Pictorial into conceptual and reverse

Hrvoje Turkovic

In his notes written during the winter of 1927-1928 S.M.Eisenstein posed himself a remarkable task: to film Karl Marx' Capital (cf. Eisenstein, 1984: 57-67).

Even today this task seem implausible for some theoreticians. E.g. Ian Jarvie in his Philosophy of the Film (1987) states flatly: "The film medium is not especially suited to the discursive mode of address" (p.150).

Though Eisenstein have never filmed Marx' Capital, he obviously considered the realization possible though difficult. He already had some expository experience with his previously done film October (1927-28).

Still, he did not felt that he had a sufficiently clear grasp of the efficient expository methods. Throughout his notes Eisenstein groped for the moves which would led him to the conceptual and expository "taming" of film. He explicitly defined his aim as: a construction of a "film treatise" (tractatus) (p. 78), a "film essay" (p. 82).

If I try to collect and systematize Eisensten's dispersed hints, the following are the basic "principles" needed for the realization of the task (all quotations are from: Eisenstein, 1973):

1. Principle of "deanecdotization" (p. 78). Elements used in film may not be apprehended from the "event-developement" point of view, they must not be "anecdote" based (p. 78). The viewer has to be prevented to be "taken in" by the possible plot content of the scenes.

2. Principle of deviation. If used at all, a "chain of action" have to be highly "trivial", (p. 80), "banal" (p. 85). It should be presented only in order to enable us to "feel" a "deviation" from it (p. 80).

3. Principle of fragmentation. Deviations are assumed to be frequently applied and meant to produce the fragmentation of the plot ("sjuzet"). The task is, according to Eisenstein, "not to present the stock-market by a stock-exchange (Mabuse, St. Petersburg) but through the thousand of 'small details'" (p. 79).

4. Principle of conflict, contrast. Eventually, the fragmentation is supported by the old eisensteinian principle of conflict, contrast: elements has to be presented not within "the trivial chain of action", but through the "conflict" of the piecemeal fragments of scenes, through the "maximal contrast" between the "maximally isolated materials", scenic details (p.86).

5. Principle of association consistency. Conflict, contrast is actually a specific connecting principle - a principle of "association" (p. 85).

The aim of the entire process of deviation, isolation, emphasis, fragmentation of elements is - a "construction" of a "chain of such stimulus" (p. 85). Only within a context of such chain there is a possibility to achieve the unity of sense.

Fragmented, isolated, disparate, emphasized elements have to be connected into the "associative sequences" permeated by the same "construction procedure" (p. 78)."The basic directorial principle is the principle that permeates the construction of the whole thing and of the tiniest detail" (p. 87). "The particular principle has to be evolved ad limitum" (p. 78). The sequence has to be "logically consequentially evolved" - "associatively consistent". (p. 89).

6. Principle of uncommon observation (p. 80; 86). What kind of "sense" is the sense that connects the disparate elements, that "permeates" the whole chain? It can be only a "nonscenic" ("extra-sjuzet") kind of sense - i.e. the abstract, general, conceptual sense.

First, the wanted effect of any deviation is the "reorientation to the uncommon observation" (p. 79). The reorientation is caused by the audiences resistance to the deviation, i.e. their reaction to something illogical from the point of view of common life (p. 79).

Undoubtedly, this is the famous formalist principle of estrangement, foregrounding. "Common observation" is the scenic bound observation, observation of the "trivial chain of action", while "uncommon observation" is the observation of those particular elements, features of the scene, that are deviant, "deanecdotized", "isolated" and therefore not interpretable from the scenic events point of view.

7. Principle of abstraction and generalization. One of the Eisenstein's goals to be achieved by deviations is - "attraction" (emotional and intellectual attraction; p. 84). Deviation effects the viewer "twice more forcefully" (p. 79), it brings emphasis to the place where it occurs. But what is of prime interest for Eisenstein here is the generalizing effect of "deanecdotized", "deviational" aspects of discourse.

