Several of Max Ophüls’ late films make elaborate use of narrators who, in different ways, tell the story, or at least major parts of it: Lisa (Joan Fontaine) in Letter from an Unknown Woman (1948), the meneur de jeu (Anton Walbrook) in La Ronde (1950), ‘Maupassant’ (Jean Servais) in Le Plaisir (1951) and the Ringmaster (Peter Ustinov) in Lola Montès (1955). In each, though with significantly differing effects, these processes of narration produce a crucial tension between what the narrator tells us and what we see and (in addition to the narrator’s voice) what we hear. These films all therefore challenge any easy assumption that the apparently controlling narrative voice can also be understood as controlling the film’s images. One effect is that the oblique relationship between verbal narration and other dimensions of the film produces differential perspectives on events which encourage us to adopt a critical, or at least questioning, stance to the claims made by the narrator, so that her/his attitudes, interpretations, authority over the story’s shape and meaning, themselves become part of the films’ subject matter. How we should understand these ‘doubled narratives’  has become a significant question in Ophüls criticism.
At the heart of any such enquiry are what George M. Wilson refers to as “epistemic” dimensions of the films: the ways in which knowledge about the narrative world is controlled and communicated. If we find in Ophüls’ late films strategies that point to the fallibility of the dramatised narrators, what is the status of the films’ wider frameworks, notably their images and the selection of narrative material that they present? It seems reasonable to think that, insofar as we can separate the perspectives developed by these wider frameworks from those of the narrator, the wider frameworks must be the more trustworthy. Certainly none of these films allows us to doubt the veracity of what we are shown and our access to events seems to imply a narrative authority that is not subject to obvious restriction. In discussions of the novel, such narration often used to be called ‘omniscient’, implying that at any moment of the narrative we could be shown any aspect of the fictional world, informed about any thought or feeling, transported freely through space and time, at the whim of the author. But apparent narrative freedom is not necessarily unrestricted, let alone ‘omniscient’. The significance of specific and systematic epistemic restrictions in the organisation of film narrative is the major subject of Wilson’s book. In an illuminating chapter on Letter from an Unknown Woman he points to ways in which, in a film which develops delicate and complex perspectives on the limitations of its central characters through a highly self-conscious visual narration, Ophüls nevertheless withholds judgement and ultimate explanation. “In these lovers’ lives, there are truths that they cannot at all discern, which the film reveals to the properly responsive viewer. Nevertheless, there are also mysteries about their lives which this and, probably, any narration will not dispel. It is a part of the affirmation of the possibility of seeing things more openly, broadly, and clearly to acknowledge as well the limits to what we can expect to see in such a case.” 
This article pursues the question of such limits by focusing on parallel moments in two of Ophüls’ late films: Le Plaisir and Lola Montès. At a climactic moment in The Model, the final story in the triptych of adaptations from Guy de Maupassant that make up Le Plaisir, Josephine (Simone Simon), rejected by her lover, Jean (Daniel Gélin), runs up a staircase towards a window and throws herself out. As she runs to the stairs, the camera moves with her in an extended, mobile shot. Without cutting from the previous framing of the couple the camera swings away from Josephine but moves with her, looking down at the stairs as she climbs and accompanying her movement to and out of the window until she crashes through a glass ceiling far below. In Lola Montès, as the elaborate circus representation of Lola’s life approaches her affair with King Ludwig of Bavaria, Lola (Martine Carol) is at the highest point of the circus tent, ready for the spectacular climax of her ‘act’, her jump without a safety net, from the high trapeze to a trampoline on the circus floor. As she jumps, the camera seems to take her optical point of view and we fall vertiginously with her. In neither film do we see the immediate aftermath of the fall. In both there is a fade to black and the film re-opens on a later scene: returning to the containing narrative framework in Le Plaisir, as we see Josephine in an invalid chair, being pushed along the promenade by Jean; and to Lola in a cage, accompanied by the Ringmaster (Peter Ustinov), extending her hands through the bars to be kissed by a queue of men lining up to pay for the privilege, in what turns out to be the final shot of the film.
Ophüls was one of the most self-conscious stylists in the history of cinema, his films marked by the elegant elaboration of relationships between camera movement, decor and action, and by patterns of ‘rhyme’, repetition and variation both within and between films. To appropriate a line from Letter from an Unknown Woman, in an Ophüls movie “nothing happens by chance”. These moments are intriguing partly because they ‘rhyme’, but also because in their use of what feels like ‘point of view’ (POV) filmmaking, they break with Ophüls’ characteristic methods. His films rarely use POV; in fact its absence is almost as central a feature of Ophüls’ work as its presence is in Hitchcock’s. A tendency to avoid POV filming is one aspect of the ways in which Ophüls’ films require us to look at his characters, seeing them within the elaborate material constraints of their worlds. Something was clearly at stake for him to choose for these two closely linked events methods close to POV, to abandon his pervasive observation of his characters and to cede the camera’s view to the falling woman.
