This the first of four undergraduate lectures in which I explore how the thought of Michel Serres can inform film studies. I embarked upon the lectures as a speculative experiment, but in writing them I became convinced that there are rich resources in Serres’s thought for generating novel and engaging readings of films that often depart from critical orthodoxy in productive ways, just as Serres’s thought does itself. I post the lectures here in their original form, reflecting the conventions and register of spoken delivery. My hope is that they can be a stimulation to scholars of film studies and scholars of Serres’s thought alike.
Introduction
The journey I invite you to join me on today is, I am pretty sure, a world first. That doesn’t necessarily mean it will be great, but it does mean that you will be living through something that no-one outside this room will have experienced, thinking thoughts that no students have ever thought before.
What we are going to attempt together over the next four weeks is to read a series of films in conversation with the thought of a French philosopher about whom I have been thinking a lot recently: Michel Serres. Each week I want to take one key aspect of a film and explore with you how Michel Serres can help us to read it differently, see it with fresh eyes.
It is an experiment. No-one in the world has discussed these films in these terms before, and no-one has used Michel Serres as a film theorist to the extent we are going to do so. You won’t find these ideas in any books, because they are fresh and new.
You might find that exciting. I do. New work is often restricted to research and publication in journals, but in these lectures I want to bring research and teaching together, and think together with you about some fresh ideas straight out of the oven. You can disagree with my readings of the films if you like. In fact, I’d rather like it if some of you do. My aim is to invite Michel Serres to think about these films in new ways, which will not always be his way.
Michel Serres was born in 1930 in Agen, in South-West France. He spent time in the French Navy before resigning because of his antipathy to the nuclear bombs dropped on Hiroshima and Nagasaki at the end of World War II. He went to the elite École Normale Supérieure in Paris, where he was friends with Jacques Derrida. He taught at universities with Michel Foucault and Gilles Deleuze. He has written over forty books, co-hosted a weekly radio show from 2004 to 2018, voiced two TV documentary series and produced one of them. He has taught at the Sorbonne in Paris and Stanford in the USA, and since 1990 he is a member of France’s Elite Académie Française, the forty “immortals” thought to represent the greatest living writers, philosophers, politicians, diplomats, scientists and clergy.
Cléo de 5 à 7
Cléo, like many New Wave films, is partly shot on the streets of Paris, at a time when films were customarily made in studio sets. As Valerie Orpen suggests, one effect of this is that New Wave films “positively revelled in the beauty and energy of Paris, and the joys of strolling, or better still, driving, in Paris”.[1]
Cléo was shot in 1962, and it tells the story of two hours, between 5 and 7 in the evening on 21 June 1961, in the life of its protagonist Florence ‘Cléo’ Victoire, played by Corrine Marchand. You have seen the film, so I won’t repeat the plot here.
Varda called the film a “subjective documentary”. The “subjective” elements include our journey with Cléo through the turbulent hours after her tarot reading and before the confirmation of her diagnosis, and the way in which aspects of the mise-en-scène reflect Cléo’s subjective mood. It is a “documentary”, however in that it shoots real people—not actors—on the real streets of Paris, filmed at the time of day when events are taking place within the film.
Space and time
In this lecture, I want to focus on how the film deals with questions of space and time.
At first blush, time and space in the film looks to work in a pretty straightforward way. Space looks very geometrical, and time looks very linear. The scenes take us from 5 to 6.30pm[2] in durations that are scrupulously chronicled by the surtitles at the beginning of each of the film’s 13 chapters.
In addition, the film contains an unusually large number of shots including clocks an timepieces, further reinforcing the importance of the passing of time.
So, ostensibly, the duration of projection is the same as the duration of the events in the film.
Actually from 5 to 6.30, and film follows it scrupulously. In fact, the film was also very unusually shot in the chronological order of the script, and at the time of day or night that corresponded to the corresponding time in the film’s plot.
There is, however, one seeming anomaly at the end of Chapter 7, as Cléo approaches the Dôme: a street clock reads 5.50pm, but the surtitle for chapter 8 announces ‘17.45-17.52’. It is possible that the discrepancy is a continuity error, and equally that the street clock happened to be a few minutes fast. We have no way of knowing which is the correct explanation.
