Cléo from 5 to 7, and the Hour She Stops Being Looked At

A woman is told to come back at seven o’clock for a result. It is a little after five. What the film gives her, and gives us, is the hour and a half in between — no flashback, no account of how she arrived at this day, no scene in which a doctor sits her down and explains what is being looked for. She has ninety minutes to fill, in Paris, in June, and she is a person who has always known exactly what to do with a room.

Cléo Victoire is a singer. She has records, a songwriter, an accompanist, a manager, an assistant who runs her practical life, a lover who arrives for ten minutes and leaves again, and a face that makes people in shops look up. Nearly everything she is competent at is a version of one competence. She knows how she appears. She knows how to be received. She can walk into a café and know, without needing to look directly at anybody, how the room has rearranged itself around her.

The argument this essay wants to make about the film is that this is a real skill, learned under conditions she did not set, and that it is of no use whatsoever in the situation she is now in. Not because she is shallow. Because the skill is directional. It runs one way, outward, from her to the people watching, and it has no reverse gear. What she needs during this particular hour and a half is information travelling the other way — from the world into her — and she has spent her working life training the opposite muscle.

So the film is built as a conversion, and it converts its own form along with her. In the first half she is looked at: by shop windows, by mirrors, by admirers, by a camera that takes part in this rather than observing it and does not pretend otherwise. In the second half she looks, and the pictures change accordingly. The hinge between the two halves is the moment her competence stops working in front of us, on screen, inside a song.

What this Library takes from it is a sentence that has nothing to do with France in 1961. Attention received and attention paid are not the same substance. A person can have a lifetime of the first and no practice at the second, and the discovery tends to arrive at the worst hour available.

The facts first, because one of them is almost always reported incorrectly, including by people who love the film.

Cléo de 5 à 7 was written and directed by Agnès Varda, produced by Georges de Beauregard and Carlo Ponti, photographed by Jean Rabier, scored by Michel Legrand and led by Corinne Marchand. It premiered in Paris at the start of April 1962 and went into general French release on 11 April 1962. It runs ninety minutes. It played in competition at the 1962 Festival de Cannes; the festival’s own retrospective record lists it among that year’s competition titles and does not list it anywhere among that year’s awards. The Palme d’Or that year went to O Pagador de Promessas.

Now the error. The title says five to seven. The film does not cover five to seven. It covers five o’clock to about half past six, on 21 June 1961, the longest day of the year, and it is built as a prologue and thirteen short chapters, each headed with the name of whoever drives that stretch and the clock time it occupies. Ninety minutes of film, ninety minutes of afternoon. The widely repeated description — two hours of a woman’s life, with onscreen titles counting down two hours — is simply wrong, and it is wrong in a way that matters, because the entire formal wager of the film is that screen time and story time are the same length.

Even that has to be stated carefully. The volume on this film in the British Film Institute’s Film Classics series, written by Steven Ungar and issued by Bloomsbury, describes the film as depicting ninety minutes in near real time. Near, not exact. This essay follows that wording. The claim available is that the film is constructed to be experienced as continuous, not that a stopwatch would confirm it shot for shot.

The title is also a joke with a trap inside it. In French, a cinq à sept is not a description of an afternoon. Reference dictionaries record it as the slot in the day conventionally used for visiting a lover, and by extension as a name for the affair and for the mistress. A viewer who knows the idiom arrives expecting a particular sort of film about a particular sort of woman. What the phrase actually names here is the interval between a laboratory and its answer. The expectation is set up in order to be disappointed, and the disappointment is the first thing the film does — to the audience, not to Cléo.

It is easy, and common, to describe the first half of this film as a portrait of a vain woman. She tries on hats. She checks every reflective surface she passes. She is superstitious, petulant, indulged. She has an assistant who manages her and a lover who is fond of her the way a person is fond of a room he visits.

That reading is available, and this essay declines it — not out of charity, but because it gets the mechanism wrong.

Consider what she is actually paid for. She is a singer whose records sell on the strength of how she is received, in a market where a career of that kind had a short shelf life and rested on a look. The hats are not a character flaw. The hats are inventory. The mirror is not narcissism; it is quality control on the single asset she has been permitted to develop. She was not offered a range of competences and then chose the shallow one. She was offered this one, and she became excellent at it, which is what people do with the one thing that pays.

The distinction is not a technicality, because the two readings produce two different films. If you call it vanity, the film becomes a moral story in which a shallow woman is frightened into depth, and the frightening is her correction. If you call it competence, the film becomes something harder and more useful: a skilled woman discovers that her skill has a boundary, and that the boundary runs exactly where her life does.

And there is a specific cruelty in the timing, which the moral reading cannot see at all. The skill fails at the precise moment she most needs a skill. Not gradually, not across a career decline, not at fifty when the offers thin out. In an afternoon, without warning, on the longest day of the year.

