NOTES ON Bogue, R.  (2003) Deleuze on Cinema.  London: Routledge.

[A very good clear read and commentary on Deleuze's Cinema 1 and Cinema 2]


Deleuze was particularly attached to cinema, although he was better known for commentaries on literature, theatre and painting.  The point in the cinema books is to develop specific philosophical concepts proper to cinema.  There are specific analyses, but also 'a general conception of cinema as a mode of thought' (2).  We need to think about time, space and movement to get there.  'Deleuze is not deliberately obscure, but his cinema texts are often difficult, primarily because they are so allusive and so highly condensed in their references'.  They assume existing knowledge of films and film criticism.  This book is an attempt to read along with Deleuze, to draw out implications, to explicate the condensed comments and trace the broad outline of the argument and its overall coherence.

We have to understand Bergson first, especially the idea that 'the things we commonly call space and time are merely extremes of the contraction and dilation of a single duree, or duration' (3).  There is an 'open vibrational whole, a flow of matter-movement'.  This is to be applied to the specific problems of the cinematic image, time and movement, down to the level of analysing frames, shots and montage.  The open whole is being demonstrated, 'indirectly present' in individual frames and also in whole montages.  Montage is particularly explicit, and Deleuze shows there are four basic approaches in early film making—'organic unity... dialectical totality...
mathematical infinite... and dynamic infinite made up of clashing intensities' (4).  Each bring with them specific techniques of 'framing, composition, lighting, camera movement and montage'.

Deleuze also attempts a taxonomy of images of cinema.  Bergson's notion of image suggest that everything is made up of images [an early version of the claim that everything can be grasped as phenomena].  Living entities are special in that they can choose to some extent how they will interact with other images, imposing their own sense of space and time.  Deleuze derives from this general argument three specialised images in every living image: 'the perception-image, whereby the living image senses the outside world; the action-image, which structures the space surrounding the living image; and the affection-image, which connects the living image's outer perceptions, inner feelings, and motor responses to other images' (4).

A more complex taxonomy emerges, via Peirce and the notions of firstness, secondness and thirdness.  These suggest three types of images in addition: 'the "impulse-image," the "reflection-image," and the "relation-image"'(5).  The six categories can be used in a taxonomy.  The signs [of images] are to be non linguistic.  There are basically three categories, 'one genetic, the other two compositional'.  Six images, and three signs each gives 18 types of signs, 'each of which is a specific kind of [specific nonphilosophical?] image'[as I suspected, the term image is being used with several meanings?].

In Cinema 1, the movement - image is discussed—'all images that are regulated by the sensory motor schema…  [Which]…  Provides the commonsense temporal and spatial coordinates of our everyday world'.  The signs of this image appearing classic cinema as simple coordinates of the commonsense world.  In modern cinema, this schema is abandoned, and new time-images appear, with new signs, which are not common sense, especially "opsigns" and "sonsigns" then '"mnemosigns"' [hereinafter M] of flashback memories, and the "onirosigns" [O] of dream landscapes; and finally in the "hyalosigns" [H] of time crystals (5).  The first two allude to time outside measurable time, and Italian neo realism is the first sign of 'this shattered temporality'.  M and O can be recuperated within common sense time, but they can also allude to a different sense, 'a bifurcating time in the flashback, a floating time in dream sequences, and the oniric dance world of Hollywood musicals' (6).  Only H suggests a full time image.

Bergson suggest that memory consists of a doubling between an actual present. and coexisting virtual past, moments.  There is a virtual past extending from the present point in time, and this suggests a non objective time.  Cinema's time crystals can bring to awareness this connection between the virtual and the actual and how they link together to become indiscernible.  This can be shown in a number of ways, for example the mirror scene at the end of The Lady from Shanghai [surely a cliche by now though?], but an entire Fellini film can be seen as a time crystal (And the Ship Sails On).  Different directors offer different '"states of the crystal": in Ophuls the perfect crystal; in Renoir, the split crystal; in Fellini, the crystal information; and in Visconti, the crystal in dissolution'.

H and time crystals can unsettle us by offering into penetrated present and past.  '"Chronosigns" [C] offer additional images, concerning both the order of time and time as a series.  The first depicts time as coexisting sheets or simultaneous points of the present, as in Resnais, and Robbe-Grillet respectively.  Time as a series 'is manifest in images that incorporate a "before" and an "after" within a "now"'.  These reveal the 'power of the false', a becoming which undermines fixed identities and blurs the
true and false, as in 'Welles's Nietzschean tales of deception and fakery…  [and]…  Rouch's ethnofictions' (7).  Godard in particular shows how categories appear in series 'engaged in a perpetual metamorphosis, each series disclosing a specific power of the false'.

C are signs of thought '"noosigns"', and signs that must be read '"lectosigns"'.  In classic cinema, images and thoughts were simply in accord, organized through the 'sensori-motor schema common to film and viewer'.  In modern cinema, the image produces alien thoughts and this engages '"a spiritual automaton", in the mind of the viewer and also displayed on the screen as a 'brain surface of a non human thought'.  There is a non naturalistic relation between images, a gap, and in the gap 'the fissure of a pure outside is made present', a split in objective time and space.  There can also be a gap between images and sounds, again requiring an active reading not a common sense understanding.  Audio visual signs can become autonomous, located in a continuum, which might separate visual from sonic and interrelate them in different ways.  'The films of Straube/Huillet, for example, reveal a stratigraphic space and an aerial free indirect discourse, the visual images and sound images communicating through a staggered back-and-forth passage between the two domains' (8), while Duras offers a flow of visual images and the flow of words 'tending towards inarticulate cries', which interacted but never dissolve into each other.

This book also relates to other analyses of the arts [and Bogue's other books].  It is intended to be general not specialist.  Other useful studies of Deleuze on cinema are cited (Rodowick, Kennedy, and Marks; Flaxman, Serrano, Fahle and Engell).

Chapter One Bergson and Cinema

The two books on cinema are about taxonomies of signs, but Deleuze also uses the concept of auteur [a contradiction?  see the excellent essay in Boundas and Olkowski by Ropers-Wuillemier].  In the second book, time is seen as 'the fundamental element of cinema'(12), which would naturally involve Bergson for Deleuze.  Deleuze denies that Bergson is a dualist, despite appearances, especially that there is no split between consciousness and its operations, and the processes going on between life and matter.  Deleuze wants to argue there is a higher monism which generates apparent dualities.

Bergson develops the notion of duration in his work.  At first, it is directed towards arguing that subjective feeling such as emotions cannot be quantified, but come in qualitative complexes.  He went on to suggest that all physical sensations are qualitative, and that single changes always produce qualitative involvement of different parts of the body.  Some are more abstract, and some more related to physical sensations.  Even numbers are a quantitative reduction of qualitative differences.  Such reduction replaces changes in time with dimensions in space [Deleuze has a good section on this, which is rather like Adorno, in Difference and Repetition, I think, which says that in order to be able to claim that you are measuring different quantities, you have to assume that they have not changed in any other significant way over time, that events are replicable, and so on].  Normally, we do think of time as a form of space, some abstract element proceeding from one point to another.  But we are also aware of duration, of links between the past and the present pointing to the future.  Bergson's analogy here is of musical melody, which can be seen as a sequence of individual notes, but which is really 'an indivisible multiplicity changing qualitatively in an on going movement…  The entire succession of notes forming a single process…  an indivisible heterogeneity' (14): each note pushes on into the next one producing qualitative change, something new.  This is one way in which duration is always indeterminate [in the sense of emergent].  In a deterministic causal universe, time has no effect [as we have just seen]. Succession considered as part of duration becomes a matter of real movement.

The psychological emphasis is continued in the early discussions of memory, which offers an internal notion of duration 'a heterogeneous qualitative multiplicity of indivisible, of inseparable terms' (15) the very opposite of the scientific objective conception of space.  In this early work there seems to be an opposition of external space and internal duration, but Bergson goes on to speak of duration in general, driven by a whole virtual past.  This can be linked with élan vital as a vital impulse affecting all living entities, driving evolution, life as invention producing divergent lines.  Here there seems to be an other opposition between the animate and the inanimate, but Deleuze argues that there is a more general process again, the virtual itself being actualized, with evolution as a matter of actualization of the different degrees [see the essays on Bergson in Deleuze's Desert Islands].  Both matter and living things are different degrees of contraction and relaxation, with the difference between the animate and inanimate as simply an emerging form. 

Deleuze focuses on the more general aspects of Bergson, including his insistence that solid matter itself is made of movement and flux, and how this should even affect modern physics [and Bogue credits him with some early insights into physics on matter and energy here].  Matter itself moves towards entropy, but life goes in the opposite direction, towards creative differentiation.  This helps us see that 'entropic matter is simply a decrease in tension in a field of energy, and creative life an increase in tension in that same field' (17) [this trope is finally starting to make sense, but why call it 'tension' -- interconnection would be better? Multiplicity?], hence the differences all depends on the degree of relative contraction or relaxation of duration.

[Another mysterious set of terms is now to be explained].  'There are different temporal rhythms in the universe and... qualities and quantities form a continuum, my most fleeting perception of the quality of red [for example] being a temporal contraction of trillions of nearly identical oscillations into a single moment' (17).  At the subjective level, we know that we can contract [here we go again -- simplify, disconnect, called relaxation above?] our experiences into a present moment in order to undertake an action, but we also know that in dreams, there is a relaxation of will and memory [but not of duration!] , and a loosened tie between present and past [meaning lots more duration].  Similarly, inert matter is minimally contracted [the upside down feeling again!], because it contains minimal connections with the past.  Our attempts to grasp objective matter also contract [simplify] temporal connections to the past, and things appear to be connected so tightly to events in the past that they can be seen as identical, and their emergence as predictable.  We also develop a notion of external effects.  [Back to the mysterious upside down nature of the term loosening—duration itself is infinitely loosened in such notions, so that individual moments can spread out so to speak into some sort of continuum.  OK, but doesn't our consciousness have to ignore duration first, so that we can eventually substitute the notion of xtension and spaces? This is called 'contraction'! ]

Yet duration can never be eliminated entirely—inertia is a tendency towards elimination of energetic duration, an ideal, as in Euclidean geometry or Newtonian physics [where other things have to remain equal, or processes reduced to the laboratory, and so on.  Similarly, pure duration separated from consciousness also offers an ideal.  In particular, 'all our thoughts and emotions are embodied and connected to the physical world' (19).  Relaxation and contraction are relative to each other.  [Back to the mysterious terms of vibration and speed...] The different forms can be seen as 'different rhythms in a vibrational whole'.  These are not vibrations in the usual sense [!], Of something material.  Instead 'movement is inseparable from that which moves', amino now there is a connection at the sub atomic level between flows and vibrations and apparent entities.  In particular, something from the past is always retained, 'a memory of some sort', and there is always an 'impulse towards the future (an élan or will)' (20).  Using terms like memory and impulse, we can argue that matter itself has 'some form of consciousness, some degree of dynamic contraction of the past into a present towards the future', and 'conversely, the most complex forms of consciousness are part of a single continuum of rhythmic flows' (20) [really a sleight of hand here, involving particular and very general definitions of memory and impulse?].  Only 'relative speeds and degrees of contraction'differ between humans and other matter, especially the intervals between events and the amounts of the past that get contracted into the present.

This reduction to vibrations makes Bergson a monist [via Deleuze, for rather peculiar philosophical reasons as we saw, mostly to be consistent and avoid having to use different concepts and so on].  This monism further explains differentiation and distinctions, since there is 'an irreducible multiplicity of rhythms'[here, a matter of rhythms of contraction and relaxation—the earlier example of red light seemed to take ordinary human consciousness and the capacity to perceive vibrations as the standard, and it is science not philosophy that has discovered the myriad of vibrations in what looks like a simple unitary colour? There is this constant rhetorical contrast between ordinary consciousness and philosophy, but never science -- and social science-- and philosophy].  At the limits of the range are inert matter and extensionless mind [relaxations and contractions respectively of duration].  Duration itself is 'a qualitative multiplicity', constantly differentiating itself. Deleuze therefore has reconciled all the apparent contradictions by seeing Bergson's thought as developing through different stages—a critique of the conventional false dualism slight mind and body, quality and quantity; differentiation between duration and external space, qualitative and quantitative multiplicities respectively; seeing the difference between duration and matter as produced by 'the rhythmic contractions and relaxations of a vibrational whole'; explaining how duration actualizes itself through the élan vital, producing apparently dual forms like inorganic and organic.

Deleuze is particularly interested in Bergson's notions of movement, which he derives from Creative Evolution.  First, it is an illusion to see movement as a matter of moving through space.  Interestingly, Bergson sees this illusion best represented in cinema, which links a series of snapshots to produce the illusion of movement.  Much of our ordinary consciousness invokes the same kind of illusion, where we sliced time into static moments, 'or immobile cuts' (21), and then link them back together again.  What we need to do instead is to see each moment as connected to an indivisible multiplicity, with its own duration as the source of movement.  However, Deleuze argues that this precisely gives insight into how cinema develops movement - images, and how we can use Bergson to talk about immobile cuts and real movements depicted in cinema.

Secondly, Bergson talks about the Greeks and their notions of movement.  One conception sees movement between privileged moments, such as infancy, boyhood and adulthood.  Galileo was able to reject this notion of periodic change and replace it with 'the sequence of equidistant, indifferent, and interchangeable instants—...  "any - instants - whatever"' (22).  For Bergson, this is still only a more precise version of the cinematic error,  but Deleuze notes that the cinema actually provided early knowledge of a different conception of movement, as in the famous early films of galloping horses.  It is more than just denying the notion of privileged moments, rather that any singular instant is produced by from the flow of any-instants-whatever.  It follows that privileged moments in films, 'images of crisis, paroxysm and intensity' as in Eisenstein, a also linked to any-instants whatsoever, that these 'remarkable instance are immanent within movement' (23).  This is something genuinely new emerging from cinema, an explanation of novelty, 'the remarkable and the singular'as a result of a flow of any-instants-whatever.  Deleuze sees this is nothing other than a suitable metaphysics for modern science—and for cinema.

The third notion in Creative Evolution pursues the notion of 'cuts' in [snapshots of] duration.  Instants are immobile cuts, movements are mobile cuts.  We can now see movement as no longer 'the shifting of positions of objects in space', but rather as transformation of one qualitative whole into another [as above, with the early work on the emotions].  Individual movements are best seen as parts of qualitative changing wholes, and qualitative changing is emergent.  Bergson apparently has a strange example about dissolving sugar in water—our psychological experience of duration helps us to grasp duration in general, a duration that's also linking the sugar and the water.  [If we go on to philosophise], we can see that the water, sugar, glass and the observer are all abstractions from the universal vibration, and that each one changes,albeit at different speeds.  Doubtless other changes also take place, and we have only artificially restricted the discussion to the one event.  Even natural science sees that dividing the universe into closed systems is an abstraction.  We have a notion of a fully open whole, and a critique of the reduction involved in determinism, objectivity and causality.  This whole is 'given'in every attempt to analyse parts of it, including deterministic ones, and, in biology, in finalist ones.  The whole prevents any simple determinism, since it guarantees new and undetermined developments.

It can seem as if matter develops into isolated systems and stable objects, but there is a difference between the whole and ensembles: the latter constitute isolated system or closed sets within a whole.  Individual items are immobile cuts, like still photographs, and the whole set can be seen as an immobile cut [immobile ensemble would be better].  The sets are not simply subsets of the whole, but 'insistent illusions' (26).  These are useful for ordinary actions and common sense, but it leads to a further illusion that movements within a set can be seen as 'unchanging bodies shifting positions within a space container'.  Instead, such movement is best seen as 'the specific local manifestation of the vibrational flux of the universe'.

How can we perceive the real actions of duration through the portion that presents itself?  Deleuze has to extend Bergson by arguing that duration is the whole of the relations, and movement to an expression of it [these are only implicit in Bergson, Deleuze argues].  [If we know how to perceive it], mundane movement inside closed sets can be seen as qualitative changes in the whole as well, so that mundane movement is an 'intermediary between closed sets and the open whole'.  We have to see this movement as simultaneously expressing duration— '"Movement thus has two faces, in a way" says Deleuze.' The issue is further discussed through the concept of expression in Spinoza, especially the idea of two directions of expression, explication where the one becomes multiple, and implication, whereby each example of the multiple implies the one.  Together, these two movements can be seen as complication.  Separately, explication is a process of development of the one, and the one [surely the entity expressed?] can be seen as a sign that points beyond itself.  Thus entities in enclosed sets are signs of the process of duration becoming actualised, and duration is 'immanent within each closed set' (27) [not really an argument though, more a series of interlocking assertions and definitions]. 