What then do these parallel moments mean? What significance should we attach to the repeated act of a woman jumping from a great height and to the ways in which Ophüls chooses to represent it? And how do they relate to the dual structure of story-telling in each film – the roles of ‘Maupassant’ in Le Plaisir and of the ringmaster in Lola Montès, and the relationship of their narration to the films’ images and sounds.
Structurally and thematically, we can relate the moments to strong motifs and central concerns in both films. Movements of ascent and descent are recurrent features of Ophüls’ work, with aspiration, striving towards a goal, invariably followed by descent, deflation – or fall. This is an emphatic part of the patterning of Le Plaisir and Lola Montès, scenes in each repeatedly staged around movements of characters up and down staircases, climbing from floor to floor in buildings and then descending, the camera often following the movements in extended takes. Lola’s climb to the trapeze and Josephine’s to the window are decisive acts in these patterns of movement, and their falls simultaneously complete and break the cycle of seemingly endless movement the motif enacts.
Thematically, these moments also seem bound up with the women’s moral and existential trajectories. Lola’s rise to the eminence of becoming the mistress of a king must be followed in the circus display of her life by a precipitous fall. This may partly explain why in Ophüls’ staging actions, which are properly described as ‘jumping’, should be experienced as ‘falling’. In English there is an evocative cluster of meanings centred on ‘fall’: ‘the fall of man’ in the Genesis story, a ‘fall from grace’, ‘fallen woman’, and the literal act of falling. Both stories in which these shots occur are about the lives of women who are ‘fallen’ according to the conventional morality of their societies. This is of course familiar territory for Ophüls, who returns constantly to the dilemmas of ‘fallen women’. To quote Susan White, citing Nina Auerbach: “The melodramatic and exemplary aspect of the fallen woman’s situation is embodied in the ‘one constant element in the myth of the fallen woman, reaching back to the Old Testament and to Milton’s epic recasting of it . . . the absolute transforming power of the fall’. Such a view, consistently at work in Ophüls’s films, runs counter to the more insistently feminist demythologising of the fallen woman’s state, in which she is seen as precisely not having irreversibly ‘ruined’ herself.” Ophüls repeatedly draws on these traditional images of the ‘fallen’ woman, while invariably developing critical perspectives on the intractable ideological formations and social attitudes within which his women are held.
The parallels between the two ‘falls’ are accompanied, of course, by significant differences in dramatic context. Josephine’s legs are broken and she is reduced to life in a wheelchair – her run to the window and her jump are her last physically independent movements. Her extraordinary action is in that sense unique and definitive, freezing her life even as it brings her the marriage she desires. Lola’s jump is unique only for the film spectator; as the climax of her act it takes place daily in the constantly replayed ritual in the circus ring. It is definitive therefore only as part of a performance, repeatedly enacting for the public gaze Lola’s acceptance of the consequences of her hubris. Nevertheless, they share a sense of finality, of the active lives of these women being effectively over. In different but closely related ways the falls dramatise that closure.
To reflect further on the epistemic aspects of Ophüls’ decisions we need to return to the detail of the sequences. Although the filming of the actual falls is very similar, both feeling like point of view shots, the camera seeming to replicate the visual experience of the falling women, there is some ambiguity in the status of each, created by the different approaches Ophüls adopts. In Le Plaisir the elaborate shot which accompanies Josephine up the stairs is deliberately not POV, Ophüls maintaining the continuity of a shot which begins with a quite prolonged static framing of Josephine and Jean in their final dialogue. Josephine is not seen during her ascent, only her shadow passing in and out of frame; while having much of the force of POV, it is clear that until she jumps, the camera’s view is close to but not to be identified with hers. The movement through the window and down seems to maintain the continuity (although looked at frame by frame it probably involves a concealed cut – perhaps to a model shot) and the momentum of Josephine’s climb and jump. The implication, therefore, is that while the framing is very close to Josephine, what we see during the fall is not POV (the shot’s status as separate from Josephine has not changed). But the experience of watching is that our view is barely distinguishable from that of the falling woman.
