Deleuze for the Desperate #5: the movement-image

Dave & Maggie Harris

The first thing you need to consider when you read Deleuze on the cinema is that Deleuze is a philosopher.  He seems to display a great deal of knowledge about film, and he probably was a considerable film buff, but his main interest is in seeing film as a kind of philosophy.  He sees a number of other art forms, like novels, painting and theater, in the same way, as attempts to think about the world and to depict reality.  With the cinema, he is particularly interested in the way it is able to show us movement and the operation of time: that's why the two books are called the movement-image and the time-image.  This makes Deleuze's account of cinema rather selective from the beginning, because he's choosing films that particularly illustrate these two sorts of image, although he thinks they are very common.  Largely, his chosen films are going to be what might be called art house or experimental film.  There are some films discussed which are more popular, but even here, these films illustrate the work of thinkers and philosophers, who also happen to be directors or screenwriters, auteurs as they are normally called.  These auteurs have written about film as well as making some classic breakthrough films.  They include figures you will have heard of, such as Griffiths, Hitchcock, Hawks or Welles, Pasolini, Rossellini and Visconti, Eisenstein and Vertov, but also Mizoguchi, whose work I did not know but who is acclaimed.

M. Harris

What you don't find in Deleuze is much discussion of the things that have interested recent media theorists: nothing on the audience; nothing on the production side of film; often very little commentary at all once the film has been fitted in to his overall scheme of work.  Indeed, you can call on knowledge of these more modern topics to criticize Deleuze's work for its obvious selectivity.  Deleuze tells us that he knows this is a selective treatment, but that he is not interested in the history of the film.  He also has a pretty low opinion of a lot of popular film.

End M. Harris

I repeat, he is a philosopher, and he is interested in film as philosophy, not as entertainment, not as an aesthetic form, and not as ideology.  The very book starts with a commentary on a particular philosopher that Deleuze admires and has borrowed from— Henri Bergson.  I feel sorry for poor film studies students who open the book and encounter that first chapter!  Ideally, you need to read what Deleuze thinks of Bergson more generally.  If you don't have too much time at present, you might just have to rely on my gloss on Bergson's concepts for now.  There are also some notes on Deleuze's view of Bergson on my website here.

In one of his books, Matter and Memory (vii--viii) , Bergson defines what he means by an image:

'a certain existence which is more than that which the idealist calls a representation, but less than that which the realist calls a thing – an existence placed half way between the "thing" and the "representation".  This conception of matter is simply that of common-sense'.  So the image is not a picture or representation of something, which is what we might normally think of as an image.  For Deleuze,  these representations would be better understood as signs.  Instead, an image is rather an idea of something, a conception of something that exists.  So, to jump ahead, a movement - image is a conception of the way in which things move.  As we will see, this conception does not involve a single representation but a whole series of signs, a whole sequence of shots.

Deleuze also thinks that the study of signs in the work of CS Peirce is going to be a good way to begin to understand the signs found in cinema.  I'm not going to discuss this here at all, however.  Bogue (2003) is the best text to read to trace this argument out.

There's a lot of references to films you may or may not know.  Again we see Deleuze's rather narrow interest in movement, and also his unconscious elitism again—no doubt all his friends in Paris had seen these films, and he is assuming that we readers will have seen them too.  There is often rather limited comment as a result.  You might be more used to discussing film in more detail,  and this will be a critical resource.  However, one of the delights in reading the books for me was the discovery of directors and films I'd never heard of before, and I was able to go off and have a look at them.  I learned a lot.  Another delight was seeing what Deleuze made of films that I had watched, and using his work to develop new ways of looking at them.  If you had the time to read more widely, and watch the films like I did, you would probably enjoy the books a lot more.  But you are pressed for time, I have been assuming throughout this series, so let me see if I can put you straight on the money.  It is not quite so simple as with the earlier topics, where we plucked key concepts out of the wild and woolly theorizing, because whole books are devoted to these concepts.  So this video can only offer a kind of initial orientation to the argument, a first grasp, which will obviously be limited and will need fleshing out.  But the assumption is we have to start somewhere, that we cannot just plunge straight into full blooded Deleuze, especially if these books on the cinema are the only ones of his you have read.