Bogue's homely example concerns an observation of an interaction between a dog and a girl beneath the tree.  Each can be seen to move, but [a skilled observer] can also see 'an ongoing transformation of relations between them, metamorphosing configurations of girl - dog - tree', shifting over time.  I can then speculate about other elements, some of which may be invisible, such as the wind touching the leaves, or the patterns of light on the scene.  The cinematographic illusion of the initial simple observation has opened on to an idea of holism and duration [but only after developing a series of special philosophical concepts to guide the whole process.  It's a form of rhetoric really, allowing Deleuze to persuade us to see the world his way].  Observing the scene through a frame adds to the illusion of a closed set and immobile cut.  Given the right perception, the closed cut becomes a mobile one, a slice of duration. As a result, Deleuze says that Bergson thinks that there are several images available to us: instant ones, in the form of the immobile cuts; movement images which are mobile cuts; time images which relate directly to duration, change, relation beyond movement.  These are to be the categories which analyse cinema. 

First, Bergson needs to be explored further.  The notion of an image is used to deny both idealism and realism.  In common sense, we use this term too—the object as it appears to us, which seems neither in our minds, nor to have a completely different nature.  Objects are seen as more or less as we perceive them, but as images which exist in themselves.  We can avoid the reduction of either idealism or realism.  Deleuze notes the similarity between this conception and phenomenology, although phenomenology still begins with consciousness: Bergson wants to deduce consciousness as 'a particular kind of image' (29).  We can see the images act and react among themselves according to the laws of nature, producing a general predictability.  However, living beings introduce a certain in determination, a gap between mechanical causes and effects, or relay between movements from outside and individual movements.  Living images perceive movements.  Perception is always linked in action, in a sensory motor schema.  We do not perceive abstractly, but always in order to further action, control space, gained some freedom for action.  This in turn involves a selective subtraction of the features of surrounding objects according to our interests and concerns.  Living beings do not 'react equally to all surrounding forces and bodies' (30).  In effect, we convert images into representations, and this involves a preliminary separation from other images, which Bergson describes as 'a diminution of light, and darkening of its contours'(31).  However, this process does not just
depend on consciousness, but on objects themselves, who reflects and emit light enabling us to perceive them, if we prioritize particular emissions.  Bergson goes on to introduce a general conception where matter in its entirety presents itself to be perceived as a kind of 'photograph of the whole', while individual human perceptions offer different sorts of screen on which parts of this image can be displayed [this is to deny the usual notions of perception as a kind of mental photograph—the photograph is already there, so to speak, and perception does not take place in individuals, since it involves luminous points in the whole].

This means that perception is not special, but another part of action, and there is no division between the material world and the inner mental reality.  Indeed, images perceive each other, and are best considered as part of the whole.  Objects are already images, while perceived objects are subtractions—the qualities remain with the object, and even the perception is the result of an 'interaction of perceiver and perceived' (32).  Bergson goes on to say that pure perception like this, without memory, is just a convenient fiction, but it is already possible to see that movement is integral—a flow of interacting images, with living images as 'a centre of indetermination', a perception as a form of action involving selection and reduction of possible interactions.  This general notion [far too general in my view -- can all objects have a sensori-motor schema shaping their perceptions?] of perception is not to be restricted to human beings, since any image can perform selective interaction—human perception 'is simply a subtractive reduced version of the perception of non living entities'.

We therefore have a 'primal cosmos of images as '" a gaseous state"…  universal rippling: there are neither axes nor centre, neither right nor left, high or low"' (33) [quoting Cinema 1].  Not only does the image exist in itself, as a matter, but there is a connection between image movements and the flow of matter.  This produces '" a source of plane of immanence"'.  We have temporarily left out duration, so this can be seen as a mobile cut, a block of space-time, one of an infinite series, '"so many presentations of the plane"' (34, still quoting Cinema 1).  Deleuze wants to go on to see this plane of immanence as featuring the identity of matter as light, which further explains the identity of the image and movement, bringing in the ability to discuss visual appearances specifically.  This does raise the issue of non visual modes of perception, however, but it helps us move towards an analysis of cinema.  The visual images in cinema are now the same as matter itself: the image in itself is 'a virtual image, a given "block" of light'(34), and it becomes visible to us through the process of actualization of a portion of it [the notion that there is an objective dimension to perception again].  Thus visual perception takes place both in the object, and when the perceiver configures light.  Consciousness itself can be seen to depend on light.  [This sharing between the subjective and the objective, their common nature as configurations of light leads to the famous phrase]: 'we may ultimately look on "the universe as cinema in itself, a metacinema"'(35). [Includes the objective nature of the camera as a non-human eye etc -- camerapersons select but they select from what is already being presented as actualized].

Consciousness is immanent within matter [and perception follows from the actions of the metacinema].  What living images do is to supply a certain indetermination, or gap in universal interaction.  The three kinds of movement image relate in a different way to this centre of indetermination.  All living images receive the movements of other things and convert them into their own movements [still using the term action, a deliberate blurring of the difference with humans?].  [Ordinary human] perception is geared towards action, and permits an analysis and selection of received movements.  Living images therefore have two sides, one to receive and one to act.  Perception images are movement-images related to the receptive side, 'the selective registering of incoming movements, a framing' (35).  They are things with aspects subtracted from them.  There is a second kind of movement-image connected to action, however—here, a perception is affected by expectations and anticipations, just as the world is arranged for us around ourselves as the centre [very like Schutz here, with things arranged around us at the centre, according to their proximity, and also their concreteness/anonymity].  This leads to the notion of an action-image, which locates perceptions in our concentric worlds: Deleuze wants to argue that this produces both the action of things on us, and vice versa.

Affection-images are based on what Bergson says about qualities.  Categories of language, adjectives, verbs and nouns, emerged from our organization of transient impressions into continuities, eventually into objects or bodies, nouns, with qualities, adjectives, and subject to repeatable action, hence verbs.  Deleuze notes parallels with perception-images and nouns, action-images and verbs and affection-images and adjectives.  The link with the latter arises because perceptions can produce internal sensations, such as pain—affections which are qualitatively different from perceptions, although the two are always copresent.  In complex organisations like human bodies, registering gets spread among specialized organs.  In some cases, the perception of the impact of an external movement on the body itself, on the surface of its organs, like intense light on the surface of the eye, can have a separate effect, including bodily reaction: here, perception and affection come together, but they are usually mixed in different combinations.  We are both centres of indetermination, and also 'extended bodies in an interactive field of forces and movements' (37).  Our bodies themselves can encounter directly external movements and there is no indetermination or gap, and this possibility can be constant when we are trying to calculate action in a more distant way—hence the inevitable mixture of perception and affection, the ability to reflect, but the necessity to absorb external movement and impacts.

Thus we can see affection as arising in the interval between perceptions and actions, registered by bodily sensations that accompany our perceptions as things impact on us.  We understand that the sensations refer to our qualities as living beings.  It also helps us distinguish between the outside and the inside.  In the interior, we absorb movements and this produces motor effects [like a reflex action].  Again we see a connection between subjects and objects. Deleuze talks about a 'receptive plate' in the interior of our bodies, and sees this as the same as the black screen which intersects the virtual photograph in perception as above.  However, affection also produces the transition between [perception of] external movement and action, not as a physical transition, but more as 'a "movement of expression, that is quality, simple tendency agitating an immobile element"'(38).  This is seen best of all in the human face, 'a relatively immobile surface on which are registered motor tendencies, expressions of pure qualities that suggest the connections between incoming perceptions and outgoing actions' [reminds me of all the bleating about faciality in Thousand Plateaus].

All three kinds of movement images are found whenever living beings with their centres of indetermination interrupt flows of matter.  All are connected to the sensori-motor schemas.  Perception-images involve a selective interpretation of the surrounding world, action-images locate perceptions in a concentric world based on the individual, affection-images produce motor tendencies following the absorption of external movements, and their representation as qualities.  Affection-images in particular unite subject and objects, sensations and perceptions, and produce movements of expression.

In Cinema 2, Deleuze offers a summary of this material.  He says there is both a vertical and the horizontal axis in ' the "system of the movement image"' (39).  The vertical axis includes 'the immobile cut, the mobile cut and the open whole'. The horizontal axis relates to the three movement images.  On the vertical axis, we see differentiation between the open whole and a closed set of actual objects, with its notion of movement as limited and spatial.  On the horizontal axis, we see a process of specification—the development of three species of movement images.  Bogue says the vertical axis also refers to the operation of duration and its constant division and differentiation, while the horizontal refers to the processes of perception without duration.  Together, we have '"a plastic mass, and a-signifying and a-syntactic matter, and non-linguistically formed matter, though a matter that is not amorphous but semiotically, aesthetically and pragmatically formed"'.  This plastic mass can be seen as a plane of consistency of the image, movement, matter and duration.  Initially, it is identified with light, but here, it becomes 'all forms of matter/flow', featuring modulations of all sorts, sensorial, kinetic, affective, rhythmic as well as verbal.  Films are made from this stuff, and so are we, as an ensemble of different images, [perception, action and affection].  Film directors can shape this material through the use of frames, cuts, shots and montage (the vertical processes), and 'long-shot perception-images, medium-shot action-images, and close up affection-images' on the horizontal dimension.

Chapter two.  Frame, shot, and montage

So matter tends to form isolated systems yet they still retain connections to duration.  Movement is the 'intermediary between closed sets and the open whole', because of its mundane and philosophical aspects, although some sets are more closed and immobile than others—an 'ideal limit to the space free of [duration]' (41).  Even here, we can still see an allusion to duration.  We can simplify by referring to closed sets and open wholes, with closed sets further subdivided into immobile cuts and mobile cuts.  We can then grasp the effects of frames, shots and montages, which are present in all films: frames close sets, the shot shows endurance over time or a unit of movement, montage displays even more clearly the open whole of duration.

Cameras always frame a portion of the world, and the results are projected on to a screen, 'functioning as the "frame of frames"' (42, quoting Cinema 1), since the screen always limits the camera frame, whether this is a close up or a long shot.  Sometimes there can this can lead to a disorienting heterogeneity, so in cinematic images 'there is inherent…  a destabilising force…  "A deterritorialization of the image"'[that is, a non- naturalistic image].  Framed images have additional features:

  • they are more or less saturated with content or information;

  • they can limit images either 'geometrically or dynamically' (43), with examples of the latter as iris shots [vignettes], or even variable screens that open and close;

  • frames both separate elements and unite them, and again this may be geometric or dynamic—'German expressionist configurations of diagonals and triangles' on the one hand, and 'images of fogs, fluids, shifting shadows, undulating shapes and metamorphosing forms' on the other.  In both senses, we are illustrating multiplicities, with the latter, qualitative multiplicities 'not divisible or indivisible, but "dividual", the closed set of elements dividing itself in each moment into a qualitatively different set of elements' [incomprehensible to me—again, the philosophical viewer does this dividing?];

  • frames have angles or framing, 'a position in space from which the framed image is shot…  [an]…  Implicit point of view', which may be naturalistic or not—some can even be unsettling, out of the narrative, involving "deframing" [not Deleuze but someone called Bonnitzer, 43], and they require reading or interpreting;

  • frames include, but also suggest an "out of field", which, according to Burch, may be 'above or below the frame, to the right or left, in-depth away from the camera [for example behind a closed door] or toward the camera and beyond it in the audience's direction [as in direct addresses to camera?]'.  Deleuze thinks that we can even establish an absolute out of field, relating to duration again: 'every framed set of elements may be included within a larger frame…  Until an ultimate frame of frames includes the entire universe' (44) [reminds me of the last scene in Men in Black] , although this would still not actually represent duration which can 'never be "given" as such', but only manifested, often 'in a disquieting way'.  Normal or relative out of field effects strengthen the conventional notion of three dimensional space and an abstract time, and great efforts are made to achieve continuity between the immediate and its out of field [realist editing].  However, some images still suggest something beyond, some absolute out of field.

Shots can do this.  It is important to realize that for the French the notion of a plan or a plane incorporates both the English sense of the shots and the take, spatial and time relations.  What we would call different sorts of shots are called different sorts of plans for them—for example 'plan d'ensemble (long shot)'.  In this way, for Deleuze a number [multiplicity] of spatial determinations can produce temporal perspectives [so it's all done by linguistic sleight of hand again], and quantitative multiplicities are seen as connected with qualitative ones, especially if the entire film can be seen 'as a single plan-sequence' (45).  Such shots can produce the two faces of movement discussed above, [mundane and philosophical as I have called them]—as the elements alter, so qualitative change becomes apparent [so the famous deep shot in Citizen Kane which shows the suffering wife trying to kill herself in the foreground changes meaning as we see the manipulative Kane in the background?].  This shot in this sense is intermediary between the frame and the montage.  Deleuze cites the sequence in Frenzy, where the camera follows a man and woman up a set of stairs to a door, then backs away as they go in only to climb the wall and look in through the window [a marvelous technical achievement, down to the crew].  This camera movement shows that something is happening in the whole as various modifications of the interaction take place: a meeting turns into a murder.  This particular change is expressed as a narrative, but narrative is not the only unifier: in general, the shot decomposes and reunites elements into local configurations and patterns, sometimes those of contraction and expansion, sometimes following some organising figure—a straight line in Rope, a spiral in Vertigo.

[French] shots or plans are movement-images, and mobile cuts of duration.  In cinema we see 'movements disengaged from bodies' (46); it's the camera that moves and movement takes place between shots and in montage.  Camera movements are themselves mechanical [can be non-naturalistic again as in crane shots], pointing to some general notion of movement, 'the pure movement [abstracted] from bodies, one that takes on an existence independent of any specific character or point of view', despite the efforts of realist directors.  Even realist editing produces an idea of movement 'that to some degree always escapes the bodies from which it issues' (47).  This is often supplemented [and normalised] by mobile cameras, which for Deleuze means that fixed shots in early film illustrate pure movement best.

In the plan or the mobile cut, elements and dimensions change and produce a [mathematical] set, 'like a cubist painting'.  There may be no common denominator [although there usually is, in realism].  Plans themselves may offer incommensurable points of view.  This ceaseless variation is what makes the film different from photography.  Films show modulation not just passive registration.  Such modulation is a temporal perspective for Deleuze.  In this way, cinema '"acts like a consciousness"'(47), developing a nonhuman or superhuman point of view [which usually matches exactly the all-knowing narrator in literature?].  This links with the insistence that consciousness is in things, including the initial 'prehensions of brute matter', and only better actualized in living things.  Again, the universe is a metacinema so that cinema shows us this consciousness in things in a version we can recognize, a camera consciousness, autonomous and able to develop non-human notions of movement [it all depends on the slippery definitions again].

Montage is already implied by the plan [the example here is the extended tracking shots of Magnificent Ambersons, where a long tracking shot accompanies one couple as they move to meet to another couple which we then follow—a tracking shot replacing an actual montage.  The whole of Slackers is assembled like this].  Montage deliberately composes the whole from movement images.  It is usual to see montage as guided by an overall plot , theme or concept, but for Deleuze it is the form of time and type of continuity ('narrative,  motivic or discursive' (49)) that unifies. There are four tendencies in early film making: 'American organic, Soviet dialectic, French quantitative and German intensive'(49).

The organic tendency is found in the films of DW Griffith.  His notion of an organism includes distinct parts, binary pairs such as 'rich/poor, male/female, North/South'.  These appear as parallels, alternating in the montage.  These parallels can converge, through accelerated alternation.   Close-ups indicate particular aspects of wholes.  Overall we have '"the alteration of differentiated parts, that of relative dimensions, that of convergent actions"'.  Deleuze argues that narrative derives from this conception, and indeed individual narratives themselves can be composed, to depict time as a whole, a series of spirals [the example is Intolerance, which apparently involved narratives from different civilisations, ancient and contemporary life—'such as that of the Babylonian chariot chase and the modern American automobile - train pursuit' (50).  There is also a depiction of time not as contracting towards convergence, but as dilating and generative, depicted by a bird's flight or a rocking cradle.]

