In the next-to-last episode of his video series Histoire(s) du cinéma (1988–98), Godard recombines familiar images from Hitchcock's films in a demonstration of what he calls ‘Hitchcock's method’, an all-powerful form of montage. In titles and voiceover, Godard makes several grand assertions: that Hitchcock is the ‘greatest creator of forms of the twentieth century’; that he is the only director, apart from Dreyer, to ‘film a miracle’; that he is the only ‘poète maudit’ to achieve major commercial success; and that during his Hollywood prime he ‘took control of the universe’ and ‘succeeded where Alexander, Julius Caesar, Hitler, and Napoleon had all failed’. We hear and read these claims while Godard manoeuvres between Hitchcock's images, each clip separated from the next by a black screen. Speaking in a low, gruff voice, Godard contends that ‘we’ don't remember the plots of Hitchcock's films: why Joan Fontaine leans over the edge of a cliff, why Joel McCrea goes to Holland, why Janet Leigh stops at the Bates Motel. Instead, ‘we’ recall a handbag, a bus in the desert, a glass of milk, the sails of a windmill, a hairbrush, a row of bottles, a pair of spectacles, a sheet of music, a bunch of keys. Godard lists these things in a different order than he shows them, but his videographic montage works to suggest that the power of ‘Hitchcock's method’ lies in its intense orchestration of details.

The basic thrust of the segment seems to be a salute to Hitchcock's fascinating command of objects through film form – simple, quotidian objects that float free from their narratives and linger in collective memory. Godard's itemising voiceover indicates as much, and so far this has indeed been a common reading of the segment by critics. Laura Mulvey, in her book Death 24x a Second, refers to Godard's ‘fetishization of Hitchcock's things’” and points out that Dominique Païni's 2001 museum exhibition Hitchcock et l'art, which displayed several of the same physical objects on red velvet pillows, carefully lit and encased in glass, took Godard's exercise in Histoire(s)‘to its logical conclusion’.1 But this view, however encouraged by Godard's own statements, gives us a limited sense of what's going on in this complex sequence. It's important to observe that we are not dealing with a highlight reel of ‘object images’ that illustrate the voiceover. Take, for instance, the Pommard bottle in Notorious (1946): when it appears in Histoire(s), first it totters forward on the shelf at a gradual, uneven pace that suddenly accelerates; then it falls, according to Hitchcock's original speed, past Cary Grant shown in close-up as he inspects the other bottles; then, in slow motion again, the bottle shatters on the floor and its granular content spills out; then, with the speed still slowed, we see a low-angle shot of Ingrid Bergman responding. What we're given is an event in four parts. Hitchcock's original découpage is left intact, but the rhythm is adjusted so as to stress not just the bottle but the ensemble of shot-units that renders it acutely memorable.

If the black screens have the effect of isolating each fragment, they also open up new connective streams. To stay with the same example, the shots of the shattering wine bottle sustain a motif of objects dropping to the feet of a performer: the eyeglasses of the strangled wife in Strangers on a Train (1951), the key nudged into a floor vent with the toe of a high-heeled shoe in Marnie (1964), the key dropped and kicked under a table in Notorious after one hand transfers it to the other with the dexterity of a pickpocket. The bottle also initiates a textural association across three scattered fragments: the substance spreading at Devlin's feet finds an echo in the black dye rinsing from Marnie's hair in the sink, then another in the blood collecting at Marion Crane's feet in the shower, all three shots slowed and given a slight turbulence with video stop–starts. As the segment unfolds, we see that ‘Hitchcock's method’ has to do with the recombination of images according to formal resonances – repetitions, contrasts – across separate films. And we see that this method depends on performance gestures as much as it does objects. Godard makes this evident when, in three consecutive fragments, he links the slashing movements of Rose Balestrero from The Wrong Man (1956) frantically raising her hairbrush into the air, the wiper blades on Marion's windshield, and the downward stabs of ‘Mrs Bates’ through the shower mist. As Marion, defenceless, tries to avoid the attack, Godard repeats the title of the episode in voiceover, ‘Control of the universe …’

The more closely we examine this segment in Histoire(s), the more difficult it becomes to reconcile what Godard is doing with what he is saying. Questions stack up without easy resolutions. How did Hitchcock achieve this degree of control? What, for Godard, constitutes a miracle, and when did Hitchcock film one? Just who are the ‘we’ mentioned in Godard's voiceover? Is this ‘method’ even Hitchcock's in the first place? Godard, after all, seems to claim it as his own: rhythmically altering the speed of motion, tinkering with the colour and texture, ‘matching-on-action’ across multiple works – all these efforts open Hitchcock's images to new forms of linkage. And yet, in keeping with the funereal tone of Histoire(s), Godard implies that the force of this method is now lost and irretrievable. This in turn begs the question, no less complicated than the others, of what Godard is trying to accomplish for his own practice through Hitchcock's images. Jacques Rancière, in a number of recent essays that concern Histoire(s), has explored this question at length and made the case that ‘Hitchcock's method’, as Godard presents it, is essentially a matter of ‘inter-expression’– as Rancière puts it, a poetic principle by which each image, as a ‘pure sensory block’ indifferent to its place and function in an unfolding plot, can be made to recombine with all other pure images that together form a boundless continuum. For Rancière, the lost power of montage that Godard ascribes to Hitchcock is, in effect, the same power Godard resurrects in Histoire(s) as a means of imposing a sense of historical ‘co-belonging’ among unrelated elements.2

Though I will eventually take up Rancière's account in greater detail, I want to contend at the outset that Godard is, in fact, profoundly ambivalent about the power of montage and the ‘control of the universe’ he attributes to Hitchcock's cinema, and that part of Godard's work in this segment is to differentiate his own montage practice so as to forge ahead with his videographic investigations. Certainly a ‘method’ is at stake, and also the manner of seeing and thinking to which it orients us. Yet this method ultimately comes down to something more specific than an ‘inter-expressive’ poetics. As I hope to show, Godard works towards a point of divergence from Hitchcock's montage, and this point hinges on the device of superimposition, which in Histoire(s) is a privileged form of discovering connections between more or less disparate fragments. Paying attention to this undercurrent of criticism on Godard's part will not bring us closer to a ‘main thesis’ that explains all the mysteries of this segment, but it will, I believe, reveal a deeper sense of how Hitchcock matters to Godard in the series and how this puzzling tribute has strong implications for Godard's own historical project.