Namely, according to him, the deviations are the "springboards" from which we "rebound into abstraction and generalization" (p. 86); they are "montage fragments for the articulation of thoughts" (p. 84).

8. Principle of the inferential unity (p. 87). Though the "generalizing" effect is achieved through deviations, its discursive effect depends on its inclusion into the context of particular associative chain too. It is actually the associative context and its "logical" consistency that establishes a "meaningful sequence".

The meaning of the cinematic discourse based on associative chain cannot be of the "narrative" kind. It can be only akin to the "associative thought" (p. 88). In order to achieve the meaningful coherence all diversified elements in the associative chain "have to be reduced to one inference" (p. 87). The whole exposition has to be uniquely "thematic" (p. 81) (in contrast to "sjuzet" organization). Film has to '"represent" one or two thoughts, to "fix" cinematically "one method" ' (p. 84).

Shortly, to become conceptual expository device.

All these procedures are, basically, procedures of diagrammatization. But what are diagrams?

The aim of a diagram is to represent a regularity, to present a generality in an immediately graspable way (cf. Gombrich, 1978: 167; Waddington, 1977: 52).

The diagram is basically a pictorial representation, but it is a "schematized picture" (Goodman, 1968: 171) or a "logical picture" (Langer, 1967: 29). The diagram does not represent the whole scene or object (Gregory, 1970: 137) but only some features of it, and at that not just some individual features but the whole class of such features (Ivins, 1970: 59) and the class of their relations (Langer, 1867: 29). The diagram is generalized, schematized, stylized.

The diagram selects to represent only relevant features from the point of view of particular generalization, dismissing the nonrelevant features and relations. It is therefore abstractive (Langer, 1967: 33-34).

But, being iconic the selected features and relations do have a relation of structural analogy (Langer, 1967: 29) to the diagrammed class of scenes. Diagrams are, like models, exemplars or instances of what they model in certain selected aspects (Goodman, 1968: 171). Differently from scenic models, relation of diagrammatic homology is not only highly schematized, stylized, but also strictly controlled relation; it is - "conventional".

It is usually controlled either by learned procedure of use, or, on occasion, by a kind of metacommunicational "determinatives" (Gregory, 1970: 138) which are, in contemporary diagrams, commonly verbal: inscriptions", "captions" or verbal "interpretations".

Now, Eisenstein wanted to make film a suitable medium for communication of the political and economic concepts as they were laid out in Marx' Capital. It was possible only through the diagrammatization of cinematic (montage) sequence.

Since the cinema is scenic by its registering nature, and therefore particular, individual, subject to manifold conceptualization, the main problem was how to narrow its possible conceptualizations and how to make the given cinematic sequence of use for conceptual generalization applicable over many scenes beside the presented one.

The procedures he devised were, as we saw, the isolation of features through deviation, fragmentation of scenic presentation, contrastive chaining of fragments, non-spaciotemporal association of them, logical consistency, and conceptual, "thematic" unity.

Eisenstein's hardly fathomed diagrammatic procedures are actually part of the common practice in instructional and scientific films of today.

E.g. in Radovan Ivan~evi}'s four minute single concept film concerned with the explanation of the concept of Iconographic perspective (1972. Zagreb: Filmoteka 16), first the painting is presented with what is called "iconographic perspective": a disproportion in largeness of painted figures according to the difference in religious importance of figures.

Since the of iconographic disproportion is just one feature among the magnitude of other scenic features of the presented painting, author use first the arrowed diagrammatic measure of the pictorial length of figures, and later on he transforms the figures into the black contours in order to eliminate irrelevant details and concentrate the viewer on the relevant question of the comparative largeness of the figures.

Animation, as an obviously artificial transformation, is also used in order to introduce abstract comparisons of relative largeness of painted figures. All the shots are combined in a non-scenic sequential manner, and the procedures are repeatedly applied on different examples in order to establish the generality and applicational variety of iconographic perspective principle.