Let's concentrate here on the philosophy of the cinema.  I have more extensive notes on the actual books on the cinema on my website which you can go to if you want: here and here. 

Why should images or conceptions of movement and time be so important?  Deleuze has a very good argument here.  First, these are important philosophical concepts to understand how the universe works, and Bergson has made a major contribution.  Second, these are precisely the images or conceptions that the cinema in particular can depict.  The cinema offers us images which are quite unlike those of any of the other arts.

Strangely, though, a lot of conventional film studies can be seen to largely ignore movement and the specifics of moving signs.  For example, in my day it was common to use analyses of photographs to try to understand film, or at least stills from film.  One example would be Barthes' (1973) famous analysis of the photograph of the black soldier standing underneath a French flag.  We would examine signifiers in this photograph, those usually understood as denotations and connotations.  Other analyses tended to take film as a form of written literature, and would use the same analytic techniques to grasp filmic narrative as they would to grasp narrative structures in a novel.  I think of analyses which I used to do of the Bond movie, which would begin with a discussion of Eco's structuralist analysis of the Bond novels.  Analysts would try to find all sorts of parallels with literature, and use terms such as signifier, syntagm and paradigm.  Sometimes they would identify particular narrative structures, such as realist ones, in the novel, and then try to find them in film, to cite a debate that raged in the 1980s, associated with the name of Colin MacCabe (see Bennett et al 1981).

M. Harris

The first shock that Deleuze offers is his argument that none of these approaches are going to be able to properly grasp movement in film.  If we look at film stills we are not looking at film movement, and if we are looking for literary forms, we will miss the particular ways in which film links elements, including those not found in literature, like visual or sonic ones.  These are linked together to form a narrative.

End M. Harris

Let's start with some very simple observations about movement.  The cinema is very good at illustrating movement, we can all agree, but it does this in a number of ways.  Very early cinema, for example, was rather like the theater in that a fixed camera on a tripod recorded the action that was taking place in front of it on a set.  We saw movements, by the actors, and sometimes by non humans like horses, but it was within the confines of theatrical notions of the set and a scene, even when shooting outdoors.  The scenes would then join together with other scenes, with straightforward cuts, where the screen just went dark, just like scenes at the theater.

However, cinema soon developed different ways of illustrating movement.  One way is by editing scenes together so that they represent a larger sequence—montage.  There are clearly different ways to do this, and you might have learned about some of the conventions of realist or continuity editing.  For example, the camera obeys the 180° rule, or the need to match shots on eyeline across cuts, or pursue shot - reverse shot techniques.  If skillfully done, the join between the shots is practically invisible to the audience, at least once they had learned how to view cinema, and the movement in the film flows across the cuts.  One great example here is Hitchcock's film Rope, which is cleverly edited it so it looks as if it is just one continuous take, or at least a couple of continuous takes (Deleuze does not mention this example himself).

The real breakthrough, however, came when cameras were able to move themselves, tracking in various ways for example.  The invention of different sorts of lenses, including the zoom lens, also enabled them to move in and out of the scene, so to speak.  The cinema developed the classic variety of shots—long shot to establish context, mid-shot to focus on action, closeup to display emotions.  Again a movement flows between the shots as montage or zoom.  Of course, with the advent of talkies, film could also include sounds and make them act as signs as well.

The cine camera also developed unusual pictures of reality, which are not just those of natural human perception.  The camera can give us a nonhuman or objective view.  To take some easy examples, it hovers above the landscape or it zooms in and out.  In more experimental cases it can show slow motion or time lapse, reverse sequences, or very unusual non natural angles (try Downside Up).  Modern cameras can stand in places where humans cannot—in front of a stampede, in a clearing in the forest in front of a dangerous animal, lowered into the sea, put on the outside of a space vehicle.  We can see the world as it is without human perceptions (Deleuze calls this whole set of points of view, including human and nonhuman ones, 'percepts').  The ability to depict the nonhuman or the objective is important, because it is going to show us how human subjects are affected by things moving in a life of their own, relating to each other without human intervention, or even human knowledge.

