Interview with Jacques Rivette


Jacques Rivette has made four films (or five if, as we shall see, the last may be counted as double): Paris nous appartient [1961], La religieuse [1966], L’amour fou [1969] and Out/Spectre [1971/1974]. The first two, to simplify slightly, belong to the classical cinema: predominant are a dramaturgy pre-existing the filming, a written text, a ‘planned’ découpage. The films that follow radically modify the function and powers traditionally assigned to the film-maker: here the script is no longer a program to be carried out, a score to be followed, but a sort of vast fictional trap, simultaneously rigorous and open, designed to orient the improvisation (by actors and technicians), to subject it to certain ‘obligatory passages’ or to abandon it to a fret flow which will acquire its order, its scansion, its proportions only during the final montage, in an ultimate interplay between the inherent logic of the material filmed (its potentialities, its resistances) and the demands of a rational critical organization. Critical in two ways: of the material filmed (concrete) and of the scheme (abstract) which provided the initial impulse. This attitude, with the principle pushed to its limits, resulted in the birth of a film-fleuve with Out, undoubtedly one of the longest in the history of the cinema: thirteen hours. Needless to say, under the current distribution system (and with television being what it is where culture is concerned), this film has remained, except on one single and highly successful occasion at the Maison de la Culture in Le Havre, virtually unseen by the public. But Out is two films. A ‘shorter’ version exists, running for approximately four hours, which is not simply a digest of the full version (as was the case, it may be recalled, with the condensed version of L’amour fou disowned by Rivette), but quite literally a reorganization along new lines of the completely transformed premises: it is another film, a splintered reflection of the first, illuminating, obfuscating, challenging it by suggesting other avenues or angles of approach for anyone who has seen both films, but also working perfectly as an autonomous experience.

A case as systematic as this (two different films emerging from the same basic material) is, as far as I am aware, unique. It goes without saying that the interview which follows is justified primarily by the intrinsic interest of Rivette’s method, since readers of La Nouvelle Critique will have seen neither Out (thirteen hours) nor Spectre (four hours). But through these two films, perforce unavailable for the moment as concrete items of reference for the interview, a whole problematic is adumbrated that is of vital importance today, with Rivette among those who have explored it furthest in both theory and practice: the problem in the cinema (following Rouch, Perrault and a few others) or in music (Cage, Stockhausen) of the relationship between premeditation and improvisation, freedom and restraint (between chance and design); the problem of another function (different, if not less important) for the film-maker and his collaborators, technicians or actors; of new fictional forms arising from this method; of ‘direct’ cinema and its possibilities, its limits, its risks; of montage; and, more generally, the problem of ‘performance’. Furthermore: Rivette’s thinking is founded on an extensive, systematic intercourse with existing cinema in its entirety, both as a heritage that is constantly revalued and restored to its proper perspective (from Feuillade and Griffith to Godard and Resnais), and as a cultural environment (the cinema developing its form). With this in mind, the interview attempts to introduce a train of thought to be pursued in terms of history and the evolution of forms. These general considerations explain the specific organization of the text that follows. The first part is devoted to Out (and the overall problematic illuminated by that film). The second, more generally, to the cinema today, and to those elements in the cinema today that constitute for Rivette what Pierre Boulez would call ‘a crest line’ (Jancsó, Fellini, Straub, Tati, Bergman, etc.).

Jean-André Fieschi


La Nouvelle Critique:What is the origin of Out (its place in relation to your earlier films)? What part is played by doubt and certainty – or premeditation and chance, for that matter – in the initial stages of a project like this?

Jacques Rivette:It’s a sort of offhand synthesis, treated deliberately offhandedly, of contradictory things I had been more or less thinking about for more or less a long time. There was the desire I’d had before La religieuse, and which was aggravated by the filming of La religieuse, to make a film which, instead of being predicated on a central character presented as the conscience reflecting everything that happens in the action, would be a film about a collective, about a group, though in what form I didn’t know exactly. One of the only things I did know was that it wasn’t going to be set in Paris but in a small provincial town.

A group of young men and women, fluctuating since it was to cover three months, six months, a year, with the theoretical notion – a little too theoretical, actually – that the point was the variations within the group, so that eventually, by the end of the film, the people wouldn’t be the same ones as at the beginning, all the members of the group had changed and their relationships had become completely different, with even the group becoming something else again. Finally the idea was left up in the air because I couldn’t find an anchoring point for it.

Another desire I had, at variance with the first, was one that came to me much later, possibly in thinking about Méditerranée [1963]1 and that type of film. It certainly also came to me in connection with About Something Else [O necem jinem, 1963]2 and other films based on this principle of parallel narratives. This was the desire to make a film which would not be made up of just two interlocking films but of several, a whole series. Although it isn’t a good film, or maybe because of that, I don’t really know, I had been struck by Cayatte’s La vie conjugale [Jean-Marc ou La vie conjugale, 1964 / Françoise ou La vie conjugale, 1964]3 . What I liked about it was that when you saw the first film, you thought it was awful, but when you saw the second it began to be interesting. Above all, seeing the second one made you want to see the first again. I very much wanted to make a series of films referring back and forth to each other.