The spatiality of the film is similarly straightforwardly represented. Cléo follows a journey through Paris that you can plot on a map, and indeed if you buy the screenplay of this film it includes a handy map of Cléo’s journey.
Steven Ungar has even suggested noticed that there seems to be a link between the arrondissements (districts) of Paris through which Cléo passes, and the film’s chapter: “When the prologue sequence is added to the thirteen chapters that follow it, the sum of fourteen temporal units matches the numerical limits of the urban districts (arrondissements one through fourteen) in which the film takes place.”[3]
So it is usual for critics to highlight the linear nature of the film’s diegesis, and apart from pointing out the odd isolated anomaly here or there, to leave it at that:
Cléo d e 5 à 7 has the ultimate linear structure with remarkably few instances of spatial and temporal manipulation. There is spatial verisimilitude; we know where we are at all times, there is no discontinuity between one location and another, we can follow Cléo’s itinerary on a map (in fact published with the screenplay).[4]
Wow, that’s interesting! What’s going on there? What do critics make of these exceptions? Usually, very little. They list them, chalk them off as exceptions to the film’s continuity, and move on. It’s like taking the bag of exceptions and throwing it in the garbage bin, and it means that the reading of the film that these critics espouse it a reading of most of the film, not all of it. One of the principles of reading a text well, however—and the twentieth century literary theorist Roland Barthes was very big on this in a lovely essay of his called ‘The reality effect’—is that your reading should account for the whole of the text, not just most of it. So to the extent that existing readings don’t do a very good job—in my opinion—of accounting for all of this film, there is some work for us to do…
This is where I want to begin to pick away at this critical consensus. Saying that there are “remarkably few instances of spatial and temporal manipulation” is a little like saying that there are remarkably few moments in a game of soccer during which a goal is being scored. If something is rare, it does not for that reason follow that it is uninteresting or insignificant.
The problem with the “linear continuity” reading of Cléo is that it leaves you with a bag full of exceptions: “It is an exceptionally linear film, but… but… but…”. So Valerie Orpen for example writes that:
there are a few other manipulations of time: when Cléo changes behind the black curtain in Chapter 7; the black screen lasts just two seconds when Cleo would surely have taken a little longer to don her black dress and shoes. And when Dorothée and Cléo are in the car after picking up Raoul’s film from Montparnasse station, the film is speeded up a little.[5]
Critics repeatedly put this down to the difference between objective and subjective time –a distinction made by Varda herself and recycled many times since in relation to the film. But this difference only accounts for some of the anomalies. Here is Varda’s explanation of the difference:
Cléo de 5 à 7 c’est la contradiction basique entre le temps objectif et subjectif: 5 heures à 5 heures 10, à 5 heures 15 ça c’est le métronome au-dessus du piano tac, tac, tac. Or, il y a une femme qui vit ses questions, ses émotions et le temps devient une perception molle qu’on ne peut pas contrôler. Elle est comme un violon—certains moments sont très vastes (on a tout le temps), et, quelquefois, ça va très très vite et il y a ça dans Cléo: le métronome et le violon.[6]
It is a beautiful image: the regular metronome beating objective or cosmic time, and the lyrical violin evoking subjective time which can drag or fly past in the blink of an eye. It is beautiful, but does it do justice to the film itself. You might think it is odd to ask this question if the image is Varda’s herself, but that is the fascination of meaning and cultural production: once the film has been made the director’s opinion does not dictate what everyone is bound to think about it. If it did, we would think that all films are wonderful!
The problem with the metronome and violin image is that both the metronome and the violin resolve to linear time: a musical score is just as linear as the ticking of a clock, but despite what most commentators say there is more going on in the film than linear progression, even if we allow that progression to be squeezed and stretched by subjective experience.
If you hold to the linear continuity thesis you end up with awkward and deflating statements like the following from Steven Ungar: “Breaks in shot continuity and in linear chronology also support the equivalence of screen time and narrative duration as approximate rather than strict.”[7] This comes over as saying something like “Varda tried really hard, but couldn’t quite pull off the equivalence she aimed for.” It is like the person who describes an oval as a failed circle. Let it be what it is, and begin your understanding with what it is; don’t assume it should be something it is not and then point out the ways it’s not quite what it is supposed to be! Don’t say “it’s a wonderfully linear film… but it’s not really linear”. Critics have a hard time telling a coherent story about time in this film.