Now, the camera can depict some very interesting kinds of movement, interesting for philosophers, that is.  We are finally beginning to see illustrations of what movement really is for people like Bergson.  It is not a mysterious force that operates somehow in addition to or between still moments, which is the classic Greek conception.  Instead, movement is a force in itself.  Everything is in motion, and still objects are seen as temporarily halted movement.  We now see movement not just in terms of fixed poses or instants, but as operating constantly, affecting all the stages in between significant moments, 'any-instants-whatever', as Deleuze puts it.

This conception of movement as the major process in depicting reality began to deliver results in maths and science too in the early 20th century, and Deleuze briefly mentions a couple of them.  One view of physics, often shown in popular documentaries (like B Cox's recent Forces of Nature episode 2), has the independent movements of forces occasionally stabilizing around attractors and slowing down or cooling down to produce matter, first gaseous then solid or liquid matter.  Deleuze cites Bergson's understanding of this as a matter of light being obstructed by matter, and we should see light as standing for all the electromagnetic forces. 

Incidentally, in the second commentary on Bergson, Deleuze gives the clearest definition yet of the term that keeps cropping up in his work—the plane of immanence.  Immanence (with an 'a') is a state where some inner potential is being realized.  The plane of immanence is that theoretical but real level of the universe in which energy is just starting to turn into matter, where light is just beginning to be obstructed, a plane where matter and energy coexist, both virtuals and actuals exist, in Deleuze's terms.  It is a bit like the level at which both ice and water coexist at 0°centigrade.

Cinema also shows this new kind of creative, active movement on the plane of immanence, or rather the image or conception of it.  This is the important development in the movement image.  One such movement that Deleuze actually talks about quite a lot in the first volume is between background and foreground, context and location, the big picture and the local picture, the objective and the subjective, what is in the frame and what is left outside it.  Parts are shown in relation to wholes by these camera movements.

You probably know that some people think there is a formula for the narratives of popular films: an initial state of equilibrium is shown, then it is disturbed by some intruding force to produce disequilibrium, then the problem is resolved leading to a new equilibrium.  Think of High Noon, my favorite western, although again not mentioned by Deleuze.  A town in the American west is just settling down at last to a normal life with law and order, then the bad guys turn up, from the past and from prison in another state, destroying equilibrium and threatening to turn back the clock, metaphorically speaking.  We hear of them off-frame long before we actually see them.  The hero has to take action, reluctantly, to restore equilibrium, partly by invoking his own violent past.  He wins, but it is a new equilibrium that results, with lessons for us all—ideological ones, you might think, suggesting that violence is sometimes necessary, even for Quaker women, that civilization is constantly under threat from dangerous outsiders, but that luckily real men are available to rise to the challenge.

Deleuze is not interested in ideology and offers a different terminology.  For him, there's a closed set of elements, which could be humans and objects, interacting in a predictable and ordered way.  We mean a mathematical set here not a film set.  Then something disturbs the set, something from the outside.  This is able to disturb the set because the set is always only a part of a larger whole, never fully closed off from the outside, even though no-one in the set realized this.  The disturbance produces a qualitative change in the operation of the set.

Again, this is not just a narrative for film, but a description of how change happens in the real world too: stable sets of solid things interact predictably, but they are always subject to the Whole, something more open, something more chaotic if we can use popular terms.  This is what produces qualitative change.  Think of a homely example: the dinosaurs were developing nicely into different species and types, fairly regularly and predictably over millions of years, and then...  The asteroid strikes and we get qualitative change, not change inside existing patterns of dinosaur development, but changes of species themselves, the extinction of one and the rise of others.  The asteroid moving through space was always part of the Whole, a cosmic system that earth was related to, although of course the poor old dinosaurs did not know that, plodding along in their seemingly closed set on Earth.