I set out from all these ideas, and when shooting began, it was with the intention of filming material inspired by quite separate characters, so that there would be four completely different ‘threads’ at the outset, without knowing – I should have known, but I’d avoided asking myself the question – how this material would be edited together. Ultimately, perhaps it would have had to be edited in the form of films independent of each other but referring back and forth, along the lines of La vie conjugale, though instead of having two films showing both sides of the same story, there would have been at least three films whose relationship to each other wouldn’t just have been positive and negative, right side and wrong side. But this wasn’t a carefully blueprinted project; on the contrary it was a sort of amorphous mixture of more or less bygone impulses that had coagulated in this way.

(1) & (2) Out 1, noli me tangere (Jacques Rivette, 1971)

Several films in one

Anyway, a week before shooting began, I was faced by the need to find some way of representing all this. The situation was becoming urgent if we weren’t to waste the six weeks’ filming provided for by the budget, so we had to have a planned shooting schedule. I spent two days with Suzanne Schiffman4 . All one afternoon she asked me questions, saying tell me everything you know. So I told her roughly what I knew, the characters, remembering what each actor had said, what he wanted to do, what we’d talked about in each case5 .

Suzanne scribbled all this stuff down, filling up thirty or forty pages in a notebook. Then we looked at each other and we said: what are we going to do with all this? We tried to take each character in turn, but nothing came of that, then suddenly I think it was she who had the great idea: we must draw up a bogus chronology – because after all a story does unfold in time – indicating an arbitrary number of weeks and days on the vertical lines, and the names of the characters going the other way. From that moment... it was very odd but this sort of grid influenced the film a lot.

The great temptation was, not to fill in all the squares of course, but when you saw from the document that Colin, say, was meeting somebody or other, to think: well now, why shouldn’t he meet so-and-so as well? Our idea throughout wasn’t so much to have one character meet another, Colin and Thomas, or one actor another, Jean-Pierre Léaud and Michel Lonsdale, and so forth. But it seemed a pity not to have this or that actor at the same time and to have them confront each other for at least twenty minutes to see whether anything happened between them. After that it became like a game, like a crossword. Actually it was done very quickly.

Whereas L’amour fou brought into play two individuals and (centering on Jean-Pierre Kalfon) a group of people particularly coherent as a group (being basically Marc’O’s company6 at that time), by contrast with this I wanted to play on a more heteroclite, more heterogeneous casting; to play on the heterogeneity, and in fact from my point of view this heterogeneity is much less flagrant than I’d originally planned.

And afterwards, when filming was completed, what was it like compared to your prior idea (or ideas)?

To our way of thinking, the diagram indicated possibilities; it was a way of starting off a film which could be even longer and which could be continued, where the various strands were brought together of a plot that could be continued, with new characters still to turn up later on. We had made up a possible list of all the actors who might be interested in working on a project like this. The ending was deliberately inconclusive. I’d asked Suzanne to plan the work schedule so that during the last few days of shooting, corresponding roughly to the last week in our chronology, we wouldn’t be trapped by the grid in everything concerning the ending; this way we could change, rearrange if need be, depending on what happened during filming earlier. In the end I suddenly felt – and Suzanne and the actors eventually agreed – that instead of leaving the story suspended, I wanted to pretend to end it, to make it (relatively speaking) a film with a conclusion. I realized I wouldn’t want to carry on with the continuation six months or a year later. It was with this in mind that I filmed the ending, whereas when we were preparing the film it was with a view to continuing it afterwards. That’s one thing that happened.

Another was that when we started the editing with Nicole Lubtchansky, faced by all these rushes, we started by sorting them out according to a rough chronology. I still didn’t know whether I might not follow the principle of separate films during editing, but then, as we were putting it together, we soon began wanting one particular scene to come after another. We quickly realized that it didn’t cut together any old how, in any order. For instance, if you put Juliet Berto’s scenes one after another, or did that with Jean-Pierre Léaud’s scenes or Michel Lonsdale’s – something we never did, actually – it was obvious that they had to be broken up by each other, and that there was in fact another continuity in the intersection of the strands which we had to follow or find.

(3) & (4) Out 1, noli me tangere (Jacques Rivette, 1971)

Some myths

How exactly does the method you used differ from the traditional conceptions of cinema that are still dominant? Several more or less established notions are shaken up in a pretty radical way: ‘the director’, ‘the script’, ‘the actorand so on. How?

Time was, in a so-called classical tradition of cinema, when the preparation of a film meant first of all finding a good story, developing it, scripting it and writing dialogue; with that done, you found actors who suited the characters and then you shot it. This is something I’ve done twice, with Paris nous appartient and La religieuse, and I found the method totally unsatisfying, if only because it involves such boredom. What I have tried since – after many others, following the precedents of Rouch, Godard and so on – is to attempt to find, alone or in company (I always set out from the desire to make a film with particular actors), a generating principle which will then, as though on its own (I stress the ‘as though’), develop in an autonomous manner and engender a filmic product from which, afterwards, a film destined eventually for screening to audiences can be cut, or rather ‘produced’.

Now, under all this there are several myths which should of course be flushed out one after the other and challenged.