Soviet dialectic as in Eisenstein or Vertov [and others I've never heard of] was discussed in Eisenstein's own theoretical writings.  He admired Griffith but saw the formal alterations between narratives as accepting social divisions through the parallel structure, above all the struggles between rich and poor.  Instead we need something more dialectic.  For Deleuze this is still organic, but it is now a matter of organic growth, with the shots as cells.  However, they have to be juxtaposed in a form of conflict or collision, the collision of opposites to produce a higher synthesis.  Eisenstein thought that this process should be guided by a notion of a 'Golden Section', an allegedly classical ratio of parts to whole, 8:13, deliberately unequal and unstable ratio.  Battleship Potempkin shows conflict in every level, between shots, between sequences, to structure the plot as a whole—five sections each with unequal parts to follow the ratio.  The overall effect is meant to show the transition from quantity to quality, an explosion following culmination, like the one between water and steam.  Apparently, the moment of culmination was called pathos, leading Deleuze to talk about a pathetic jump, where the second instant still incorporates the first one.  The Golden Section can be seen as a spiral, and pathetic leads are straight lines that short circuit the arc, like strings to a bow. 

Thus we have continuous organic growth, yet through discontinuities of forces that produce qualitative changes.  In opposition to Griffith and his controlled montages [in convergence and the like], Eisenstein offers qualitative leaps.  Close-ups can also be used to indicate these leaps because of their markedly contrasting size and their ability to illustrate pathos in the sense above [the sudden realization of the destructive intention of the guard on the Odessa Steps registered in close-ups?  The sudden realization of their political power on the faces of the sailors?].  The spiral itself does not just relate parallel montages, but generates a synthesis of conflicting parts and qualitative leaps.  Deleuze says that in this structure time appears differently, in the interval between the shots as well as in the whole.  The same ideas are found in other early Soviet films, despite stylistic differences [one example is Pudovkin, who also shows dawning political and class awareness of the subjects].  Vertov makes a genuine contribution by adopting a more radical affirmative notion of the dialectic of matter, found in machines and landscapes and buildings themselves.  He particularly develops the 'camera consciousness'(53), as in The Man with the Movie Camera.  This depicts an alternative to organic composition, a more mechanical one as the interrelation of movements, captured in the eye of the camera. [see also Kino Eye]

There are French and German alternatives to organic composition too [as usual, they include lots of early directors whose work I do not know, although I have come across Jean Renoir, Jean Vigo and Jermaine Dulac for the first group, and Fritz Lang for the German expressionists].  Deleuze sees the difference as one of quantitative or intensive approaches to movement respectively—when considering the open whole, it is a matter of 'Kant's mathematical and dynamic sublime, the one infinite in number, the other infinite in power' (53).  For the French tendency, the quantity of movement and the metric relations between movements produce a mechanical
composition of movement images [the examples are obscure to me, but one turns on the depiction of a Spanish cabaret dance scene by L'Herbier, where the dancers lose their personal differentiations to allude to dancer in general and dance in general—'the dance becomes a machine' (54)].  Indeed, there is a centrality of machinery in the French school, both the automaton and steam or combustion engines, linking energy heterogeneity, the mechanic and the force.  In Vigo's L'Atalante, the puppets'movements extend the patterns of action beyond, with the camera looking from behind the puppets to the pedestrians and to the puppets' reflection in the shop window.  There is no dialectic unity as in Soviet cinema, rather the engineer appears as a kind of '"soul" of the machine', an expression of passions (55).  There is also 'a fascination with water', the mechanics of fluids. 

Apparently, the early directors were particularly interested in what they called '"photogeny"', things that look particularly good on film: they decided the above all the issue was mobility, and they conceived this again quantitatively, as a combination of different factors.  It might be a rhythmic motion, it might be a matter of optimal speed, which may be acceleration or slow motion.  The relevant factors include 'lighting, space, camera angle and distance, the objects/events filmed, the interrelated movements of objects and camera' (56).  Montage was to be decided by the optimal rhythm for a particular configuration. We can understand this in terms of the maximum quantity of movement relative to a particular set of elements, but this relative notion also alludes to the maximum movement in duration, giving the movement image two faces or two sides, one facing relative the other absolute movement.  This is like Kant's notion that mathematical numbers do the same, alluding to infinity, and sublime is linked to beauty in the same way.  There is even a mathematical sublime 'something beyond measure of the senses but graspable by reason' (56), and this was to be the inspiration for the French school, illustrating material movement, but the alluding to 'a conceptual, mental whole' (57) [the depiction of something emerging from the interactions of individuals in the dance?].  The aim is to show 'simultaneous co-presence of temporal movements', leading to a general interest in French art in simultaneity, and resulting in triple screens [at the time?] or 'superimpressions'—apparently, Gance shows Napoleon's face over layered images of a schoolboy snowball fight.  Images are present at the same time rather than linked in time, alluding to a different time, 'spiritual simultaneities' (58).

With German expressionist cinema, Deleuze acknowledges the usual features, like sharp contrast of light and shadow, diagonals, oblique angles, themes of madness, hallucination and so on, but he argues that all these 'arise from a single conception of movement and light', based on some 'Gothic line of non organic life and Goethe's theory of colour'.  The break with organic system leads not to mechanism but to 'gothic non - organic vitalism', [almost an élan vital?], that pervades all life and even animates objects—hence the depiction of automata, or buildings that seem to have a life of their own, or the nervous energy that seems to be reflected in diagonals or zigzags.  This principle 'constructs space rather than merely describing it'(59), and produce a prolongation of normal limits and new convergences, 'junctures of accumulation'.  This is seen best in the way in which light is managed, a matter of particular interest for Deleuze, as we saw above.  In the French school, there is an allusion to some luminous matter, the luminous grey, around which light oscillates, while in German cinema, light becomes more intense, with light and shadow separate, 'infinite forces in perpetual conflict'.  For Goethe, light is invisible and can only be grasped when it encounters shadows, so that all the colours of the visible world should be seen in terms of opacity.  In German cinema, like varies in its 'intensity in relation to darkness'(60), with solid objects revealing it as a presence.  Beneath these contrasts, 'the intensity of non organic life moves in zigzag patterns, delineating trajectories that guide the montage links between shots'.  Light and shadow are 'vectors that interconnect objects…  atmospheres, and people' which seem to assume a life on their own.  Again, it is duration that is being expressed in the various intensive movements or degrees, expressed in particular in effects of light.  Going back to Goethe, colours like 'yellow and blue are movements of intensification', a matter of adding or subtracting shadow and light.  The midpoint is 'a reddish - purple', seen as some essence of colour or brilliance, the most excessive of the visible colours, so brilliant that it provides excess sensations. 

In German film, the effects of brilliance are displayed as 'scintillation, glistening, sparkling, fluorescence, phosphorescence, shimmers, auras, halos' (61).  [The only example are recognized are the burning circles in The Golem, but there are apparently phosphorescent demons heads, blazing heads, or silhouettes as in Nosferatu].  These depict '"a terrible light...a flash of the infinite"', and 'intimate the presence of an infinite non organic force'[which apparently can also be destructive].For Deleuze, this is an example of the dynamic sublime produced through overwhelming force, the same effect as witnessing volcanoes or waterfalls, which both frighten and thrill, because we can overcome our fear and realise that we are superior to nature.  We might also be able to get some notion of spiritual destination, where everything is consumed, but also we become aware of nonhuman life, usually rendered as the divine part in us [shades of Deleuze and stoicism].  This is not the emergence from mechanical movement as in the French, but an notion of uncontrollable dynamic force emerging in such a way as to disengage from all previous degrees, culminating in a single point, detaching itself from normal time as well.

Although Deleuze discusses different kinds of montage and shocked in considerable detail, his main point is always that montage refers to duration, available in the present as a contracted moment, but alluding to the infinite.  This may be an organic infinite built from an accretion of parts, it might be a qualitative leap through the dialectic, and numerical unity hinting at infinity, or accelerating intensity.  Deleuze is specific discussions should be seen as leading to 'the implicit conception of the whole that informs them' (64) [as we saw best in the discussion of the specifics of German expressionism].  Intention is everything, and this helps us distinguish differences even when similar techniques are used.  For the same reason, the usual claimed differences between shots' and montage are less relevant.  Overall, the shots' or plan is particularly special, both offering specification by the development of different images [perception action and affection] and also alluding to the closed se/t open whole relation on the vertical axis.  The next step is to see how particular signs develop from these early analyses.

Chapter three.  Eighteen Signs (more or less).

The uncertainty arises because Deleuze means different things by a sign [!].  Deleuze admires Peirce as the one who classified images best, particularly signs, although this is only a general influence, and Bergson is more central.  Peirce at least develop some non linguistic theory unlike Saussure, and instead of signifier and signified, Peirce has 'a triad of representamen - object - interpretant' (66) [and we know that Deleuze wants to oppose linguistic semiotics and its privileging narrative -- good discussion in Cinema 2, and part of the general rant against linguistic imperialism in Dialogues].  Narrative depends on the common sense space time based on effective human action, the pragmatic world, the sensory motor schema.  Deleuze uses Lewin's term 'hodological space' [interpersonal space as I recall].  Modern cinema has abandoned the schema though.

Peirce still privileges the linguistic and the cognitive, however, seeing signs as permitting associations of one sign with another so that knowledge can develop.  Also, the triad is not completely useful in explaining the three kinds of image in Deleuze, which shows Bergson's continued influence.  Nor does Peirce referr to images, but rather to '"phaneron" for "that which appears"' (67).  Deleuze discusses Peirce's notions of firstness, secondness and  thirdness in terms of three kinds of images—'"something that refers to nothing but itself, quality or potential, or pure possibility"…  "Something that refers to itself only through something else, existence, action-reaction, effort-resistance"…  "Something that refers to itself only in relating one thing to another thing, relation, law, necessity"' (67, quoting Cinema 1).  However, these qualities are supposed to be displayed by all phenomena and all things, and they coexist in even the simplest perception for Peirce [all perception includes perceiving qualities, oppositions and 'a minimal degree of abductive inference (Thirdness)'.  Nevertheless Peirce can be used to extend Bergson's types of images, and Deleuze notes parallels between the affection image and firstness, the action image and secondness: to tie up with thirdness, Deleuze has to construct the notion of  'relation - image', while seeing the perception - image as something outside of Peirce's classification, '(a Zeroness)'(68) [as in a basic starting point?].  We now have two additional types of movement images as well: 'one midway between the affection-image and the action-image (the impulse-image…), the other between the action-image and the relation-image (the reflection-image…)'.  So we now have six types of movement images. 

It seems that the insistence on perception as a starting point would have helped Peirce in deducing the three types of his images through an analysis of perception, rather than just announcing them.  However, Bogue says the point of the six images is to bring Bergson in more centrally—the notion of consciousness as a centre of indeterminacy leads to a notion of perception in living images as a filter and a selective response, compared to a notion of universal perception, the action and reaction of non living things.  Thus perception images arise from this field of universal perception, and it also functions for living beings to provide an interval or gap—thus it has two aspects or poles, depending on whether we relate it to a universal perception and relations, or to filtered ones.

The perception image actually has a triple quality—a genesis, a composition and function in consciousness as an interval, and the function in relating to the whole.  It is also more basic than the other movement images, in a way because we need to perceive affections actions and relations as well.  This involvement produces the same kind of triple quality in all the other movement images.  This lies at the bottom of Deleuze's notion of the sign, which is quite different from Peirce's—signs relate to types of images either in terms of their dual composition or in terms of their genesis [the discussion on p.69 is pretty dense, but still MUCH better than the original].  So we can now tabulate the six movement images and their related signs [and two useful diagrams appear, PP. 70-71].

table 1

table 2
In classic cinema, perceptions appear as subjective points of view, contrasted with objective images, things viewed from the inside and from the outside, with a pov inside the action or outside.  However objective images often shift in and out of the setting, and thus sometimes sub seemed entirely within subjective pov.  In this way, the cinematic image can be semi subjective, representing an anonymous pov [the classic all-seeing narrator].  Deleuze wants to borrow this notion for his discussion of free indirect discourse [apparently originating with Pasolini].  This term is also used to describe the insinuation of the narrative voice into the reported or direct voice of the characters.  Deleuze follows Bakhtin in seeing this as introducing heterogeneity [not a hierarchy as MacCabe would argue?], producing an assemblage of enunciation.  [I recognized one example, Antonioni and Red Desert where the director immerses himself into the protagonist, reproducing her world but also bringing in his own aesthetics, even making the camera noticeable: 'characters enter and leave the frame but the shot seems to continue too long…  betraying an insistent, obsessive camera consciousness' (72).  Similarly, the shift of perspectives can reveal the action of the camera, especially if the subject's pov is supplemented in some way, or contradicted.  In free indirect discourse, we have the characters and a narrator, to represent this subjective and obsessive camera pov, but they are not strictly separated, and both can be indicated by a single shot—there is no autonomous pov, but rather '"a correlation between a perception-image and a camera-consciousness that transforms it"'(73, citing Cinema 1).  Where we have a perception in the frame of another perception, we have a dicisign.  Such signs reveal the basis or structure of every perception image, as perceptions of perceptions, affections, actions and so on.

Deleuze alters terms to make subjective perceptions those where 'images vary in relation to a central and privileged image', as when living images introduce indeterminacy, while objective perceptions are ones 'in which images vary all in relation to all', as in constant unmediated interactions.  This brings about the special usage, so that what we normally think of the subjective images, like dreams or hallucinations are really to be seen as objective perceptions, where we see 'undulations of a single vibratory flux'.  The dicisign therefore is reserved to a perception image that belongs to a single living image or centre of determinacy, while that which perceives things themselves becomes  a reume (based on the Greek term which refers to that which flows), one which is no longer confined to the living image but which can show the flow of matter—often depicted in French cinema as images of water or fluids.

A third type of perception image, the gramme, engramme or photogramme goes further, developing the notion of the cinematic eye able to perceive all the points in a space, all actions and reactions.  It is a 'gaseous perception'(74).  Showing how 'universal matter flow' relates together all the molecular images [specific objects].  Is therefore seen as a genetic sign, compared to the sign of composition in the dicisign [God this is dull OCD type philosophy, Deleuze at his worst. Like R-W in Boundas says --who has ever actually botherd to try to apply any of this stuff? The whole thing is ridiculously arbitrary and unreplicable], and these are opposed in that the first one is universal, and the second one related to the subject.  The reume lies between, closer to the dicisign because it still is a sign of composition, although it offers a more free floating pov, but one which can still be grasped by common sense: the gramme breaks with common sense.

An example of the gramme is Vertov's kino-eye, which intended to escape all commonsense notions of time and space,and perceive without limits.[So Vertov claimed himself, especially for TMMC].  Deleuze thinks significant the discovery of the genetic element of the image, or the photogramme, the single still image becomes 'a generative cell of movement', especially when Vertov reanimates stills, makes them run on, or combined slow motion with normal speed or even reverses.  It represents 'movement compressed into a micro interval'(75), and this is how the film can go beyond human perception to the genetic, to focus on the single element of energy, a molecular perception, suggesting that molecules can interact fully and freely—akin to the difference between solid, liquid and gaseous states, with the dicisign as the solid, reume as liquid, and gramme as gaseous.

[Pause before going on to affection images...]

Affection images live in the interval between perceptions and actions, often located in specialised functions such as particular body parts—in humans, the face.  Deleuze borrows Bergson in seeing affection as containing a motor tendency following sensation, as in facial movements of expression, produced by contrasting the immobile surface and the active features.  In portraiture, it is possible to emphasise either of these, and Deleuze goes on to discuss a philosophical notions of 'the passions of admiration... and desire' (77)—admiration offers a maximum of expression with a minimum of movements, while desire offers a series of small movements which can escape the calm appearance of the face, and threaten a qualitative shift.  Hence intensive desiring faces express the power to change qualities, while reflective faces display qualities common to several objects [this tedious and waffling discussion culminates in another table, page 77.  I have not reproduced it for fear of losing the will to live].