Expository powers of all this diagrammatic procedures are obvious, successfully defying the sceptics.

The mentioned single concept film was conceived to be projected silent, without any registered sound accompaniment.

But, in spite of the extreme cleverness and clearness of exposition, in spite of the fact that the film is almost selfexpanatory to the well educated western movie viewer, the Ivan~evi}'s single concept films do presuppose a verbal comment. Not a registered one but a free chosen verbal comment by the lecturer who is using the film as an instructional device in a lecturing situation. The single concept films are commonly distributed together with booklets explaining the main points of the given film for the lecturer, and assuming that the lecturer will comment them according to his ability and adjust them to the audience he or she is addressing at the time.

The inclusion of verbal comment in film is quite common feature of all instructional and scientific films. And though it is obviously possible to construct a film without any verbal comment save the title of film, it is regarded more as an extravagancy than as a desirable goal.

Moreover, there is a hidden and persistent conviction that the verbal exposition is a kind of the ideal paradigm of the expository film discourse. The most frequently heard comment on the expository "power" of film discourse is that the film expository discourse is not sufficiently efficient without verbal expository accompaniment.

Actually, Eisenstein's dream was not to substitute the Marx's book Capital by film Capital, but just to approximate its verbal exposition through the cinematic pictorial exposition. The Marx's verbal exposition was considered all the time as an ideal kind of exposition.

But, if verbal discourse is the most suitable for conceptual expositions, why bother with pictorial media at all? Why bother with any non-verbal media at all?

The only possible response to this question is that, after all, the verbal discourse cannot be so ideally suitable for conceptual exposition if non-verbal demonstrations are used such commonly.

Actually, seen from the side of theology, philosophy, natural sciences and modern logic, verbal discourse has never seemed the best kind of medium for the presentation of conceptual structures.

One reason for a distrust of verbal discourse was convincingly presented by Gombrich in his study "Icones Simbolicae" (1978). He quotes a passage by Pere le Moine (from 1666) which goes as follows:

"Were I not afraid of rising too high and of saying too much I should say that there is in the Device something of those universal images given to the Higher Spirits which present in one moment and by means of a simple and detached notion what our minds can only represent in succession and by means of long sequence of expressions which form themselves one after the other and which more frequently get into each other's way rather than help each other by their multitude." (Gombrich, 1978: 161)

The similar support to the advantage of pictorial presentation over verbal one was given earlier by Leonardo da Vinci who pointed out that picturing anatomic parts of the neck from different viewpoints "gives a truthful information about the forms of it, and it is impossible that the writer can ever achieve it without enormous and boring and muddled delaying in writing and in time." (Leonardo, 1981: 42)

As a matter of general contemporary opinion, David Crystal, in his The Cambridge Ecyclopaedia of Language (1987) points to the same thing:

"A large part of scientific expression consist of representations that are wholly or partly nonlinguistic in character - such as physical models, charts, pictures, maps, graphs, and diagrams. The immediacy and economy of presentation achieved by these methods is self-evident. It would be impossible to provide a coherent account in words of all the interrelations found on a map, graph, or tree diagram, for example, and verbal descriptions of formulae and equations are often highly complex and ambiguous." (p. 381).

Obviously, the sequential necessities of verbal discourse tend to complicate the grasp of the essentially "synchronic", steady state nature of conceptual structures. On the other side, due to their pictoriality, diagrams have the capacity to bring conceptual patterns to immediate and overall grasp (cf. Waddington, 1977: 52) and make them available for a quick checks. This is one of the reasons they have become customary support of the expository discourses of modern theory.

Even the logical symbolism in modern logic has been promoted not only because it has contributed to the avoidance of ambiguity and vagueness of natural languages (cf. Langer, 1967: 51-57) but also because it has helped in condensing the arguments, it has schematized the presentation of logical structure and made it more easily graspable at glance ("Our eyes shows us a structural difference" - Langer, 1967: 52, speaking of symbolic presentation of propositional structuree).