What makes the film an argument rather than an observation is that its form takes part.

In the first half the camera behaves roughly the way the people around her behave. It frames her. It gives her a moment to arrange herself. It finds her in glass, in mirrors, in the polished surfaces of the shops she moves through, and it keeps finding her there, so that a viewer accumulates the impression that she exists most fully as a returned image. It hands her the compositions that a woman of her profession would recognise as good ones. We are not watching somebody being looked at. We are doing the looking, and being flattered for it.

Mirrors in this film have been written about at enormous length and this essay has no interest in adding to that pile for its own sake. Only one part of it is worth keeping, and the useful part is structural rather than symbolic. A mirror is not primarily an instrument of self-love. It is an instrument for occupying somebody else’s position. To check a reflection is to stand briefly where the observer stands and audit what arrives there. To do it continuously is not self-absorption. It is being absorbed, continuously, in the problem of another person’s experience of you.

Which is why the charge of narcissism, when it is made, is almost exactly backwards. The person at the mirror is not thinking about herself. She is thinking about a stranger, and she is thinking about him accurately, because inaccuracy costs her money.

The camera’s collusion is therefore not a lapse. It is the film making its audience complicit early, at low cost and with real pleasure, so that when the form changes later the audience registers the change as the removal of something it had been enjoying. An argument you have been made to participate in is harder to step out of than one you have been told.

The hinge is a song, and the way it is staged is the most precise thing in the film.

Her songwriter and her accompanist come to the apartment with new material. She turns down the light, flirtatious numbers and settles on a serious one, given the title Sans toi on the film’s soundtrack recording, with words by Varda and music by Michel Legrand, who also appears on screen as the pianist. She begins to sing it in her own flat, in an ordinary realist scene, with furniture and daylight and two men present.

Then the film does something it has not done before and will not do again. As she sings, the camera circles her and then settles, so that she is singing directly out at us; a black backdrop rises behind her and the apartment ceases to exist; and a string orchestra that is not in the room, and is visibly not in the room, comes up underneath her voice. Inside the space of one verse she is lifted out of a place and set down inside a presentation. She becomes, completely, an image.

This is the purest available version of the thing she is good at, and the film gives it to her at full strength and full beauty. And then she breaks. She stops, she is crying, she turns on the two men and tells them the words are horrible, and she asks what she is for. She pulls off her wig, puts on a plain black dress, and goes out into the street.

It is worth being exact about why the song destroys her, because the obvious answer is not quite the right one. It is not simply that the lyric is about loss and she is frightened of dying, although both are true. It is that the song requires her to perform abandonment as material while she is in the middle of it, in front of two men who wrote the abandonment for her and who are pleased with how it is going. Her competence has just been used, in front of her, to convert her situation into a product. And at the instant she is most completely an image, she is most completely alone, because an image cannot be met. There is nobody on the other side of it.

Varda, asked about the film decades later, put the change in blunt terms: at the start of the film Cléo is there to be looked at, and it is when she takes off the wig and puts on the black dress and goes out that she becomes the one who looks. The remark comes from an interview published in cléo journal in April 2018. It is offered here as what the director said about her own film, which is a different thing from what makes a reading correct, and the counter-reading below returns to exactly that difference.

A great deal of writing about this film treats the removal of the wig as a liberation, and it is one. It is also a demotion, and the film is honest enough to show both in the same twenty minutes.

Out in the street without the apparatus she is not more free in any simple sense. She is less legible. People do not rearrange themselves around her. The compositions stop being flattering. She goes into a café, puts her own record on the machine, and the room does not look up. That scene is usually described as a small humiliation, and it is, but the useful part is what it exposes about the arrangement she has been living inside: the attention was never hers. It was a service the environment performed, under conditions, and the conditions had been met by her work. Remove the work and the service stops, immediately, with no notice period.

This is the point at which the film stops being about a singer. Anybody whose standing in a room has been maintained by continuous effort — by being pleasant, by being composed, by being pretty, by being useful, by being easy to be around — will recognise the particular vertigo of the hour in which the effort is suspended and nothing arrives to take its place.

And she is now undefended in a way she was not ninety minutes earlier, which is not the same as saying she has lost something valuable. The thing she has put down was doing structural work. Being looked at had been organising her day, her sense of her own standing, and her method for finding out whether she was all right. With it set aside, the only remaining way to find out whether she is all right is to ask the world directly and listen to the answer, and at that she has no practice at all.

What she does in the second half is not glamorous and is easy to undersell. She walks. She listens to a friend talk about that friend’s own work and does not steer the conversation back. She sits in a park. She overhears strangers. She watches a street performer do something unpleasant with live frogs and does not look away. She registers that the city has been carrying on without reference to her, which is a thing she has always known and has possibly never once felt.

