By Pascal Bonitzer
An attempt to assemble critical pieces approaching cinema “from the standpoint of sound”. This article was first published in Cahiers du cinéma, issue 256, February-March 1975, and with revisions in Pascal Bonitzer, Le Regard et la voix (Paris: Union Général d’Editions, 1976). Translated by Philip Rosen and Marcia Butzel, as published in Narrative, apparatus, ideology : a film theory reader, ed. Philip Rosen (New York : Columbia University Press, 1986).
Ah, how repugnant imposing my own thoughts on others is to me!
— Nietzsche
The “Minimum of Commentary”
We know that, politically and ideologically, the issue of the point of view a film reflects is a crucial question, and often becomes the arena of violent controversy. For example, Lacombe Lucien (Louis Malle) and The Night Porter (Liliana Cavani) recently gave rise to contradicting interpretations of their ideological content: fascist, progressive, reactionary? Then there was the case of Antonioni’s China: certain “friends of China” analyzed its point of view as progressive and pro-Chinese, while others (not to mention the Chinese themselves) as anti-Chinese and reactionary. This leads us to questions on three points:
(1) A film produces a discourse.
(2) This discourse is, to a greater or lesser extent, implicit, veiled.
(3) And it is the spectators who, “in the last instance” utter (contradictorily) its truth.
Of course, it is true that the fiction film gives rise to diverse interpretations. Is not ambiguity the element in which all fiction is immersed? But a documentary or a montage of documents – on the contrary, isn’t its aim to shed light on the real with which it deals? to disengage from that real a readability and hence a point of view? Insofar as it deals with the real, isn’t documentary a matter of truth? Camera-views, montage: reality is seized and worked through at a certain angle in order to render something on the screen – to the spectators.
It is therefore necessary to envisage the work by which a film organizes its point of view, articulates its discourse. How is it done in the case of a film like Mai 68 (Gudie Lawaetz), which is precisely a film that brings together different points of view from divergent political extremes onto the same object? Does this imply that the film in its entirety produces no point of view? But why does this question have to be asked?
Mai 68 is connected to a type of film and a type of montage with its own traditions and codes, those of the “free confrontation” (produced essentially by television, but which cinema can take up, as is shown, for example, by the success of The Sorrow and the Pity (Marcel Ophuls)). The free confrontation is that of contradictory images and witnesses, a mise-en-scène of the multiple “facets” of an event by means of interviews and archival documentary footage. (The latter’s difference from nature is a sign of the richness of the investigation and of the information and so should be marked: in this case the interviews are shot in color, which has the additional effect of emphasizing the temporal gap 1968-1973, the “receding” of the testimony and of the positions taken.) The lure for the public is that of mastering a dossier – by leafing through it.
A formula credited by Gudie Lawaetz to Sartre rather well summarizes this genre of historical digest, whose success testifies to its seductiveness: “let the event speak.” This is an interesting formula not only because in it can be read the elision of (the author’s) point of view toward the event in question, but also because it displaces this “question of point of view”-which is so important for “politics”-to a problem of speech. It is because it is inscribed there that “it speaks,” that the just vision of an event depends on what the latter says, that the eye is carried by the voice, and under the circumstances, a voice which, if not silent, is at least without subject.
Without subject – this is central to the question. It is true that the event belongs to no one (and this is more true of May 1968 than any other event). Gudie Lawaetz’ film shows this well: if it witnesses something, it is the failure of those who appropriate the movement to themselves (see the points of view the film compares, which are all perplexed or worn down). So, is it “the event itself” which actually speaks in this film? Of course not. What speaks is its descent into the past, into journalistic information and into History in the old sense of the word: linear history, flat chronology, trifles of the recent past and the derision of senescence. Of course all the actors of May have aged. Of course the images of May have aged. But what is this? Is this the meaning-this aging, this slackening? It is like the final scene of Rentrée des usines Wonder, but taken out of its militant context, for which is substituted an “open ending”: Mai 68 repeats the end of May.
This film has the past in its eye. It is indeed a “point of view” which is thus organized, a discourse which is woven in silence. Is it without subject? But this aging, whose image the film offers across its impoverished informative chronology, is indeed that of a subject. Who is this subject? Who if not us, the spectators, in whom “the event speaks” silently. To “let the event speak” is to let it be spoken by the spectators, according to the paths traced in silence by the film and guided by the signposts of montage. It is thus important to analyze the structure, the montage, which conveys the “event” and what historical capacity is distributed to it-here, pretty much that of a Mirror of History. It is important to recognize that what speaks in a film is never the “event” (and what is an event, if not a nodal point of redistributed historical intensities?), but rather the subject who is supposed to know it.