In Lola Montès, the build up to the jump alternates medium close-ups of Lola, showing her evident distress, with shots of the circus ring far below. A cut to close up shows her raising her arms and falling forward, the camera seeming very briefly to tilt with her. There are several blurred frames before the clear framing of the fall towards the circus floor. The feeling of POV is very strong, though without the status of the shot having been unambiguously established.
In their contexts, then, the shots are immediately striking and involving but their effects are paradoxical. It is noticeable, for instance, that Ophüls does not use the conventional ABA structure of the POV figure, so that although we to some extent share the character’s visual field we do not alternate views of and views with her. In neither case does Ophüls introduce the ‘falling’ shot by means of a conventional shot of the character’s look out of frame or return to the character during or at the end of the fall. This is very clear in Le Plaisir, but even in the apparently more conventional build-up in Lola Montès the effect of the editing is very odd, as though Ophüls wants to evoke, without embracing, a familiar editing strategy. The overhead shots of the circus ring seem at first sight to be POV yet in the shots of Lola her eyes alternate between moving in panic and being actually shut. None of the three static overhead shots is unambiguously attached to her looking down, as it would be in a more conventional handling of POV. When she finally falls forward, her eyes are closed. If we take the film to be using the POV figure, the handling seems clumsy, precisely because Lola’s eye-line does not emphatically cue the POV shot, and clumsiness is not something we tend to associate with Ophüls. It is at least possible that what he is doing is precisely to refuse the familiar pattern. The avoidance of these conventional frames for the POV shot, normally used to bind the shot into a flow of action and a familiar dissection of space, makes the effect of the final shot in each case doubly unusual – almost as though the filmic effect of subjective vision were being alluded to rather than fully embraced. This is to suggest that in more than one way the use of these shots feels – and seems intended to feel – highly self-conscious.
These thoughts make it difficult to associate the shots at all straightforwardly with ideas of ‘identification’. We might be tempted by the vividness of the effect of falling to override the details of Ophüls’ treatment and to assert that we ‘identify’ with the women as they fall but Ophüls seems to want to withhold the familiar experience of closeness which the conventional POV figure often provides. His camera retains vestiges of its usual separation from its subjects while nevertheless taking on what is for Ophüls an unusually close visual association with a character’s experience. If the details of Ophüls’ decision-making allow us to make and maintain this distinction, they may also provide small but telling examples of a hesitation in the methods Ophüls uses to construct his fictional world about what can be known and what can be shared – in this case an unwillingness to pretend that the gap between character and spectator can be closed, even at those moments at which Ophüls wants the spectator’s view to be most closely associated with the women.
A less ambiguous consequence of Ophüls’ decisions is noted by V.F. Perkins in a discussion of Lola Montès: “… while [the subjective shot] involves us vividly and immediately in Lola’s plunge to earth, it also means that we do not see Lola falling.” We are denied the view of Lola’s descent enjoyed by the circus spectator. In fact in both films, while visually we share the fall with the women, one effect of Ophüls’ method is that they themselves entirely leave the visual field at these climactic moments. At their moments of greatest intensity and danger we are not allowed to see them either during or after they fall. These seem important perceptual and epistemic gaps, as though in some strange way their acts make them, momentarily, unrepresentable. This seems a significant issue in films which are so dominated by women as objects of the camera’s look and which analyse the extreme social limitations within which women are able to act.
The perceptual gap (we do not see the women fall) takes on additional resonance in films which both seem preoccupied in various other ways with what can be known, what claims about our knowledge and understanding can be sustained. To pose this more directly in terms of the telling of stories, the making of fictions: Ophüls seems to be exploring one of the ground rules of most fiction, the authority of the story-teller.
In an earlier discussion of Le Plaisir  , I argued that in La Ronde and Le Plaisir the layering of narration which foregrounds and problematises the relationship between narration and fictional world is further complicated by Ophüls’ use of adaptation – the events and characters already existing in their literary originals, their trajectories defined in advance. In these ways Ophüls detaches himself from responsibility for the stories themselves – almost as though he was quoting – while as a result of his processes of adaptation the act of storytelling, of bringing a fiction into existence, becomes part of the material the films present and work on. Although not based on a distinguished literary original, in Lola Montès the historical figure of Lola allows for a parallel distance on the film’s material, which is presented as variations on episodes from a notorious life, within an even more emphatic double narrative. One way of describing the result might be that the stories are presented, not just represented – the events, their mode of telling and even the interpretative frameworks they contain, inviting sceptical attention. The subject of the films becomes not just three stories by Maupassant, or the story of Lola Montès, but the telling of these stories.