The first aspect one very often stumbles up against, and about which there is enormous confusion, is the mythology, the myth of what has been called direct cinema, of what has even been called, in an even greater misusage, cinéma-vérité. Well, I don’t think there is any point in putting cinéma-vérité on trial all over again: it’s a word, a formula, which has been applied completely mistakenly, because even the films for which this formula was invented were far from corresponding to what the term implies. It isn’t worth pursuing, let’s just stick to the term ‘direct cinema’: it’s more ambivalent, it’s more adaptable. People have long been aware now, though they weren’t always, that ‘cinéma direct’ ... we won’t say equals ‘cinema lie’, but that it has at any rate nothing to do with notions of real or false. It’s a technique, not just like any other but a technique all the same, which produces artifice by means other than those of traditional mise en scène, but which does, through its very function, produce artificiality. This artificiality is simply designated differently from other cinematic artificialities. But there is no innocence, no candor, no spontaneity attached to direct filming.

Knowing all that, you use this method anyway. Why?

Precisely because that’s where direct cinema becomes exciting; from the moment you realize it’s a creator of artifice (and not, I repeat, of lies). But one which, by comparison with the traditional method, is more directly in touch with that particular artifice which constitutes the act of mise en scène, of filming.

When I say this, I’m probably exaggerating. It is quite clear that the borderline between direct cinema and the cinema of mise en scène is a false one, like the old Lumière-Méliès border. What you can say, in fact, is that there you have the two ends of the chain, but it doesn’t part anywhere in the middle; you get from one end to the other through a whole series of detours, surprises, circumstances demanding adaptability. In any case, in the two films I’ve made ‘since’, I’ve never felt any desire to use a technique in any purist sense. On the contrary, what interested me – and if I make any more films this is the direction in which I’d like to go – was to see how, within the direct cinema method, it could be used more ‘impurely’. Because, in my case anyway, it never is direct cinema properly speaking; it remains a technique very closely akin to mise en scène. To begin with, I’ve always worked with actors, with a comparatively precise canvas-scenario as a starting-point, and with a normal technical crew; not at all the ‘wild’ conditions of someone like Perrault7 , or Rouch when he made La chasse au lion à l’arc.

(5) & (6) Out 1, noli me tangere (Jacques Rivette, 1971)

New fictions

Isn’t there an attempt to effect a renewal of fiction here, in the shifting of responsibilities from you to the actors?


Nevertheless, a threshold of intervention still exists...

The myth in this sort of film-making is of a creative collectivity in which everyone is happy and spontaneous and everybody ‘participates’. I don’t think this is true. Quite the contrary, the atmosphere is usually relatively tense, because nobody knows where they are, everyone is exhausted, film-maker, actors and technicians are all in a muddle. Nobody really knows what is going on. I think the only possible attitude in situations like this – it’s what I’ve always tried to do, at all events – is to adopt a perspective that is beyond good or bad. You must virtually refuse, for the time being, to judge what is being shot. There are moments when you feel you’re letting everything go by passing on to the next bit, and others where I suddenly find myself sticking on a detail anyway: it has to be just so, at this particular moment the character has to say such and such.

Out seems to be constructed contrary to any established dramaturgic principle. The strictly fictional elements, for example, take a long time to appear.

We asked ourselves this question. Nothing simpler, if we’d wanted, than to do what is almost always done in films, which is to kick off with strong dramatic elements, after which you can get the exposition out of the way. It’s the old device the cinema has used almost since the beginning: any narrative activity requires the prior disposition of a certain number of elements before, by interlacing these elements with each other, crisis situations can then be reached which constitute what is traditionally called the story proper, the motor element. This is a necessity dating from the origins of all dramatic or narrative forms, and each period, each means of expression, has resolved it in a different way.

Even in L’amour fou where it was time that created the action, which was the action – more so than in Paris nous appartient – we started in a small way, on a minor crisis: the sequence in which Bulle Ogier walks out of the company. We deliberately made the sequence rather flat, however, not dramatizing it at all. Coming back to Out, Suzanne and I decided we wouldn’t use the good old method and that we’d start... in documentary fashion would be putting it too strongly, but at any rate without any dramatic elements. And since the Histoire des Treize8 was used in the film, we thought: all right, we’ll have an exposition in the manner of Balzac. Comparatively speaking, of course, but the dramatic interest of the first three or four hours is purely in the description, not so much of the settings as of the actor-characters, their variously interesting or uninteresting jobs, their different social spheres. And within this pseudo-documentary (almost documentary in certain sequences involving Lonsdale’s theatrical group, which were shot as reportage), the idea was that the fiction gradually proliferates. We start off with the reportage – it’s phony, of course, set up, but presented more or less as reportage – with the fiction slipping in very stealthily at first, but then beginning to proliferate until it swallows everything up and finally autodestructs. This was the principle governing the whole of the end, where we linger on the remains, the refuse left, you might say, after fiction has been at work like this.

(7) & (8) Out 1, noli me tangere (Jacques Rivette, 1971)


The film makes use of a large number of pre-existing texts...

This was one of the main ingredients of the thirteen-hour version, one of the few things mentioned in the five pages we gave to the Commission du Centre in requesting our advance9 . Actually, it remained more an intention than a fact in the film as it turned out. A lot of texts vanished en route. But others came to us along the way, ones I hadn’t thought of, that I discovered during the editing.

Among the pre-existing texts there was Balzac, there were the two Aeschylus plays10 , which produced results, more or less. In theory they were to proliferate. There was Tasso [Torquato Tasso], of whom little remains. The one who came along very late but became very important was Lewis Carroll, The Hunting of the Snark.