In this “linear” reading time is imagined like a concertina that can be stretched and contracted, but I want to argue that Varda’s own manipulation of time and space in the film is subtler than this metaphor allows. Time towards death and truth can’t be mapped onto the metronome/violin duality. In the rest of the lecture I want to develop these intuitions further and propose a reading that accounts both for the linearity and the exceptions.
Michel Serres on space and time: topology
When he writes about space and time, Michel Serres sets his own understanding against what for most of Western modernity has been the dominant view. Space is “Cartesian” (after the French philosopher René Descartes). “Cartesian space”—or sometimes “metrical geometry”—refers to the way of mapping space in terms of coordinates on two or three axes. Every point in space has a coordinate, and distances between any two points can easily be measured. The corresponding notion of time is that time is linear, like an arrow travelling at a constant rate in a constant direction. Cartesian space and linear time are the default settings of the modern West.
But there’s a problem. And the problem is that we don’t actually experience space or time this way, as Serres points out:
Ce qui est dit toujours du temps et de l’espace l’est constamment au singulier. Or, que savons-nous, aujourd’hui, de l’espace ? Rien, en toute rigueur. L’espace comme tel, unique et global, est, je le crains, un artefact philosophique. Et, de nouveau, que savons-nous du temps ? Rien, en toute rigueur. Le temps, comme tel, unique et universel, est lui aussi, un artefact.[8]
So how should we think about space and time then, Michel? Think of space like a handkerchief:
If you take a handkerchief and spread it out in order to iron it, you can see in it certain fixed distances and proximities. If you sketch a circle in one area, you can mark out nearby points and measure far-off distances. Then take the same handkerchief and crumple it, by putting it in your pocket. Two distant points suddenly are close, even superimposed. If, further, you tear it in certain places, two points that were close can become very distant. The science of nearness and rifts is called topology, while the science of stable and well-defined distances is called metrical geometry…. As we experience time—as much in our inner senses as externally in nature, as much as les temps of history as les temps of weather—it resembles this crumpled version much more than the flat, overly simplified one.[9]
The “topology” Serres refers to here is a branch of mathematics that deals in transformations, rather than in fixed forms. So as the critic Steven Connor describes it in an essay on Serres, “Thus a triangle is topologically equivalent to a circle, a cube is topologically equivalent to a sphere and, less intuitively, perhaps, a doughnut is equivalent to a teacup and a two-holed doughnut to a teapot.”[10] In order to get from a triangle to a circle, or from a two=-holed doughnut to a teapot, you don’t have to fill in any holes or create any new ones; you have stretched and squeezed, but you haven’t torn or separated:
Topology is very good at finding similarities between things that at first sight may look very different from each other, which is also a characteristic move of Serres’s own thought.
And what about time? To start with, its flow is much more complex than the linear model would suggest:
Time does not flow according to a line […] nor according to a plan but, rather, according to an extraordinary complex mixture, as though it reflected stopping points, ruptures, deep wells, chimneys of thunderous acceleration, rendings, gaps-all sown at random, at least in a visible disorder. Thus, the development of history truly resembles what chaos theory describes. Once you understand this, it’s not hard to accept the fact that time doesn’t always develop according to a line and thus things that are very close can exist in culture, but the line makes them appear very distant from one another. Or, on the other hand, that there are things that seem very close that, in fact, are very distant from one another.[11]
Serres urges us to think of time as dough that is kneaded and re-kneaded:
Le temps entre dans la pâte, prisonnier de ses plis, ombre de ses rabattements, il pénètre la théorie des ensembles, car les partitions se recouvrent les unes les autres, le temps peut se marquer au feuilletage de la pâte, évaluez l’immense temps de nos tissus, […] il plonge dans le mélange.[12]
He refers his reader to what in topology is called the “baker’s transformation”, so named because the mathematical operation it described is similar to what a baker does when dough is kneaded. As in the figure below, one first flattens the unit square making it twice as wide and half as high. This is then split into left and right halves and the right half is placed on top of the left half. Repeating the operation many times “mixes” the unit square into itself.