M. Harris

That is a spectacular example, but it does not too far from what Bergson thinks normal evolution involves: some element from a complex multiplicity outside the apparently closed world of species of plants and animals intrudes and sets off a qualitative change.  Bergson describes this process in terms of a life force, an élan vital, and the concept is connected to his major term, duration, which we will come to in the video on the time-image.

End M. Harris

So, back to film.  Elements of the Whole, a set of multiple possibilities, intrude and spark off qualitative change and the film shows this happening.  This can be done in many different ways, and Deleuze explores some.  There are different conceptions of the Whole, for example, in French and German film and in different genres.  However, the thing about classical film is that it still feels it should place human beings at the centre of this process.  It is all tied to humans, and the ways in which they respond to the world and then react, in the 'sensori-motor schema' as Deleuze calls it.  One characteristic of modern experimental or art cinema is that it abandons this organizing schema, as we shall see in discussing the time-image.  You might be able to see some possibilities already by thinking of how we discussed the nonhuman perceptions of cameras just now.

How does this work in normal films?  First someone, or the camera itself, perceives that something is happening out there, outside the normal set of stable life, from somewhere that is normally kept at a safe distance, on the horizon.  It has the capacity to change everything.  We see the smoke from the Indian fires, the gathering snowstorm, a new threat to world order in the rise of dangerous politicians or gangsters, the initial signs of an alien invasion, or whatever.  This has to be shown on screen in a manageable way, in a perception-image.  We might have subjective perceptions by the character, or initially objective ones.  Some objective ones can rapidly becomes subjective, or sometimes the reverse, as in interesting cases where the camera does not adopt the standpoint of any one character, but moves among them and can come to tell their story in a particularly important development—the 'free indirect discourse'.  These interplays between subjective and objective are possible because of Bergson's initial insight that we started with: an image has an existence placed half way between the thing and the representation, and is thus capable of acting as either.  Deleuze modifies Bergson in fact, but let's leave that for later.

After perception, everyone takes appropriate action, requiring an action-image.  Again, there are different possibilities here including different sorts of links between action, situation and subsequent action, in what Deleuze calls 'large' and 'small' forms, and this is a particularly interesting section of the book.  There are different types of action too, like impulsive or reflective types, and this is where Bogue (2003) says Deleuze draws on Peirce to help him identify the classic signs for these different types.

We also see what motivates this action.  We see the impact of external events upon the people, usually in the form of some facial responses shown in close up—a frown, a doubt, a tear or whatever.  Deleuze particularly admires actors who can let micromovements play  across their features, and takes as an example the amazing classic 1928 silent film La Passion de Jeanne d'Arc.  Incidentally you can now watch this free,  online, on the superb  website Open Culture  here, together with many other classics.  This film tells the whole story, more or less, in a series of large closeups of the actors' faces—the slyness of the prosecutors, the surprise, hurt, but saintly innocence of Joan, the sadistic voyeurism of the torturers.  This is the affection-image, and 'affection' here has a special philosophical meaning: it means the human responses to external events, often, but not always, emotional responses.  In Bergson (19), this is clear: affections refer to all the human impulses for initiating action, not just emotional responses.

Deleuze also sees 'affect' not in modern terms, as just emotion, but as any influence acting on humans.  The term is much discussed in his other work, but there is a particularly neat argument here in the second commentary on Bergson.  Human perception is selective, and we perceive what we want and need at the time.  But the things left out by the selectivity can still affect us, without us knowing at the time.  We still receive affects from events, and we can be strangely affected by things like the weather,  in a favorite example in Deleuze and Guattari (2004), even if we do not notice it or perceive it at the time.  Anyone trying to teach young kids, or students, during a snowstorm will know that.