The film changes shape to match her. The reflections thin out. The camera stops arranging her and starts going where she goes, at the speed she goes. Long stretches are given to what she is looking at rather than to how she looks while looking at it. Radio bulletins about the war in Algeria, which had been furniture in the first half, become audible as somebody else’s catastrophe running underneath her own.

This is the substance of the claim in the title. Attention received is a condition; attention paid is an act. They are made of different material and they train different capacities. Received attention tells you how you are landing. Paid attention tells you what is there. It is entirely possible to be world-class at generating the first and to have, at twenty to six on the worst day of your life, no equipment whatever for the second.

And the reason this is not a moral failing is that the first is what she was rewarded for. Every incentive available to her — professional, romantic, social — paid out on the received side. Nobody pays for the other one. Nobody notices when it is missing. Its absence is invisible right up until something happens that only it can address, at which point the absence is total and there is no time left to fix it.

This is not new ground here, and the honest thing is to say how much of it was already laid before this essay arrived.

Days ago this Library published an essay at the slug restraint-that-reads-as-acting, which took an award record from Cannes in 2000 and argued that when a woman does difficult work by appearing to do nothing, the description that arrives first is composure — a quality of her, or of the image she is standing in — rather than an act she is performing. That essay is about the reading. It is about what happens in other people when a woman holds herself together well, and about how work that is not perceived as work cannot be credited, cannot be priced and cannot be refused. The present essay is deliberately the other half of the same argument, and it runs from the inside out. The earlier piece asks why the work goes uncredited. This one asks what becomes of a person for whom that uncredited work has turned into the only fluency she has.

The essay published as looking-sexy-and-feeling-sensual established the mechanism underneath both of them: attention has a single location, so modelling how one appears and inhabiting what one feels trade against each other one-for-one, and a woman can do the first beautifully for an entire evening and come away with no information about the second. That essay described the trade inside one evening. What this film supplies is the same trade run for a career, until the outward position is no longer a choice made on a given night but the only room in the house she has furnished.

The essay at being-seen-not-available separated attention that treats a woman as a specific person from attention that treats her as an interchangeable instance of a category, and located the whole difference in authorship. The essay on Portrait of a Lady on Fire, published as portrait-gaze-without-possession, drew the line in the right place: being looked at is not passive and is not intrinsically diminishing, and what diminishes is being looked at inside an arrangement in which you cannot look back, cannot comment, and have no standing to correct the account being formed of you. Both of those describe an arrangement from outside it. Neither asks what a long enough residence inside such an arrangement does to the resident.

That is the only new thing here, and it is enough for one essay.

Now the case against all of this, made as well as it can be made, because there are at least four ways this essay could be wrong and two of them are serious.

First, and structurally the strongest. The near-continuous time scheme may be a formal problem the director set herself rather than a psychological argument she was making, and this essay may be converting a constraint into a meaning. A director who decides to run story time against screen time for ninety minutes has to fill those minutes with movement; a person who is waiting must therefore be sent out of doors; once she is out of doors she will encounter things; and the encounters will accumulate into something that looks from the outside like change. On that account the second half looks like an awakening for the simple reason that a walk generates more material than a wait. It is a complete alternative explanation, it requires nothing of anybody’s inner life, and an essay that passed over it would be dishonest.

Second, and this one cannot be waved off. The turn in the second half is completed in the company of a man. Antoine is a soldier on leave, due to return to the war in Algeria; he talks to her, he is kind, he looks at her without appearing to want anything from the looking, and it is after him that she says her fear has gone. A reading that celebrates her arrival at her own subjecthood has to account for the fact that the film hands her the last step through a man’s attention. Critics have made exactly this objection for decades, and its strongest form is not that the scene is badly done but that it undercuts the thesis from underneath: if what she needed turned out to be a better quality of being looked at, then attention received is not hollow at all, and this essay has argued its way out of its own conclusion.

The narrowest answer available is that the exchange is two-way — she asks him things, she attends to his situation, which is objectively worse than hers, and the film gives her the asking rather than only the receiving — and that a first act of paying attention arriving inside a conversation is not the same as being rescued by the other party to it. But that is a narrowing, not a refutation. The film does give her a man at the end. A reader who finds the ending sentimental, or who thinks the film loses its nerve in the last chapter, is not making a mistake, and this essay is not going to pretend to have dissolved the objection.