Randolph Nelson, Probability, Stochastic Processes, and Queueing Theory: The Mathematics of Computer Performance Modelling (Dordrecht: Springer Science & Business Media, 1995) 519.
As Serres describes it,
Or un point dans cette pâte n’est qu’un point. Le point le plus voisin du précédent suit un trajet tout à fait différent du premier, tout aussi fou, tout aussi compliqué, illisible. Or le temps n’est pas le trajet d’un seul point, mais au moins de plusieurs, et au plus de tous.[13]
Time doesn’t flow, it percolates:
it [time] passes, and also it doesn’t pass. We must bring the word pass closer to passoir—“sieve.” Time doesn’t flow; it percolates. This means precisely that it passes and doesn’t pass. I’m very fond of the theory of percolation, which tells us things that are evident, concrete, decisive, and new about space and time. In Latin the verb colare, the origin of the French verb couler, “to flow,” means precisely “to filter.” In a filter one flux passes through, while another does not.[14]
One consequence of this, as Bruno Latour points out in conversation with Serres, is that there are no temporal distances; everything can be contemporaneous with everything else.
For Serres there are in fact three times folded together:
- Clock time, where one second follows another always at the same rate.
- The globally entropic time of thermodynamics: entropy in the universe as a whole is always increasing, never decreasing. Everything is wearing out and growing old.
- Local negentropic time: information and complexity increase, for example in evolution.
These three temporalities are not merely abstract or theoretical; all three of them are lived:
Nos organismes vivants connaissent, eux aussi, la synchronie de plusieurs temps : newtoniens, ils se lèvent et se couchent avec le soleil, portent en eux des horloges qui s’affolent lors de parcours rapides traversant les méridiens, meurent, épuisés, usés, couverts de rides, suivant le second principe de la thermodynamique, mais, imprévisibles, bergsoniens ou darwiniens, parfois se reproduisent en petites filles améliorées.[15]
This positions Serres’s notion of time very differently to the subjective/objective dichotomy that governs the discussion of Cléo in a number of ways. For instance, it refuses to locate clock time in an abstract depersonalised domain but inscribes it in the body, and it recognises that “subjectivity” artificially divides mind from body and it refuses to consider human experience as artificially disembodied. Furthermore, each of these three temporalities has subjective and objective aspects, so we cannot think of time in the film (or outside it, for that matter), as a simple division between the subjective and the objective.
In the rest of the lecture I want to show how Serres’s account of topological space and time help us better understand Cléo as a whole (rather than saying it is linear… but …but …but), and how it might also be a helpful tool in understanding and talking about the medium of film as such.
Reading Cléo, and reading film, through Serres’s topological space and time
A Serresian account of time in Cléo would begin by insisting that there isn’t just one time in the film, and neither are there only two. There is the time of the disease and of death: a slow, inevitable entropic decay; there is the negentropic time of making new relationships: Cléo’s relationship with Antoine; and there is clock time. It would go on to argue that saying that time in the film boils down to the difference between subjective and objective time (between clocks and waiting for result of the test) does not do justice to complex temporalities within this film, including the prologue, the songs, and the short film featuring Anna Karina and Jean-Luc Godard en abyme within Cléo.
Take the prologue, for example. It folds a cosmic temporality (Tarot) into the “clock time” of the film (the time of the prologue is counted in the 5-6.30pm span; chapter 1 starts at 17.05). It also fragments time and space with jump cuts and mirror reflections. Time and space are being pulled and kneaded from the first five minutes of the film.
Or consider the surtitles. It is not enough, not nearly enough, merely to state that they are precise. This would be like praising the precision of Robbe-Grillet’s description of a tomato in Les Gommes, and leaving it at that. The surtitles are absurd in their precision (Cléo de 17h05 à 17h08, Angèle de 17h08 à 17h13), as if Cartesian precision had a truth and a revelation of its own. Furthermore, by announcing durations of time they also fold the future into the present, almost like a prophecy: we know how long each scene will last as it begins, how long it will take the drama to play out, with a prescient clairvoyance of our own to rival the Tarot reader of the prologue. Cléo’s future is inscribed on the film’s celluloid just as certainly as it is inscribed in her diseased body.