At stake in this type of film is knowledge. What is desired from it is a knowledge of an object (history, May 1968, etc.), a knowledge that is in some way specifically cinematographic, obtained by the eye and the ear, optical and sound recording, the scene which their montage composes.
In this knowledge, it is important to allow the spectator to enjoy — that is, not to take the spectator’s place by uttering it. The principle “let the event speak” also signifies “Let the spectator enjoy (know).” Of course one can protest that this is a false knowledge, an impression of knowledge (in the sense of an “impression of reality”) obtained through the specular lure produced by the cinematographic apparatus [l’appareil]; whereas, in truth, knowledge requires something other than this specular impression, for it only exists as the effect of a labor of inquiry and theoretical elaboration, of placing the object in perspective and investigating its form. Undoubtedly.
But the enjoyment – the jouissance – of the spectator? A difficulty arises here, and one which is nodal for a cinema which, more than any other, utters its knowledge. This is the cinema whose principles are opposed to those practiced by Gudie Lawaetz – namely “militant cinema.” The latter has often been reproached for ignoring jouissance (of the spectators), and to a great extent this reproach is well-grounded. The difficulty thus revolves around a certain number of notions put in play in a specific way by the documentary or the compilation film: discourse, subject, knowledge, jouissance. The place of the spectators in the apparatus [le dispositif] – their role – is there implicated and acted out.
In the Lawaetz system (which is to say, in a classical system), one sees how jouissance under the form (if one likes) of the “impression of knowledge” is put into operation. It is from an absence in the body of the film that it is offered to the spectators. This absence is neither that of a discourse nor that of a subject- both of which, as we have seen, can be implicit and veiled, but never lacking. It is the absence of commentary, which is to say of a voice. More precisely, what is absent is the voice-off, that voice of knowledge par excellence in all films, since it resounds from offscreen, in other words from the field of the Other. In this system the concern is to reduce, insofar as possible, not the informative capacity of commentary but its assertive character and, if one likes, its authoritative character – that arbitration and arbitrariness of the voice-off which, to the extent that it cannot be localized, can be criticized by nothing and no one. This is a system of “no commentary” or “minimum of commentary.” This is, we maintain, a classical system for the documentary, the archival film, and the compi- lation film. In appearance it is a democratic system: it puts restraints on the arbitrariness of the voice-off, of commentary which does violence to the real and to the spectators (by not allowing them to think about the event). What is this system in reality?
The Voice Which Keeps Its Silence
In Mai 68, when Séguy speaks of May, the student movement, and the workers’ movement, his words are inflated by his face and have to be evaluated in connection with that face, with that simultaneously embarrassed and cunning air, those fleeting glances, etc. – that collection of traits which, with a light regional accent, connotes the “personality” of a union leader for cinema and television. For the spectators, having the image of Séguy (or any politician) imparts a relative critical power. Doubtless, this criticism is greatly limited and very equivocal (how could anyone honestly judge a politics by a physical image of those who reflect it? racism is not far away.) But it is still criticism, and spectators never fail to make more or less use of it. If he speaks facing the camera, one can laugh at what he says or insult him (and this happened in movie theaters of the Latin Quarter when Mai 68 was shown). Enclosed in the frame of the (large or small) screen and in some way visibly calling for our complicity, he is, in a certain fashion, delivered to us. Prisoner of a perceptible appearance, doubly mastered by the lens of the camera and the eye of the spectator, the authority of the voice is thus encountered as being submitted to criticism, or at least to a criticism-the most rapid one, that of the look. The image plays the role of fixative for the voice. It restrains the power in it (resonance, amplitude, ambience, its disquieting power.) It is completely otherwise when the voice absents and uproots itself from the image, and returns to haunt it from outside, from space-off, from off screen. It then secures a hold on the image and, through the latter, on the real which it reflects; and this hold can no longer be countered by an easy criticism from the look. So instead of looking, it is necessary to think-but this is precisely the problem: is there time to do so? Was Merleau-Ponty wrong when he wrote “a film is not thought, it is perceived?”
Voice-off. There are at least two types of voice-off, which refer to at least two types of space-off. Take the case of a narrative fiction such as a detective film, when the voice of a character comes from off screen. Even if there is no reverse shot to suture the gap opened by this voice, we know that it comes from a place homogeneous with that of the scenographic lure offered by the filmic image. This is a homogeneity in accord with realist physical space. Such a voice can be disquieting, as, to a greater or lesser extent, voice-off always is; however, it is so only within the dramatic frame of the fiction, which is to say, in a restricted way. Thus, in Kiss Me Deadly (Robert Aldrich), if the criminal whose identity and terrifying secret is sought by Mike Hammer remains off screen until the last sequences, it is unnecessary from the point of view of the detective enigma; for as he himself says, nothing will be learned from his face by the detective (or the spectators). But keeping it off screen (we only see his lower legs and only know him from his blue suede shoes) gives his sententious voice, inflated by mythological comparisons, a much greater disquieting power, the scope of an oracle- somber prophet of the end of the world. And despite that, this voice is submitted to the destiny of the body: the one institution of the narrative to whose law it is submitted renders it decrepit and mortal. It suffices for the subject of this voice to appear in the image (and thus it suffices that he could appear there) for it to be no longer anything but the voice of a man, in other words, of any imbecile. Proof? A gunshot, he falls—and with him, but in ridicule, his discourse with its prophetic overtones.