Across the triptych that is Le Plaisir, Ophüls develops increasingly acute questions about the narrator and his understanding of events. Each of the stories begins with an event that requires explanation: the strange masked figure dancing with the professional dancers in ‘The Mask’; the mysteriously closed brothel in ‘Maison Tellier’; the invalid woman being pushed along the beach in ‘The Model’. We are then led, conventionally enough, into explanations that make up the central events of each story. But Ophüls’ invention of ‘Maupassant’ as the individualised narrator creates the basis of the film’s ‘double narrative’, providing a dimension missing in the original stories.
The narrator is amiable, even playful, but detached, anonymous but authoritative. As a piece of story-telling on the part of the narrator, ‘The Mask’ seems fairly conventional, presented to us as a moral tale about desire and age. While Ophüls’ extraordinary mise en scène cannot be identified with the relatively sparse commentary of the narrator, the voice-over seems in control of the unpeeling layers of explanation. It seems to guide us into the dance-hall, to the back-stage area to which the masked dancer is carried, then with the doctor as he takes the old man home to his tenement, where we hear from his wife the story of his life and his compulsive dancing. As the doctor prepares to return to the dance, chilled by the story but not deterred from his own pursuit of sexual pleasure, the narrator observes that he left reflecting on the “eternal drama that is acted out daily”.
There are no overt indications here that we should be sceptical of the narrator’s authority or his interpretation of events. Yet his thematic summary and detached commentary lack the extraordinary exuberance, the dramatised details of social life and individual character of the film’s images. When he reduces what we have witnessed to an example of “eternal drama” we should possibly hesitate to accept his words at face value. Have we seen no more than that? At the same time, although the story is centrally about the compulsions of male desire and is told by a male story-teller, the wife’s role as a kind of embedded narrator is crucial to our understanding and our point of view. The narrator’s view of women seems, implicitly, little different from the doctor’s or the dancer’s – they are reduced largely to objects of male desire, pursued and discarded. The wife’s narrative cuts across the dominant male voices and attitudes of the episode to give voice to another story (hers) and an alternative point of view, that are articulated by the film but which seem of little significance to the narrator.
The world of ‘Maison Tellier’ is much more complexly created and the issues of narration are correspondingly elaborated. ‘Maupassant’ again leads us from the initial mystery of the closed brothel to the explanation, claiming both knowledge of events and the ability authoritatively to interpret them. Here, he is, at times, also strikingly in sympathy with the mood of the characters, as when he evokes the atmosphere of the countryside at night and describes the congregation’s experience of transcendent spirituality during the confirmation service. In close analysis it is possible to identify a number of ways in which the narrator’s words may be qualified by the mise en scène but the issues of epistemic limits and the two levels of narration are most obviously raised by the famous visual presentation of the brothel itself. As I argued in Movie nos. 29/30, nothing in the verbal narration suggests that ‘Maupassant’ regards his knowledge of the story or its world as limited. There is no reason in his terms why the interior of the brothel should not be shown. Yet Ophüls’ camera remains insistently outside, craning around the walls and offering glimpses through windows, but never entering the building.
This strange and striking decision can only be fully investigated within the overall systems of the episode. For the purposes of this much more limited discussion, what is immediately significant is the disparity between ‘Maupassant’s’ assumed ‘omniscience’ and the emphatic refusal of the images to claim the same unlimited access to the film’s world. The film can move us through space and time, following the narrator as he tells the story; it can go into other houses, into the church, the camera can move independently of characters or with them, but it will not take us into the brothel. In marking limits to its own overall access to the fictional world that has been created on the basis of the original story it implicitly challenges the claims to knowledge that ‘Maupassant’ makes. I associate this difference between the two levels of the ‘doubled narrative’ with the film’s developing subject of men’s pursuit of women and their reduction to objects of desire. However nuanced and humane ‘Maupassant’s’ response to the events of ‘Maison Tellier’, he is shown to be complicit in these attitudes. The film’s external view of the brothel marks its hesitation in sharing this confident male view of how the world is. The emphatic sense of the women’s enclosure or entrapment that is one effect of these views is among the ways in which the film (as opposed to ‘Maupassant’) marks its recognition of the price at which the celebration that ends the story has been bought. Simultaneously, the film firmly but discretely sets limits to what it claims to know of these characters’ lives.
‘The Model’ makes these matters much more explicit. The ‘Maupassant’ figure, who remains simply a voice in the first two stories, is now embodied in the character of Jean’s cynical friend. By ‘lending my voice’ to a character in the story, in effect ‘Maupassant’ identifies his viewpoint with that of the character, while the film allows the spectator a more detached and critical relationship to the narrator. He appears both in the framing ‘present’, watching as Jean pushes Josephine’s invalid chair along the promenade and telling the couple’s story to his companion, and in the story itself. In this episode alone he has in common with Ustinov in Lola Montès the dual role of narrator and participant in the drama, the duality which brings to the foreground of each film the question of these men’s reliability as tellers of the women’s stories.