And Feuillade11 ?

Feuillade, of course, but I wasn’t thinking so much of film texts. I’d reread Les Misérables a year previously, so what I had in mind was the 19th century popular novel, Les Mystères de Paris, rather than Feuillade. Even what we borrowed from Balzac hinges on three or four archetypes of the popular novel. There is for instance the idea of people living on the fringes of society, whether because they are labeled as ‘artists’ or because they indulge in ‘marginal’ activities like Jean-Pierre Léaud or Juliet Berto, or because they really are marginal, like the secret society. And the other old standby element is the secret message. As we used it, it’s Jules Verne rather than Edgar Allan Poe. What I wanted was a message that could be subjected to successive readings, as in Captain Grant’s Children. It’s all very reminiscent and not very serious, but at the same time the mainspring of the film, the desire to make it, lay in the amusement of writing these messages. It’s one of the few things I did in the film. I didn’t write the dialogue, but I did take great pleasure in writing these12 .

They are motor elements in the film, not just accessories...

Yes, but they’re decentralized motors which you don’t see.

Rather like your own position?


What is the film about, ultimately?

In fact, it’s about too much rather than too little. To begin with, play in all senses of the word was the only idea: the playing by the actors, the play between the characters, play in the sense that children play, and also play in the sense that there is play between the parties at an assembly. This was the basic principle, implying a relative interdependence between the elements, and a relative distance maintained by the actors between themselves and the characters they were playing. I discussed this from the start with Michèle Moretti, with Bulle, with Lonsdale, in order to counteract the spuriously ‘lived’ aspect of L’amour fou: each actor had to play an extremely fictional character, and theoretically maintain a considerable distance between himself and that character. In the event, there really was ‘play’ between the actors and the characters they were playing, and at the same time they revealed a hundred times more things about themselves than if they had been identifying with these fictional characters or were supposed to be playing their own ‘characters’.

Actors, one constantly feels, could never speak a written text in this way...

One great difference between the thirteen and four-hour versions is that in the thirteen-hour version, and not the other, we deliberately retained a number of fluffs here and there in the final cut, because there were things among these fluffs that we liked, that we found moving, that preserved the slightly perilous aspect of the project, the feel of walking on a tightrope. We left in some passages where the actors repeat themselves or get muddled, which we could have cut out. It wouldn’t always have been easy, but we could have: you always can. Similarly with the very long ‘reportage’ sequences on Lonsdale’s group: either you kept them integrally or you cut them out entirely. There are several I didn’t include at all, but those we did put in are there in their entirety.

(9) & (10) Out 1, noli me tangere (Jacques Rivette, 1971)


In a normal length film, the fluffs would have looked like nulls in any case, but the film establishes other criteria for judgment. If the generative element in the thirteen-hour version is the duration itself, what is it in the four-hour version? This shorter version appears, by contrast, to be very much edited in the classical sense of the term.

I wanted, without knowing how, to make a film quite different from the other. I asked Denise de Casabianca13 first of all to spend a fortnight alone with the 12-hour 40-minute film, getting to know it a little, because there was no other material. Having done that, she made a first rough cut. And it was immediately apparent that you were still held by the fictional center, which proved to be much tighter, much more compelling than I’d thought, and that there weren’t umpteen solutions, there were only two. Either we could do something extremely arbitrary, with flagrant ruptures in time, breaking up the chronology, a sort of Robbe-Grillet montage. Or we could play the game of a seeming narration, which was after all the game played by the material, thus keeping a seeming chronology, no matter how patchy and wobbly it might sometimes be.

At which point the whole center of the film dug its heels in completely; and this four-and-a-quarter hour version was edited from the center outwards. We couldn’t really touch this center, because there is a moment, one single shot even, in which almost all the fictions intersect, as if all these lines had to pass through a ring. This shot we put squarely in the middle: it comes just before the intermission. Having done so, we then had to keep everything relating directly to this shot before and after. After that, of course, we were a little more at liberty for the two ends: the first and last hours.

The interesting thing with the four-hour-fifteen-minute version was to use this material which incorporated a good deal of improvisation in as precise, as tight and as formal a manner as possible. To try to find as many formal principles as possible, visible or invisible, for getting from one shot to the next. Some are very obvious, others happened by chance and we only noticed afterwards.

How should one envisage the spectator relationship in Out, and what is your opinion concerning what you were saying about the impossibility or ‘distanciation’ in the cinema, with reference to Out and the very strong identification factor that operated – partly in a negative way – in L’amour fou?

The relationship to the thirteen-hour film is a relationship totally falsified at the outset by the fact of the performance. Even in L’amour fou this came into play for the spectator to a certain extent – the idea of going into an auditorium and getting out four and a quarter hours later – but at least it kept within reasonable limits, it was still feasible, only a little longer than Gone with the Wind [1939], though without the bonus of the Civil War. Whereas twelve hours forty minutes... it may not be the first time a film has run for so long, but at any rate the only equivalent in my opinion – and even then it isn’t so long – is when Langlois shows a Feuillade serial at the Cinémathèque, starting at six in the evening and going on till one a.m. with three little breaks. It is obvious that the first two hours of Out, for example, are bearable only because one knows one has embarked on something that is going to last for twelve hours and forty minutes. We impose three-quarters of an hour of hysteria from Lonsdale’s group on people, something that can be done only under these conditions.