The contrary case is when the voice is inscribed in a space which is not in proper interaction (not homogeneous) with that of the image. This voice intro- duces a division of the filmic field that is much more enigmatic. In fiction films this division is rare and produces an effect of strangeness (an example of its utilization for this effect of strangeness is Mankiewicz’s A Letter to Three Wives). But as rare as this is in fiction films, it is common and seems natural in documentaries. And to the extent that this is the case, it makes for no problems in them.
In Kashima Paradise ( Bénie Deswarte, Yann Le Masson) the voice-off intervenes with a certain brutality. It forces the image from its discourse of class struggle, and counters naturalness and the perceptibly obvious. But who speaks, and from where and when?
Another practice can be found in China, whose commentary is rarefied and barely assertive. Antonioni does not inform the image with any theory (and especially not with Marxist theory), but rather principally with a circle of information about what is not shown, namely the conditions of the filming, the specifics of what is off screen, etc. For example, he describes, off, not what the filmic image reflects-the Chinese peasants are extremely fidgity, clearly uncomfortable with the camera, almost tormented by it – but, en abime, that which is supposedly seen by these peasants who are being stalked by the apparatus [appareil] (and which is precisely what they avoid looking at): Antonioni himself and his crew, in the disquieting strangeness of their Western disorder (beards, long hair, faded jeans, etc.) But if this permits him, ever so little, to situate himself (unlike the authors of Kashima), from where does he get the right, in their place, to speak the look of these peasants? And what effect is he seeking?
The secret power of commentary and voice-off is that these (and other similar) questions are practically never posed to spectators during the time of projection. The conventional realist homogeneity of narrative space calls up identification by means of the image, and thus all which intervenes from offscreen immediately causes questioning (at least of an anterior identification by means of the play of the shot/reverse shot, reframings, etc.). At the inverse of such narrative space, in the divided, heterogeneous space of documentary, the voice-off forbids questioning about its enunciator, its place, and its time. The commentary, in informing the image, and the image, in allowing itself to be invested by the commentary, censor such questions.
This is not, one surmises, without ideological implications. The first of these is that the voice-off represents a power, namely the power of the disposition of the image and what it reflects from a place which is absolutely other (from that inscribed on the image track)-absolutely other and absolutely undetermined. In this sense it is transcendent; hence, incontestable, uncontested, and supposedly knowledgeable. Insofar as it issues from the field of the Other, the voice-off is presumed to know: this is the essence of its power.
Power indeed.
For if the voice knows, it is inevitably for someone, who will not speak. To say it is for someone is to discuss both its addressee and its place. This someone is undoubtedly the spectator, but not only him. In the sequence from China cited above, Antonioni (the voice) speaks from the place of the Chinese peasants and addresses the Western spectators; and the inverse is also in operation (for in a certain way the film addresses itself to the Chinese as well, to see how the latter have grasped it; and speaks of China from the place of the spectators, to see precisely the position of Antonioni and of the spectator – privileged). The voice speaks from the place of the Other, and this also must be understood in a double sense. It is not charged with manifesting the Other in its radical heterogeneity, but on the contrary with controlling it, with recording it (that is, with suppressing and conserving it), with fixating it by means of knowledge. The power of the voice is a stolen power, stolen from the Other; it is a usurpation. In his article “Who Says What, but Where and When?” (Cahiers du cinéma, no. 250), Serge Daney gives a striking example of the use which Power can make of this appropriated power of the voice-off. It is the daily use which television makes of it:
A short time ago, O.R.T.F presented a short film on prisons. While the camera fluidly slid along the white walls of a model prison, the commentary-off reprised in its way and in its language a certain number of demands and problems posed elsewhere by the prisoners (elsewhere: everywhere, except on television – for example, in the C.A.P.). A “contentist” criticism could satisfy itself with this and see in it, deservedly, the effect – reading through it — of the real struggle of the prisoners, without which television would never have been obliged to make the film. But who does not see a difference in nature between a film like this and Attica?