The close up is the classic way to show the face in cinema, and for Deleuze this reveals the face in abstract, it 'deterritorializes the face' and changes it from its conventional functions of individuating, socialising and communicating between two persons (78).  The face becomes autonomous, a communicating surface, and thus an affection-image possessing both an expression [an expressing capacity] and that which is expressed.  Faces become icons, and each icon has both 'trait'and a 'contour' pole.  Icons of both types [limits presumably] both indicate that they are signs of composition, revealing intervals between one quality and another on particular faces, or between multiple elements on a reflective face as above.  Affection appear as in a pure form, and this is what Peirce refers to as firstness, pure qualities without context.  Deleuze goes on to distinguish the actualization of real connections as opposed to 'virtual conjunctions', beyond normal space and time.  Although faces are the most obvious examples, close ups of any object can reveal its pure qualities, so once more 'affect…  is not strictly human' (79), and we can see, for example in a close up of the knife of Jack the Ripper, qualities that are interacting while impersonal.

Dreyer has developed a close up to become a flowing close up [like a zoom out], still displaying a lack of depth and perspective coordinates, flat images.  This is another way of making affect visible for Deleuze, and he also finds this in Bresson [Pickpocket, which I have not seen]—apparently, a space is depicted in a fragmented form, with incomplete parts of objects and no common measure, which disorients.  This is the 'any - space - whatever' (ASW)—a '" singular space"' which has '"lost its homogeneity", opening the possibility of an infinite number of linkages or connections, a virtual conjunction (80).  Affect appears via the decontextualizing space and raising 'yet - to - be - actualised possibilities' [delirious and overheated, a kind of Stendhal reaction to film instead of tourist sites].

ASWs have their own signs to express their quality-power—the qualisign '(or potisign)'(80).  These correspond to the gramme in making visible quality-powers in a decontextualized space, but are fully beyond common sense and the sensori-motor schema.  Again these are genetic, showing how local configurations are actually composed.  ASWs can be constructed by using shadows to break the contours of objects, through '"lyrical abstraction"' (81) as in Dreyer and Bresson, or where the goal is to contrast actual and virtual to illustrate limits of the one and the possibilities of the other.  In Bresson, the actual space is normally white, black represents impotence and grey our indecision, but sometimes the protagonist can '"choose to choose"' and the colours become components of a new world [must track this through Jeanne D'Arc]: this spiritual decision constitutes affect.  Colour is the third possibility, and colour qualities can absorb objects which can decontextualise them—the deserted landscape as an absorbent.  This leads to two kinds of qualisigns—'of disconnection and of vacuity', although the one always implies the other.  However, overall, the ASW '"no longer has coordinates, it is a pure potential, it exposes only pure Powers and Qualities independently of the states of things or milieus that actualise them' (82).

[Pause for a while again, dear reader...I made a cup of tea and let out the cat]

Between the virtual ASW and the actual milieu of action, is another domain, one of impulses, drives or instincts.  Hence the impulse - image, usually associated with naturalism and found in films by Von Stroheim, Bunuel, and Joseph Losey.  It can be seen as if human beings are being portrayed as animals driven by natural instincts, but for Deleuze, these directors showed that 'the primordial world of drives and forces' is 'immanent to the real world of concrete particularities', an originary world formed of rough unfinished forms or fragments, or basic energy dynamisms, 'a kind of primal swamp from which the material world arises and an ultimate garbage dump into which all matter eventually passes' (83).  Impulses develop entropically [as in The Exterminating Angel—marvelous film!], as a kind of law of gravity.  That originary world becomes apparent in swamps, deserts, or bourgeois drawing rooms when they can't escape.  [As we might expect], actualized spaces are both mundane and pointing to the virtual, and films can depict these in a kind of continuum as in the degradation of the drawing room.  Thus originary worlds mediate between ASWs and actual milieus, tied to the actualised world, but capable of showing how it is derived [and the example is the surreal bits in naturalistic films as in Bunuel].  Impulses mediate between affects and actions, embedded in real situations  as 'protoactions'[the example is the fetishized sex drive of the old guy in Diary of the Chambermaid that manifests itself in continually making the same proposition about boots to the different maids].  Humans can display animal qualities as a result of their impulses, so can fetishism: these in turn can become good or evil, either relics or '"vults or voodoo objects"'(84, quoting from Cinema 1).  Thus impulse-images are represented as symptoms or fetishes, both representing fragments, although Deleuze goes on to argue that symptoms belong to the originary world, while fetishes appear as fragments tied to or found in the actualized world.  Whatever the actual specification, the originary world becomes 'the genetic elements from which the impulse-image arises' (85), just as the ASW is  for the affection-image.

[Another pedagogic pause before we move on to action-images.. This is hard bloody yakka.]

The action image 'has the greatest affinity with narrative' (85), but again narratives presuppose 'configurations of movement images'.  Here, we have not originary or virtual contexts, but 'determinate milieus and actual behaviour'.  The domain here is realism, actualised qualities and powers, and affects and impulses actualised in behaviours.  Indeed, this is how Deleuze defines realism—incarnated [sic] milieu and behaviour, and the various relations between the two.  Actualised milieu also curve around living images, just like their consciousness does, so there he is in realism a notion of global synthesis, all those forces configured together that affect the living image and instigate actions and reactions.  This corresponds to Peirce's secondness, where some things appear through reactions with other things.  Deleuze preserves the notion that the actual therefore implies two elements, 'dyads, oppositions, reactions and resistances', between elements in the milieu and with one's self.

The large form becomes the SAS [see my notes on Cinema 1]—the structure of situation, action and modification of the situation, an hourglass convergence of broad situations narrowing into specific actions and then expanding again.  The milieu in question is designated by a synsign, an ensemble of forces ('qualities-powers') actualised in a milieu.  Actions are associated with the binomial sign, to indicate the duel, oppositions and so on within the self.  Together these compose the large form action image.  Examples include documentaries such as Nanook of the North struggling with his hostile milieu, westerns where the landscape is the milieu within which actions take place, but there are several other examples including King Vidor, Hawks and Huston.  There are five laws which describe the workings of the large form:

  • The depiction of the milieu such as configuration of the landscape as a series of impinging forces and settings, and how this curves around particular characters, the 'rhythms of the milieu's "respiration''' (87, the alternation between interior and exterior, 'primary and secondary situations, panoramas and close ups'
  • The passage from milieu to action, as synsigns become binomials, a 'climactic duel', often through 'an alternating montage of interrelated actions' [the example is the final hunt for the killer in M]
  • The '"forbidden montage"', where convergent forces eventually meet in a single shot, a scene where 'the conflicting forces are simultaneously present', the final shootout in the western
  • The '"nesting"'of one duel within another [my example would be the series of conflicts their sherriff needs to undergo in High Noon—with his neighbours, with his wife—before he confronts the bad guys]
  • The 'law of the "great gap"', between the surrounding milieu and the eventual action, as the hero gradually realises that he has to do something.

There is [almost inevitably] an additional third sign, the stamp or impression—the milieu 'impregnates' the character and this finally explodes into action.  The impregnation is done through certain emotional objects, which can be introduced in various sequences.  Deleuze admires Method acting which show this link between internal emotions and the milieu, through a 'discovering an inner emotional analogue to an external situation and establishing a link between the two by manipulating and milieu object (or glove, knife, ball, etc.)'(88).

As well as SAS there is also ASA, and this produces the notion of the small form, as an equivocal action clarifies the situation that permits a new action.  The differences can be seen in the detective film as opposed to the crime film, which starts with fragmented actions and tries to explain them in terms of an underlying situation [the example is The Maltese Falcon].  There are other reciprocal versions of the costume drama, or some of the early documentaries say of Grierson as opposed to Flaherty.  Comedy also depends on 'the equivocal nature of the characters actions'[examples include misunderstandings based on misleading identities, or ambiguous actions' being revealed to be something different, not shaking with grief, but shaking a cocktail in a Chaplin example].  Such signs are called indexes, following peirce—'a sign of the existence of recurrence of some concrete singular entity or state external to the signs—smoke, for example by an index of fire' (89).  Sometimes what is indexed is a lack or a hole in the story, something that needs to be gradually filled in.  There are also indices of equivocity, which point to different directions as it were, with shaking, above. 

Both indexes of lack and equivocity are the 'two signs of composition of the Small Form action- image'.  The genetic sign is the vector, and we see this in the small form's 'skeleton space', the framework of elements, the underlying wire form [to use a modern notion].  The example, apparently is drawn from Chinese painting [with which I am fully familiar, of course].  The fragments and appearances communicate with each other even now they are heterogeneous. Deleuze obviously admires the zigzags and indirect connections, which he clearly wants to line up with all that tosh about the line of the universe.  The examples include the construction of 'a skeleton - space' (90) in Mizoguchi, with its slow pans and tracking, through an across houses and neighbourhoods, sometimes with a dissolve with a flow across to a field of forest.  The camera movement is the 'line of the universe' joining heterogeneous elements, constructing a series of spaces each with their own dramatic intensity, each affecting the character.  One example familiar to me is Ugetsu, where the camera follows the bewitched potter and the demon princess from a palace room into a forest, then into a poor, then it tracks across rocks and over open ground dissolving to a shot of the soil of the garden, and this, apparently links three spaces with different 'rhythms, trajectories and moments of dramatic climax'(91) [not all how I read it I must say!  I could see some connections being suggested, but I read these as mundane episodes in an overall depiction of a relationship.  Left to their own devices, non philosophical readers will always impose their own naive common sense realism]

In the small form of the action image, actions themselves are linked to situations but also to themselves.  We may experience this through narrative and story, but some filmmakers avoid smooth transitions [the example is the westerns directed by Anthony Mann], and focus on the intensity of present situations and follow 'the shortest path'[not the straight line as we know] to link two actions in ASA.  Vectors usually traverse skeletons spaces, but not always [there is not always a spatial linked, but sometimes a direct one between action in different local spaces].

[Pause again before encountering the reflection image]

The reflection image is also a transformation image, between action images and relation images, and allowing , for example the transition between the large and small forms of the action image.  Reflection produces signs called figures which show these transitions and transformations.  The reflection image can offer an indirect link between large and small forms, and can be considered as a subspecies of the relation-image, although it remains tied to action images, and thus takes on an intermediary form between action and relation.  Examples are drawn from Eisenstein's Ivan the Terrible, part 2—Ivan's bodyguards act as an assassination as a pantomime, and this 'prefigures a future action'(93), as a necessary link between the situation, the real plot to assassinate Ivan, and the subsequent action where Ivan tricks one of the plotters into impersonating him.  This performance can be seen as  an index of
equivocation, in the sense above.  The whole sequence links the large form SAS to the small form, since the action in the middle indexes the small form and introduces a small form into the large.  [Another Eisenstein example follows, but I have not seen the film—again, a small action acts as a kind of metaphor or index, so that the small form produces a larger sequence, or an action becomes an index in order to trigger off subsequent action in the small form].) Deleuze wants to discuss these possibilities as two kinds of sign—'sculptural/plastic and theatrical/scenographic', indicating these two forms of relation as above, the jerk.

There are two additional reflection image signs—'figures of inversion and discursive figures (or figures of discourse)' (94).  We thus have 4 signs which can be categorised, playfully, Bogue assures us, by linking it to someone else's taxonomy of figures.  Thus the plastic figure becomes the same as metaphor metonym in and synechdoche; the theatrical figure is akin to allegory, hyperbole and irony;  inversion in literature is often done for dramatic effect [the example is 'to the battle he came'], and the same goes for cinema; the figures of thought 'entail no change in the words themselves, but a change in the way one thinks about them' (94, with the bizarre example of prosopopoeia, where dead or imaginary people are represented as speaking).  In the second cinema book, these four classifications becomes three, divided into 2 signs of composition, plastic and theatrical become a figure of attraction, and the
inversion figure remains, and 1 genetic sign, the discursive [for the purposes of philosophical tidiness, no doubt].

Inversion is illustrated with reference to Hawks, where landscapes cease to be englobing milieus and become mere functional settings, non organic and inactive, and where to disruptive challenge comes from the interior.  Outside and inside our inverted.  Other inversions include male female, and adult child, high speech low speech, and love and money.  These are persistent enough, Deleuze thinks, to take on the role of an autonomous figure.  The discursive figure is illustrated through Chaplin's later comedies, or where the burlesque of the small form, as in the equivocal action discussed above persists in, say, similar moustaches between the Jewish barber and the dictator, but the barber's discourse at the end of the film makes explicit that it is the social worlds that are all important.  As a result, the situations in the film take on a new dimension and significance, even small ones, such as the twitching of the moustaches.  In this way, the small form of the burlesque takes on the larger form explaining how social worlds are all important in the creation of dictators or liberty. The discursive figure is also genetic, because it directly refers to or reflects its proper object, even where this is the relation between action and situation.  The discursive figure makes possibilities manifest.

The obsessive categorisation is less important [!] Than the general processes of transformation within action images, which can include a transformation between the forms, but also of internal transformations, sometimes in the form of an emerging question that must be solved in order to be able to act [the example is The Seven Samurai, where the enduring question is what is a samurai these days].  If the hero doesn't recognise the question, but simply response the immediate situation, he usually perishes [Japanese Macbeth misinterpreting the witches prediction].  If he does respond, he can emerge into the larger milieu, becoming an imprint, linking action and situation.  The question remains tied to the situation, however, and Kurosawa specialises in showing how situations develop into these big questions.  An alternative is to stretch the small form, as in Mizoguchi, constructing skeletons spaces and using the camera to link them in a way that suggests 'a vision of an unlimited cosmos with a "very special homogeneity"' (99), had joined by the line of the universe.  In his films, it is the women that transgress and links the spaces, and often get punished for that, and this is the limit of the action image, when hierarchy reasserts itself and breaks the line of the universe [although it seems that Deleuze is arguing that order is never fully restored, and some disorientation remains—if the sensori-motor schema breaks down, movement images can dissolve].

[A pause before the final bit on the relation-image—bear with me, nearly there...]

The relation-image is akin to Pierce and thirdness, 'the category of continuity, regularity, habit, rule, law, interpretation, representation, and thought,, where first and second are related (99).  In this way, relations become intelligible, through a process of the sign mediating between objects and interpretant and this is the basis of all intelligible relations, all notions of law or of development.  There is a universal tendency to regularize relations, moving from indeterminacy to regularity, of generality, a kind of 'habit forming'(100).  The origin of habit lies in mind, so that matter is simply rigid habit that seems to have escaped the ability to form and reform.  Scientific laws similarly are 'rigidifying habits', part of a general process of generality or evolutionary development.  This generality requires connecting two things by a third, for example some rule or formula.  All signs display these three part characteristics, for Peirce, [and active semiosis develops them as understanding?].  The sign itself is best understood as a relation between objects and interpreter, or actually between 'representamen (the sign vehicle), an object and an interpretant (that which interprets)'.  Every interpretant can become a sign itself, pointing to another sign, leading to an open process of interpretation or '"infinite semiosis"'.  This open process is guided by the generalising tendencies of habit, however, signs help to establish these habits [by being applied to different objects via analogy and so on?].  There is therefore always a mental element connecting actions, and that is what thirdness is, the notion of a relation.

This helps Deleuze develop the notion of the relation-image, as 'the mental image, a "figure of thought" in which the mental is introduced into the image'[at last, human consciousness is given a proper role!] Mental activity is in place in the other images as well, as in consciousness [awareness?] in an affection-image, or the intentions in action-images, but relation-images directly refer to '"relations symbolic acts and intellectual feelings"'(101).  We can now treat perceptions as interpretations, signs as mediated by other signs in a whole chain, actions as necessarily including symbolic elements, of law for example.  [And we come close at last to the idea that this is what separates human action from behaviour].  Even affections are interpreted by using logical conjunctions, or other intellectual notions of relation.

Hitchcock develops relation-images best, since everything has to be interpreted or deciphered, as signs 'force themselves on characters'.  This is registered on the faces of the characters, where expressions disclose intellectual notions of relations, such as 'the affects of "if", "hence", "although" and so on'.  All the action in Hitchcock is structured around these mental relations, what Hitchcock apparently called postulates, sets of relations which then have to unfold.  This is what made Hitchcock refuse the title of the whodunit, which plainly has an author of the action.  What is distinct to Hitchcock therefore is a representation of thirdness, where a criminal act emerges from a set of relations, including accidental meetings on trains, or young men in Rope trying to impress their teacher, and accidental involvement in a murder in Rear Window.  The characters can unravel these relations, but above all, the camera does it for the audience—the audience must be engaged, in the famous 'suspense'.  The relation itself is indicated in a relation-image, displayed by figures of thought as signs.