If the embodied story-teller now becomes an obvious part of the film’s subject matter and subject to its critical view, he can also be seen as paralleling Ophüls, the controlling figure behind the film. Ophüls presents Maupassant’s story being told to a listener within the film’s world. The fallibility of ‘Maupassant’ dramatises issues of selection, knowledge, ideological viewpoint, that are inherent in any act of story-telling, including Ophüls’. Part of the fascination of ‘The Model’ lies in the way Ophüls develops these reflections on narration through an extraordinary tension between complicity with ‘Maupassant’ and rejection of his viewpoint.
Distance is immediately established by the persona of the narrator – isolated, detached, sardonic. His initial commentary asserts an emphatically superior, masculine, view of the world – women are irritatingly un-knowable, their emotional honesty baffling male logic. This is a conventional, dismissive, attitude to women based on the familiar opposition between male rationality and female emotion. While acknowledging the difficulty of knowing women it emphatically registers male superiority. His telling of the story will become an illustration of this world view, a sad example of the immutable way things are between men and women – a kind of moral tale, told from a position of superior wisdom.
In turn, the film presents this telling and analyses something of the conditions which produce ‘Maupassant’s’ attitudes. The opening of the embedded narrative establishes the reflective distance from which we will view the events, the camera moving from Jean and the narrator in the Museum to follow Jean’s pursuit of Josephine up one staircase, and adjusting to frame the second staircase on which the couple reappear. The movement also firmly enacts the deflection of Jean’s gaze from the statues he was sketching to Josephine. What might initially seem like the escape from tedious academic drawing via desire into love is shown in the next sequence – Josephine posing in an elaborate studio setting – to be nothing of the sort. Josephine is defined by her position as model and Jean’s desire is inseparable from producing the woman as the subject of his painting – which is to say as an object for his gaze.
This is a perspective that the film provides visually and develops by showing us the art gallery, where Jean’s ‘new way of painting’ produces Josephine’s image as a commodity to be sold. Her human reality is reduced to a body to be sexually possessed and an appearance to be captured on canvas. The commentary offered by the narrator over the early scenes is significantly different in that it is about the folly of desire being mistaken for love. The film does not deny the narrator’s cynical reading of the inevitable parabola of male desire, but it is the film which offers the spectator an analysis of romantic love, from the captured gaze in which the woman becomes an object of desire, to idealised contemplation, to sexual possession and disenchantment. And it is the film, not the narrator, which shows the consequences for the woman of this destructive process – her apparent helplessness within a system in which she is condemned to have no power of action independent of male desire. The story-teller’s narration, becomes, in fact, part of the prevailing ideological context which the film presents for our scrutiny.
The role of the narrator within the story’s social world is crystallised in the scene of Josephine’s visit to the apartment to confront Jean. Acting as Jean’s friend, he takes Josephine aside to explain to her the ways of the world: she must not make a fuss because, in the way these things inevitably go, Jean’s family is insisting that he marries a suitable young lady. We have no reason to believe that this is anything more than a convenient fiction, designed to make Josephine accept her fate, but, whatever the validity of the story, it aligns the narrator with a view of the world as intractably ruled by bourgeois values. There is, he implies, no escape from the demands of family and respectable marriage; to accept this is simply to accept reality. He offers this homily to Josephine as he begins to play the piano. When Ophüls cuts to the outer room on Josephine’s exit, the piano playing continues and, in a remarkable decision, breaks into the music which accompanied the dance in ‘The Mask’ and the final moments of ‘Maison Tellier’ – music associated specifically with the ‘pleasure’ of the film’s title – the sexual pursuit of women by men. In ‘Maison Tellier’ the distance of the film from the festive atmosphere pervading the brothel is carried by the exterior viewpoint the camera maintains. By playing the music, the narrator, who presents himself in the containing framework of the story as detached and superior to the follies of the world, is here definitively revealed as wholly complicit with the reactionary and destructive forces governing the social world that Ophüls presents to the film’s spectator.