Ideally, I still hope the film can be shown. It wasn’t in any way intended to be a difficult film, except perhaps in its length and the fact that there are moments one can call longueurs. Otherwise – perhaps it’s hypocritical, but it’s a hypocrisy I cling to – everything that might loom as an obstacle or flaunt its difference is rejected. Perhaps this is also a weak point, since these are often the things that certain spectators latch on to. In theory I can very well imagine the film being shown in cinemas; but precisely because it is so long, there wouldn’t be any point except in suitable cinemas where people can be not too uncomfortably seated, where they can breathe, and that are big enough to house a sizeable audience. It is 16mm, but it was made with the big screen in mind: it has a meaning on the big screen which it wouldn’t have on the small screen. Even visually it is composed of elements implying a massive image – a monumentality is putting it too grandly but that’s it nevertheless.

This struck me at each of the screenings we had for television with five or six people present. Even if people liked it, I felt the relationship to the film was wrong, because it is first and foremost theatre. It is performances up there on the screen, mise en scène in the theatrical sense of staging rather than in the film sense.

How could one define the particular narrative form that interests you here? Or to put it more generally: how does this film contribute along with other films (which ones, in your opinion?) to the attempt, fairly widespread in spite of everything even though still very limited, to bring about a renewal of fiction, and indeed of the cinema as a whole?

What I’d like is to discover a cinema where the narrative element doesn’t necessarily play the driving role. I don’t say it would be completely eliminated, I think that’s impossible: if you throw narrative out by the door, it comes back through the window. What I mean is that in the cinema I have in mind it wouldn’t be in the driving seat, and the principal priorities on the screen would be purely spectacular ones, in the strict sense of the word. That’s why, when I say I’d like this film to be shown on a big screen with an audience of five hundred people, it isn’t at all because it tells a story where spectators are caught up by a plot, as in a Hitchcock film; my motive is purely plastic, allied to a certain status of the image and sound.

But let’s talk in general about the cinema I’m after, of which there are better examples around at the moment: examples of films which impose themselves on the spectator through a sort of domination of visual and sound ‘events’, and which require the screen, a big screen, to be effective. In the final analysis, all the films that have impressed me recently are films in which, in very different ways, this fact of a narrative spectacle comes into play: Fellini, Jancsó, Werner Schroeter’s Salome [1971]. And for me it’s very important even in films where this ‘spectacular’ quality seems less obvious, as in Othon [Les yeux ne veulent pas en tout temps se fermer, ou Peut-être qu’un jour Rome se permettra de choisir à son tour, 1970]14 , or Le moindre geste [1971]15 , or Tati’s films of course. These are films that impose themselves visually through their monumentality. I’m using the word monumental simply because I can’t think of another offhand. What I mean is that there is a weight to what is on the screen, and which is there on the screen as a statue might be, or a building or a huge beast. And this weight is perhaps what Barthes would call the weight of the signifier, though I wouldn’t go to the stake on that...

These are films that tend towards the ritual, towards the ceremonial, the oratorio, the theatrical, the magical, not in the mystical so much as the more devotional sense of the word as in the celebration of Mass. Technique and Rite [La tecnica e il rito,1972], as Jancsó has it, is a good definition. These words should be explored to try to see what lies behind them: rite or ceremony or monumental. One would probably first find what Barthes and Ricardou have been pointing out for years: in films, as in texts and in theatrical performances, the accent should be placed on the elements in which the spectacle itself (or the fiction) is represented.

But this isn’t enough to account for the element of violence, of affirmation without evidence, of erotic power, which I’m trying to express when I talk of monumentality and when I think back on these few films.

The spectator at Out finds himself faced by a vast, very unusual ‘machinery’. What does this machinery set in motion?

A film is always presented in a closed form: a certain number of reels which are screened in a certain order, a beginning, an end. Within this, all these phenomena can occur of circulating meanings, functions and forms; moreover, these phenomena can be incomplete, not finally determined once and for all. This isn’t simply a matter of tinkering, of something mechanical constructed from the outside, but rather – to refer back to what I was saying at the beginning – of something that has been ‘generated’ which seems to entail biological factors. It isn’t a matter of making a film or a work that exhausts its coherence, that closes in on itself; it must continue to function, and to create new meanings, directions and feelings.

Here one comes back to the Barthes definition. I refer to Barthes a good deal, but I find that he speaks more lucidly than anyone else at the present time about this kind of problem... and he says: there is a text from the moment one can say: things are circulating. To me it is evident that this potential in the cinema is allied to the semblance of monumentality we were just talking about. What I mean is that on the screen the film presents a certain number of events, objects, characters in quotes, which are closed in on themselves, turned inward, exactly as a statue can be, presenting themselves without immediately stating an identity, and which simultaneously establish comings-and-goings, echoes, among one another.

(11) & (12) Out 1, noli me tangere (Jacques Rivette, 1971)

Tati, Fellini, Jancsó...

Is there a connection between these elements and the fact that these films lack a central character, a protagonist?