Given that ‘method’ and ‘control’ are pivotal terms in the segment – and in the episode more broadly – it is worth remembering that Godard's intense engagement with Hitchcock's cinema itself has a complex history that traces back to Godard's formative stages as a critic and filmmaker. That topic warrants a thorough examination in its own right, but here I will only stress that Godard, across his many dealings with Hitchcock's work, has wavered on the issue of methodic control. Initially, in his writings for Cahiers du cinéma in the 1950s, Godard turns to the work of Hitchcock as a way to challenge the aesthetic notions handed down from André Bazin, in particular the ‘evolution’ towards greater realism that spans silent and sound eras and reaches its high point with the long-take, composition-in-depth styles of Welles and Renoir, and also with the neorealists De Sica and Rossellini, who likewise ‘respect’ what Bazin calls the inherently ‘ambiguous’ nature of reality.3 Godard took interest in Hitchcock's films because they had no part in this trajectory (as Bazin's sceptical reception of Hitchcock would confirm): they blended aspects of the two main traditions that Bazin's evolution rejected – Soviet montage and German expressionism – and they showed that classical découpage, far from being a mechanical, uniform style that had run its effective course by the 1940s, was still amply equipped to seize and accentuate the emotional realities that a long-take aesthetic would play down.4 In his reviews of Hitchcock's films, and in two of his most notable essays of the period, ‘Defence and Illustration of Classical Construction’ (1952) and ‘Montage My Beautiful Care’ (1956), Godard admires and defends the way in which Hitchcock structures events with a view to controlling their affective impact. Hitchcock, he argues, gives reality ‘the style it lacks’, and if he employs ‘clever and violent effects’, he does so only to ‘transmit the drama to the spectator at its highest level’ (pp. 24–5).

But Godard, in his last stretch of articles for Cahiers leading up to À bout de souffle (1960), becomes more and more interested in chance and spontaneity, and he makes a basic distinction between ‘the cinema of freedom’ and ‘the cinema of rigour’:

Broadly speaking there are two kinds of film-makers. Those who walk along the streets with their heads down, and those who walk with their heads up. In order to see what is going on around them, the former are obliged to raise their heads suddenly and often, turning to the left and then the right, embracing the field of vision in a series of glances. They see. The latter see nothing, they look, fixing their attention on the precise point which interests them. When the former are shooting a film, their framing is roomy and fluid (Rossellini), whereas with the latter it is narrowed down to the last millimetre (Hitchcock). With the former (Welles), one finds a découpage which may be loose but is remarkably open to the temptations of chance; with the latter (Lang), camera movements not only of incredible precision in the set but possessing their own abstract value as movements in space.

(p. 79; trans. modified)

In his lengthy 1962 interview with Cahiers, given not long after the release of Vivre sa vie (1962), Godard elaborates on this general split. There, he claims that for Demy, Resnais, Eisenstein, and Hitchcock, making a film is just the ‘practical application’ of a concept already fully imagined and worked out in the planning stages. ‘The others, people like Rouch’, he continues, ‘don't know exactly what they are going to do, and search for it. The film is the search. They know they are going to arrive somewhere – and they have the means to do it – but where exactly?’ Citing the example of Renoir, he allows for the integration of these approaches, and he surely has his own practice in mind here as well (p. 180). From his early films to his late videos, Godard has brought ‘rigour’ and ‘freedom’ into convergence. Accounts of his production methods, such as Alain Bergala's Godard au travail: Les années 60, and the materials compiled in the Documents book released in connection with Godard's 2006 exhibit at the Centre Pompidou, give the lie to the view that the French-Swiss director merely improvises on the spot.5 And yet for Godard the film is still resolutely ‘a search’, an attempt, a rough sketch: less the application of preconceived ideas than an open-ended continuation of the thinking process.

This deep tension between ‘rigour’ and ‘freedom’ is very much alive in Godard's Histoire(s), which oscillates in mode from sequences that come to us with the fragments precisely (not to say unambiguously) organised by a set of ideas, events, and propositions to sequences that seem half-willed, half-accidental and operate more in a lyrical register of feeling and imagination. The issue here is not just how the work is created but how it impresses on and involves the spectator, and Godard's ambivalence towards Hitchcock's ‘control of the universe’ is tied to this question. To be sure, the segment is a reverential tribute, entirely genuine in the way it situates Hitchcock as a supreme formal innovator, and in the way it implicitly reaffirms his importance to the New Wave figures at Cahiers. In fact, the segment is an hommage not just to Hitchcock but to the kind of criticism his work inspired. The claim for Hitchcock as the century's ‘greatest creator of forms’ is a rephrased line from the conclusion to Rohmer and Chabrol's 1957 book-length study on the director, the first of its kind: ‘Hitchcock is one of the greatest inventors of form in the entire history of cinema.’6 While it's often glossed as a one-note thesis on the ‘exchange of guilt’ theme, Rohmer and Chabrol's Hitchcock is nothing if not a nimble and diligent exercise in mapping out the formal architectures in Hitchcock's work, within and across films. This meticulous attention is what Bazin finds so intriguing in his 1958 review of the monograph: not being a Hitchcockian, he is unconvinced by the argument, but what strikes him as a major critical achievement is the manner in which Rohmer and Chabrol, through their ‘micrometric’ description of patterns and transferences, summon up for the reader a spectacle that is just as intricate as ‘the best Hitchcock film’.7

Godard's handling of Hitchcock's images in Histoire(s) alludes to and animates this diligent criticism, its acuity of perception, its impulse to achieve ‘material intimacy’ with the cinematic work.8 Yet Godard's attentiveness has a more contentious edge to it. Even as he exalts the power of Hitchcock's cinema, Godard works carefully against the logic that underpins it. We can sense this ulterior project in the two moments in which we hear Hitchcock speak in the segment. First, over a succession of black-and-white stills of other esteemed directors – Bresson, Lang, Cocteau, Rohmer, Truffaut, Rivette, Visconti, Garrel, Fassbinder – we hear Hitchcock assert: ‘Sometimes you find that a film is looked at solely for its content, without any recourse to the style or manner in which the story is told, and after all this basically is the art of the cinema.’ Godard's voice abruptly intrudes on Hitchcock's last few words. ‘Images first’, he says, ‘but the ones Saint Paul mentions, which are a death, therefore a resurrection.’ This interruption seems to be a correction of sorts, since Godard's claim – contrary to Hitchcock's – is that we forget the content and remember specific details, gestures, objects, patterns, moments. But the variance here is not simply an issue of form and content. Hitchcock's remarks concern public perception of the film during its phenomenal unfolding, whereas Godard's claim has to do with the film as it hangs in our heads afterwards, dissolved into remembered fragments. This tension resurfaces a few moments later, when Hitchcock says:

We have rectangular screen in a movie house, and this rectangular screen has got to be filled with a succession of images. That's where the ideas come from. One picture comes up after another. The public aren't aware of what we call montage, or in other words the cutting of one image to another. They go by so rapidly, so that they are absorbed by the content that they look at on the screen.