There is a difference between 'natural relations and abstract relations', however.  Leading to a new classifications of relation signs.  Natural signs emerged from habit, 'customary or ordinary mental relations'.  Abstract signs involve non habitual connections, and the deliberate construction of a whole rather than a series.  The signs of composition are 'the mark and the demark'(103), the first one being the natural relation, located in the customary series so that they can be easily interpreted by being members of a series [same old objects, easily classified].  Demarks break with series and unsettle habit—the crop duster inexplicably turning to attack the hero.  However, the disruption depends on a series of ordinary marks [doesn't seem very different from the ambiguous object discussed above].  The genetic sign of the relation image is the symbol, 'the sign of an abstract relation that constitutes a whole', as when a ring symbolises a marriage [looks like a metonym to me].  Terms are to be compared in a nonhabitual natural way, and initially, any term might be included: Deleuze argues that this indicates the possibilities of all mental relations, so the mark and demark become special limiting cases.

Overall then, we have 14 signs of the movement image, or possibly 23 [one difference is whether you distinguish between icons, all the different kinds of qualisigns or fetishes].  Whether or not there are specific differences doesn't matter [! forgiving Deleuze's OCD] , because the taxonomy is supposed to just generate new terms for talking about cinema and seeing.  It is the underlying concepts and logic that matter, the mesh of dyadic and triadic relations, which together form 'acentred, rhizomatic combinations' (104).  It is not the sign that interests Deleuze primarily, but rather images, which are not simply neatly divided into three types of sign, but which possess a wider potential as '"signaletic matter"' (105), that directors [only? not crew, actors etc? Leads to seeing films as over designed, containing only one readings etc?] shape like sculptors: 'Deleuze's taxonomy is merely a tool for inventing a language adequate to those sculptures and the creative processes that generate them'.  We have seen that the main movement images related to particular kinds of signaletic matter—'gaseous perceptions,espaces quelconques [ASW], originary worlds, respiration -/skeleton-spaces, metaphysical respiration -/skeleton-spaces, mental "relations spaces"', and each involve new ways of seeing.  For example 'gaseous perception is seeing from all perspectives at once' [meaningless? All possible perspectives?], with the other kinds of seeing it as more limited.  In ASWs, we see affective spaces, qualities and powers.  Originary worlds help us see 'impulses and energies'.  In respiration spaces we see 'rhythmic contraction and dilation of milieu and action'.  In skeleton spaces, we can see 'both an action world and an Idea imminent within it', such as situations and questions, or limited vectors and unlimited lines of the universe.  In mental relation spaces we can see relations 'within a concrete, tangible world'.  The point is to open possibilities to the imagination, to go beyond what commonsense cannot see—'affects, energies, rhythms, vectors, ideas, and mental relations', and to illustrate that these can be seen in film.

Chapter four Hyalosigns: crystals of time

Movement images relate to matter and its 'signaletic' qualities, and the signs express these qualities.  There is a conception of time also offered, but cinema provides a different notion as well in the form of the time image.  We see this first in 'pure optical and sonic images that break the sensory motor schema (what Deleuze calls opsigns and sonsigns)' (107). We also have memory images and dream images that connect these signs together— mnemosigns and onirosigns.  But there is also the time crystal or crystal image and hyalosigns, which reveals time directly.  There are others too, which are discussed in later chapters.

In classic Hollywood films, there is a notion of 'the seamless and continuous presentation of action within a single time and space' (108).  After the war, the sensory motor schema collapses, and we start to see films like Nashville, with a much looser connection between actions and milieus, gaps in the action with weak linkages and connections [the examples are Cassavete's films], 'an aimless wandering'[Taxi Driver], and the appearance of '"the stroll, the balade [ramble, jaunt] and the continual round trip journey"'.  We also find the creation of deliberate clichés and parody, or 'diffused conspiracies anonymous plots and ubiquitous technological surveillance'.  In the absence of the sensory motor scheme, integration is provided either by 'a network of circulating clichés or a conspiratorial system of surveillance'.  This is still only a negative critique, however and the emergence of a positive time image needs
more than critique and mockery.

The time image emerges in Italian neo realism, with the depiction of optical situations, not linked to action images, such as 'an every day series of motor actions' which lead to the characters suddenly seeing something, 'pure seeing', unrelated to the action—the realities of poverty and misery, or the lives of others in postwar reconstruction.  There is still a narrative, but the characters almost become spectators rather than active participants.  These are opsigns, found in Deleuze's favourite Italian directors.  Distinctions such as those between objective and subjective, real or imaginary are rendered indiscernible.  For example in Fellini there may be memories or dreams, yet these contain 'a practical staging', making the mental world like a rehearsal, while the real world becomes a spectacle [8 ½ is apparently the best example] this is how opsigns can subvert conventions like narratives.  They represent something that exceeds our normal capacities to manage.  They are the opposite of cliché.  This links with Bergson's notion of managing memories through selection, and this can be disrupted when new associations for objects, or ignored characteristics appear.  Opsigns have to be combined with '"growing powers"' (111) that force the entire image to be read, that appear as thought, that open them to revelations.  Opsigns can then develop into chronosigns, lectosigns and noosigns.

Deleuze shows how opsigns appear in memories and dreams, then crystals of time, then 'full fledged chronosigns'(111).  Opsigns are detached from sensory motor schema and have to be connected together differently.  Memory helps us see how this is done, and Deleuze relies on Bergson: present objects are recognized by drawing upon past memories, and this is often automatic, appearing in an action, not necessarily a representation.  Such recognition is guided by  sensory motor patterns that are acted out in ordinary life.  Here, 'the perception image and memory image occur in the same instant' (112).  We can attend more explicitly, to summon up remembered images, through reflection, and mould it to the present object.  The process here is akin to reading a text—assembling meaning and anticipating [which incidentally provides the '"the illusion of being there"', 112, quoting Bergson].  It is the same when we walk through familiar streets.  When we pay full attention, we add additional memory images from wider zones of the past, forming larger circuits to which present objects can be connected.  The same processes of automatic and reflective perception apply in cinema, and indicate that the sensory motor schema can be engaged or relaxed respectively—in the 'pure optical situation', (114) it is suspended altogether, producing a virtual image. 

We can see the circuits between the present and memory as well, sometimes represented on the faces of the characters.  Sometimes the image of the past is ambiguous, however, and can become 'strange and unreal, dreamlike, hallucinatory'.  This can produce confusion, or, more generally, indiscernibilty.

We can see what goes on explicitly in the flashback, say where every action is linked to a flashback.  Again this can be controlled by a conventional narrative, but sometimes, flashbacks become more genuine time images, with a non realist logic, something inexorable, and destiny that structures the past, something beyond determinism and causality.  The examples are Mankiewicz's films, which illustrate for works of time, temporal labyrinths.  All About Eve can be seen as a conventional narrative, but also alludes to 'a bifurcating time', with memories held by multiple narrators, and the depiction of branching paths.  The conventions of Hollywood forbid that these paths should be mutually contradictory, but at least they produce 'unpredictable breaks' (116) [There is a definite sense that history repeats itself, I thought, in All About Eve, but apparently, each of the events in the narrative is an 'improvisatory
moment, a zigzag movement towards stardom', which Deleuze saw as a break with causality—the flashbacks are the only way we can understand Eve's unpredictable [contingent] career path.  Again the conventional reading would be that this is an ambitious, opportunist and ingenious individual, like an entrepreneur]. 

There are conventional flashbacks, which stick to normal chronology, and are selected for their practical value, but we all have experience of dreams, where our sensory motor scheme relaxes, and our ability to access images expands.  It's not surprising that we find dream sequences in films, conventionally indicated by special effects [like dissolves].  Sometimes, the world seems to become the personalised, acting in ways that the subject themselves can no longer do.  Usually, however they are fully integrated back to the normal time space—except in musicals.  In dance sequences, 'narrative time is suspended, objects and people are conjoined in improbable combinations and configurations, and space is frequently metamorphosed as one setting flows into another' (117).  The movements of the dancer become parts of this the wider world, as if in a dream.  In the best musicals, for Deleuze [those directed by Minnelli], normal and dream worlds alternate, producing an indiscernibility between the real and imaginary.

Usually, convention wins out with flashbacks and dream sequences. However sometimes we see the point of indiscernibility itself, time crystals, which are represented by hyalosigns [apparently from the Greek word for glass].  Bergson discusses deja-vu as arising from time perception and memory themselves.  The usual view is that memories are fainter versions of perceptions, but Bergson suggest that in fact a memory image coexists with each perception image, so the actual existence is doubled by virtual existence—hence the idea that the present produces one stream which falls back towards the past and the other of which aims at the future.  We can see this when we find ourselves suddenly unable to do something that we used to do, so that we experience ourselves as two individuals, one performing, while the other one observes, as a spectator.  The performing self is often experienced as being an automaton, playing a role, and this is often associated with deja-vu.  Actually it is produced from the difference between automatic perception, which is largely unconscious, and our activities when we remember, which is reflective, but also passive [experienced as arising from somewhere else].  This also shows us that memory is not just a matter of personal subjectivity, that the past  'preserves itself by and in itself…  as a single domain, and hence as a kind of gigantic memory…  Each mind is inside memory, like a fish in the ocean' (119).  In Deleuze's terms, '"time is the interiority in which we are, in which we move, live and change"'.

The present is double, both 'an actual present perception and a virtual memory of the present', a coexistence.  Usually, the difference is indiscernible, producing an "'objective illusion"'(119, quoting Cinema 2).  This is a real doubling, not something that just goes on in consciousness.  The same combination of actual and virtual images can be found in cinema, a process that Deleuze says is like a photographic image coming to life and becoming independent, even while it took its place back in the photograph in '"a double movement of liberation and capture"'(119-20). Robbe-Grillet has noticed this double movement.  It is a break with the usual view that the object exists independently of the description.  Instead '"crystalline"' descriptions refer to the '"purely optical and sonorous situations, detatched from their motor continuation"'.  This sounds just like the opsign, although there is a deliberate bringing together of the virtual and the actual: this is the genetic element of opsigns and their composition, reflected in the more specific opsigns.

Deleuze seems to confuse objective illusions like mirror images, and the issue of representation.  However, representation only takes place within sensori-motor schemas, and when this collapses, there is no difference between presentation and re-presentation, objects and appearances—all become 'simply images'.  It is like the difference between a conventional optical illusion and a deliberately artistic illusion aiming to combine the virtual and the actual.  The effect is that we are not just like actors but that we become actors, that once the sensory motor schema collapses, all that is left is theatre or spectacle, or play of images.  Thus crystal images in film are reflections 'in the broader sense of the term' (121), not just the usual actual mirror images of photographic images or whatever.  We find crystal images in painting and theatre as well.

The simplest model is when we see an actress looking at herself in the mirror and becoming the actual image, playing a role that is actualized in a film.  The actress and her character are co-present.  The role being played its part of 'a fictional world that reflects the real world'.  Although the fictional world initially becomes a virtual in relation to the actual real world, it becomes actual.  All this is normally understood from within a sensory motor schema, but once that collapses, all these distinctions become indiscernible, 'unassignable in the sense that one can no longer determine definitively the category to which a given image belongs' (122).  When directors deliberately play with these relationships, they are re-presenting reality, offering a world as 'the proliferation of reflections', indicating precisely the 'simultaneity of the virtual and the actual'.

Usually, the directors simplify things by sequencing the link between actual and virtual, so we see a shot of the actress's face, and then the mirror image.  However, this convention can be broken, so that both our images, and one reflection leads to another, in a whole 'proliferation of reflection images', as in the sequence in The Lady from Shanghai in the hall of mirrors.  This is what makes the crystal image differ from a simple mirror image.  There are other differences, in that the surface of the crystals can also be transparent, or filter light, with 'varying degrees of limpidity or opacity' (123), which are affected by their surroundings.

We can also see a genetic process, of crystallization itself, around a seed crystal, perhaps.  There may be different types of crystallization.  Again the seed serves as a virtual that is actualized in a milieu of potential crystallization.  Overall then, there is  '"crystalline circuit"'involving the actual and the virtual, the limpidity and the opaque, the seed and the milieu.  Deleuze intends this to describe entire films, where different facets are explored, and the process of crystallization revealed in an overall 'giant crystal image' (124). [ The example here is Fellini And the Ship Sails On, where the world of appearances on board appear as a newsreel, and cameramen themselves appear in the footage, opera soloists break into song and the crown joins in, grey and sepia tones give way to colour images, cooks and waiters on board dance, while the diners synchronize, and correspondences are established between aristocratic passengers and proletarian crew, between dining room and backstage kitchen, between opera singers in the engine room, between performers and audiences.  The arrival of some rescued refugees becomes a seed crystal for further socio political connections between the ship and the historical world, as a warship appears to demand them back, 124-5]. Fellini uses all the techniques of blending art and life—films within films, films being made, the notion of the world as theatre or as narrative, as a ritual or as dance.  Reflection and image change positions. 

As well as the three figures mentioned above [the actual/virtual, the limpid/opaque, the seed/the milieu] producing three hyalosigns, as components of all crystals, there are also several states of the crystal, indicated by crystal images themselves, again of several types just as real crystals are.  There are four crystalline states in particular, and each refer to time, and the processes of actualization of the virtual in different ways.  Each of these can be indicated by looking at the whole oeuvre of a director [suspiciously neat and showing the real importance of classification based on auteurs]:

  • The perfect crystal, and Ophuls.  The world is a spectacle or theatre, and theatrical sets merge into streets, narrator's play parts, scenes from the past come to life, circular motifs appear, including time as circular [the examples are La Ronde and Lola Montès].  This is the perfect crystal because the actual and virtual images coexist in the same overall development.

  • The cracked crystal in Renoir.  There is the same emphasis on spectacle and the theatre [the example here is The Golden Coach, where travelling actors parallel the real world plot, and there are several shots where theatres merge with real sets.  However, there is still some life beyond, and divisions are restored, producing 'a crack, a line of flight, in the crystal of time' (128).  The Rules of the Game seems to offer a perfect self contained crystal, where the amateur stage links with the larger social stages, living beings are reflected in automata and humans and animals, the aristocrats parallel the servants and vice versa: no escape seems possible, and people cannot leave except by dying.  All are trapped by the rules of the game, past conventions and customs.  In the Grand Illusion and The River, however, some characters do escape convention.  For Deleuze, it is the past that is crystallised, yet it is still split, in the same sense that, for Bergson, time splits into the past and the future.  Despite the fixity of the past, sometimes characters are able to emerge into a genuine life beyond the theatre, leaving the crystal and escaping '"the eternal recurrence of the actual and the virtual"' (129, quoting Cinema 2).

  • The growing crystal in Fellini.  There is the usual material on the world is a carnival spectacle, tableaux and clown shows, where each scene offers an entry to a different world.  The actual entrance to different pavilions 'is the seed crystal in the process of expanding into a milieu', and the scenes themselves accumulate into a larger crystal.  Each film itself shows this accumulation all formation.  In Fellini, it is the present not the past that offers some compelling proportion, and the virtual past offers an escape—childhood memories, for example.  Deleuze insists that Fellini does this 'to create a dimension of coexisting pasts' (130).  So the compulsive present and the conserving past interfere with each other.

  • The crystal in dissolution in Visconti.  Here we see the self-enclosed aristocratic domains, which include offering life or spectacle, dances, the family dinner and theatricals and ceremonies.  The musical score emphasizes the 'ceremonial nature of this rarefied world' (131).  However there is a process of decomposition that undermines them, decline and decadence indicated by events outside, such as wars.  There are suggestions that these worlds might be reconnected to real worlds, but this always happens, sometimes as a moment of clarity, too late, and it is this notion of too late that provides the dimension of time.  According to Deleuze, only art can grasp this, just as in Proust.