For Attica speaks with, for, and by means of the voices of the prisoners. It is necessary to note that the film described by Daney is representative of a great deal of news reporting and magazine programs produced by television. Now, what is in operation in that smoothness, that fluidity? First of all the image, that image which shows nothing properly speaking (the blank walls of a model prison), only a metonymy (part for whole) of the carceral world, whose pure, abstract presence is made to shine forth by the small screen. This metonymic usage of the image is very frequent in television news. It fills in the hole of real information, the lack of serious inquiry, and functions as a kind of sign of information (it is necessary to fill up the image box, to show something). But above all, this radical metonymy, this near-emptiness of the image, makes the real shine forth (this paucity of things, these deserted shots, are indications that the camera has touched the real, which can, in turn, touch your eyes); and the commentary-off is then able to seize the real. It is the visual and perceivable support of the commentary – if one likes, its flesh. The action of this kind of film is thus a double reduction, a double mastery of the real, which is to say of that which burns (here, not only carceral space, but the entire social crisis which it reveals): mastery by the image, which manifests the real while denying it, and by the voice, which speaks the real while imposing silence on it.
Indeed, what is this voice which “reprised in its way and in its language”? Whose way and whose language? But these are precisely the questions which are barred, obscured by space-off. What speaks is the anonymity of “public service,” of television, of information in general-and, extending the circle of connotations, perhaps the compassionate and grief-stricken Law, Democracy enunciating its wrongs, Man… A little of all these silently speak in the anonymity of the voice-off, there vaguely tracing the great veiled and abstract subject in whose name it speaks [ça parle]. And this it is the homogeneity of the social order, which in reprising “in its way,” is straining to reabsorb the heterogeneous, burning discourse of the prisoners.
What the anonymous voice-off of “public service” accomplishes is in fact a double suppression of the voice of the prisoners. First, it suppresses their voice by not allowing them to speak, and second by substituting for them. The purpose – television also serves this purpose, and one could demonstrate that a program such as Dossiers de l’Ecran has no other rationale – is that the burning voice of revolt (and through it the burning fact of revolt) give way to the cold voice of order, normality, and power.
In its function as commentary, the voice-off neither is supposed to be nor can be a burning voice – ephemeral, fragile, troubled by revolts when they have for once managed, at great price, to break through the wall of silence. The atony of commentary recloses that wall.
True, it is not always atonic.
Indeed, this is what leads us back to the stake pointed out by the technique of the “minimum of commentary”: the jouissance of the spectator, the right of the latter to knowledge. The submission of the spectator to the master voice indeed has limits. Those limits are encountered when the voice is obviously deceptive – a privilege of the master, but a risky privilege – and when partisan passion is perceived in it. This experience is amusingly illustrated by Chris Marker’s short documentary Letter from Siberia (Chris Marker).
In it the same sequence is seen three times, in continuity, but each time accompanied by different commentaries. The sequence shows workers in an unremarkable Siberian city (Yakutsk) occupied with paving a street; passing through the foreground is an Asiatic whose squinty look crosses the monocular one of the lens. What Chris Marker wishes to demonstrate is that such an image, in itself, is not neutral but is at least ambiguous and is susceptible in its referential function (Siberia, the Soviet Union) to antagonistic, contradictory readings; and that the role of the commentary can be precisely to wrest from the image a strong, univocal meaning (the image opens a field of multiple connotations which the commentary closes in denoting it, in bestowing names on things). Thus, the first commentary for this sequence is a discourse of the sanctimonious Stalinist type: the meaning extorted from the image by this discourse is that of the joyous construction of socialism (“In the joyful spirit of socialist emulation, happy Soviet workers, among them this picturesque denizen of the Arctic reaches, apply themselves to making Yakutsk an even better place to live”). The image track presents the same series of shots and the commentary changes: the voice now extracts from it an eloquent expression of Soviet misery, as would be seen by a professional anticommunist (“Bending to the task like slaves, the miserable Soviet workers, among them this sinister-looking Asiatic, apply themselves to the primitive labor of grading with a drag beam”). For the third and last time, the film returns to the same shots and the end of prosopopoeia: Chris Marker stops parodying the discourse of others – of official joviality or of primitive anti-Sovietism (which, from a certain angle, are the same thing) – and finally enunciates what can be properly said of the material: a commentary which is neither black nor white and which is objective and nuanced (“With courage and tenacity under extremely difficult conditions, Soviet workers, among them this Yakut afflicted with an eye disorder, apply themselves to improving the appearance of their city, which could certainly use it.”)
In reading these (Commentaries by Chris Marker has been published by Editions de Seuil), one cannot avoid a vague sense of deception the third time the sequence is repeated. Is it a matter of nothing being said? Not completely. To begin with, in its entirety the sequence does say something, but in a negative form: it says that commentary should not do violence to the image. To do violence to the image and to impel its referential reality is to extort a surplus meaning from it, to interdict its ambiguity, to congeal it into a type, symbol, metaphor (paving a street = joyful spirit of socialist emulation; an Asian passing = by yellow peril, etc.). This is what makes for propaganda. The ethic here suggested is close to that professed by André Bazin, when he denounced the rape of the image (of its “ontological realism”) by montage, that power of montage to orient the haphazardness of events snared on film toward any meaning (toward a meaning intended a priori). Like Bazin, Marker denounces the reduction of the signifying field opened by the filmic image. He understands it as a terrorist reduction to the benefit of interests exterior to and transcending the cinematographic experience.