Here Godard works counter to Hitchcock's statement by decelerating and reshaping the fireworks-embellished kiss from To Catch a Thief (1955): enclosed in a diamond-shaped frame, Grace Kelly and Cary Grant, in a profile two-shot, gradually lean towards each other. Godard re-edits the shot so that it's bracketed by explosions of light and colour situated elsewhere in the original. After a black screen, we see, as if an extension of the previous action, James Stewart and Kim Novak (as Judy made over as Madeleine for a second time) already entwined in a kiss in Vertigo (1958), this image irised and slowed more drastically, her arm reaching up and around his shoulder in short, staggered spurts. Precisely when Hitchcock says audiences don't notice the workings of montage due to the speed of succession and narrative absorption, Godard takes measures to enable the viewer of Histoire(s) to concentrate on each image and the resonances between images. In fact, Godard transforms these scenes into explicit ‘embodiments’ of montage, of two elements converging and igniting a ‘spark’. In other words, the intra-image content, the action depicted, corresponds to what Godard is doing in formal terms (and he repeatedly uses human gestures in this fashion in his video work, as corporeal figures of montage9). Godard thus undoes Hitchcock's hypnotic power even as he celebrates it. Hitchcock isn't bothered that audiences are ‘absorbed’– for him it seems a point of pride. His montage, its regulated rhythm, its particular organisation of stimuli, depends on this condition for its subliminal effects. Godard, however, needs us to notice and notice acutely the binding forces and structures, the material intensity of the process.10

Immediately after the slowed embrace taken from Vertigo, we see two separate superimpositions involving the sequoia forest scene of the same film: in both fragments (originally structured on Scottie's look, in shot–countershot alternations, as ‘Madeleine’ recedes into the landscape), a flickering black-and-white still of Hitchcock with his hand raised merges with the graphic contours of the composition. His face (the sides of which are blackened with a magic marker) fits congruously in an opening between trees, where sunlight spills into the forest. From one fragment to the next, it looks as though he makes ‘Madeleine’ vanish behind a tree, then makes her reappear. The titles breaking up these images announce: ‘the only one, with Dreyer, who knew how/to film/a miracle’. This leads directly into the last image of the segment – the superimposition from The Wrong Man in which the face of Manny Balestrero (Henry Fonda) overlaps with and dissolves into the facial features of his double, the ‘right man’. But Godard adds a third layer to the ensemble: the same still photo of Hitchcock, flashing as before, his hand now seeming to call into existence this unlikely revelation of truth. In Histoire(s) the shot is given an even stronger touch of the epiphanic, as it coincides with the one moment of harmonic uplift in the sparse, volatile music that plays throughout the segment, the Georgian composer Giya Kancheli's Abii ne viderem (1992) for viola and string orchestra.11

The shift from successive to simultaneous juxtaposition towards the end of the segment is less abrupt than it might seem. Godard prefigures this shift with two earlier, short-lived superimpositions that are less conspicuous: first, the composite image that begins his demonstration of ‘Hitchcock's method’, the eerie shot of Marion's rearview mirror framing a police car on the barren highway behind her (which in Histoire(s) calls to mind Godard's own irising techniques); and second, the cross-fade from the carpet of Alex's bedroom, where Alicia has just hidden his key to the wine cellar, to the virtuosic overhead shot of the ballroom – but Godard cuts to a black screen before the initial layer of the cross-fade disappears and before the shot swoops down into a tight close-up of the key in Alicia's hand, as in Hitchcock's film (and this in turn allows the chandelier lights to rhyme graphically, across the black screen, with the fireworks from To Catch a Thief). Neither of these two moments stands out as a superimposition in its original situation: the rearview reflection, a point-of-view shot in Psycho (1960), is integrated into the series of events concerning Marion's flight; and the cross-fade in Notorious indicates the passage of time and softens the change of scene. But in Histoire(s), where their narrative function is diminished, they figure primarily as composite structures, and they work to position the double exposure from The Wrong Man as a culminating shift into simultaneity.

Here again there is a deeper historical context for what Godard is doing than he makes apparent in the segment. This superimposition in The Wrong Man held a special significance for the French Hitchcockians at the time of the film's release. As Jonathan Rosenbaum has pointed out, this image constituted ‘a miracle in relation to the struggles of French Hitchcock criticism of the 1950s’ because it so strikingly confirmed their claims about the transference of guilt and the motif of the double.12 Truffaut called it a gorgeous summation of Hitchcock's entire oeuvre, and Rohmer and Chabrol said it put The Wrong Man in company with other magisterial explorations of the miraculous in postwar cinema (namely Rossellini's Viaggio in Italia (1954), Dreyer's Ordet (1955), and Bresson's Un condamnéà mort s'est échappé (1956)).13 Godard, for his part, wrote a detailed review of the film in Cahiers entitled ‘Le Cinéma et son double’, and there he describes not one but ‘two miracles happening on Fonda's face’. To get a clearer sense of how Godard revises Hitchcock's superimposition in Histoire(s), we need to consult this 1957 article, which is already concerned with matters of succession and simultaneity. Godard, in fact, directs us back to the article not just by revisiting the film and its miraculous revelation, but also by repeating a certain line from the earlier piece, a quotation from André Malraux's Les Voix du silence: ‘What is art if not that through which forms become style?’14

In ‘Le Cinéma et son double’, Godard observes that Hitchcock's film is intricately composed on the principle of repeated shots: ‘Each crucial shot in The Wrong Man has in effect its respondent, its “double”, which justifies it on a narrative level while at the same time “redoubling” its intensity on the dramatic level’ (p. 53). Godard identifies a number of repeated elements (‘two imprisonments; two handwriting tests at the police-station; two conversations with Rose in the kitchen; two hearings …’) and argues that these narrative echoes are bound up with other repetitions at the micro-level of shot design. He describes how the shot of Manny finding out he's to be released on bail – where the camera at first ‘pushes him’ into his prison cell, watches him through the rectangular eye-slot, but then suddenly retreats along the same path – doubles and negates a previous moment in which Manny, seated in a cramped automobile and surrounded by police officers, sees the driver coldly observing him in the rectangular rearview mirror. ‘A first miracle enters the lists’, says Godard. ‘The film seesaws completely’ (p. 52). Here we should bear in mind that this ‘miracle’– as Hitchcock handles it and as Godard describes it – is more nuanced than the mere staging of Manny's release, or the graphic resonance between the eye-slot and the rearview mirror. The moment in the car is structured as a series of point-of-view shots as Manny passively and helplessly takes in his situation, the impersonal cops on either side of him, the driver watching him in the mirror. Together the shots impart a dreadful sense of enclosure and entrapment. The prison cell image counters this moment by temporarily reversing the grim, mechanistic force implied in its shot construction.