Overall then, the sensory motor schema collapses, at first in terms of clichés or conspiracies, and then pure opsigns and sonsigns emerge, a first in things like Italian neo realism.  Normal cinema tries to use flashbacks and dream sequences as attempts to depict these signs, but only time crystals fully illustrate the possibilities, and only when they develop do points of indiscernibility emerge fully, and the three hyalosigns appear.  Then various crystalline states or crystal images also develop.  The films discussed are also conventional in some ways, having coherent narratives, for example, but Deleuze does not see narrative as important compared to overall visions such as the world as reflection, spectacle, theatrical and so on.  This is a whole way of seeing, and, for Deleuze, it 'issues from a particular conception of time' (133).  A few directors have particularly seen the possibilities, and depicted both reflections and also particular ways that time links the virtual and the actual in the present.  'The narratives of their films issue from this conception of time'.

Chapter five.  Chronosigns: the order of time and time as a series.

Hyalosigns show images of time, and its differentiations into two streams.  In crystals, there are circuits of exchange between the actual present and virtual past, rendering the difference indiscernible.  With chronosigns, we find time images that combine 'past and present,  virtual and actual', but in order to develop a commentary on the true and the false, rather than on the real and imaginary.  There are two kinds, that depict the order of time and time as a series, and in both cases the true and the false are shown to be 'undecidable or inextricable'(135), through showing that different times can coexist, or showing a sense of becoming, as a series of powers, respectively.

For Bergson, the virtual past is 'a single dimension in which all past events coexist' (136), and each actual present has a virtual double.  We can think of this as a cone, remembering that 'each present moment is a contraction of the past, the concentration of the entire cone in the point of its apex'.  We locate ourselves in this virtual past and find their different planes of consciousness, cross sections of the cone or sheets of memories.  On those sheets, there are some points with particular affective tones depicted with particular brilliance.  Deleuze refers to sheets of the past as geological layers or strata, each one with its '"tones,…  aspects…  singularities…  brilliant points ... dominants"'(137).  As a result, we find in the virtual past notions of a non-chronological time—the '"pre-existence of the past in general" (the cone as a whole), "the coexistence of all the sheets of the past" (the cross sections), and "the existence of a most contracted degree" (the apex)'.  The virtual past therefore contains all the aspects of time—the dilated past, the contracted past, and the future projected past.

However, the present also has some interesting dimensions, and apparently Deleuze refers to the work of the phenomenologist, Groethuysen, who sees the present as a meeting point of future actions and past facts, 'a "dialectical nunc [Latin for 'now]'"'.  However, there is also 'an "intuitive nunc"', outside the notion of succession, 'like a dramatic scene... an indivisible whole'.  This scene can take place in a short period of time, as in a sudden event, or a long one, including an entire childhood, but it appears to us as having no dimensions, but being a' unified lived experience'(138).  The discussion goes on to say this is a bit like a narration when boring bits can be skipped, an experience of time when nothing happens. When we relive this experience later as a personal memory, we have to disengage from the present, or de-actualize it [in the sense of standing back from the actuality].  This again gives us a sense of empty or dead time when nothing happens.  Deleuze likes this idea of deactualization, but not to describe how consciousness works [Bogue suggests the dialectical and intuitive nuncs can be better understood as 'an actual present and a virtual present', 138].  Instead, this temporary suspension of time helps us see that the present is a point on a timeline, integrated in a sensory motor schema, which we become aware of when we are acting pragmatically.  We realize that we can grasp the present from within, through an optical vision, not in the usual horizontal or chronological way, but rather vertically.

The event has variable dimensions, incorporating an episode or a whole life, but it also shows non-chronological time as well.  For example, single events in the present often allude to 'the simultaneous coexistence of multiple presents' (139), when we realise that past events are also included [the example is the event of finding a lost key, which can be considered as 'three different present moments within the same event…  a "present of the past" (having the key), a "present of the present" (losing the key) and a "present of the future" (finding the lost key)'].  Each of these presents are implicated or folded in the same event, and can be seen as individual peaks or apexes. 

However, in actuality each of them must be compossible: in time as a whole, however, [in fiction] there can be incompossibles, rendering events inexplicable '(unexplainable in rational terms)', and incapable of being unfolded.  This gives us two ways of looking at time, 'as "the coexistence of sheets of the past," or as "the simultaneity of peaks of the present"' (140)

This gives us two sorts of chronosigns, one showing the aspects or layers of time, and the second showing the accents or peaks.  Robbe-Grillet shows the accents.  He produces 'generative fictions', where we see the process of generating the story [there is a good example on Ubuweb], variations of the same scene, or in other words so that 'incompossibles abound' in the same story.  These are not just comments on the fictional nature of the works for Deleuze, but 'evidence of a different conception and presentation of time'.  The various scenes are the peaks of the present, the different sorts of present as in the lost key example, all folded up within the film as a whole which is 'a single simultaneous present'. 

In Last Year at Marienbad, we see these peaks of the present, but also sheets of the past. Robbe-Grillet himself says that the film is about the reality which the hero himself is constructing, and which he attempts to persuade the lady is correct, with the different scenes best understood as 'various story options' (141).  However, for Deleuze, the lady 'seems to leap from peak to peak in a perpetual present, whereas [the male] explores multiple sheets of the past, seeking out the "brilliant points" of each memory space', and these two conceptions have been produced by the collaboration between Robbe-Grillet [script] and Resnais. Robbe-Grillet apparently agreed afterwards that Resnais imposed more continuity, to please the audience by not depicting abrupt shifts.  Deleuze says two conceptions of time were responsible for their differences.  We can see the differences in the different reactions of the
male and female characters as well.  Resnais's continuities are not chronological but depict a 'malleable, non personal virtual past' where they interlock and overlap.

Conventionally, sheets of the past are taken to be psychological variations of a chronological time '(personal memories, fantasies, dreams, etc.)' (142), and this is the basis of the conventional reading of Welles's films.  Deleuze again sees something different.  Citizen Kane can be seen as a standard biography with understandable variations in personal memory, but Deleuze sees it as an 'extended search within a transpersonal memory space for the elusive Rosebud', which belongs to no individual, and is revealed as a unity only by the camera shot at the end [I think the shot imposes a meaning as well privileging Kane's].  The stories of the past of Mr. Arkadin can also be seen as partial reminiscences of encounters on different sheets of the past.  The Trial could be either a coherent narrative of a dream, but also 'a quest within the past for the source of K's guilt, each site a strange sheet of the past'.  The deep focus shots show characters in regions of time, showing interaction without cuts, and forcing the audience to interpret 'actions transpiring simultaneously in the various planes of the image' (143).  Welles's deliberately makes foreground middle ground and background communicate and interpenetrate, in order, according to Deleuze, 'to make time visible'[the example is the suicide scene, where Kane in the background indicates Susan's immediate past, and the pills in the foreground Susan's future, in the form of "an invitation to recollect"].

Deleuze sees in the deep shot Bergson's notion of what happens when we remember: we leap from the present into the dilated past of memory, and then we explore specific planes on which specific recollections are located.  Welles's shows us not memory images themselves, but the process of summoning up memory by exploring these planes.

Welles anchors sheets of the past around a single contracted point, but Resnais lets the present itself floats and become uncertain and indistinct, with no secure mooring for memory.  Thus Marienbad shows a 'memory world' for two possibly incompatible memory spaces, while in Hiroshima, the characters are struggling to 'construct a hybrid…  memory space'(144).  [In other examples, or group memory emerges showing undecidaball alternatives, or the juxtaposition of different "ages"]. Thus for Resnais, memory is not just psychological and personal, but composed of sheets of the past, 'a supra- personal memory cone of past events' (145), traversed by planes of variable integrity and flexibility.  Individual characters may have elaborate biographies, and Resnais was in the habit of drawing diagrams of the environs and movements before shooting, but he was actually mapping and combining these individual sheets, according to Deleuze, using objects to indicate mental relations, or rather mental worlds 'that are apersonal' (156).  This involves showing not only multiple sheets of the past, but continuities between them, not just parallels, but perpendiculars and obliques.  Resnais apparently hoped that depicting the characters' feelings and thoughts would indicate these continuities [not just generating a straightforward empathy from the audience].  Sometimes the music establishes a 'continuity of mood', but also a disjunction with the visual images [as a further distancing device?], and there are sometimes voice overs.  Resnais apparently claimed that it was the feelings not the characters that mattered, enabling him to produce 'mental figures, elements available for a thought that interconnects various sheets of the past' (147).  The technique can be seen as 'hypnotic dedramatisation', to reveal feelings and then thoughts.  The feelings map the sheets of the past, and thought enables us to transverse them: the feelings correspond to ages of the world, while thoughts depict non- chronological time, for Deleuze, a continuity across elements on different planes [maybe]

Chronosigns blur the distinctions between the true and the false.  Conventional narratives follow sensory motor schema and it is this that helps them claim to be true [realistic].  If we abandon common sense and the sensory motor schema, however, 'time appears directly', in the form of '" deactualized peaks of the present…  virtual sheets of the past"', and normal conceptions of space and time are 'immediately subverted'.  There has always been a relation between [conventional sequential]  time and the notion of truth, as where future events can come to falsify present predictions [in philosophical hands, this shows that 'either the impossible proceeds from the possible (since what was possible yesterday becomes impossible today) or the past is not necessarily true (since yesterday the [event] could not have taken place) [after all]' (148).  This is what leads Leibniz to develop the idea of compossibility].  If we can depict incompossibles in the form of coexistent peaks of the present or sheets of the past, the problem reappears [this time as art] , especially in the form of '"not-necessarily true pasts"'.  In those cases, narrative can be false [or a 'power of the false'].  We can describe the possibilities of each peak clearly, unlike when we use hyalosigns which blur the real and the imaginary, but 'the peaks cannot all be true at the same time —and yet they are so enfolded with one another that they cannot be separated', and the same goes for interconnected sheets of the past.

Narratives [that do not obey commonsense rules about time] have a power of the false, and this can appear in another kind of chronosign.  This one indicates the power of time, by experiencing events from the inside, as participants, where we experience the simultaneity of past and present, for example [haven't we done this above with the nuncs?] .  In these circumstances, the succession of moments appear as 'a single becoming', and Deleuze says this indicates not the sequence but a series.  In a series, the power of time flows through empirical moments and transforms them in turn [connects them as a series?].  This power can appear in the chronosign that shows becoming or series.  Again, becoming can transform events, removing their "true" identities [from the point of view of common sense] and showing the power of the false.

Again Welles's films show this power of the false for Deleuze, through the construction of 'Nietzschean figures' (149)—'the man of truth, who in the name of the ideal judges the world of appearances guilty; the man of vengeance, who no longer believes in the ideal but negates the world of appearances out of self hatred; and the artist, a joyous forger or falsifier who creates value by affirming the becoming of the world'.  For Welles the man of truth 'is at best a naive dupe', or 'an obsessed agent of the law'(150).  The man of vengeance has to adopt trickery and deception and embrace 'the becoming of shifting, metamorphic appearances', but does so negatively out of self loathing [I didn't recognize any of the characters in The Lady from Shanghai].  Only the artist fully embraces becoming [the example here is a Welles film I have never heard of F for Fake], about a real forger and someone who faked a biography of Howard Hughes: what these show for Deleuze is that art is not just fraud, but rather 'the full expression of the power of the false': while forgers are actually limited by having to copy a particular form, artists can depict real transformations.  In this way, 'the transforming metamorphic power of the false produces truth'[I love it how philosophy becomes so witty!]

Deleuze argues that Welles's characters are not just representations but 'forces in action, images engaged in a process of becoming that transforms them from mere sequences into series' (151).  The same sort of transformation of characters is found in the films of Rouch, used by Deleuze to describe how the power of the false gives us direct images of time.

Rouch's documentaries are inseparable from his anthropology, and the films are always produced after prolonged ethnographic study.  Rouch knows that his presence is a disturbance, but he also invites the people being studied to explain what they're doing, 'as teachers initiating the ignorant anthropologist into the wisdom of their ways'.  Apparently, Flaherty had shown rushes to Nanook, and Rouch did the same [as in respondent validation], but then went on to involve the people being filmed in the process of filming itself.  Sometimes, people appearing in a film have proposed that they make another film, and this is the origin of 'the "ethnofiction" Jaguar', depicting the adventures of three young inhabitants of Niger [Nigerians] as they migrate to Accra [see  a clip here].  The three involved play the leads but they also helped 'plan and stage the action', although the shooting was 'largely improvisatory', taking advantage of chance encounters.  Sound had to be recorded separately in those days, so Rouch discussed the images with the participants as a source for the commentary.  Rouch had already done fieldwork on Nigerien migration, and the performers enact their own past, helping to invent 'a story that sums up the truth of a group experience' (152).  Deleuze discusses this as an example of 'the "function of fabulation"'(152, quoting Cinema 2), neither fictional nor factual, but rather a particular kind of becoming real, which involves telling legends about one's self and inventing one's people openly and explicitly on camera.  In Jaguar, a fictional story combines with historical experiences to form a 'new collectivity that emerges through the process of making the film'.

This is also evident in Moi, un Noir [clips here, with English subtitles – Part 1, Part 2], again a suggestion from a research assistant who had viewed Jaguar and wanted to make a film about real migrants like himself.  This time, the African players adopted pseudonyms: 'Edward G. Robinson, Eddie Constantine - Lemmy Caution, Tarzan, and Dorothy Lamour'.  The performers watched the rushes and developed free-form commentaries, combining 'narrative descriptions, reminiscences, anecdotes and fictions'.  Again we have participatory ethnography, collaborative effort, but not grounded in a traditional past, so we can see clearly  a 'process of self - invention, melding western cinematic personas and Nigerien roles in a hybrid urban identity', echoed by the soundtrack which shows the performers constructing selves. [It is a welcome counterpart to the view of African youth as alien to ourselves. One thing though -- was it really necessary for Dorothy Lamour to get her tits out for the camera?]

For Deleuze, this is a matter of '"becoming - other", a metamorphic passage between identities' (153).  This is clearly depicted in Les Maitres Fou [full film,another example, and a short biography of Rouch  here, written English commentary here].  The particular cult, Hauka, is based on past rituals, but the Nigeriens who migrated to Accra developed it.  We follow the mediums from their daily lives into a compound of the high priest where they become possessed.  The spirits are 'parodies of colonial authority', and they mock the British.  They speak 'a mixture of pidgin English and broken French'.  They end up back in Accra in their apparently normal lives.  Rouch argues that he is not just recording or representing activity, but provoking it, provoking a new truth, outside of normal reality.  'In front of the camera, people adopt unnatural roles, become uncomfortable, lie, and invent'.  Roche also says that he distorts time, contracting or extending it, speeding things up, distorting the people all the story, and then makes the usual claim that this delivers a deeper reality.  In Maitres, the camera provokes the becoming-other, in Jaguar and Moi, it encourages self invention.

Rouch admits that he also becomes other, trying to leave behind his western self.  He also tries to immerse himself in shooting the film, trying to give the camera itself some autonomy.  'In this "cine-trance" the filmmaker is 'no longer just himself but… a 'mechanical eye' accompanied by an 'electronic ear'"'(154, quoting Rouch in Principles of Visual Anthropology, ed.  Paul Hocking, 1975.  The Hague:  Mouton, pp 83--102) [shades of Vertov discussed earlier].  He deliberately employs native speakers to record the sound, and together, this produces 'a kind of machine-component of a transcultural sight-and-sound apparatus, immersed through a cine trance on the reality he is provoking and producing'.

More blurring of the real and the fictive can be found in improvisatory films, in Shirley Clarke, and John Cassavetes.  Clarke combines drama, improvisation and documentary film in The Connection, and combines actors and real people in Portrait of Jason [sounds a bit like Ken Loach].  Cassavetes builds from improvisation, or combines it with scripted material [the examples are Shadows, Faces and others].  Deleuze sees this work as demonstrating another power of the false, offering a corporeal image of time, where the body itself forms a series from gestures and attitudes combining before and after an producing 'an entire "cinema of the body"'(155), which includes work by a number of experimental filmmakers including French new wave and post new wave directors [I'm afraid I have never heard of Akerman, Eustache,Doillon or Garrel].

The body displays attitudes and gests [Christ what an obscure discussion that is in the original!]  It is good at depicting befores and afters, as the effects of fatigue or anxiety, for example.  Ordinary bodies can be filmed to display movements, and so can stylised bodies, taking part in ceremonies, and displaying 'the gests, organising patterns, that interconnect its various attitudes' (156). The term gest comes from Brecht, and refers to both gesture and gesticulations, which convey character and sum up social relations [the example is seeing a man chasing away a dog as somehow conveying the idea of workers escaping the employers' watchdogs].  Deleuze develops this to mean a tie or knot of attitudes which are coordinated together and show the development of attitudes themselves, as '"a direct theatricalization of bodies, which is often very discreet, since it takes place independently of any role"'(156).