Although affects are often registered by movements of the face, or even parts of the face, or a montage of faces, Deleuze says that nonhuman things or processes can affect us as well.  Here, he notes that spaces or sections of spaces can also offer affects—the bits of wall or stairs almost out of shot, caught in the closeup of Joan's face.  In another film on Joan, The Trial of Joan of Arc, directed by R. Bresson, we see the mundane spaces of Joan's cell with a glimpse of a busy corridor crammed with English guards just outside, or we see shots of Joan's manacled ankles as she sits on a nondescript prison bed.  All these refer to the depressing and alarming indifference and impersonality of these mundane ordinary bits of space, an awareness that they have witnessed imprisonment and executions before as a routine.  Deleuze says such shots indicate any-space-whatever in a clear link to the any-instant-whatever discussed earlier.

Together, the perception image, action image and affection image are components or 'avatars' of the overall movement image, the overall conception of a force from outside, from the whole range of possible forces, affecting the actors and objects in the limited set.

I will let you go off to examine the very interesting ways in which Deleuze develops these points and illustrates them with films, very often with just a simple couple of comments.  The books are not at all easy to follow at first, and are spattered with references to films you may not know.  I coped by running through the books doggedly waiting for something or someone I did know, while going off to learn more and watch more films.  Quite a few of these films are now online, doubtless because fans of Deleuze have uploaded them to YouTube or Ubuweb.  As I said before, Bogue's excellent commentary is very clear and helpful. 

Finally, you can also turn to the Glossary at the end of volume one.  It is in alphabetical order, but I have picked out the main terms, and I think it makes more sense organized slightly differently.  I have also added some comments.

So the movement image is defined in the glossary as 'the acentered set [ensemble] of variable elements which act and react on each other'.  An acentered set means that movement is not immediately centered on human perceptions or actions but is objective.  As we saw, the most interesting movements are those that come from the outside and cause qualitative changes.  In classical cinema, centered on human beings, the objective reality has to be given human significance, provided with an image center: 'a gap between a received movement and an executed movement, an action and reaction (interval)'.

This center is provided by connecting variable elements to human action, mimicking ordinary consciousness by organizing everything around the individual's body.  First this is done through the perception image, 'a set [ensemble] of elements [either subjective or objective]  which act on a center, and which vary in relation to it'.  The effects of these perceptions are shown in the affection image , filling the gap between an action and reaction, 'that which absorbs an external action and reacts on the inside'.  The overall result is an action image, an image of 'the reaction of the [human] centre to the set.

I must say I don't find these definitions terribly useful.

Finally, Deleuze argues that this whole way of making movement image films changed, as the result of a number of factors.  There were social crises like world war two, which disrupted normal perceptions and actions and their settings.  The whole process of depicting movement images became rather artificial predictable and clichéd.  There's also a new philosophical thinking about movement, with particular emphasis on time.  The result was a series of time- image films, which we're going to cover in the next video.

STOP PRESS. I forgot to mention the useful if condensed discussion of cinema in
Deleuze, G. (1995) Negotiations. Trans Martin Joughin. New York: Columbia University Press. I'll maybe mention it in the next video.  I have some notes on the whole book here


Barthes, R., (1973) Mythologiques. London: Fontana.

Bennett, T, Boyd-Bowman, S., Mercer, C. and Wollacott, J. (eds) (1981) Popular Television and Film, London: BFI Publications.

Bergson , H. (2004) [1912] Matter and Memory. New York: Dover Publications Inc

Bogue, R.  (2003) Deleuze on Cinema.  London: Routledge. My notes on

Deleuze, G (1989) Cinema 2 -- the time-image, London:  The Athlone Press. My notes on

Deleuze, G. (1991) Bergsonism, New York: Zone Books  My notes on:

Deleuze, G. ( 1992) Cinema 1: The Movement Image, London: The Athlone Press. My notes on

Deleuze G and Guattari F (2004) [1987] A Thousand Plateaus, London: Continuum. My notes on:


Downside Up, 1984,Dir. UK Tony Hill. Available from

High Noon. 1952. USA Dir. Fred Zinneman. United Artists.

The Trial of Joan of Arc. 1962. Dir. Robert Bresson Agnes Delahaie Productions

La Passion de Jeanne d'Arc. France. Dir Carl Th Dreyer. Société générale des films.

Rope.  1948.USA. Dir. Alfred Hitchcock. Warner Bros.