Yes, it is almost always groups that are involved, and this is a further link to the idea of ritual, of ceremony. If, at a pinch, there is an individual at the ceremony, it can only be the priest, someone who is never anything but the delegate, the representative of some community. But I don’t think this tendency in the cinema entirely obviates the possibility of a central protagonist. For example, I put Straub’s Othon in the same category (yet it’s a play which plays more than other Corneille dramas on the impossibility of saying who is ‘the’ hero; there is no one interlocutor, one purveyor of truth in Othon, which isn’t Le Cid or Polyeucte – the role circulates freely). Another example I include in the cinema of monumentality is Le moindre geste by Deligny and Daniel. On screen there is a protagonist who is present for nine-tenths of the film, but this protagonist is someone with whom identification is strictly precluded, because he is by clinical definition a mental defective, and he is there on the screen purely as a physical presence, or on the soundtrack purely as an utterance detached from the physical presence, disconnected, first of all because these are fragments of tapes recorded independently of the visuals (and therefore not synchronous), but also because this utterance is itself aberrant in the literal sense of the word. Another film I’d place in the same category, perhaps mistakenly, is Tati’s Traffic [Trafic, 1971]. Now there is a protagonist in Traffic, and it’s Hulot; but at the same time it’s quite obvious that the process which had already started in Monsieur Hulot’s Holiday [Les vacances de monsieur Hulot, 1953] is accentuated here, whereby Hulot no longer runs the conveyor belt, he is swept along by the fiction in almost precisely the same way as Fellini is swept along by Roma [1972]. You see Hulot wandering by from time to time in the background of his film, just as Fellini passes by suddenly wandering around in his Roma. Well, is one thing a consequence of the other? I have no idea. And one of the things I’m interested in now is to remake a film which has a strong, totally present central character – therefore one who is proposed as a vehicle, as operating the conveyor belt – and see what this contradiction would do to him.

I think it might be a pity, within a cinema of ‘signification’ – to employ this neologism – to abandon completely something that was so exciting in that traditional cinema: this play with the protagonist, the so-called central character, the Hitchcockian-Langian play on the phony central consciousness and all that this allows. But perhaps they are incompatible: it remains to be seen.

Another common factor in all these films – to my mind the only ones of any importance for several years – is the categorical refusal in practically all of them to use written dialogue. A refusal which doesn’t always happen in the same way, however, and doesn’t produce the same results. Broadly speaking, what I see in common is their refusal to write a text themselves which the actors will then interpret, the refusal to have the actors become interpreters of dialogue written beforehand which one has either written oneself or at any rate is responsible for. As to how to proceed from here, there are different solutions...

I won’t say they’re equivalent, some are undoubtedly more powerful than others. These consist either of asking the actors to find their own words, or of giving the actors pre-existing texts for which one isn’t oneself responsible: the author of the text is antecedent, the author of the text is challenged in a sense, because either it’s Corneille (though as ‘matter’) in the case of Straub, or, in the technique that Godard has used more and more systematically, the author of the text is multiple, this being what has been called the technique of quotations. But these aren’t quotations, because the important point in the sequence of Jean-Luc’s films came when he began removing the quotation marks and the names of the authors, thus not wanting to be the author of his scripts and wanting these texts coming from all over the place to lose their authorship. This method has I think been taken up again in part by Jancsó, in so far as Jancsó uses songs a good deal – pre-existing material, of course – and I think that in the moments with dialogue, or so I felt while watching Red Psalm [Még kér a nép, 1972], he draws extensively and very systematically on real militant texts, historical or contemporary. And I believe that Hernádi’s16 job is increasingly becoming a matter of collages, on the same principle as Godard though perhaps not quite so extreme.

Other solutions to this rejection of scripted dialogue: it can be Tati, unintelligible dialogue where all one hears is snatches of speech; it can be the Taviani Brothers17 ; it can be Daniel, who takes previously recorded tapes, which moreover carry speech having no fixed reference in the fiction and being completely erratic as speech. The only great exception I can see is Bergman. All the same I feel very inclined to put The Rite [Riten, 1969] with the cinema of monumentality and signification, and nothing could be more scripted than that.

Or even over-written, like Resnais.

Resnais, actually, is someone who has always worked with written texts, very highly written even, but with the purpose of not writing them himself. For Resnais, as for Godard, there is purpose in using someone else’s text, a text of which he isn’t the author, the signatory. One could perhaps salvage Bergman simply as a sort of monster, a perfect schizo. Bergman is dual, like Jekyll/Hyde. There is a Bergman who writes a script, then a Bergman who films the script, and they aren’t the same. In The Rite, the words carried by the images are not filmed for their meaning but rather for their materiality, as events and not as meanings. The same is true of the ‘strong beat’ moments in Duras’ films. Yes, I think that’s the basis for everything: to treat the text as material which plays a role exactly similar to the other materials in the film: the actors’ faces, their gestures, the photographic texture.

But can the signified carried by this text-material be a matter of indifference?

I think one can say of the signified what we were saying earlier about narrative: it’s something that inevitably reappears anyway. Knowing it will reappear, one might as well try to have it circulate as much as possible, to use Barthes’ phrase. I think what all these film-makers are trying to do is to have the signifieds that are present be caught up and carried in the general movement of the signifiers. This seems to me glaringly obvious in Jancsó’s films. I know too little about the Hungarian context of these films to be able to tell whether some things may not well have a more precise function there than here, with respect to the situation in Hungary.