Symbolic strings have been more akin the diagrams than verbal expositions, and this was also a reason why they were considered a better "language of thought" (Boole), better "Das Begriffschrift" (Frege) then the verbal language itself.

The second reason for the preference of diagrammatic presentation of conceptual structures over verbal exposition is of the related but different kind.

Namely, every generality and regularity is formed over a number of particular instances (usually of scenic, spaciotemporal, perceptual nature) and applicable to them. The communication of the conceptual generalities commonly requires to be supported by descriptions of those particular instances on which the generalities are based. Because of that, when some generalities are to be communicated there is usually a need to demonstrate their applicability to the particular, scenic events.

In both cases a description is needed, and, as we saw previously, a verbal description of the scene and of scenic events is condemned to be much more cumbersome than the pictorial, model presentation. The problem is especially sharply faced by those who are forced to produce things according to the verbal instructions.

"A purely verbal description demand of the tool maker that he take a series of arbitrary, abstract, vague, word symbols arranged in a linear order and translate them into concrete forms of material in a three-dimensional space in which there is no linear or time order and in which everything exists simultaneously." (Ivins, 1970: 58)

In such cases the demonstrative pictures, models, diagrams are required in order to enable the craftsman to understand and to apply the verbal generalities adequately.

Being pictorial and combinatorial, the film offers just that: the demonstrative presentation of a number of scenic instances over which the given generality applies. And at that, as demonstrated previously, the presentation of scenes and objects in film can be of the chosen and specifically articulated level of abstractness, as required by the communicative situation.

The verbal discourse evidently has its limitations in the presentation of conceptual structures.

But still, let us remember, it is predominantly and commonly used. And it is still almost obligatory accompaniment even in the case of predominantly pictorially based discourses, such as film discourse.

What is it with the verbal discourse that makes it such a useful and almost obligatory accessory of diagrammatic exposition? Or, from different angle: what is it in the diagrammatic discourse that is in a need of verbal support?

First of all, diagrammatic presentation do have a conceptual capabilities close to the verbal ones, as was demonstrated, but their use is comparatively limmited. In the words of William M. Ivins, Jr.,:

"At one end of their gamut they are completely abstract diagrams, such as that of the drawing for the instrument maker about which we have just been talking. There is an infinite number of ways of making any such drawing and in the end they all come to the same thing - for very much the same reason that sentences written from dictation by a number of persons, once they have been proof read and conformed, all have the same symbolic meanings, despite their marked differences in particular shapes and general forms. As a matter of fact, practically all drawing of that kind are actually proof read, but the man who does it is called a 'checker'. the only reason he can do his work is that the lines in the drawing are mere representatives of classes o lines and from that point of view can hardly be called lines at all. Certainly they do not function as particular lines." (Ivins, 1973: 59)

But still:

"The competent and honest observer and recorder, however, had his very distinct limitations. (...)

"Thus while there is very definitely a syntax in the putting together, the making, of visual images, once they are put together there is no syntax for the reading of their meaning." (Ivins, 1973: 60,61)

In spite of the unhappy use of the term "syntax" instead of the term "convention" ("I shall frequently refer to such conventions as syntaxes.", ibidem, p. 61), what Ivins points out at is the lack of generally socially accepted convention in the reading of even the most formalized and schematized diagrams.

The diagram conventions are usually socially local, and commonly made up for the occasion. There are very few diagrams or diagram class features that are of "standard shape", that can function in a manner of standard language. In order to be taught, spread, their conventional side made understandable, some more standard and publically spread communicative medium has to be used. And that is mostly the medium of standard verbal discourse. That is the reason why verbal interpretation of diagrammatic features are used so commonly.

Of course, there is no principal impossibility in establishing some standard "diagrammatic system" akin to language system which would enable a purely diagrammatic discourses without the use of verbal aids. The history of pictographic and ideographic writing systems supports this claim (cf. Gregory, 70; Crystal, 1987). And, the existence of "sign language" of deaf (Crystal, 1987) and computor languages do support such claim also.