In one sense, this is not wrong, but only in one sense. In his brilliant caricature, Marker’s critique runs together without distinction the enunciation and the enunciated, voice and meaning. So one cannot avoid the impression that some- thing (in the functioning of commentary) is escaping notice.
There are two aspects of partisan passion: passion and taking sides. It is wellknown that passion is blind. It enables one to cast doubt on truth and is the very foundation of one’s partiality (retaining the equivocation between partial as meaning in part and partial as meaning biased and unfair). Moreover, it is violent and can settle nothing. (And it is intolerable when it occupies the place of power-“off.”) Marker is against this passion and blindness; he does not judge from a position of involvement. But is it necessary always to prefer detached reason, the “scientific” or ironic gaze? Chris Marker poses the question well (“But objectivity isn’t exact, either. It does not deform the Siberian reality, but stops its movement for the time of a judgment, and thereby deforms it all the same”). But he responds quickly and vaguely: “What counts is the élan and diversity.” Another response, which comes closer to what is in the power of the voice, is sketched out in another short documentary which is much older than Letter from Siberia: Bunuel’s Las Hurdes [Land Without Bread]. We will return to this.
What is certain is that the cold voice is imposed (see above) where the accents of passion give birth to mistrust. Ultimately, is this not what Chris Marker is saying in the sequence of Letter from Siberia? Isn’t he ultimately denouncing not the false character of passionate commentary, but its inoperative character? Here we refind the technique of “minimum of commentary.” What Chris Marker enunciates is the modern knowledge – Western and bourgeois, if one wishes – of what is the reserve of commentary that defines the regime of its mastery, its terrorist opacity. He does not enunciate solely the democratic, liberal ethic of the commentary of political information, but also and perhaps especially, the cynicism of a science, of a mastery of the commentary: it doesn’t say too much, it doesn’t devour the silent, open meaning of the image, it doesn’t counter the “obtuseness,” ambiguity, or power of silent assertion of the image. It is a layer protecting the filmic image, lubricating it and not forcing it.
The countertest of this “science” of the commentary is given to us as much by travelogues as by certain militant films, and also by prewar newsreels of the Pathé-Journal type. These experiment with the descent of the voice into falseness and ridiculousness. One laughs, one cries, one doesn’t believe one’s ears. It is so silly that it is comic, or odious, or both at once. What is evident in this is precisely an accent: the surprising twang, the forced note, the comic rapidity of the voice-off in Pathé newsreels; or the syrupy sentimentality of commentaries in travelogues; or the epic bombast, obligatory class hatred, and insufferable optimism of the commentary in certain militant films. What one hears, then, is something like the body of the voice – and its body is its death to meaning. The voice is no longer drawn for us from the image (presuming that the latter is not itself perceived as grotesque and abusive, which happens but in which case the problem is not posed), nor a fortiori from the real which is mirrored there. It detaches itself and becomes part of an ensemble which does not hold together. This is what permits aesthetes to love film in spite of itself by finding in it the charm of the old, an aspect of kitsch and nostalgia.
The Voices Which Are Heard
The voice ages. Its signifier (that in it which is heard) “labors.” In it is perceived an accent which is not that of a region of geography but rather a region of meaning, the accent of an era, a class, a regime – and this accent neutralizes meaning, defuses the knowledge imposed by that voice. This is why it is necessary for it to speak as little as possible. No commentary is the wisdom or the prudence of the master, of power. The voice expounds. It expounds more than the image, for one can play with the image, but then the ridiculous risked is not that of the ultimate, mortal ridiculous – of meaning.
That which has the “right to speech” (in the sense that it has the power to do so) should use it as little as possible. This is a curious rule. Does it hold for all films in the register of the real, of the document and of the documentary? Is it evaded in the limit case (such as Mai 68) when commentary is abolished and scansions of montage substitute?
To this second question one can respond that the accent – since it is this asperity of the voice which catches critical hearing (as a flood of light can catch the grain of a surface which is the “material” of a painted canvas) – the accent is also perceived, though less directly, beyond the voice in the shocks of montage and the frequency of cuts which are the means by which montage makes discourse. One does not thus escape the hindrances of discourse or truth (even though this is a platitude).