(13) & (14) Out 1, noli me tangere (Jacques Rivette, 1971)

Recreation and Terror

All the same, I think that Jancsó’s prime ambition is not to play cat and mouse with the State system as a whole: it seems to me a very partial view to see them as films playing on ambiguity, playing on the questioning of certain political values, inasmuch as he in fact spends his time playing on the rearrangement of labels. What struck me about both Sirocco [Sirokkó, 1969] and Technique and Rite, in so far as objectively they are perhaps the least ‘successful’ though not the least fascinating of his recent films, is their element of juvenile play. These are ten-year old children playing at spies or at war just like cops and robbers (or like Le petit soldat [1963]: Godard again). In all these Jancsó films it is really recreation time: the children are in the playground during break between classes, dividing up into groups, forming into rings, it’s the political game to the letter: politics as a game, a game as politics, with the whole arsenal of revolutionary signifieds congealed, put back into circulation. And when I say juvenile, it isn’t meant pejoratively in any way, this may be a partial view of Jancsó’s films, but it’s more and more how I see them. Recreation time, but in the widest sense of the word; as Cocteau said, “When a child leaves the classroom, we say it is for re-creation”, I think this is the value of Jancsó’s films: within a revolutionary state, he plays the role of re-creation.

Aren’t we coming back to the problem of pleasure in the cinema?

Yes, but pleasure has never been absent.

Hasn’t there been a tendency to minimize its importance on pseudo-scientific grounds? Now the idea that it is an important factor is being rediscovered.

But this idea of a game, of pleasure, is also found in Renoir, it’s found in Rouch, it’s found in Godard.

Isn’t it to be found in Brecht too, but oriented towards the spectator? And during the period of the formation of the Berliner Ensemble, particularly...

Yes, the whole last part of the Short Organum is concerned with this. Anyhow, pleasure is contrary to what I call journalistic films – in the derogatory sense of the term – whose only merit is to provide information that is already out of date, and very often don’t even do that.

One must be careful, though: the inflated notions about pleasure and entertainment over recent years have been very confused, because this can very quickly lead to the attitude that anything goes. All it takes is to smoke a little, acquire the right euphoric state, and one can get pleasure out of looking at or listening to absolutely anything. Some films seem to me to be made purely with a view to narcissistic pleasure, totally without productivity: if one doesn’t bring along one’s own euphoria, the films themselves produce nothing. So I am inclined to continue defending films which are themselves the producers of pleasure.

Then again this pleasure or – why not? – this ecstasy18 of the spectator isn’t necessarily connected with euphoria; it can tend more in the direction of... let’s not say work – which is a large word that has been much abused (and one mustn’t confuse the work of the spectator or of the signifier with other forms of work) – but this pleasure in fact passes through certain stages, certain periods, which can equally well be attentiveness, perplexity, irritation or even boredom. For me, the most powerful pleasure in cinema – and this is something that interests me more and more, and I don’t know if it can be related to this cinema of signification, of monumentality, that we were talking about – is connected with terror and anguish. For some years now I have been re-fascinated by horror films. And in Out this was something I hadn’t planned at all at the outset. Initially we thought it was going to be very jolly, and we started out with the actors by criticizing L’amour fou for its element of anguish, of psychodrama – psychosis, even – saying well, it won’t be like that this time but just a jolly game with serial-type fiction; but very soon an element of anguish crept into the film (rather than the actual shooting). So even in a film where anguish hadn’t been planned, it reappeared, to such an extent that my editors said to me: “Now you should really make a horror film...” But perhaps they were hoping that might be fun to edit...