But there is another more inherent problem of diagrammatic presentation.

No matter how stylized and schematized they may be, very oftenly they are still predominantly of analoge kind, their features are oftenly dense and replete (cf. Goodman, 1968: 228-232), of the shifting differentiation value. There are usually many different and even conflicting "readings" of the diagram possible in particular case.

To overcome such possible vagueness and ambiguities in diagram readings, usually an "outside" imposition of the limitations on possible readings is needed. And, again, the verbal discourse "imposition" is the most handy one.

When the two sets of discourse limitations and advantages are faced, their clever merging can be of extraordinary use. It is, obviously, the very combination of diagrammatic and verbal exposition that "does the trick": helps in multifaceted analytic/syntetic presentation of conceptual structures.

But still, it is seldom that they are combined on equal footing. They are never assimilated one into another. They are kept distinct even within the same exposition. And, there is always one discourse type that is the basic, and another that functions as accessory.

This "inequality" is usually a functional one: "accessory" type of discourse, be it a verbal or diagrammatic one, usually serves as a metacommunicative regulator over the basic discourse. Occasional verbal informations serve for better orientation within the diagrammatic course of cinematic discourse, and the diagramatic presentation within the predominantly verbal discourse serve as the globaly orienting points in the course of the discourse.

The coordination of the two kind of discourses is not just supportive in the way the two blind men are to each other. It is a specific hierarchic communicative coordination, which begs for better and more detailed description and interpretation.

Bibliography:

Crystal, David (1987). The Cambridge Encyclopaedia of Language. Cambridge: Cambridge U.P.

Eisenstein, S.M. (1984). "Kapital"; in: Branko Vu~i~evi} (ed.). Avangardni film 1895-1939. Beograd: Radionica SIC (original: S.M.Ejzen{tejn, 'Iz neosu{~evlennyh zamyslov, /"Kapital"/'. In: Iskusstvo Kino,. Moskva, No. 1, 1973, pp 57-67)

Gombrich, E.H. (1978). "Icones Symbolicae; Philosophies of Symbolism and their Bearing on Art". In: E.H. Gombrich (1978). Symbolic Images, Studies in the Art of the Renaissance II. Oxford: Phaidon.

Gombrich, E.H. (1982). "The Visual Image: Its Place in Communication". In: E.H.Gombrich. The Image and the Eye. Oxford: Phaidon Press.

Goodman, Nelson (1968). Languages of Art. Indianapolis, New York: The Bobbs-Merrill Comp. Inc.

Gregory, R.L. (1970). The Intelligent Eye. New York, St. Louis, San Francisco: McGraw-Hill Book Comp.

Ivins, William M. Jr. (1973). Prints and Visual Communication. Cambridge, Massachusetts, London, England: The M.I.T. Press.

Jarvie, Ian (1987). Philosophy of the Film; Epistemology, Ontology, Aesthetics. New York and London: Routledge & Kegan Paul.

Langer, Suzanne K. (1967). An Introduction to Symbolic Logic. New York: Dover Publ. Inc.

Leonardo da Vinci (1981). Quadrifolium. Zagreb: GZH.

Waddington, H. (1977). Tools for Thought. Frogmore: Paladin.

Hrvoje Turkovic Biography:

Born in Zagreb, Croatia, 4.11.43. Graduated in Philosophy and Sociology, Zagreb University; M.A. in Cinema Studies, NYU, New York City; Ph.D. in Philology, ZU. Teaches on Theory of editing, Types of film discourse and Film analysis at the Academy of Dramatic Art, Zagreb. Publishing since 1965. - criticism, essays, studies. Editor in several magazines an in Film Ecyclopaedia; vice-president of Croatian Society of Film Critics. Published books: Movie Options (1985); Metafilmology, Structuralism, Semiotics (1986); Understanding Film (1988); Film Theory (1994).