Thirdly, consider the film within the film. It was shot at 16 frames per second but played at 24, squashing time in a technologically assisted way that cannot be chalked off as “subjective”. It also contains a homage to Lumière brothers’ L’arroseur arrosé (The Hoser Hosed), folding into itself another time and another space, making two distant things contiguous in a way that Michel Serres has made his own in folding together Lucretius or thermodynamics, or Zola and the steam engine. The film within the film also folds the time and space of Cléo into itself in a mise en abyme not only of the film but of the film’s self-understanding and self-aware manipulation of time. It is both a metonym and a metaphor of the film’s temporality.
Then consider the songs. Un cri d’amour (37 minutes into film) folds an abstract space of performance into the film (an extradiegetic full orchestra is heard swelling from a single piano as the shot of Cléo singing is framed against a black background, abstracting her from the determinate setting of her boudoir at the same moment that she stops reading the lyrics and sings straight to camera. This is a moment of transfiguration. Art transposes her subjectivity to a different space and a different time, not located in terms of the same physical coordinates (i.e. “over there” or “in a minute’s time”), but a time and a space that cuts across Cartesian norms.
Consider the jump cuts. Phil Prowie notes that “Varda’s directorial style is characterized by an obvious penchant for linking elements in a scene via jump cuts”,[16] and we can understand these in a Serresian way too, folding absence and loss into the fabric of the film. A jump cut is a way to make an absence present, to create a distance between the contiguous in the opposite way to how folding and kneading make the contiguous distant.
Consider, once more, the preponderance of fabrics, clothing and mirrors in the film, and their repeated foregrounding. Time and space in the film are fabrics too: folding, pleating, stretching, covering… Mirrors similarly fold and duplicate (think of the hat shop again, or the prologue). Critics acknowledge that clothing is important in the film (see for e.g. Nelson, 738) but they can’t join the dots to the broader themes of time/space. Serres’s topology allows us to offer a coherent reading of time, space, fabric and mirrors. This reading also embraces the theme of the gaze, with the eye acting like a mirror (Cléo looking at people looking at her). In short, Serres allows us to see the film as a whole, and make sense of it as a series of topologically related elements which increase and decrease in distance from each other over time.
Thinking beyond Cléo to film as a medium for a moment, we might consider that the time and space of film is itself topological: time is kneaded into film, caught in the folds between shots and folded into the seemingly straightforward linearity of its running time with flashbacks, parallel editing or homage. Similarly, through editing any space or place can be “next to” any other, just like in topology. Film is a way of folding, squashing and stretching space.
If I were to walk across this lecture theatre all the way to the back, it would necessarily take me a certain amount of time. Every inch of ground has to be covered. I can’t skip the second and third rows of seats or instantaneously vanish here and appear there. But those are, of course, exactly the sort of things that can happen in film.
We tend to think of both space and time as empty receptacles with their dimensions already mapped out, waiting to be filled with some content: this lecture theatre would exist next to the neighbouring room, and it just depends what furniture is in them; this hour would exist anyway, and it just depends what fills each minute of it, one after the other. But that is not how film works. In film, time and space are folded, not just extended and condensed. There is no receptacle. In cinema, any space can be juxtaposed with any other space, and any time with any other time. Space can, so to speak, be folded in on itself like a handkerchief so that any point can be next to any other point.
So what should we call this sort of space and time that the cinema allows us to explore? It is not geographical or geometric space-time, but nor is it adequate to call it “subjective” space and time either. Mathematicians have a word for this space and time, and so does Michel Serres: it is topological.
Thank you for your attention today. Next week we will continue to bring Serres’s into conversation with Varda’s work as we consider her 1985 Sans toit ni loi.
NOTES:
[1] Valerie Orpen, Cléo de 5 à 7. French Film Guides (Champaign, IL: University of Illinois Press, 2007) 57.
