THE IMAGINARY SIGNIFIER what is occasionally called the plot pure and simple(inaccur ately however, for this manifest level also includes the charac ters, their social positions, the diegetic locations, the indicators of period and many other factors that go beyond the action). The instance of the script, in the wide sense i am giving it, now shifts to the side of the signifier, since it is no longer related to the codes of expression that communicate it, but to the interpretations on to which it opens. In order to show, as a recent doctoral thesis does, that one of the meanings'of Howard Hawks' s Red river is to present a justification of private property and the right of conquest, and that another of its meanings is to be found in a mis- ogynist variant of male homosexuality, it is the film script above l that has to be examined for the corresponding clues(signi- fiers). The script has ceased to be a signified: nowhere in the film is what I have just said plainly inscribed To study the script from a psychoanalytic or more broadly semiotic) viewpoint is to constitute it as a signifier In this the script is like a dream, as are many human products The manifest dream, i. e the dream as such-'dream content for Freud, in opposition to 'dream-thoughts'-is a signifier for the in- terpretation, and yet it has itself had to be established(narrated and to begin with communicated to the conscious apperception of the dreamer)as the signified of different codes of expression including that of verbal language (one does not dream in languages one does not know). There would be no manifest dream and hence no interpretation, if for example the dreamer failed to identify any of the visual objects in the dream, i.e. ifhe was unfamiliar with the code of socialised perception; it is because he recognises some of them that others he is unable to identify take on their true value of enigmas or that composite objects(= condensations) appear as such, presupposing some idea, or suspicion, of the different objects superimposed into one or at least of the fact that there are several of them. The imagin ary itself needs to be symbolised, and Freud noted"that without secondary revision( without codes) there would be no dreams, for the secondary process is the condition on which access to percep tion and consciousness is possible. For the same reason, the script, a conscious and perceived instance, must first be a signi- fied before it can be made to signify anything else at all Analyses of scripts, then, are not studies of the signified. what strikes me is rather the fact that they are sometimes considered as
THE INVESTIGATOR S IMAGINARY 31 such. An error which is part of another more general one; it is easy to forget that every signifier itself needs to be a signified and that every signified, in turn, can but be a signifier (this constant back and forth is precisely the work of the symbolic; it is not poss- ible to constitute some elements as pure' signifiers and others as ure'signifieds the adjective 'pure'is a very tiresome one anyway-except with respect to a precise code, a given investi- gation; to one segment and one only in the indefinite chain of sig nification). The result, for the script, is that two different things are confused its place in the text, in the film(everywhere a mani- fest instance), in which it is the apparent signified of apparent signifiers, and its place in the textual system, a non-manifest instance whose apparent signifier it is(or at any rate one of its apparent signifiers). At this point the suggestions of linguistics and those of psychoanalysis coincide perfectly. Relying solely on the former i had carefully distinguished between the text, an attested progression, and the textual system, which is never given and does not pre-exist the semiologist's work of construction this was already to define the gap between them as that between a manifest content and its interpretation. The textual system shares the status Freud sometimes called latent and which includes both the unconscious and the preconscious Certain studies of scripts are directed above all at unconscious significations, and thus correspond to what is normally expected of a psychoanalytically inspired approach. Others work princi pally at the level of preconscious significations: e. g. most of those studies described as ideological'(i gave an example a moment ago vis-a-vis Red river), turned towards layers of meaning which do not feature directly in the film but constitute its implicit rather than its unconscious part. It does not follow that studies of the preconscious of the script are necessarily 'less psychoanalytic: it all depends on the way they are conducted, and psychoanalysis includes a theory of the preconscious (Freud did not regard this instance as an accessory one, he showed great interest in it, notably in The Psychopathology of Everyday life; he reckoned that a kind of 'second censorship' dynamically linked to the first divided the preconscious topographically from the conscious nor does it follow that ideology is an exclusively preconscious production: more probably, like many other things, it has its pre- conscious strata and its unconscious strata, even ifthe latter have hardly been explored as yet(not at all before Deleuze, Guattari
32 THE IMAGINARY SIGNIFIER and Lyotard) and differ more or less from ideology in its classi- cal conception. It is just that ideological studies have tended so far(as a point of fact, not one of principle)to limit themselves to preconscious ideology. As is well known, it is for this reason that Deleuze and Guattari criticise them, anxious as they are to trace the imprint of history in the unconscious itself and hence rather sceptical about the very notion of ideology, at least in its current ors PSYCHOANALYSIS OF THE TEXTUAL SYSTEM At any rate, once analyses of scripts are at all profound and do more than state the obvious, they posit the perspective of a latent'region which they wish to approach. That is surely why in certain cases they are hard to distinguish from another-fourth possible kind of psychoanalytic studies of films: those that have the textual system and interpretation as their aim, like the preceding ones but set out for it from the manifest filmic material as a whole(signifieds and signifiers), not from the manifest signified (the script)alone. It is the