The film’s distance from those views is inherent in Ophüls’ treatment of the rest of the scene. Josephine’s confrontation with the totally immovable Jean leaves her with only two choices – to leave and accept the situation, or to carry out her threat to kill herself. To leave would be to accept that she is powerless and defeated. Suicide could equally be seen as an admission of defeat but here it is turned into an extraordinary assertion of independence, even as it defines the appalling limits within which she is free to act, and the dreadful consequences of such action. Up to the moment of her decision, Ophüls has framed the action in a static, observational take. In moving with her, the camera in a sense embraces her act, abandoning its observational mode. To this extent, at least, it ‘identifies’ with Josephine’s action. But in doing so it also allows Josephine to leave the frame. Her movement motivates the camera’s but the camera can no longer see her.
Lola Montès embeds the dramatised narrator at the heart of the film’s world and makes him a central participant in the drama. In a sense, the ringmaster combines aspects of the roles of Jean and the narrator in ‘The Model’ – he is in love with Lola but also tells her story, and in telling it, makes her body a spectacle to be commercially exploited. This is the most extreme version of the compromised and fallible narrator in Ophüls’ work – an extraordinarily corrosive view of the relationship between the storyteller and his material, even as it acknowledges the pathos of the ringmaster’s position.
The circus itself is realised with great density and imaginative power, its images more resonant and memorable than those of most of the ‘flashback’ sequences of Lola’s life – a thought that suggests which aspects of the film most engaged Ophüls. But its spectacle has none of the connotations of anarchic pleasure and freedom from everyday restriction that circus can readily embody. Here the film creates, in Andrew Britton’s words, “a quite extraordinary image of capitalist society in which economic, ideological and social relations are beautifully and complexly co-ordinated” , at the centre of which is the largely immobile figure of Lola, contained and displayed as the focus of this hyperbolic spectacle devoted to the idea of female transgression and scandal. Everything in the circus is eloquent of entrapment – most obviously Lola’s, but also the ringmaster’s. He may dominate and control the performance but he is an employee, a functionary of the circus, the owner of which, himself a clown, we see on a number of occasions ‘back-stage’. As a place of spectacle and storytelling, the circus is also a metaphor for cinema, and the ringmaster – in his economic as well as his authorial position – not merely a storyteller but a remarkably grim and self-critical portrayal of a film director.
The circus makes overt the constructedness of the spectacle and the story, so that telling and artifice are constantly foregrounded – its actors held within pre-determined roles in the unfolding performance. At the same time, the basis of the circus presentation of Lola is to promise the paying public access to Lola herself and to the truth about her life. Yet it is made repeatedly clear that no such immediate access and no such authenticity are available to the public. The ringmaster selects only the audience questions he wants to use; he speaks repeatedly for Lola; the audience sees only the factitious and selective reconstructions of her scandalous life.
Because these issues are so central to the circus framework and the circus seems to invite a reading as metaphor for cinema, the status of the non–circus sequences takes on particular importance. Triggered quite explicitly by Lola’s memory at the outset but offering perspectives which cannot be identified with her subjectivity (the needle and thread episode is a major example), how do they relate as representations to the circus and to the ‘reality’ of Lola’s life? This is also to ask how the authority which offers us the flashbacks is linked to and/or differentiated from the ringmaster/’director’.
In a way we might see Ophüls as teasing the film spectator almost as the ringmaster teases the circus audience, with the promise of the truth. The flashbacks appear to give us direct access to episodes in Lola’s life. But even as we seem to be offered more authentic representations than those of the circus, in a number of ways what we meet is a series of refusals to offer a clear or coherent understanding of that life. The flashbacks are selective both in terms of the episodes shown and what is seen within these episodes. Such selectivity is of course conventional and inevitable but Ophüls makes of this inevitability a significant principle – our view is limited not simply because a film can only contain so much narrative, but because not everything can be known. The way the flashbacks are shaped does not claim unlimited access to Lola or the events of her life. We are often not shown events that we would like – and might expect – to see, such as Lola’s first meeting with King Ludwig.
In a parallel way, ‘Lola Montès’ as a character is unified by the body of Martine Carol but we are given not a psychological portrait so much as a series of partial views. Each episode offers us a perspective or perspectives on ‘actual events’ – insights of various kinds which are both suggestive of what might have driven Lola and at the same time deliberately fragmentary and inconclusive. We see her acting the man’s traditional part in sexual matters as she ends her affair with Liszt. She loses her father and is distressed by her mother’s love affair on the ship; her mother attempts to marry her off but she elopes with her mother’s lover; she escapes from an abusive marriage and becomes a performer. These sequences all hint at explanatory frameworks – her childhood; the desire to be loved; rebellion; the attempt to be free of control – all of which, once articulated, seem simultaneously plausible as material factors and hopelessly banal as explanations.