[2] The French term “un 5 à 7” refers to an engagement undertaken between work finishing at 5pm and arriving home at 7pm. It is often understood as the time when married men would visit their mistresses. The term is used euphemistically to mean an afternoon lovers’ tryst.
[3] Steven Ungar, Cléo de 5 à 7. BFI Film Classics (London: British Film Institute, 2008) 36-7.
[4] Valerie Orpen, Cléo de 5 à 7. French Film Guides (Champaign, IL: University of Illinois Press, 2007) 29.
[5] Valerie Orpen, Cléo de 5 à 7. French Film Guides (Champaign, IL: University of Illinois Press, 2007) 25-6.
[6] Jean Decock and Agnès Varda, “Entretien avec Agnès Varda sur Jacquot de Nantes”, The French Review 66:6 (1993) 948. ‘Cléo from 5 to 7 shows the fundamental contradiction between objective and subjective time: 5pm to 5.10pm, 5.10-5.15, that is the time of the metronome on the Piano: tick, tick, tick. But there is a woman experiencing these questions and emotions, and time becomes a soft perception that you can’t control. It is like a violin—some moments are vast (we have all the time in the world), and sometimes it goes very quickly and that’s what we have in Cléo, the metronome and the violin.’
[7] Steven Ungar, Cléo de 5 à 7. BFI Film Classics (London: British Film Institute, 2008) 33. Ungar himself writes there are not two but in fact three temporalities in the film: chronological, subjective and astrological (p. 36).
[8] Michel Serres, Le passage du Nord-Ouest. Hermès 5. (Paris: Éditions du Minuit, 1980) 67-8. ‘We always speak about space and time in the singular. So then, what do we know of space today? Precisely nothing. Space as such, unique and global is, I fear, a philosophical artefact. And once again, what do we know of time? Precisely nothing. Time as such, unique and universal, is another such artefact.’
[9] Michel Serres and Bruno Latour, Conversations on Science, Culture and Time, trans. Trans. Roxane Lapidus (Ann Arbor: University of Michigan Press, 1995) 60.
[10] Steven Connor, “Topologies: Michel Serres and the Shapes of Thought”. Available at: http://www.stevenconnor.com/topologies/
[11] Michel Serres and Bruno Latour, Conversations on Science, Culture and Time, trans. Trans. Roxane Lapidus (Ann Arbor: University of Michigan Press, 1995) 57-8.
[12] Michel Serres, Rome: le livre des fondations (Paris: Grasset, 1983) 106. ‘Time enters into the dough, a prisoner of its folds, a shadow of its foldings-over; it penetrates set theory, for the partitions cover each other; time can be marked in the layering of the dough; estimate the immense time of our tissues; […] it plunges into the mixture.’ Michel Serres, Rome: The First Book of Foundations, trans. Randolph Burks (London: Bloomsbury, 2015) 69.
[13] Michel Serres, Rome: le livre des fondations (Paris: Grasset, 1983) 108. ‘A point in this dough is only a point. The point nearest the preceding one follows a path quite different from the first one, just as crazy, just as complicated, illegible. Time is not the path of a single point, but of at least several and at most all.’ Michel Serres, Rome: The First Book of Foundations, trans. Randolph Burks (London: Bloomsbury, 2015) 70.
[14] Michel Serres and Bruno Latour, Conversations on Science, Culture and Time, trans. Trans. Roxane Lapidus (Ann Arbor: University of Michigan Press, 1995) 58.
[15] Michel Serres, Le passage du Nord-Ouest. Hermès 5. (Paris: Editions du Minuit, 1980) 67. ‘Living organisms also know the synchrony of several different times: Newtonians, they rise and go to bed with the sun, carrying inside them clocks which go haywire when they cross the meridian at speed. They die, exhausted, used up, covered in wrinkles, according to the second principle of thermodynamics but, in an unexpectedly Darwinian or Bergsonian way, sometimes make grand-daughters better than they are.’
[16] Phil Prowie, “The Years of the French New Wave”, in By Rémi Fournier Lanzoni (ed.), French Cinema: From Its Beginnings to the Present (London: Bloomsbury, 2015) 232.
Michel Serres and film 1: Agnès Varda’s Cléo de 5 à 7 by Christopher Watkin is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.