But who is it that is afraid of deception, aging, and dying? Is it everyone? Perhaps everyone, to the extent that, as it is said, he aspires to mastery, he defends his power, his knowledge, his memory. This is to say to the extent that he is one. Voice off: the voice of his master. And there can be only one master. This is why, generally in this system, there is only one commentating voice on the sound track (and most of the time it is a man’s voice). Sometimes, but more rarely, there are two alternating voices (man, woman) so that the superior unity of discourse arises from their discrete conjugal harmony. But if the voice is found to be geared down and with this the unity of the discourse is found to be broken, the system and its effects change. Space-off then stops being the place of the reserve and the interiority of the voice (that place where it “is heard to speak”); it is itself divided, and from this division takes on the dimension of a scene, is dramatized and is peopled. Something happens there, parallel to the image or interlacing with it, but unfixed from it. The voice is no longer simply planted in the image when there are many voices and they struggle with one another (as in Godard, notably during the period of the “Dziga Vertov Group”), or when they desire and love one another (as in Duras). But what is found to be subverted is the very principle of the documentary, its principle of reality (both as unique and as objective).
To encounter the “body” of the voice (what Barthes calls its grain), this refuse from meaning, is to encounter the subject of the voice with its division, the subject fallen to the rank of object and unmasked: thus, in Chris Marker’s exercise “the paranoid anticommunist” and “the jovial Stalinist” were carnival figures. On the one hand meaning is naked, neutralized, laughable (“the emperor has no clothes”); on the other hand, there is its “body,” its cadaver: its noise and cacaphony… Why is it that when it becomes perceivable, this bursting of the unity of commentary and of voice, this scission of sound and meaning, is accompanied by manifestations of revolt, by laughing and crying in the theaters? Against what is a revolt occurring? Against humbuggery. But this humbuggery does not give rise to a special reaction if it does not claim to dominate with its discourse the real that crosses the image. To recognize the humbuggery in the imperturbable voice-off by laughing is to lift the oppression of the commentary. The laugh explodes, its splits open: it breaks through the voice of the commentator.
If the unity of voice and meaning in the commentary-off defines a regime of mastery or of oppression, it is perhaps starting from its scission that one could begin to define another politics (or erotics) of the voice-off. Something of this order was devised by Buñuel in Las Hurdes: the commentary-off is cold, but the image shrieks. Over the image, it cracks, it rots, it cruelly grimaces; and by this, the circumspection, the reserve of the commentary becomes strange, becomes the atony of the disquieting voice, as if the abyss between the silent cry of the image and the discourse of the voice-off were imperceptibly traversed by a silent laugh which belies what this voice says. In the end, this voice is heard. Starting from the moment when it is heard, its discourse – which is never that of mastery – is found to be menaced, the function of the commentary is lacking. and that of documentary is placed on trial. What is devised in Las Hurdes is a radical testing of the mastery based on the commentary, on imperialism, and on colonialism, which are deep-seated in documentary.
All modern cinema since, let us say, Godard on the one hand and Bresson on the other, is instituted by simultaneously placing on trial the filmic image as a full, centered, and deep image, and the utilization of the voice as homogeneous with and in harmony with the image. It has often been said that what cinematographic modernity puts into play – no matter what claims it uses to promote itself – are effects of rupture, disalignings, “noise” in the filmic chain. A tearing in the effect of the real of the image and in the effect of mastery of the voice is brought about. The relation among voice, sound, and silence is transformed and musicalized.
This is what provides the site for limit experiments, such as those in the margin of militant cinema of the “Dziga Vertov Group,” or those of Straub or of Duras. Straub is certainly the filmmaker who has played the most richly, musically, and dramatically with plurality, with the capacity of voices in filmic space; it does not always know a lot, but in the end it will be heard. In Duras (who recently declared that she can no longer synchronize – she said “screw” – voices into mouths) this is, among other things, an experiment with silence and with the subversion of the voice with silence. This is true from Destroy, She Said and Nathalie Granger (Nathalie Granger is inhabited by silence not in order to reinforce, from an effect of reserve, the power of speech which is that of man, but to paralyze and to cast a spell over the latter and to return the voice as lapse, as trouble, as a liberty of the unconscious – to return the voice to women) to Woman of the Ganges and India Song (and perhaps Vera Baxter) where the silence of the image provokes the sonic peopling of space-off. Duras thus introduces the spark of desire and sends the question back to the spectators.