If there is a thread, it may be the desire for independence. It seems significant that the one major break with chronology comes in the first flashback, initiated by the question of whether Lola remembers her past, when, in a long dissolve from her face, she remembers the end of her affair with Liszt and the active control she was then able to exercise over her life. What the film sees clearly is that her assertions of independence had the effect of transforming her unwittingly, in a logic familiar in female centred melodrama, into a sexually scandalous figure. It is significant, too, that the question about the past triggers memory but does not initiate the circus display of her life. It is sex that drives the fascination of Lola as spectacle, as the ringmaster acknowledges in making the parade of lovers an early scene in his show. But the film reduces this almost entirely to a quantitative matter, like the enumeration of Don Giovanni’s conquests in Mozart’s opera. In another refusal on the film’s part, sexual desire is constantly alluded to but is barely a presence in the scenes from Lola’s life. As in La Ronde, sex is the relay that carries the narrative forward, but Lola is less its instigator than its occasion. It is telling that when she reaches what the circus presents as the pinnacle of her sexual career – her affair with King Ludwig – their relationship is shown as companionable but not erotic. Paradoxically, in terms of how she is seen by the public in Bavaria and in the circus representation of her life, Lola is most publicly scandalous when she is privately least the femme fatale.
This is one of many perspectives which Ophüls’ construction of the flashbacks makes available to us, even as he refuses to offer a key to understanding her life. In ways that parallel what Wilson argues about Letter from an Unknown Woman, each episode develops delicate and complex perspectives on Lola’s situation through a highly self-conscious visual narration, charting for us “truths that [she] cannot at all discern”, but Ophüls nevertheless withholds judgement and ultimate explanation. The vexed question of Martine Carol’s performance is relevant here. Many writers have rightly regretted the extreme limitations of her portrayal of Lola, particularly in contrast to the wonderfully responsive and nuanced performances of, say, Joan Fontaine or Danielle Darrieux in other late Ophüls films. The limitations are striking and at times painful but Martine Carol’s relative blankness is also a factor that pushes to the foreground of the film the refusal to develop Lola as a psychologically realised character in the mould, for instance, of Lisa in Letter. Ophüls makes eloquent thematic use of the performer imposed on him.
It’s difficult, though, to come up with confident arguments about these matters. There is an almost vertiginous fascination in attempting to unpeel the layers of narration and the levels of storytelling authority, but some of the fascination comes from the sense of being held in a fictional equivalent of Chinese boxes or a hall of mirrors. In exposing processes of narration to such sustained scrutiny, the film pushes towards the very limits of the narrative conventions that it employs. Part of the force carried by the implicit parallel between the ringmaster and the director is to question the role of the male narrator and the very act of narrating. But then, what remains possible?
The circus creates a potent and emphatically critical metaphor for what happens to women in the process of representation. If Ophüls is recognising his kinship with the ringmaster, he is also recognising the inherent destructiveness of telling the woman’s story. Yet that is also what his film is doing. One response to that dilemma is to find ways of distancing the film’s view from that of the ringmaster and the circus. For instance, the refusal of the flashbacks to offer us the ‘truth’ about Lola is consistent with the revulsion we are encouraged to feel about her treatment in the circus. The film also creates views for us which are distinct from those of the circus audience, who remain distant, anonymous and visually obscured. The camera observes the show from inside the ring, not from the position of the audience. We see Lola’s distress and are given access to information about her illness: we learn that participation in the circus, her status as spectacle, is killing her. But all these accretions of distance, together with the film’s exposure and analysis of the mechanisms within which Lola is held, still leave the film representing her.
If Le Plaisir takes as one of its subjects the relationship between narrative, male voice, and the representation of women, Lola Montès attempts to push this further through its metaphor of the circus and its dramatised ringmaster/narrator. Susan White writes “One can feel ‘Ophüls’ being torn between the laying bare of cinema’s ability to show, to exhibit, and the more occulted, fetishizing powers of the camera unmediated by a ‘director-figure”. I am less inclined to put the quotation marks around Ophüls and more inclined to give that extraordinary historical individual credit for the achievement that Lola Montès represents, but otherwise my sense of the tensions in the film parallels White’s. The doubts about narrative and representation which the film embodies push it to the limits of ‘classical’ filmmaking. In the circumstances of the film’s production, Ophüls could not explicitly ‘lay bare’ the mechanisms of cinema even if he had wanted to; but it is also unlikely that in those terms he would have wanted to. His work demonstrates his devotion to the possibilities of symbolic/realist film narrative, even as, in his last films, he pushes them to breaking point.