We're dealing here not with foreshadowing and fulfilment but with ‘positive’ and ‘negative’ pairings that collect in progression (or, as Hitchcock might prefer, that pile up subliminally). Godard points to a second ‘miracle’ that crystallises this logic of doubling, the scene in which Manny heeds his mother's advice to pray not for help but for strength. Getting dressed for work, he looks at a portrait of Christ on the bedroom wall. In a point-of-view shot, we see the painting superimposed with his heavy shadow, and the camera pushes in, enthralled by its object. In a countershot we return to Manny in close-up as he mouths the words of a prayer. Suddenly, in a double exposure, we see a long shot of the ‘right man’, the actual criminal, strolling down the street into the foreground: the distance between the two figures collapses and their faces merge in a composite of close-ups, until Manny dissolves and we are left with a robbery scene in which the double is captured.

Godard, unlike Rohmer and Chabrol, is uninterested in the Catholic overtones of the scene. He regards this shot as a miracle because it ruptures the chain of false sightings and false evidence –‘the machine grinding inexorably on’ (p. 49) – that has suppressed the truth. As with the earlier miracle in the prison cell, it spares Manny from the narrative of entrapment, and it does so through a gesture of doubling and negating, only this time the juxtaposition is simultaneous. ‘The transition here’, Godard asserts, ‘is no longer a hinge articulating the story, but the mainspring of the drama whose theme it paraphrases’ (p. 54). In Hitchcock's film, the shot is confirmed as a miracle by the ensuing events: the double is caught while holding up a grocery store, and his arrest absolves Manny of all criminal charges. This turn of fortune seduces us into thinking that Manny's trials are over, but as Godard observes in the review, this is no simple case of Hitchcockian guilt transference. ‘In The Wrong Man, the transference no longer resides in the innocent man's assumption of the real murderer's crime, but in the exchange of Manny's liberty against Rose's’ (p. 52). Indeed, Manny's misfortune, in spite of being absolved, is that he cannot share his freedom with Rose. ‘That's fine for you’, she says to him at the sanatorium. Manny, dejected, tells the nurse, ‘I guess I was hoping for a miracle.’ Yet Godard's notion of the miraculous in the article has little to do with this narrative logic by which a miracle is confirmed or not confirmed. In a sense, the double exposure, as he describes it, is less a miracle for Manny than it is for the spectator. As an ostentatious gesture of revelation, it makes visible what no look internal to the story world has seen. It divides us from the witnesses in the fiction whose failures of vision have led to Manny's predicament. It offers us a composite image that has only a weak equivalent in the diegesis: two photographs printed in the newspaper side by side, above the headline ‘Suspect in Holdups Cleared by Double’.

When Godard reworks this superimposition in Histoire(s) and again considers it as a miracle, the basis for this designation is not the narrative turn of events that follows. After all, if we follow Godard's general claim in the segment, we don't actually recall the circumstances of Manny's misfortune anyway. Godard doesn't remind us by showing the double's failed holdup and arrest. Nor does he show the painting of the Sacred Heart to which Manny is praying – for Godard, the divinely providential aspects of this moment are not sufficient grounds to call it miraculous. In Histoire(s), we are witness only to the convergence of facial features, with the still of Hitchcock flashing at an erratic rate, then fading along with the close-up of Manny. The addition of Hitchcock to the ensemble has two distinct purposes. It celebrates the control of the director, the creator of forms, and in this sense it fittingly concludes the segment. However, the flickering third layer is also a creative act of montage on Godard's part, and it signals a crucial point of difference from Hitchcock's use of superimposition. As with the rearview shot in Psycho and the routine cross-fade in Notorious, Hitchcock rigorously integrates the composite of figures into a narrative scheme. But Godard ‘remembers’ the same moment in Histoire(s) as a stunning gesture of rapprochement, of two separate elements brought together for inspection: the dramatic causes and effects are stripped away so as to situate the ‘miracle’ as a potential one, dependent on a process of looking and judging (fig. 1). Godard is drawn here, above all, to the possibility of an unforeseen discovery through montage. And this is why the diegetic complications in The Wrong Man– from the possible innocence of the double (since we only see him attempt a single robbery, we can't be sure that he is guilty of the crimes for which Manny is accused) to the psychic collapse of Rose – have no immediate bearing on his description and appropriation of this moment as a miracle.

Godard recasts Hitchcock's double exposure by bringing it into alliance with his own investigative methods in the series. As an image formed of heterogeneous elements, it becomes an elegant demonstration of what Godard, following Pierre Reverdy's notion of the poetic image, calls ‘the rapprochement of two more or less distant realities./The more distant and just the relations between these realities, the stronger the image will be – the more emotional force and poetic reality it will have’.15 In Histoire(s), Godard carries this principle over into history, using it as a means to detect affinities between disparate materials. And by aligning the superimposition in The Wrong Man with this practice, he also reaffirms his own particular investments in miraculous disclosure – a possibility he raises in the first episode of the series, where, over the much-discussed superimpositions involving the young Elizabeth Taylor, corpses at Auschwitz, and Giotto's fresco Noli me tangere (1304–6), he says in a half-whisper, ‘O how marvelous to be able to look at what one cannot see. O sweet miracle of our blind eyes.’