But these are fictions, which is to say works supremely unconcerned with that “reflection of the real” by which the truth of a documentary is measured and which imposes on the voice of the latter that specific discretion which is our problem. It would be necessary to discuss militant cinema, where the voice finds itself invested with a precise function which is, moreover, variable. For example, in Oser lutter (made in Flins in May-June of 1968 by established militants) there is that confused mixture of voices, over black leader, from which emerges in bits and pieces the “truth” of the struggle: “and above all, no negotiations…” To this mixture and burning confusion on the sound track is opposed, as the clarity of revolutionary knowledge, intertitles where that knowledge is all written, white on black. Or in Shanghai au jour le jour, there are two voices-off of women in dialogue, but they do not directly comment on the real which the image reflects (as, for example, does Antonioni, discussed above); but rather they comment on the image track and clearly are speaking in an editing room. Here there is something which blocks the terrorist indetermination of the voice-off-two women who speak to one another, who have a dialogue before the images…
If, in a general way, the problems here evoked are rarely resolved in an interesting fashion (and rarely even envisaged) by militant cinema, it is nevertheless from its place and starting from the problems posed by militant film makers that it is possible to begin to think and to realize a subversion of documentary and of that which secretly operates there: the discourse of power, the discourse-form of power. Indeed, militant cinema cannot be something like classical documentary plus rage and great, fine-sounding words. It has to be something else completely, something which organizes otherwise the relation to the real, the look, and the voice.
Here or there, in varying degrees this is being done, for what is called militant cinema never begins except from where classical documentary ends, in that which the latter smothers and erases: the speaking subject. The stake being risked is that of the subject. This is the reason classical documentary represses that stake, most often in the dogmatism of a voice-off without subject. The greatest difficulty in the order of the document and the real which there makes itself known is indeed that of not effacing the subject, the I and the you. Perhaps this leads toward a new way of approaching the real, a way which is more supple, and also more hazardous, more open to chance. Hazard, discovery – this is the chance of the new.
Hence, militant cinema loves the hazardous…
N.B. 1. It will have been understood that commentary is not point of view. Its problem is precisely how to silence a point of view. Not that the commentary exists before an organized point of view, which could otherwise say everything. But commentary is not for saying all: its role is to legislate on images of the real, to inject knowledge into them.
Rather, the question of point of view has to do with truth. It is about the way in which a subject is implicated in a process and what emerges from it. In discussing commentary, I have thus spoken of the signifier of the voice. I have placed hardly any emphasis on commentary which is written and which in film is called “intertitles.” The procedure of speaking through them has been singu- larly rare, so that today one barely finds it except in the cinema of Godard and, since May 1968, in a somewhat sophisticated sector of militant cinema: le Peuple et ses fusils (distributed commercially), Oser lutter…
One immediately sees the difference between spoken commentary-off and commentary through intertitles: beyond the necessarily fragmentary, discontinuous, and compiled character of the discourse of “intertitles,” in them knowledge is designated as such, which is to say as didacticism. To the interiority of the voice-off is opposed the exteriority of the text presented for reading, which solicits the eye and appeals to the “consciousness” (in the political sense of the word) of the spectators. In comparison to voice-off, intertitles involve a kind of effect of distanciation or – as Philippe Ivernel has proposed to translate the Brechtian Verfremdungseffekt – of “disalienation.” This is why it is essentially in militant cinema that intertitles are utilized.
It has to be added that they rarely make for pleasure. They will rather provoke irritation, even hatred. The spectators say they are offended that the dotting of the i’s is thus placed between the images for them. They would prefer instead to let the images speak for themselves; and with the images, the real.
With words and intertitles it [ça] is indeed interposed between the spectators and that which filters the real in the image. This is the appearance. The appearance is not disturbing when you are injected unawares from behind, with the voice-off. But in contrast, when on the screen, that appearance frustrates the eye and provokes resentment (or rather, one’s resentment is toward the authors of the film).
N.B. 2. In classical and modern cinema (with the exception of certain under- ground films), but specifically in documentary, filmic space, and even the ensemble of the cinematic apparatus [dispositif], is polarized by two concurrent orderings-the look and the voice. The look and the voice are not equivalent to image and sound. They are the two complementary and concurrent orderings to which the work of the image track and the sound track (shooting, editing, sound mixing, color and tonal grading of the image, etc.) are submitted. This prevalence of the look and the voice is a relation of the subject’s “flickering in eclipses” which regulates the filmic chain (the alternation and disposition of shots: the 30-degree rule, shot/countershot, etc.) and of which the spectator is the support (see Jean-Pierre Oudart, “La Suture” in Cahiers du cinéma nos. 210 and 212). This prevalence points toward that which constitutes the spectator as such: spectator of his or her desires (insofar as desires are constituted by the Other). It thus also involves the division, the identificatory alienation of the spectator, in the filmic chain.
