33] Deleuze cites Descartes’ The Passions of the Soul in Cinema 1: The Movement-Image, p. 230, n.2.

 

Notes

For archival purposes, the augmented and new time stamped version of the transcription was completed in February 2021, with added notations about the Gallimard CD in June 2021. Additional revisions to the transcription as well as the translation were completed in February 2022, with additional updates in January 2024. [NB: The transcript time stamp is in synch with the WedDeleuze recording link provided here.]

Lectures in this Seminar

square
Reading Date: November 10, 1981
right_ol
square
Reading Date: November 17, 1981
right_ol
square
Reading Date: November 24, 1981
right_ol
square
Reading Date: December 1, 1981
right_ol
square
Reading Date: January 5, 1982
right_ol
square
Reading Date: January 12, 1982
right_ol
square
Reading Date: January 19, 1982
right_ol
square
Reading Date: January 26, 1982
right_ol
square
Reading Date: February 2, 1982
right_ol
square
Reading Date: February 23, 1982
right_ol
square
Reading Date: March 2, 1982
right_ol
square
Reading Date: March 9, 1982
right_ol
square
Reading Date: March 16, 1982
right_ol
square
Reading Date: March 23, 1982
right_ol
square
Reading Date: April 20, 1982
right_ol
square
Reading Date: April 27, 1982
right_ol
square
Reading Date: May 4, 1982
right_ol
square
Reading Date: May 11, 1982
right_ol
square
Reading Date: May 18, 1982
right_ol
square
Reading Date: May 25, 1982
right_ol
square
Reading Date: June 1, 1982
right_ol