Histoire(s) thus picks up a loose thread from Godard's late films, an aesthetics of the miraculous beginning with Je vous salue, Marie (1985). There the event of the virgin conception is dispersed into sounds and images of nature, gestures of daily life and work. It is manifested (if at all) as a mysterious commingling of forces: sunlight on rippling water, gusts of wind in the grass, spates of birdsong, mutters of thunder, flare-ups and fade-outs of both secular and sacred music, seasonal shifts into winter and spring, a jet soaring over bare tree limbs and powerlines, a relay of spheres linking Marie's stomach and biological rhythms to the sun and the moon. Similarly, the ‘miracle’ of Marie and Joseph's reconciliation after Joseph's initial doubt and distrust takes material form in tentative gestures of affection – his hand placed gently on her bare abdomen – and in a fleeting, off-screen remark as Joseph cites the famous last line of Bresson's Pickpocket (1959): ‘O Marie, what a strange road I had to take to reach you.’ If Godard installs the divine firmly within the profane, he also raises the problem of how to discern a miracle in the world of human affairs – and the problem of how to proceed once this transformation has taken place, how to sort out its lasting implications. Joseph's Bressonian comment, suggestive of ‘grace’ and ‘conversion’, is swiftly followed by Marie's off-screen voice inquiring, ‘Now what's wrong?’ Christ has entered the world by virgin birth; Marie and Joseph have become reconciled by a strange path. Where to go from here?16

Godard's use of video superimpositions in Histoire(s) mounts a response to these lingering difficulties. The double exposure in The Wrong Man is one of several reminders in the series of cinema's power to show us miraculous incidents: in other episodes we see Johannes clasping hands with his young niece and raising Inger from the dead in Dreyer's Ordet; Christ healing a leper (between a shot and its countershot) in Pasolini's Il Vangelo secondo Matteo (1964); Michel and Jeanne pressed affectionately to one another through prison bars in Bresson's Pickpocket; Karin overwhelmed and transformed on the volcano in Rossellini's Stromboli terra di Dio (1950). Godard invokes the category of the miracle not to make a theological argument but to reflect on the conditions by which a miracle of seeing can emerge through cinematic form. These examples serve to orient us to cinema's capacity to bring into sight – with a force of conviction – events and relations that would otherwise remain obscure, and that spring forth to challenge our usual ways of perceiving and knowing the world. By appealing to the miraculous, Godard points up three entwined features of the discoveries he makes through superimposition: their emergence is sudden and contrary to the known course and order of things; they furnish, for our blind eyes, the possible grounds for belief in cinema's singular power to reveal; and they elicit a sense of wonder (miraculum being bound conceptually to mirari, ‘to marvel at’). The scene from Ordet is such a strong figuration of the miraculous for Godard because it entails both the joining of hands (evocative of montage) and the naïve readiness to believe that a miracle can come about; Johannes, distressed by the ‘lukewarm faith’ of the adults in attendance, carries out the miracle only at the urging of his niece. Godard sounds a similar note in the Hitchcock segment when he states, over the superimposition from The Wrong Man, ‘the childhood of art’. A recurring phrase in Histoire(s) that refers variously to cinema's place among the older arts of painting, music, and literature, and to the ‘youthful’ experiments of cinema prior to the arrival of the talkies, ‘the childhood of art’ also entails a child-like intensity of perception, of seeing and making links without immediate recourse to reason. To play along when Godard brings together more or less distant fragments, we have to be willing to step outside of mature habit and to put our faith in the resulting images.

If these are requisite conditions for the bearing-forth of a miracle, they are in no sense guarantees. Nothing intrinsic to the use of montage as rapprochement can assure a miraculous result; the relation between elements might be ‘distant’ without being ‘just’, and for this reason the moment of child-like openness must be followed and balanced by critical assessment. We find Godard calling for these conditions in his earlier video essay Scénario de ‘Sauve qui peut (la vie)’ (1979). As he tests out some of the same techniques that shape and punctuate Histoire(s), he tells us in voiceover:

What I'm trying to show you is how I see things, so that you can judge whether I am able to see, and what I have seen. I want to show you the relationships between images and then you would be as in a court of law where you are both the defendant and the prosecutor … and you can see if I see something. I show if there is something to see and how I see it. And you can say, ‘No, he's wrong, there's nothing to see.’ So what I would like to show you is a way of seeing – for example, superimpositions, cross-fades, and slow motion.

Godard has often spoken of rapprochement as the ‘scales of justice’, a critical procedure for detecting whether there is balance and justness between combined terms.17 Here he is clear in suggesting that the stakes of this process, this weighing of material evidence, are more than private. Godard alone cannot affirm whether there is something to see. Hence, each act of montage bears with it an appeal to dialogue and to the critical judgement of the observer. The ‘way of seeing’ he demonstrates, with great virtuosity, is effective in so far as its discoveries can be shared. The miracle – if a revelation does occur – is of our blind eyes. As Richard Neer has recently argued, Godard's investment in ‘public aesthetics’, in the ‘sharedness’ of perception between monteur and audience, is central to the project of Histoire(s).18 To be sure, Godard's video montage is historical not only in the sense that it attempts to show us ‘just’ affinities between fragments taken from the archival record, but because it investigates the terms and procedures by which a community of spectators ‘seeing more or less the same thing’ may be constituted – because it boldly espouses and assays the ‘lived conviction that we can see together’.19

Given this stress on public reception, we can understand why Hitchcock is such a significant figure for Godard. His interests in the director have never been purely formal. What Hitchcock's cinema embodies is the possibility of formal innovation in a decidedly popular context. As Godard puts it in a recent dialogue, Hitchcock ‘achieved success by doing difficult things, which is rare’.20 The segment in Histoire(s) underscores the point that Hitchcock's method derives its force from the worldwide reach of his films and their immense public appeal. Godard, again, phrases this claim in terms of collective memory. ‘Perhaps there are ten thousand people who haven't forgotten Cézanne's apple’, he says in voiceover, ‘but there must be a billion spectators who will remember the lighter of the stranger on the train’. In this sense, Hitchcock's ‘control of the universe’, while achieved through style, is entirely dependent on cinema's status as a popular medium.

Yet there is a broader sense of the term ‘universe’ at work here. The title of the episode, Le Contrôle de l'univers, also encompasses an extended recitation of Denis de Rougemont's 1936 essay Penser avec les mains (‘to think with one's hands’) that leads into the Hitchcock segment by reflecting on questions of individual creative thought (as ‘manifested’ in actions for which the agent is responsible) and friendship in response to brutal interferences of the state (de Rougemont's main cause for alarm is the ascendance of National Socialism). During this recital we encounter two fragments from the work of Hitchcock: first, a slowed shot of a schoolgirl under attack and crying out for help in The Birds (1963), her face streaked with blood; then, a few minutes later, six shots of Manny Balestrero pacing in his jail cell, looking at his open hands, then clenching them into fists. These snippets look ahead to the section on Hitchcock's method and illuminate its stakes: more than stylistic innovation, more than popular triumph, the question is one of human community and its articulation through the resources of montage. In the de Rougemont recital, images of hands recur and highlight the work of joining and separating, but here montage takes on an ethical dimension: it figures as ‘a hand held out’, an ‘act of love for one's neighbour’, a creative thought extended into action. It harbours a power to transform, to mend the damage done to human relations by abstract ideologies and ‘laws born of the abandonment of thought’, but this power is itself violent and potentially dangerous, even to the person who exercises it. Viewed in this light, the raised hand in the recurring photo of Hitchcock suggests not merely demiurgic creation but public responsibility for what is created (the first time it occurs it is superimposed with a detail of Caravaggio's Madonna del Rosario (1606–7), a scene of outstretched, receptive hands). The section on Hitchcock and his ‘control of the universe’, obtained by montage instead of imperial violence, takes up and inflects this principle of ‘thinking with one's hands’, and thus it enables Godard to acknowledge the public stakes and responsibilities of his own montage practice.