Antonioni shows the ordinary body ['the quotidian body'], usually fatigued, disquieted or despairing. Bene [pass] shows ceremonial bodies with parodic movements.  And it's possible to show the transition between the two kinds, where ordinary attitudes or postures become gests.  The example is Clarke's depiction of the addict's body [in The Connection], which clearly displays past present and future behaviors and traces of them to embody addiction, but then bodies merge into body parts, 'the anonymous hands of the addicts', and this develops into a pattern or gest, to indicate 'an apersonal junky' [a lot of fuss about the conventional notion of symbolisation in film for my money].  Cassavetes' films are semi improvised into relations between the actors, where spaces become less important: the actors themselves generate the space determined by their activities.  Actors construct themselves as well [I think this is on about getting into role, and then taking on some pedagogic function].  The interactions embody attitudes [which just about says all you need] and also changes in attitude and emotions.  For Bogue, the patterns that emerge become 'the kind of "mega- gesture," or gest' (158), as scenes become cohesive and emergent.

Godard shows the fourth power of the false, to impose a series on any aspect of the image, the characters, their states, their positions, colours, categories and so on.  Godard first establishes the categories then shows how they relate.  The categories are often based on classic genres—war movie, B movie, musical comedy, crime story.  These genres are sometimes subdivided or combined, and Godard wants to draw attention to the conventions and reflect on the genre [the example is the strangely accidental and emergent dance in Une Femme est une femme, which apparently shows that dance images become a series that point towards non dance and draw attention to the moments before and after.  Goddard explores and reflect upon all the categories in cinema that help us to locate images.  Sometimes the categories appear as chapter headings, sometimes they are implicit [Les carabiniers apparently is about the categories of war , the ideas, feelings and so on].  Multiple categories might appear, and the audience is then challenged to see what links them together.  Categories get new meaning after insertion into series.  To reflect on images, we have to make these categories visible or audible, and that's what the series does, showing how conventional depictions of fear, for example turn into anger or excitement.  The use of colour is the same: it's not just a symbol or theme, but we are to read them as a series tending towards a limit.

 Godard also offers a cinema of the body, because all these categories get grounded on the body and the way it moves in space.  Here, there is a particular kind of gest linking categories of the body and categories of the mind.  The example is Prenom Carmen.  Godard draws upon the original story, the opera and the film musical to highlight generic categories of fiction, opera and musical and put them into a series.  Certain genres are parodied.  The notion of film itself, as a category, is depicted by contrasting it with video, and there are, apparently, allusions to notions of life and death, language and prelanguage.  Yet we can also read that film through the bodies of the characters and how they link together and form patents, including corporeal gests.  Sounds and music play a role here in including non corporeal elements—we see performing musicians where postures and movements correspond to music but also postures and movements of the other characters, producing 'a trans- personal corporeal - musical gest' (162) which gets extended as additional sounds are added, sometimes aligned with visual images, sometimes not.  The mixture of sounds suggests a link between the bodies and the category of music, and encourages reflection on what music is as it moves towards its limits of the non musical.

Overall, there are chronosigns that refer to the order of time and those that refer to a series, and this in turn relates to two interpretations of time as a sheet of the past or as simultaneous peaks of the present.  Directors vary in their depictions, as we saw between Resnais and Robbe-Grillet.  Chronosigns invoke the power of the false, which reveals the potency or force of time as becoming.  Images become series, and series depict a becoming.  There are four different kinds of manifestation of the power of the false—characters in Welles, the process of fabulation in Rouch, the becoming of the body in Clarke and Cassavetes, especially the passage from attitudes to gests, and Godard who merges the cinema of the body with a cinema of categories, whereby any element can become part of a series, and this '"introduces reflection into the image itself"'(163, quoting Cinema 2).  This notion of reflection will lead into the discussion of thought signs in the next chapter, but all chronosigns are also noosigns in that they call for thought, and also lectosigns that require a non commonsensical interpretation or reading

Chapter six.  Noosigns and lectosigns: image and thoughts, sight and sound

Chronosigns are also noosigns and lectosigns, because the time images makes us think about and actively read images.  In this section though, the questions are explicitly raised about the relation between cinema and thought, how images affect us and how our mind affects images—and then what is thought, and then how do we produce thinking images.  We also ask how time images should be read, and how sight and sound relate in the cinema, especially when their signs 'become autonomous' [that is depart from realism.  Again I suppose all this was really radical once, when we only had naive realism and no-one had really criticised it, at least on popular film courses?]

Faure had already seen that cinema can present evolving images that head towards 'anonymous and collective modes of production'(165-6), because it seemed to offer a kind of mechanical art, a nonhuman eye which went beyond human perceptions.  He [was among many who] thought that these images would directly affect our mind and present some immediate union between the material and the mental, displaying a new universe imposed on '"our intellectual automatism"'(166, quoting a French edition of the Faure's 1964 work) [so Deleuze wasn't original here?].  The idea was that generated moving images will directly shock the mind and produce new ways of thoughts, destroying the automatic nature of our usual mental habits [Benjamin thought something like this too—and Adorno was very skeptical --see The Essential Frankfurt Reader ]. For Deleuze, our own more mundane automatism produces another kind—the spiritual automaton.  This is taken from Spinoza who argued that the soul also has its own laws of cause and effect and acts as a kind of immaterial automaton: this is a response to Descartes who argued that animals are automata, mechanical bodies affected by fixed physical laws, completely unlike human beings.  Spinoza thought that mind must also display laws of causality equal to physical laws.  Deleuze stresses the involuntary nature of thoughts in response to images, as in the bit about not being able to escape the shock that arouses thought when in the cinema [Cinema 2] .  There are also suggestions that this thought aroused by the image is alien, an other 'as remote from our ordinary human world as a wandering mummy or a robotic machine' (167)  [which is why most normal people simply reject the encounter, as an affront to commonsense in Bourdieu's terms, quite the opposite of being forced to think]

We can see this effect in Eisenstein, where montage shows collision and conflict between and within shots, but something emerges as a result—a concept, the organic totality.  Seeing these collisions shock us into becoming aware of thought and also trying to think this totality.  There is also a movement from thought to image, as in Eisenstein's pathos, where we invest images with 'emotional and sensual intensity', what Eisenstein calls sensual thinking.  This is some kind of residual from myth but also an undisclosed structure of logical thinking, 'a kind of "inner monologue"' (167), depicted best in the inner monologues of [Molly?] Bloom in Ulysses.  This thought has a logic of its own as it works with the sensual images, including 'synechdochic, metonymic and metaphoric substitutions of one image for another' (168) [compare with Freud on the dream work?].

The sense of a whole has to emerge because it 'cannot be specified except through its parts', although it can be felt and thought in non-logical terms, including pathos.  One image associates with another as in the logic specified above.  In Eisenstein's Old and New, 'an image of the gush of milk is followed by an image of a fireworks explosion', and the viewer is required to experience the shock and also attempt to link these two images, through unconscious concepts, producing a secondary affective shock.  In this way, colliding images 'impels thought to form a concept of the whole, some notion of the single entity of which the individual shots are parts -- say the concept "cosmic celebration of the workers triumph"'(168-9—Bogue invents the quote himself). [In practice,the culture indiustry does these links really well and we mimic that?]  The audience has to see what looks like arbitrary sequences as 'expressions of some preexisting whole…  as much felt as thought, a kind of dreamlike, intuited affective totality'(169), and this gives pleasure to the film viewer [as we would say these days]. So the films have a circuit that go from sensory shock produced by images to conscious thought, then thought that operates through figures, which then produce an additional emotional charge.  Although this is usually represented as a temporal sequence, in fact the movements coexist.

Eisenstein actually wanted to create revolutionary consciousness and for Deleuze this arises when image and thoughts are identical.  Worker-viewers should be able to experience some aesthetic relation between humans and the world, and realize the importance of their labour as part of the natural world.  In Potemkin, for example, correspondences are established between nature and humans, as in the shot of the mist in Odessa Harbour which 'expresses the collective mood of sorrow and mourning'.  Eventually, workers will see themselves as collective subjects, and nature '"the objective human relation"'(170, quoting Cinema 2).  Images of nature produce notions of collective consciousness, action in thought.  The ordinary sensori-motor schema, where individuals relate to their environment, become elevated to 'the utopian level of the collectivity's oneness with nature as a whole'(170).  Although Eisenstein advocates a particular kind of dialectical montage, these ideas can be found in classic cinema generally, so the generation of critical thinking is widespread: audiences experience critical thoughts as they experience the impact of images, and 'hypnotic thoughts' going the other way: both merge to produce an action thought.  All this is coexisting in the movement image [but isn't it all acted out and thus domesticated in ASA and SAS sequences that solve the puzzle for the viewer by asserting that real men have to do what they have to do? Deleuze only looks at arthouse or radical films, of course -- the Seventh Seal, not Bill and Ted].

When the unifying sensori-motor schema collapses, the world becomes intolerable, especially its '"quotidian banality"' (170, quoting Cinema 2), that the world is clichéd and hollow, that life is a parody [the ennui of the philosopher between books] .  Opsigns and sonsigns appear and this represents the unthinkable.  Normally, in cinema, an inner monologue links the images in a series of figures [including the visual equivalents of metaphor, metonym and synechdoche], but this also depends on the organising power of the sensori-motor schema.  Now we become aware of the gaps between images, where images end or begin.  The interstices are explicable by the images in classic cinema, but not in modern cinema. 

Modern filmmakers deliberately start with an image and then produce an interstice, not associating but differentiating, establishing a difference of potential in order to produce the next one in the sequence.  It is a process of addition [the much admired 'and' sequences].  The point is to select images that are different  then to produce a generative gap.

Sometimes the interstice can be a black or white screen, sometimes deliberate discontinuities between sound and image, sometimes 'false continuities..." Irrational cuts"'(172).  Irrational is used here in the sense that a number can be irrational if it can cut a rational number line in two 'without itself being represented by a discrete point on that line', hence Deleuze sometimes describing it as  "point-cut". Images joined by such a cut are re-linkages 'subservient to the cut, instead of cuts subservient to the linkage'.  Such discontinuities can exist in classic cinema, but they are always subsumed back into common sense coordinates. And not all images in modern cinema are linked in this way, as we saw with discussions of connections establishing the various signs.

In classic cinema images are linked along a horizontal axis, using association, resemblance, contrast [and the general movements of classical semiotics], separated by a rational cuts, 'a coherent space time'.  There is also a vertical axis of integration and differentiation, connecting parts to wholes—the whole can unfold itself in a sequence of different images '(differentiation)'(173), or be expressed in those images (integration).  Here some 'unifying, internal self consciousness' provides integration, and the 'coherent external world'provides differentiation. We saw above that classically [audience?] thought moves from image to concept (critical thought) and then from concept back to the image (hypnotic thought) ending in an identity (action thought).  In modern cinema, images are independent, and not expressed in a chain following a sensory motor schema and are then linked 'according to their differences from one another'.  Normal self consciousness cannot integrate them, and although they conform to the differentiations of the usual notion of the external world.  Any connection has to come 'from a pure outside', and they 'instigate' thoughts beyond common sense.  In Deleuzian terms, irrational cuts replace the conventional horizontal axis, while the vertical axis now includes an outside beyond experience, and an inside deeper than usual thought.

The pure outside reveals itself in the interstices between images, offering a non commonsensical notion of 'and' or of constitution.  The interstice expresses both horizontal and vertical axes, therefore.  The open whole that is alluded to in classical cinema becomes powerful outside 'passing into the intgerstice'[and see two helpful diagrams below, PP. 174, 175].  This power of the outside 'has a direct effect on thought'.  In classic cinema,  the shock of the images can force us to think of the whole, but eventually, we can recuperate the shock through the sensory motor schema, as in the action thought.  This is not possible in modern cinema because logical rational thought itself is rejected and we become aware of a new power, one that now has to be grasped with 'genuine thinking'(176).  We have to think of the outside as a fissure or crack [Deleuze likes those], something distant and so far ungraspable by normal thought—there is '"unthought inside thought"'[silly philosophical way of saying what we normally think of as thought is insufficient], and we must develop 'a "thought outside itself"'(177).  As a result, we lose the sense of ourself as a monologic thinking self.

axes of modern

axes of modern

All this suggests some alien apersonal thought, 'the thought of a spiritual automaton' (177).  We could represent the spiritual automaton is a robot or computer, or but Deleuze chooses the mummy, as in films by Dreyer [Gertrud].  What counts is what the spiritual automaton actually does, though. Apparently in Deleuze's examples, [Dreyer, Rohmer's Perceval, Bresson] characters can sometimes recite their own lines of flatly, as if in the third person, 'enunciating a kind of free indirect discourse' instead of their own particular speech.  More generally, modern cinema aims at a visual equivalent, '"free indirect vision"'.  Normal monologue is disconnected and so is the vision which breaks from the normal points of view and can become 'a nonhuman seeing—a floating eye, a prismatic eye, multiple eyes'.  Different ways of seeing are also represented in framing or composition of images.  These are
the activities of the spiritual automaton.

The spiritual automaton is inside the viewer, but also in the images, depicted on the screen with these various multiple points of view and so on.  What we see depicted is some notion of mind imainent within images, a new 'topological space like that of a Klein bottle…  [Or]…  a Moebius strip' (178).  Deleuze wants to draw on 'some recent brain research' to suggest that neural circuits also form a topological space, connecting the surfaces of the brain's folds, but it would be a mistake to see the spiritual automaton as an individual, rather we need to think of 'a topological brain world…  A single "noosphere"'.  This is the sense in which the brain is the screen, made 'patent' by Deleuze's discussion of the image as '"the point cut, the relinkage, the white or black screen"'.  The point cut is a manifestation of the outside, appearing in the interstice, the relinkage is the new connection, the 'and', unpredictable, and non commonsensical.  Sequences can be seen as 'a Markov chain, a sequence of chemical reactions in which each reaction is a probabilistic event that affects the next reaction without specifying which one of a limited set of possible reactions it will induce' (179).  The black or white screen makes the interstice visible, and it is topological because it is a space in which the outside and inside merge [and there are some blather about how it merges the opposites of conventional cinema as well, the negative and the positive, the full and the void, the past and the future, the inside and the outside, and '"the brain and the cosmos"'].

When sensori-motor schemas collapse, the normal relation between humans and the world also collapses.  In the Eisenstein action thought, a new relation is established, but otherwise the world seems alien, unbelievable.  The task for modern cinema is 'to make possible a belief in the world'(179) [that is the world as philosophers, and great directors who are also philosophers, see it].  This introduces a notion of beliefs into the new kind of thought that is required, and Deleuze goes on to discuss beliefs as a matter of problem and choice.  Thus some of devices already examined, such as chronosigns of sheets of the past or the powers of the false can be seen as 'a kind of problem…  [not]…  a theorem ... [which is]... a closed set of elements capable of an axiomatic systematization…  [But]…  an event that intervenes from the outside and thereby constitutes the terms of a subsequent analysis' (180) [that is not a mathematical problem but an empirical one?].  Deleuze gives as an example explaining the passage of a plane through a cone, which we can do so by describing in terms of geometrical figures, although he prefers to see it as an event, creating various figures as it passes.  This is how we get choices—which figures should we generate to explain this passage?

Beliefs involves choosing to choose, living as one who can choose, and exercise freedom, and this can be experienced as a risk.  It involves a leap of faith, but also a stoical acceptance of the results [we choose whatever fate provides us with, which Bogue sees as Nietzschean] (180).  Chronosigns often depict events which appear as problems, and they imply a choice 'and the affirmation of that choice is an act of belief'.  These events are interventions of the outside, the irruption of non rational thought [although obeying the rules of deeper thought].  This is how they establish belief in this world, although we have to see this world in non rational terms in the first place, as 'the free indirect vision of the spiritual automaton'. Thus:

  • Classic cinema leads us from image to thought and then to a concept of the unifying whole, still seen as inner monologue or sensori-motor schema.  Modern cinema, however proposes new relations to thought, linking to an outside that inserts itself between images, and replaces interior monologue by free indirect discourse and vision.  This breaks the normal unity between humans and the world and must be replaced with a set of beliefs [in the ultimate external world as really beyond rational control, but which we can choose to accept as a kind of stoical freedom, so, by sleight of hand, necessity becomes freedom].