Third: the director’s own account of the film does not straightforwardly support this reading. Varda described the work in terms of fear — of death, of the city, of losing oneself alone in it — and as a portrait of a woman laid over a documentary about Paris, and spoke of setting objective time against the subjective time her character experiences. Those are not descriptions of an argument about the gaze. The crisp statement about being looked at and then starting to look comes from very late in her life, more than fifty years after the film was made. Directors are under no obligation to be consistent, and a late self-description is not evidence of an early intention; it may be an artist accurately reading her own film with the benefit of everything that was written about it in between. This essay is a reading of what the film does, not a claim about what its maker meant, and where the two are set side by side that is what is being shown rather than what is being proved.

Fourth, aimed inward, and the sharpest of the four. A business that sells attention has an obvious commercial interest in an essay claiming that attention received is not enough — because the next sentence, the one this essay does not write and does not need to write, is that what is needed is better attention, available here, by appointment. The argument is convenient for us in a way that should make a reader slow down. Convenience is not disproof. But a reader who suspects that the conclusion was already in the building before the film was screened is entitled to the suspicion, and there is no reply to it that does not also function as an advertisement, which is why the reply is not attempted.

This house sells private companionship: bounded, scheduled, paid for. Before anything else in this section, the part that runs against us.

A large portion of what is bought here is the experience of being attended to. That is the received side of the transaction. It is the side we are good at, the side we can schedule, and the side we are paid for. If the argument above is right, then the thing we sell is the less durable of the two goods, and the more valuable one — the capacity to pay attention rather than to receive it — is precisely the one that cannot be purchased, because purchasing it would turn it into something else.

There is a second problem and it is worse than the first. Part of what is sold here is a woman’s skill at being received: reading a room, arranging herself, managing a mood, being easy to be with, being pleasant to look at, being received well by somebody who is paying to receive her well. That is Cléo’s competence, named exactly. We recruit for it. We train it. We photograph it. We bill for it. An essay from this address arguing that such a competence can leave a person undefended is describing an occupational condition of the people who work here, written by the party that benefits from the condition.

Only two positions are honestly available and neither is a claim about outcomes. The first is to say this in the body of an essay, where it costs something, rather than in a line at the bottom where it costs nothing. The second is to decline the flattering account of our own product: an evening here is an arrangement in which attention is directed at somebody for a fixed period in exchange for money. It is not a relationship, it is not a remedy, and it is not evidence about anybody’s life.

We will not get either of those reliably right. They are stated intentions rather than achievements, and a reader who concludes that this section is a way of collecting credit for having a conflict of interest is reading it accurately enough for the purpose.

This essay is cultural analysis. It is not clinical, diagnostic or therapeutic. It describes no treatment, promises no outcome and diagnoses nobody. Nothing in it is a statement about any reader, and a reader who recognises nothing of it in her own life is not being described.

It does not claim that the film is exactly real time. The volume in the British Film Institute’s Film Classics series describes it as near real time and this essay follows that wording. No chapter-by-chapter clock table is asserted, because none could be verified from a reliable reference; what is asserted is the film’s overall span, its runtime and the fact that it is built as a prologue and thirteen chapters.

It does not claim that the film covers two hours. It covers approximately five o’clock to half past six. Descriptions to the contrary circulate widely and are corrected here rather than repeated.

It makes no claim about what Agnès Varda intended in 1961 or 1962. Her remark about being looked at and then looking is reported as a remark from an interview published in 2018, and her descriptions of the film in terms of fear, of Paris and of objective against subjective time are reported as her public statements. Several quotations attributed to her from around the time of production could not be traced to a source that could be read directly, and none of them is asserted or quoted here.

It does not claim that any award should have gone differently, or that the film was wronged at Cannes in 1962. The festival’s own retrospective record lists the film in competition that year and does not list it among the awards; nothing is inferred from that beyond the fact itself. No other award or prize for the film is asserted, because none could be verified.

It does not present any character’s inner life as established fact. Cléo is a fictional character and what is said here about what she feels, needs or has been trained in is a reading of a film, not a report about a person. No performer’s experience, opinion or inner life is presented as established fact, and nothing here is a statement about Corinne Marchand.

The distinction between attention received and attention paid is this Library’s formulation and an argument rather than a measured finding. No statistic, proportion or prevalence is claimed for it anywhere, it describes no individual, and it is not offered as an account of how anybody must live.

It does not claim that beauty-work is futile, that appearing well is a mistake, or that a woman who maintains her appearance is thereby neglecting something. The essay argues the opposite: that it is real competence under real conditions. What it argues has limits is the direction of the competence, not its worth.

Nothing here should be read as an endorsement of, or by, any named director, performer, composer, publisher, festival, journal or publication. None has any connection to this house and none has been consulted. The plot of the film is given in outline, no song lyric is reproduced, and no extended passage of any copyrighted text is quoted.

This house’s commercial interest in the subject is stated in the body of the essay rather than only here, because a declaration that arrives after the argument has finished working is not a declaration. Stating it does not neutralise it.