While these stakes become more evident when ‘Hitchcock's method’ is seen in relation to the larger structure of the episode, they also become more contestable. In his essays concerning Histoire(s), Jacques Rancière teases out this communal dimension of Godard's montage and regards it with suspicion. For Rancière, this aspect comes to the fore in what follows the Hitchcock tribute, a section where Alain Cuny recites a passage from Élie Faure's Histoire de l'art, substituting ‘cinema’ for ‘Rembrandt’. Cuny speaks of elemental details that ‘plunge into light’–‘a shoulder, a face, a raised finger, an open window, a forehead, a child in a cradle’. We briefly see Cuny, his figure in deep shadow except for his chest and a sliver of his face as he continues: ‘Thought, gaze, word, action link this forehead, this eye, this mouth and hand with shapes half-hidden in the darkness, heads and bodies leaning over a birth, a sickbed or a corpse.’ Rancière characterises this section as a ‘perfect demonstration’ of Hitchcock's method for two principal reasons: it isolates images and converts them into pure, inter-expressive blocks capable of endlessly binding and unbinding anew; it describes cinema as a compendium of ‘humanity's daily gestures and archetypal poses’. The second point is critical in that it assumes an intimate link between ‘artistic forms’ and ‘shared forms of life’. For Rancière, this ‘co-belonging of forms and experience’ goes ‘by the very specific name of history’. The claim here is that Godard aligns Hitchcock's method (and by extension his own) with a certain kind of history painting of which Rembrandt is the key example and Faure the key art historian. Rancière also traces this notion of historical co-belonging to early German Romanticism, in particular Friedrich Schlegel's concept of ‘progressive universal poetry’, which both regards fragments of ancient poems as ‘metamorphic elements’ for the creation of new, modern poems, and ‘ensures that the speeches and images of art are interchangeable with the speeches and images of common experience’.21

It's the framing of collective life by a logic of pure, recombinable image-matter that Rancière finds suspicious in this poetics that he maps onto Godard's videographic style. As he sees it, the sense of co-belonging in Histoire(s) arises not from the peculiar attributes of the works swept up in the montage, but rather from the self-shaping and self-affirming rhythm of the montage itself, the figurative relays it imposes, its ‘fraternity of metaphors’ attesting to a ‘shared world where heterogeneous elements are caught up in the same essential fabric’.22 He goes as far as to call Histoire(s) a Gesamtkunstwerk, his suggestion being that it generates a sense of community through its all-embracing formal system, its capacity to weave each of its elements into an ecstatic totality.23

The problem with this description, however, is that it misconstrues the dynamics by which Godard's video montage operates. When Godard modifies the bits and pieces he takes from Hitchcock's films and establishes new links between them, he does exhibit the power of what Rancière calls ‘inter-expression’, but as we have seen, he is also out to distinguish his use of superimposition as a critical instrument and to orient us to the kind of attention and perceptual effort it demands. What prevents this manner of linkage from being subsumed into a general inter-expressive mode – in which the common measure is supplied by the rhythmic force alone – is the process of weighing the on-screen evidence and judging whether there is something to see, whether the combined elements produce a ‘just’ resonance. I don't mean to suggest that this investigative process is strictly a matter of simultaneity, or that it's at odds with the links that have a serial structure. Speaking of montage in Reverdian terms as rapprochement can give a thin impression of Histoire(s), the way relations take shape in torrential build-ups and dispersals that allow us little time to reflect – it's not as though each superimposition halts this intensity and joins only two things at once. Rather, the simultaneous combinations tend to gather up elements within a developing sequence and to compress ongoing thoughts and motifs.

To take one example, early in episode 1B, Une histoire seule, we see a three-part composite of a black-and-white photo of Vivien Leigh, a reeling strip of celluloid (irised and placed over her right cheek, appearing and vanishing twice, speeding to a coloured blur and then slowing just enough for us to make out the discrete frames: two men with drawn pistols in Rio Bravo (Hawks, 1958)), and a speed-adjusted shot of Jean Marais in Cocteau's Orphée (1950), searching with his hands around Leigh's mouth as if trying to grasp the evasive strip (fig. 2). This image springs from a growing chain of associations that Godard triggers a few moments earlier when he shows us Glauber Rocha standing at the crossroads of political cinema and aesthetic ‘adventure’ in Godard and Gorin's Vent d'est (1970), to his right a fluctuant film strip (of Ricky Nelson and Angie Dickinson in Rio Bravo) that vanishes to reveal a woman approaching. Godard intones in voiceover: ‘Sometimes at night someone whispers in my bedroom. I shut off the television but the whispering goes on. Is it the wind or my ancestors?’ This line gives rise to a ‘history of wind’, as titles tell us, citing Joris Ivens's Une histoire de vent (1988), a frame of which appears on-screen. We hear a strong current blowing. We see Lillian Gish assailed by a dust storm in The Wind (Sjöström, 1928), then Dorothy Malone tossing a stone into the river in Written on the Wind (Sirk, 1956). The titles change from ‘written on the wind’ to ‘gone with the wind’, which leads into the still of Leigh. When the photograms appear, they figure as a tenuous stream that Orphée can't quite embrace as he inspects the surface of the screen (the strip replacing Cocteau's mirror-entrance to the Zone).