  • There are horizontal and vertical axes that link together images in rational sequence and integrate them into a whole, in classical thought.  We thus have two classic noosigns, indicating rational linkage and rational integration and differentiation.  These are different in modern cinema, we now have a direct a time image showing irrational cuts between non linked images and how they are relinked, and on the vertical axis, or contacts between an outside and inside which is asymmetrical and not grasped by common sense.

  • Modern noosigns form a single noosphere, 'a brain world with three cerebral components: the point cut, the relinkage, and the black or white screen' (181). Cuts or irrational relinkage are probabilistic, and the screen is topological.

  • Chronosigns are also noosigns.  Vertically, they depict general problems in the form of the time image—'the inexplicable and undecidable problem of peaks of the present, or of sheets of the past, or of a particular power of the false'.  Horizontally, time relations indicate 'leaps from present to present, or sliding from sheet to folding sheet,  or becomings within series that combine before and after in a single event' (182).  The two axes are combined in modern cinema in a way which represents the inadequacies of common sense conceptions—the power of the outside appears in the interstice as '"the direct presentation of time or continuity"'; and this agrees with the irrational linkages between images which appear as '"the continuity of simultaneous presence, coexisting pasts, or metamorphic becomings"'.  Both together show that we are looking from a point outside or beyond the external world, 'a  vertical irrational point of the problem of a particular kind of non chronological time'.  This appears in film or as something inexplicable (Robbe-Grillet), undecidable (Resnais), or incommensurable (Godard) [the amusing series of objects in Six Fois Deux indicates a {Marxist} point of view outside the normal world?].  [So to vulgarise], the gaps between images show the eruption of an inexplicable problem from the outside, while the relinkages try to explain this in terms of 'paradoxical continuities' in order to force us into thinking, or offer are some theoretical take to help us break with ideology.
All  chronosigns are also lectosigns.  Deleuze here 'is somewhat elusive' (182). Lekton in Greek apparently  means something expressible and incorporeal—expressions can be independent of objects, images when they cease to relate to a clear external object.  First we have to see it as an opsign, disconnected from the sensory motor schema, but we also have to understand the internal relations between the elements, including sonic and visual.  These relations endow the lectosign with power [the power to have an effect that is].  Reading the relations requires a pedagogy [film directors learn how to do this? Critics do? What about viewers though?]. There is more than just interpretation or coding at stake: we have to grasp the relations between the elements. There is also a broader sense referring to relations between sight and sound in general.  So even the classic cinema had lectosigns, even in the silent era, where sounds were indicated through visual signs like steam escaping from a whistle or through intertitles.  However, this remains within a naturalistic frame of reference, even if this is 'an implicit indirect speech' (183), [which we reconstruct from seeing the characters talking and interacting]. 

When sound arrives, speech can take on a dimension of its own, and this denaturalizes and can even demote vision: 'a dimension of "sociability" and "human interaction" [becomes] a distinctive feature of discourse, and this dimension is what sound films articulate aurally and render visible' (184).  We can arrive at phatic communication—'speech meant only to confirm that speech is going on'.  This also helps discourse emerge relatively autonomously from its contexts and situations. The model here is the open-ended conversation, 'taking turns, initiating topics,…  Ending exchanges…  Free form succession of topics, now logically related to one another, now without any apparent connection'[ethnomethodology!].  These are now directly presented, compromising the visual images dominance, and introducing an unnatural element, which takes on a life of its own.  The rumour is perhaps the best example, as in the film M, where we see how rumours develop and are taken up by different media and different speakers.  Freewheeling conversation like this also appears in 'the great comedies of the classic cinema…  Especially Hawks' (185).  We can now also see that the visuals of interaction can become ambiguous, as distortions or lies, or hints.  Nonverbal expression now becomes potentially artificial and deceptive.  Dialogue can also dominate camera angles, montage and so on following the glances of the partners, or extending the visual space as off camera voice or ambient noise—clearly relating to the notion of the out of field world.  We have narrative voices, not clearly connected to visual images, by developing 'the voice-over consciousness'.

Music has a particular affect because it can express the open whole, through a sonic continuum, which might include 'dialogue, sound effects and music in both the relative and absolute out - of - field' (186).  Music also appears as an autonomous element or power, perhaps with a musical performance on screen which continues off screen and develops into the film score.  In silent films, the score is external and acts as 'a kind of running programme - music commentary on the images', but in sound films, it can be more autonomous.  It can still be connected, for example through Eisenstein' s pervading rhythm between visual images and music, while other directors, like Brecht aimed at deliberate non correspondence.  Of particular importance is the way in music can relate 'the open whole and the individual images'(186).  Music indeed can directly present the open whole, better than the visual image [and Deleuze is apparently impressed by those who see the world as embodied music].  Music as an '"immediate image"'can then interact with visual images, not always reinforcing them, expressing '"the living concept which goes beyond…  The visual image, without being able to dispense with it"' (187, quoting Cinema 2).

The last bit of this phrase shows that music is still tied to the visual images in classic cinema, still dominated by the sensory motor scheme.  In modern cinema, sight and sound become fully autonomous, and thus have to be read, and discourse can become not indirect but 'free indirect'.  We have discussed some other techniques, ritualizing conversation, making speech distanced, or engaging in fabulation in ethnofiction.  In all these cases, the act of speech itself gets framed, becomes a pure act, disconnected from visual image and standard usage.  This also leaves us with an autonomous visual image, are not explained by speech or other linguistic categories [and the example is the ASW, in Bresson, the empty deserts in Antonioni and Pasolini].  In particular, we can now get visual images of '"the strata, those mute powers…  of the before or after of speech, of the before or after of humans"'(188, Cinema 2).  This produces what Deleuze calls archaeological or stratigraphic images.

Thus we can see images that are non commonsensical, that can even seem '"badly joined"', like 180° cuts in Ozu or Straub [or, irritatingly, modern coverage of sports] which seems to turn round the image and break the usual ways of connecting images, indicating reverses of perception.  These factors are most apparent in Straub and Huillet—long slow pans of exterior landscapes, often with a buried past, attempts to show the forces at work, 'to materialise sensation in the .landscape' (189), which becomes 'the fundamental aesthetic of the modern visual image'.  Again this liberates landscape from the sensory motor schema  and reveals an inner power.  In his work on Bacon, Deleuze argues that forces are similarly visible in paintings which break the standard visual codes [fair enough], but also engage in 'bypassing the brain and working directly on the senses through images that define assimilation'
[highly debatable] (189).  [You can break with visual clichés but it's debatable whether these directors 'thereby manage to materialise sensation'.

So modern visual images archaeological, stratified and capable of various links with other images, each of which is ' forcing'  the viewer to read the image (190).  Such visual images interact with speech and with other images [an example is Straub/Huillet Fortini/Cani, where a poet reads his own work, on camera and off, and reveals his own relative distance from it, while 'slow pans of various countryside landscapes intervene', including a shot of a village identified with Nazi brutalities—although the poem is actually about the Six -Day War].  We have discussed irrational cuts, and sound can also be involved, an act of speech or a sound effect breaks of visual continuity, or lyrical music appears together with a shocking image.  Sometimes this is done to reinforce irrational cuts, as when music appears between scenes.  Marguerite Duras is hailed as the best example here [La femme du Gange], where we have  in effect two films, one of images and one of voices, both of which are autonomous although related.  [Apparently, the film 'juxtaposes a fluid, oceanographic vision and words of love and desire', alluding to the eternal which is what links the strata, and which acts as the limit or point of dissolution of both visual and voice, as it moves towards speech and noise].  [There are You Tube clips of India Song here, which gives the idea.] This tending towards a limit is found 'always' with modern visual images (191).  In the same way, free indirect discourse points to the limit of conventional codes: fabulation is 'a metamorphic movement beyond the limit, offering new variations between 'all components of language, phonic, syntactic, and semantic'.  At least we should read these sounds and visuals as metamorphic.

There is also a movement towards the limit which differs seeing and speaking, and a grasp of new relations.  In modern cinema, these limits are what we see in the visuals.  [The actual bit of Deleuze refers to the vision of "a blind man, of Tiresias", and the speech of"an aphasic or amnesic"].  The visual can indicate what is unstable, and the speech what can only be grasped through clairvoyance—this is the non arbitrary but also irrational link between the two.  These links emerged in modern films because they centre on sets of problems which call into question the conventional links.  It might be political problems as in Fortini/Cani [the links between an Italian communist poet supporting Israel, while being reminded of 'buried Nazi bloodshed' (193), where the visuals note what is unspoken in the poetry and vice versa [there is another example 193, where we see current images of festivity, while voices tell of past festivities and future abandonment].  The examples of strangeness or inarticulateness of speech show a relation between voice and music.  Godard draws from sonic material in general (Prenom Carmen).  In fact most modern cinema separates speech and visual image, and also includes non linguistic sounds, and thus faces the challenge of re-chaining in a non arbitrary way [always a terrible problem, since we often get private languages, or rather elite dialects].

Deleuze also discusses the relations between cinema, theatre and television.  Contrary to the usual view, film is not close to the theatre, since it can capture something emergent in the form of '"conversation for itself"' (194).  Open-ended discourses show rudimentary sociability.  Given the greater possibilities of depicting the visible in cinema, this sort of conversation can become important in structuring events.  Theatres have to work with stable spaces, with the 'no out of field'[surely there offstage activities?—Bogue insists that 'they never suggest a prolongation of the stage world into a surrounding space'].  Film also 'fuses voices music and sounds in away the theatre can not', constructing 'the seamless sonic/visual space [extending] from the shock to the out of field and into the absolute out of field of music and/or voice over commentary' (194-5).

Television has a different relation.  Deleuze thinks that modern cinema would not have developed without television which is where we first find separation of audio and visual constituents.  The electronic images of television also depict time [apparently as noted by Nam June Paik], a constant electronic scanning as 'fundamentally a type of time image' (195).  And digitalisation introduced much more mutability.  Electronic images can be seen as 'transformable emissions of "immedia", without clear origin or final destination'(195, not quoting Deleuze this time, obviously, but a certain Edmund Couchot].  This changes the notion of out of field, and replaces it with '"a
perpetual reorganisation, whereby a new image may arise from any point whatever of the preceding image"'(195, this time quoting Cinema 2).  So digital television anticipates cinematic reversals of perception andrelinking.  [And then the usual excess—'the video screen is the brain screen, and in this sense a version of the modern cinema's noosphere', or in Deleuze's terms, 'the video screen becomes "a table of information, an opaque surface on which <data> are inscribed, information replacing Nature, and the city - brain, the third eye, replacing the eyes of Nature"']. 

However, cinematic innovation has not just been caused by television, since an additional '"powerful will to art"'is also required to use technological innovation (196).  Thus information technology generally only provide a possibility for creativity, and cinematic concerns are what counts, an aesthetic, an interest in time images already—and we might cite Duras or Resnais here as not just imitating television.  In fact, the potential of television has not been exploited, for Deleuze, except in the direction of social control [why should that be,Gilles?], and cinema is actually the more creative—and film critics have done more to understand the nature of images, including televisual ones.

Conclusion [actually an excellent summary]

Plato and the Greeks thought of movement as 'a passage from one ideal pose to another'(197).  Bergson argued that in the Renaissance, motion was seen in 'terms of a  uniform space and a metrical time in which no moment is privileged over another', yet this still did not grasp motion adequately.  For Bergson, 'movement must be grasped as the transformation of a whole, in which any given moving entity, its motion and all surrounding entities form an open, constantly changing totality'.  Bergson went on to suggest this could be grasped as a universe of images, flows of matter-movement, interrupted by specialist living images.

This provides Deleuze with the terms that can be used to analyze movement image.  In classic cinema, there is a vertical axis of differentiation, where the frame, shots and montage indicate elements of the open whole through individual images.  On the horizontal axis of specification, there are various kinds of movement images, starting with the Bergsonian trio of perception-, action- and affection-image, then weaving in bits of Peirce to get at the relation image, the impulse image and the reflection image.  Each of these has two signs of composition and one genetic sign.  In classic cinema, these signs are displayed and regulated by the sensory motor schema.

When we analyze the perception image we start to see how perception works, and we find it grounded either in a fixed perspective or a floating liquid one, or even a gaseous one.  Affection images appear in the form of facial signs, where the face reflects qualities and registers the passage of these qualities.  [Somehow that leads to] the ASW.  Impulse images point to an originary world of drives, and its signs are symptoms or fetishes.  Action images can be large or small forms, appearing in sequences like SAS or ASA.  Reflection images turn these two forms into one another, appearing in the signs of various figures, including metaphors, inversion and figures of discourse.  These can be marks of natural relations, or demarks of disrupted natural relations or symbols of conventional relations 'formed by law or habit' (199).  Although this is a better model of movement and time, movement still dominates. 

In modern cinema, movements appear that are no longer linked within a unified time - space, but are aberrant or uncentred.  We see these developments in the form of opsigns and sonsigns not regulated by a sensory motor scheme: these can appear first in the form of flashbacks or dream worlds.  Finally, hyalosigns depict time in its own right, in the form of time crystals, where time is split 'into an actual present and coexisting virtual past' (199), and we see fragments reassembled in new circuits of images.  Chronosigns depict time as sheets or peaks.  The powers of the false develop to produce linkages beyond common sense, including 'incompossible combinations of coexisting past times' (200), or coexisting presents.  Paradoxical forms mean that the division between true and false can no longer remain.  In Welles's cinema, the powers of the false are personalized as characters like truth seekers or betrayers, while in Rouch it is a matter of deliberate fabulation [then a bit on Clarke and Cassavetes], and in Godard, classifications are metamorphosed and series are created from 'incommensurable yet interrelated categories'.

The sensori-motor schema collapses, and time is prioritised over movement.  Relations of image to sound change, and conventional thought is forced to change too.  Classic cinema offers image shock leading to a concept of the whole, which then explains the inner monologue as the characters organise their thoughts through the sensori-motor schema.  In modern cinema the images shock us into grasping a new thought of the outside, and inner monologue gives way to free indirect discourse.  [Bogue argues even more clearly here that the Deleuzian notion of encouraging belief in the world is not to be read in terms of some Baudrillardian anxiety about reality leaching away, as I had read it, but more in terms of making believable the philosophical picture of the world depicted in modern cinema. This is the challenge, of course -- to disrupt commonsense but not to let people see the result as just incomprehensible].  The normal vertical and horizontal dimensions above are replaced by a notion of the outside appearing in interstices, and deliberate disjunctive differences on the horizontal axis [as in the diagrams].  The interstices becomes important in modern thought, as do the splits in the modern lectosign.  Visual and sonic images become autonomous and indicate their limits, and we are required to read them as they are linked together in unconventional ways.

The distinction between movement image and time image can be blurred in practice, since all cinema tries to indicate movements and processes of thought, forming them from a plastic mass [signaletic matter].  The possibilities have always been indicated even in classic cinema, especially the images depicting the impact of time.  Now these images have been particularly foregrounded, in the 'brain world of the spiritual automaton' (201-2).  This automaton appearing in the cinema requires viewers to develop new modes of reading and thinking, but in the form of actualization, by filmmakers.

Deleuze insists that he is not imposing these philosophical concepts on cinema.  Standard structuralist approaches based on Saussurian linguistics are inadequate, so is psychoanalytic terminology.  Filmmakers working with cinema or and its potentials become philosophers or theorists, even when they despise philosophy or theory [as did Hawks and Godard].  Filmmaking produces 'the concepts that belong to film', and watching film pushes us into philosophy as we work with those concepts [as philosophers work with those concepts].  Philosophers might make these concepts into 'a coherent cinematic theory' (202).  Thus Deleuze [denies authorship of his taxonomy, and pretends to almost be compelled to adopt it] sees his work as like a classification of living forms, or the development of the logic of the cinema, and sees the work as an example of a positive relation between philosophy and non philosophy, philosophy and the arts.  This relation becomes the identity between the questions 'what is cinema?' And 'what is philosophy?'

What a tour de force!

back to Deleuze page