  • 1Méditerranée: a medium-length film by Jean-Daniel Pollet with a text by Philippe Sollers. A series of images, filmed in various countries around the Mediterranean (a Sicilian garden, a Greek temple, a fisherman, a young girl on an operating table), reappear throughout the film, in a different order each time, held for different lengths of time, constituting a sort of mythical narrative in which each image serves as something like an ideogram. The order of these shots was not predetermined before shooting but established, by trial and error, in the montage.
  • 2About Something Else: a Czech film by Vera Chytilova, comprising two parallel stories (one about a gymnast training for a competition, the other about a housewife). These two narratives never meet but are articulated together by a very complex formal interplay of oppositions and connections of various kinds (sound, gesture, rhythm, etc.).
  • 3La vie conjugale: two films by André Cayatte released simultaneously, each telling the story of a couple, one from the woman’s point of view, and the other from the man’s.
  • 4Associate in various guises (script-girl, assistant director, production assistant) of Truffaut and Godard; played an important role in the preparation and shooting of Out.
  • 5The actors contacted were left a very considerable margin of freedom in the choice of their characters and how they developed.
  • 6A theatrical group, experienced in the techniques of improvisation and the psychodrama, whose productions included Les bargasses and Les idoles. Among the members were Bulle Ogier, Pierre Clementi, Jean-Pierre Kalfon.
  • 7French-Canadian film-maker (Pour la suite du monde [1963], Le règne du jour [1967], Les voitures d'eau [1972]), working on confrontations between real life characters and more or less arbitrary situations. The unusual ‘fictions’ deriving from these confrontations are the result of an enormous amount of work at the editing stage, classifying, selecting and rearranging a considerable quantity of raw material.
  • 8A secret society founded in Paris under the Empire by ‘thirteen men all haunted by the same feeling, all endowed with sufficient energy to adhere to the same opinion, all mutually honest enough not to betray each other, even though their interests lay in opposing directions, profoundly political enough to hide the sacred ties that united them, strong enough to place themselves above all laws, bold enough to undertake and fortunate enough to have almost invariably succeeded in their designs [...]; having achieved mutual acceptance of each other exactly as they were, irrespective of any social prejudice [...] and recruiting only among men of excellent merit [...], all fatalists, men of great heart and tender poesy, but bored with the dreary lives they were leading’ (Balzac, preface to The History of the Thirteen). This principle of the secret society is one of the motors constituting the fiction of Out, the relationships between the characters in the film being defined in terms of their membership or otherwise (or their supposed membership) of this society.
  • 9The Commission du Centre national du Cinéma grants to a certain number of films on submission of the scenario, an advance that is in principle supposed to be repaid out of receipts when the film is released. In fact this advance is decisive in determining the fate of a number of projects which would otherwise never reach fruition.
  • 10In Out, two theatre groups, one directed by Michel Lonsdale and the other by Michèle Moretti, are working on plays by Aeschylus: Seven Against Thebes and Prometheus Bound.
  • 11French film-maker specialised in serial films (Les vampires [1915-16], Fantômas [1913-14], Judex [1916], etc.), especially prolific during the years 1915-1920.
  • 12Secret messages circulate in the film which one of the characters, Colin (Jean-Pierre Léaud), attempts to decode with a view to establishing contact with the Thirteen and their organization.
  • 13With the exception of Rivette himself, naturally, the teams who edited the two versions were different.
  • 14A film by Jean-Marie Straub, d’après Pierre Corneille. Cf. La N. C., n° 43.
  • 15A film by Jean-Pierre Daniel and Fernand Deligny.
  • 16Regular scriptwriter on Miklós Jancsó’s recent films.
  • 17Paolo and Vittorio Taviani, Italian directors, notably of Sotto il segno dello scorpione [1969].
  • 18Cf. Barthes’ discussion of plaisir and jouissance in Le plaisir du texte (The Pleasure of the Text). [ed., J. R.]

This interview originally appeared in La Nouvelle Critique, n° 63 (244), April 1973.

It was translated by Tom Milne and published in this form in: Jonathan Rosenbaum (ed.), Jacques Rivette: Texts and Interviews, British Film Institute, London, 1977. Sabzian has slighthly revised the translation for publication on this website.

A big thank you to Bernard Eisenschitz, Marthe Fieschi and Jonathan Rosenbaum


Images (1), (2), (3), (4), (5), (6), (7), (8), (9), (10), (11), (12), (13) and (14) from Out 1, noli me tangere (Jacques Rivette, 1971)

In Passage, Sabzian invites film critics, authors, filmmakers and spectators to send a text or fragment on cinema that left a lasting impression.
Pour Passage, Sabzian demande à des critiques de cinéma, auteurs, cinéastes et spectateurs un texte ou un fragment qui les a marqués.
In Passage vraagt Sabzian filmcritici, auteurs, filmmakers en toeschouwers naar een tekst of een fragment dat ooit een blijvende indruk op hen achterliet.
The Prisma section is a series of short reflections on cinema. A Prisma always has the same length – exactly 2000 characters – and is accompanied by one image. It is a short-distance exercise, a miniature text in which one detail or element is refracted into the spectrum of a larger idea or observation.
La rubrique Prisma est une série de courtes réflexions sur le cinéma. Tous les Prisma ont la même longueur – exactement 2000 caractères – et sont accompagnés d'une seule image. Exercices à courte distance, les Prisma consistent en un texte miniature dans lequel un détail ou élément se détache du spectre d'une penséée ou observation plus large.
De Prisma-rubriek is een reeks korte reflecties over cinema. Een Prisma heeft altijd dezelfde lengte – precies 2000 tekens – en wordt begeleid door één beeld. Een Prisma is een oefening op de korte afstand, een miniatuurtekst waarin één detail of element in het spectrum van een grotere gedachte of observatie breekt.
Jacques Tati once said, “I want the film to start the moment you leave the cinema.” A film fixes itself in your movements and your way of looking at things. After a Chaplin film, you catch yourself doing clumsy jumps, after a Rohmer it’s always summer, and the ghost of Akerman undeniably haunts the kitchen. In this feature, a Sabzian editor takes a film outside and discovers cross-connections between cinema and life.
Jacques Tati once said, “I want the film to start the moment you leave the cinema.” A film fixes itself in your movements and your way of looking at things. After a Chaplin film, you catch yourself doing clumsy jumps, after a Rohmer it’s always summer, and the ghost of Akerman undeniably haunts the kitchen. In this feature, a Sabzian editor takes a film outside and discovers cross-connections between cinema and life.
Jacques Tati zei ooit: “Ik wil dat de film begint op het moment dat je de cinemazaal verlaat.” Een film zet zich vast in je bewegingen en je manier van kijken. Na een film van Chaplin betrap je jezelf op klungelige sprongen, na een Rohmer is het altijd zomer en de geest van Chantal Akerman waart onomstotelijk rond in de keuken. In deze rubriek neemt een Sabzian-redactielid een film mee naar buiten en ontwaart kruisverbindingen tussen cinema en leven.