So much is condensed in this short passage. We have Godard's personal histoire bound up with the histoire de vent– his cinephilic attachments to his ‘ancestors’ (Hawks, Sirk, Cocteau), his own films at radically different moments in his development. After he mentions the whisper in his bedroom at night, he says that he had the ‘lover's chauffeur’ in Cocteau's film (François Périer as Heurtebise) speak the same phrase in a film called ‘A Place on Earth’, the subtitle of Godard's Soigne ta droite (1987). But this histoire de vent is potentially our histoire as well. Godard acknowledges this by interspersing a still taken from Ingmar Bergman's Fängelse (1949) of a man and young woman (Thomas the alcoholic writer, Birgitta the prostitute) side by side behind a film projector and focused on the spectacle before them. Godard uses this still – splitting it into singles, adorning it with text (‘histoire(s) du cinéma’), layering it with a film strip in procession – to situate the couple as spectators of the fleeting images in Histoire(s), an association that extends to the ‘couple’ of Leigh and Marais (which soon fractures into single units, shown to us within an iris in the centre of the Fängelse still). Far from imposing a fraternal coherence via inter-expression, Godard confronts the difficulty of conducting his investigation with an elusive substance – images that come and go in perpetual variation – and the difficulty of finding and presenting a formal connection that will in turn bring together, perhaps for a moment only, the spectators taking in this process. In Godard's videographic adventure by strange paths, superimposition entails more than a plastic technique – it's an operation of thought that attempts to grasp an ongoing multiplicity (‘cogito ergo video’, declare the titles at the start of 1B, following a page of Beckett's L'Image). We have to practise this form of seeing, too. The justness of Godard's montage depends on it.

If Godard attunes us to this practice by removing Hitchcock's images from their narrative contexts, his point isn't simply to liberate formal details from dramatic action. The deciding impulse, rather, is to abandon Hitchcock's schematism. Recall that in his Cahiers review of The Wrong Man, Godard describes as miraculous those moments that temporarily escape or negate the ‘machine grinding inexorably on’– the rigid structures of entrapment that register not just in the film's plot but in Hitchcock's very découpage. Godard's use of the double exposure in Histoire(s) reasserts this sense of the miraculous, but the liberating force of the image – as Godard reframes it – is such that it breaks with Hitchcockian design altogether, its degree of methodic control, its a priori patterning of events and relations.24 Godard's miraculous findings through superimposition are neither guided nor assured by mechanism. He sees if there is something to see without knowing up front, risking noise and incoherence at each juncture. He proceeds in essayistic leaps, never reaching a final position: no matter how pristine or emphatic, the images he forms are tentative, incomplete, and open to still more division and synthesis. Rancière neglects the essayistic impulse in both Godard's montage and the Schlegelian poetics he names as its precursor. For Schlegel as for Godard, each link is provisional, the poet-experimenter has limited sway over the process, and the possibility of failed communication is built in. The fraternal tide can be disturbed at any moment by what Schlegel calls ‘the glimmer of error and madness’.25 The ‘seeing together’ at the heart of Godard's project is something that must be striven for at the limits of mutual comprehensibility.

The first episode of Histoire(s) begins with titles telling us that the task at hand is difficult, ‘hoc opus/hic labor est’. The first sight we encounter is of James Stewart in Hitchcock's Rear Window (1954), his eyes shifting behind his telephoto lens at a slowed and gently uneven rate. More titles assert that we'll have to ‘negotiate’ for ourselves the relations among the manifold fragments since Godard, now reciting a Bresson aphorism, will refuse to show us ‘all sides of things’ and will ‘leave a margin of indefiniteness’.26 Godard thus stresses at the very outset the active and synthesising role of the viewer. He addresses us as diviners of significance,27 our integral function embodied in Hitchcock's vigilant, imaginative Jeffries (who compulsively projects his desires and anxieties onto the happenings across the courtyard and constructs a fuller picture from gleaned details). Throughout the series, Godard entwines the work of the montagist with the work of the viewer by assuming a spectatorial position. We frequently see him in his study conjuring up ‘visions’ with the phrases he types at his desk, or recites from a book he takes off the shelf. In these moments of self-portraiture, he often raises his eyes and stares attentively, holding his glasses in place with one hand, as if inspecting the images on a screen out of frame, or in the smoke gathering from his cigar, or in his mind. This isn't to say that we see precisely what he sees, or that Histoire(s) is a record of his own connective steps that we retrace. While our charge is to weigh the material combinations that Godard exhibits, we also have to respond to the gaps, the elisions, the blind spots in the ensembles we are given. Histoire(s) burdens us with its incompletion, and no formal logic of pars pro toto leads us towards a grand unity. Godard insists in his dialogue with Youssef Ishaghpour that he uses the device of superimposition ‘Not all the time, but to remind, to show that it's there’.28 In other words, superimposition involves a constant interplay of the patent and the latent – it enacts a process of thought that is not reducible to what coalesces on the screen. Each short-lived, concrete image is shadowed by the multiplicity of possible manifestations into which it might extend or recede. Each formal kinship is conditioned by a hesitation between the moment of synthesis and the fluctuating process that renders it unstable, between the achieved and the as yet unachieved. Nowhere in Histoire(s) does Godard suppress this uncertainty. There could be no miracles without it.

  1. I wish to thank Colin MacCabe for inviting me to write this essay, and also Clint Bergeson, Adam Lowenstein, and Dan Morgan for their advice and encouragement.

  2. 1 Laura Mulvey, Death 24x a Second: Stillness and the Moving Image (London: Reaktion, 2006), 145–6.

  3. 2 Jacques Rancière, ‘Godard, Hitchcock, and the Cinematographic Image’, in For Ever Godard, ed. Michael Temple, James S. Williams, and Michael Witt (London: Black Dog, 2004), 214–31; Rancière, Film Fables, trans. Emiliano Battista (Oxford: Berg, 2006), 114–17, 171–87; Rancière, The Future of the Image, trans. Gregory Elliot (London: Verso, 2007), 30, 33–67. Rancière does not specifically discuss the Hitchcock segment in his essay on the ‘sentence-image’ (cited here last), but he draws on the same underlying aesthetic and political framework.

  4. 3 André Bazin, ‘The Evolution of the Language of Cinema’, in What is Cinema?, vol. 1, ed. and trans. Hugh Gray (Berkeley: University of California Press, 1967), 23–40. In his early Cahiers writings, Godard is often at pains to show that Hitchcock's films confound the ‘reality’ and ‘image’ trends that Bazin keeps separate: Hitchcock is an expressionist one moment, a neorealist the next. And Godard tends to use Bazin's preferred directors in comparisons that bring out the non-Bazinian aspects of Hitchcock's style. In his review of The Wrong Man, for instance, Godard writes that Hitchcock, much like Murnau and Dreyer, heightens the drama with sudden, piercing close-ups; and like the Welles of Mr Arkadin, Hitchcock offers a ‘rapid, frenzied montage’ not simply to show the event of the protagonist being fingerprinted by a police detective but to effect an impression of it ‘with terrible immediacy’. Godard (Godard on Godard, ed. and trans. Tom Milne (New York: Da Capo, 1972), 49–50, hereafter cited parenthetically by page number in the text).