Recently the question has been posed whether this “alienation” is irreducible, or whether it is possible to lift it by some perversion of the apparatus [dispositif]: for example, by inscribing in the film its own means of production; or by making the screen opaque through the multiplication of iconic or sonic traces to the point of nondifferentiation and unreadability, so that the scene of the phantasm becomes blinding and literally puts the spectator back in his or her place. One can have doubts about the interest of such an operation from the viewpoint of systematicity. It certainly mobilizes the look differently, makes it scan the surface which is the wrong side of the décor, and opens it to other adventures. It is not so certain that the latter would fundamentally be something other than tributaries of “representation”; nor, especially, that they would come socially – on the paths crossed by desire and the social – from a fundamental subversion. Beyond the facile conflicts of imaginary depth and real surface subsists the question of the object: what to do – what is to be done – with the look and the voice?
Notes
1. “It speaks” is a translation of “ça parle.” This is a phrase from Lacanian psychoanalytic thought. “Ca” (it) is the French word used to translate the Freudian term es (also literally it), which is rendered in English-language usage by the Latin id. This is, of course, one of three terms in a famous Freudian psychic topology, along with the ego and superego. In Lacan, the phrase “ça parle” emphasizes that speech is not controlled by the coherent subject (1, ego) but something else (it, id) which is potentially disordering and disorderly with respect to that subject; language is not a product of reason, but has its roots in the unconscious, the repressed, what is “other” to consciousness.
In this, its first appearance in the article, Bonitzer puts the phrase “it speaks” in quotation marks, which emphasizes the derivation from Lacan. However, there are other times when Bonitzer engages in the ambivalence enabled by the fact that in French a word which can simply mean it might also denote the id. Sometimes he chooses to associate the antecedent of it with the Freudian id, for example at certain points when the antecedent is the voice-off (speech from elsewhere than the realist space of the image). Since English-language usage maintains the distinction between the two terms, it has not been possible to translate the article’s employment of this ambivalence when it occurs. However, where it is especially important to the sense of a sentence, it is noted.-TRANS]
2. “Jouissance” is a term notoriously difficult to translate. “Enjoyment” does not capture the full range of intensities covered by the French term. “Pleasure” ignores the distinction-some- times exploited by theoreticians such as Roland Barthes-between plaisir and jouissance. There- fore, where it appears in the text, it has been left in French.
What might minimally be said about its significance is first that its meanings can range from enjoying something in the sense of legal possession to pleasure in the sense of sexual climax. In addition, there is a tendency in recent French theoretical formulations to give the term a special, privileged sense. For example, Stephen Heath outlines the distinction in Barthes: “on the one hand a pleasure (plaisir) linked to cultural enjoyment and identity, to the cultural enjoyment of identity, to a homogenizing movement of the ego; on the other a radically violent pleasure (jouissance) which shatters-dissipates, loses-that cultural identity, that ego.” Other thinkers, such as Lacan and Kristeva, have their own special uses of the term which are related but not necessarily identical to that of Barthes. Roland Barthes, Image-Music-Text, trans. Stephen Heath (New York: Hill and Wang, 1977), p. 9; cf. Leon S. Roudiez’s comments on the term in the introduction to Julia Kristeva, Desire in Language: A Semiotic Approach to Literature and Art, ed. Leon S. Roudiez, trans. Thomas Gora, et al. (New York: Columbia University Press, 1980), pp. 15-16.
In Bonitzer’s text, the paragraph before the introduction of the noun jouissance includes two uses of the verb jouit, translated as “enjoy”-TRANS]
3. It has been pointed out to me that “end of prosopopoeia” is not completely exact. The discourse which Chris Marker opposes to two others (thesis-antithesis-synthesis) may not be an outright caricature; however, neither is it entirely that of Marker, whose dialectic is more subtle. It remains no less true that if “objectivity isn’t exact, either” (as all left-intellectuals even sightly tinged by Marxism know), Marker does not say why, or responds to the question obliquely, for want of envisaging wht is arbitrary in discourse-off (he is known, with Rouch, to be one of its most brilliant practitioners). See below.
4. [There is a linguistic play in this paragraph between generalized collective identity and linguistic gender. “Everyone” (tout le monde) is a masculine noun phrase, which is usually pronominalized by the impersonal “one” (on, often translated into English as “we,” but which, interestingly, still takes masculine complements.) In this paragraph it is pronominalized by the masculine il (“he”).
So in the text when the author writes, for example, that “he (il) aspires to mastery” the result is to emphasize, Ina somewhat unusual way. the masculine gender of the antecedent “everyone.” And. when the author writes to the extent dut he is one.” the point is not only the pulling together of “everyone” into “one,” but also an association of this social homogenization with a hierarchy of gender. because the masculine linguistic gender of ‘-everyone” has been so stressed. — TRANS.’
5. (On the peculiar use of “it” here. see note I.—TRANS I
6. Here I think especially of militant cinema and of the Dziga Vertov Group; the earlier films of Godard and Numéro Deux make use of intertitles and letters in ways which are not reduced to didacticism. but cause fantasies of the unconscious, permutations of language, witticisms, etc. to intervene.