film as a whole that is now constituted as a signifier. Thus what is striking in the both ideological and psychoanalytic interpretation proposed by Jean-Paul Dumont and Jean Monod for Stanley Kubrick, s film 2001 a Space Odyssey 2o is that the elemen ts they take as clues come not only from the the- matics of the film but also from the specifically cinematic signi- fier, or at least from the use the film studied makes of it in the structural grids of occurrences and co-occurrences that the authors establish, we find for example(p. 149)the items Track forward, Track back,, as well as 'Weight,, 'Weightlessness Spaceship,, 'Bone thrown in the air by an ape, etc. This approach raises difficulties insofar as the cinematic elements and the script elements are not revealing in the same way(the authors are too inclined to 'diegeticise' the signifier), and what is most revealing, what gives us the most central access to the textual system, is no doubt the relation between the two sets of elements rather than one or other of them or their addition but in one point, and a very important one, the method has to be approved the filmic signifier is as indicative as its signified of the latent sig nifications of the film, the entire apparent material is open to a symptomatic reading(here we recognise the banal but true ob
THE INVESTIGATORS IMAGINARY 33 servation usually rather badly expressed as 'the"form"of a film tells us as much as its“ content” about its“ true meaning”). Raymond Bellour's impressive study of Hitchcock's North by North-West, in Communications, shows i believe what is to be ex- pected of a psychoanalytic approach to films which is at once attentive to the signifier, to the script, and to their mutual articu- lation; the Oedipal structure this analysis brings to light informs the script(as it were on the large scale) but also the editing schemes in the sequence, on a smaller scale, so that the (non manifest)relation between the manifest signifier and the mani fest signified of the film is that of a mirror reduplication or an insistence a metaphor of microcosm and macrocosm: the latent is doubly anchored in the apparent, it can be read in it twice over in two sizes in a spiral movement. with this movement, which is not precisely cinematic and does not only concern the story told but is installed just between the two(and would be different in other films), we are really getting close to the order of the textual system as i understand it Investigations of this kind (the fourth kind on my list)are thus studies of textual systems. Those of the third group, for which I shall retain the name studies of scripts are also studies of textual systems, but from a narrower angle of incidence: the script is part of the apparent data, but not all of it, one of the elements that leads up to interpretation, but not the only one. artificially iso lating it from the others one runs the risk of falsifying the textual system overall, since the latter forms a whole, and this might be enough to invalidate analyses of scripts on principle. But for many films, and not always uninteresting ones this disadvantage is less than it might seem, since the script dimension plays a con siderable part in them while the work of the cinematic signifier is not very great. With other films, on the contrary(I shall return to this point), the analysis of the script is from the start an inad- equate approach, inadequate to different degrees moreover, as different as the relative importance of the script in the textual system, which varies greatly from film to film PSYCHOANALYSIS OF THE CINEMA-SIGNIFIER The more I unfold (for the moment only at the preconscious level) the scientific imaginary initially expressed in a single sen-
34 THE IMAGINARY SIGNIFIER tence, the closer i get to the problems presented to a psycho- analytic study by the signifier of the cinema as such by the level of cinematic specificity. It is a characteristic of analyses of scripts that they disregard this level: the Freudian inspiration finds a place in them as it would in aesthetic studies outside the cinema (and with the same difficulties, which i shall leave aside here because they are so familiar). These investigations may be psychoanalytic but they are not basically cinematic (though very well suited to certain films for that very reason). what they choanalyse' is not the cinema but a story which happens to have been told by it(and there are many such stories). The script of a film may be treated just like a novel and the already classical dossier of the relations between psychoanalysis and literary criti cism could be introduced here en bloc(the absence of the couch and hence of real transference, the precise scope of the method proposed by Mauron, etc. ) To sum up, what distinguishes studies of scripts (and also nosographic or characterological studies) from the approach I am trying to define here, and towards which I am slowly making my way via every thing which is not that approach, is, as i have tried to show by this traversal not so much that they are indifferent to the signifier as that they are indifferent to the cinematic signifier It will perhaps be thought astonishing that in this lengthy dis- cussion of studies of scripts I have not even thought of recalling that they become impossible in the case of certain films, films without scripts, abstract films. the 'avant-garde films of the 1920s, current experimental films, etc. Or else(an attenuated version of the same observation) that when these studies are poss- ible, their interest declines to the extent that (even if it retains a plot)the film they apprehend escapes the full regime of narration-representation (every possible intermediatecase exists; Eisenstein's films are diegetic, but they are less so than most Hollywood productions). What is peculiar to these films is the fact that in them to one degree or another the cinematic signi fier abandons the status of a neutral and transparent vehicle at the direct behest of a manifest signified which alone is important (the script), and that on the contrary it tends to inscribe its own action in them to take over a more and more important part of the overall signification of the film, thus more and more invali- dating studies of scripts, to the point of making them impossible As for studies in the fourth category(complete studies of the