  5. 4 Brian Henderson offers a valuable discussion of Godard's investments in Hitchcock's cinema in ‘Godard on Godard: Notes for a Reading’, in A Critique of Film Theory (New York: E. P. Dutton, 1980), 82–105.

  6. 5 See Alain Bergala, Godard au travail: Les années 60 (Paris: Cahiers du cinéma, 2006); and Nicole Brenez, David Faroult, Michael Temple, James S. Williams and Michael Witt (eds), Jean-Luc Godard: Documents (Paris: Centre Pompidou, 2006). I should add that the image of Hitchcock as a filmmaker who avoids improvisation at all cost is shown to be false by Bill Krohn in Hitchcock at Work (London: Phaidon, 2000).

  7. 6 Eric Rohmer and Claude Chabrol, Hitchcock: The First Forty-Four Films, trans. Stanley Hochman (New York: Frederick Ungar, 1979), 152.

  8. 7 André Bazin, The Cinema of Cruelty from Buñuel to Hitchcock, ed. François Truffaut, trans. Sabine d'Estrée (New York: Seaver, 1982), 171–80.

  9. 8 ‘Material intimacy’ is a phrase I borrow from Raymond Bellour, who has traced a lineage of critical analysis – a ‘between-the-two’ of criticism and cinema – back to the French reception of Hitchcock. See Bellour, ‘A Bit of History’, trans. Mary Quaintance, in The Analysis of Film, ed. Constance Penley (Bloomington: Indiana University Press, 2000), 1–20.

  10. 9 Godard and Anne-Marie Miéville rework the kissing scene from To Catch a Thief in a similar manner in The Old Place (1998), in their gestural exercise concerning the ‘logic of images’.

  11. 10 Let me clarify that I am not making a general argument about Hitchcock's work and the kind of viewing it invites or imposes; I am examining how Godard uses Hitchcock's images and voice in this segment to work through an abiding tension in his own practice. Certainly Hitchcock's cinema solicits multiple modes of engagement from spectators. For an incisive look into this matter, see Raymond Durgnat's A Long Hard Look at ‘Psycho’ (London: British Film Institute, 2002).

  12. 11 ‘Abii ne viderem’, which translates as ‘I turned away so as not to see’, appears in the episode as an intertitle between the image from The Wrong Man and a short stream of stills in which figures are ‘blinded’, covering their eyes, or averting their gazes (Helen Keller in Arthur Penn's The Miracle Worker (1962); the mother and child embracing in Garrel's silent, experimental Le Révélateur (1968); Frankenstein's monster hoisting the young boy in Son of Frankenstein (1939)). The suggestion here is that cinema ‘turned away’ from the form of seeing that Hitchcock's films forcefully demonstrated, a form that Godard is trying to reanimate, by very different means and with a modified purpose, in his video montage. Seeing, blindness, miraculous revelation, ‘the childhood of art’ (a concept I address later): these crucial tropes are all condensed in this transition out of the Hitchcock segment.

  13. 12 Jonathan Rosenbaum, ‘Le Vrai Coupable: Two Kinds of Criticism in Godard's Work’, Screen, 40:3 (Autumn 1999), 3212.

  14. 13 Rohmer and Chabrol, Hitchcock, 148.

  15. 14 André Malraux, The Voices of Silence, trans. Stuart Gilbert (Garden City NY: Doubleday, 1953), 272.

  16. 15 Pierre Reverdy, ‘L'Image’, Oeuvres complètes (Paris: Flammarion, 1975), 73–5.

  17. 16 That Godard looks to Pickpocket and its notoriously ambiguous last scene is highly significant. For Bresson as for Godard, the miraculous turnabout is less an act of divine grace, or the culmination of a story arc, than a ‘mystery’ at once spiritual and material, leaving in its wake a complex set of questions as to its causes and effects. For a shrewd analysis of Bresson's cinema along these lines, see Adrian Martin, ‘Estilo y significación en Robert Bresson’, in ¿Qué es el cine moderno? (Santiago: Uqbar, 2008), 57–74.

  18. 17 See the entry for ‘montage’ in Godard's written piece ‘ABCD … JLG’, in Jean-Luc Godard: Documents, 330.

  19. 18 Richard Neer, ‘Godard Counts’, Critical Inquiry, 34 (Autumn 2007), 13571.

  20. 19 Ibid., 153 (my italics).

  21. 20 Godard in discussion with Youssef Ishaghpour, Cinema: The Archaeology of Film and the Memory of a Century, trans. John Howe (Oxford: Berg, 2005), 64.

  22. 21 Rancière, Film Fables, 175–9. For a contrasting view of Godard's relation to Jena Romanticism, see Nicole Brenez, ‘Jean-Luc Godard, Witz et invention formelle (notes préparatoires sur les rapports entre critique et pouvoir symbolique)’, CiNéMAS, 15:2–3 (2005), 1543.

  23. 22 Rancière, The Future of the Image, 57–67.

  24. 23 Rancière, ‘Godard, Hitchcock, and the Cinematographic Image’, 230–31.

  25. 24 Dan Morgan has pointed out that Godard thus ends up taking a Bazinian view of Hitchcock's ‘metaphysical determinism’, as Bazin had, in his review of the American The Man Who Knew Too Much, noted Hitchcock's expression of an ‘a priori vision of the universe, a predestination of the world for certain dramatic connections’ (Morgan, ‘The Afterlife of Superimposition’, paper presented at the conference ‘Opening Bazin’, Yale University, 4–7 December 2008). For Bazin's review, see The Cinema of Cruelty, 166.

  26. 25 Friedrich Schlegel, ‘Dialogue on Poesy’, in Theory as Practice: A Critical Anthology of Early German Romantic Writings, ed. and trans. Jochen Schulte-Sasse et al. (Minneapolis: University of Minnesota Press, 1997), 186 (trans. modified).

  27. 26 Robert Bresson, Notes on the Cinematographer, trans. Jonathan Griffin (London: Quartet, 1986), 94 (French original modified by Godard).

  28. 27 ‘Accustom the public to divining the whole of which they are given only a part’, Bresson writes in a note on fragmentation (not used in Histoire(s)). ‘Make people diviners. Make them desire it’ (Bresson, Notes on the Cinematographer, 98).

  29. 28 Godard with Ishaghpour, Cinema, 34. Godard's comment is more subtle in French: ‘Pas tout le temps, mais en laissant penser, en montrant qu'elle est là…’ (Archéologie du cinéma et mémoire du siècle: Dialogue (Tours: Farrago, 2000), 27). ‘To remind’ doesn't quite capture the sense, at once passive and constructive, of bringing an image to thought, of letting it be mindful.