This may be at least part of the significance of the way the women’s falls are treated in these films. In carrying out her threat to attempt suicide, Josephine challenges the forces that have attempted to define and confine her, in the only way that seems available to her. As she runs up the stairs, Ophüls lets her escape the gaze which has defined her in the film’s world, and in doing so tacitly acknowledges the complicity of his story-telling with the view of women in these worlds. In acting, the woman makes herself, for a moment, unrepresentable. As in ‘Maison Tellier’, but more dramatically, the film refuses to claim omniscience. It is like a tear in the fabric of the film – a sudden use of a different method, which acknowledges a brief but crucial shift in representation, away from looking at the woman to moving, and then looking, with her.
The context of Lola’s jump is very different. It is the climax of the show and the moment of Lola’s greatest humiliation as well as danger. She has literally to enact her ‘fall’. But here Ophüls refuses to show what the circus public have come to see. Our views of the circus have been consistently separate from the circus spectators’ but we have still looked at Lola. Here Ophüls goes further: instead of continuing to look at Lola, his camera asserts a kinship with her, a sudden sharing of her view, so that for a second she is no longer the object of the camera, defined by being looked at. The implication is surely that to show her would be to remain complicit with the ways of seeing that the circus markets. Refusing to show her is an act of discretion but more than that. It might be discrete to look away; Ophüls makes us look with Lola so that our view effectively becomes hers. Yet as in Le Plaisir the single, brief shot also embodies a refusal to imply that this kinship can be maintained and that the gap between character and camera can be closed. The situations of Lola and the women in Le Plaisir which the films analyse so rigorously, cannot be changed by falsely claiming access to their interior lives.
This is a central part of what is remarkable about Ophüls’ handling of the falls. In adopting such oblique relationships to POV conventions he emphatically refuses to indulge the film spectator or himself by a more extended association with the subjectivity of the women. Equally, this refusal points to a deep understanding of the modes of fiction Ophüls was working with. The double narratives of both films embody profound doubts about his own role as director of films about such repressive worlds. The women’s actions at these extreme moments intensely focus these doubts, the carefully qualified ‘identification’ implicitly acknowledging, in their contexts, the limits and flaws of his position as director, and yet recognising that as the director he is still held by his traditions. The brevity of these moments is not only to do with the split second it takes to fall; it also implicitly acknowledges that Ophüls cannot escape from the conventions of narrative and representation which his last films subject to such intense scrutiny.
This piece first appeared in French translation in the Ophüls issue of 1895, nos. 34/35 (2001), and subsequently in English in CineAction, no. 59 (2002). Reprinted here with kind permission of the author.
© Douglas Pye 2001
 This phrase is borrowed from Robin Wood’s article, ‘Letter from an Unknown Woman: The Double Narrative’, in Sexual Politics and Narrative Film (New York, Columbia University Press, 1998).
 See, for instance, in addition to Wood, George M. Wilson, Narration in Light (Baltimore, Johns Hopkins University Press, 1986); Susan White, The Cinema of Max Ophüls (New York, Columbia University Press, 1995); Deborah Thomas on La Ronde and Douglas Pye on Le Plaisir in Movie, nos. 29/30 (Summer 1982).
 Wilson, p. 124.
 White, pp. 5, 6.
 Writing in Movie, nos. 29/30 (1982), I misdescribed this as a subjective shot.
 Editorial Board discussion of Lola Montès in Movie, nos. 29/30, p. 111.
 Movie, nos. 29/30.
 This bears strongly on the key question of determinism in Ophüls’ films – the levels at which it is legitimate to identify the chain of events, the lives of the characters, as ‘determined’ by forces outside their control. Characters’ in the films often believe their lives to be governed in this way (Lisa’s ‘I know now that nothing happens by chance’ is one example among many), although the film’s presentation of events invites a more complex view of human agency in these matters. But in the late films determinism is a palpable fact of the narrative trajectory – these are pre-existing stories, the main structure of which is respected by the film adaptation, but which are transformed in the manner of their telling.
 Ibid p. 111.
 For parallel arguments within different analytical frameworks, see Masao Yamaguchi, ‘For an Archaeology of Lola Montès’ and Paul Willemen, ‘The Ophüls Text: a Thesis’, in Willemen (ed.), Ophüls (London: BFI, 1978). Yamaguchi suggests that ‘… any literal rendering of reality is mocked’ (p. 64); and Willemen that ‘… the woman is explicitly and directly put on show … But what the look finds is a mask, the woman as masquerade, as screen. The film’s narrative and diegesis fragment under the pressure of penetrating beyond that mask …’ (p. 71).
 White, p.300