“Semiotics Of Visual Language” in “Semiotics of Visual Language”
If the decoding of visual language can only rest on specific syntactic hypotheses, the actual apprehension of a visual work which reveals how a particular producer has used the potentialities of this syntax also requires a particular method. Just as verbal linguistics has found it expedient to do, we borrow from scientific methodology the instrument of “analysis.” This method is defined, however, in a unique manner according to the particulars of our object of study.
This analytical method aims at bringing to light the interrelations between elements—rather than their hypothetical essence—in the totality that is the visual work. Whether it concerns a painting, a sculpture, a photograph, or an architectural edifice, the work considered as a totality “does not consist of things but of relationships,” as Hjelmslev has already proposed for the analysis of verbal language. In this context, “not substance but only its internal and external relationships have scientific existence” (1963; 23). We will also borrow from this linguist a definition of the process of analysis:
The analysis thus consists actually in registering certain dependencies between certain terminals, which we may call, in accordance with established usage, the parts of the text, and which have existence precisely by virtue of these dependencies and only by virtue of them. The fact that we can call these terminals parts, and this whole procedure a division, or analysis, rests on the fact that we also find dependencies of a particular kind between these terminals and the whole (the text) into which they are said to enter, dependencies which it is then likewise the task of the analysis to register. (28)
In a manner that is again more apparent here than in verbal language, it appears that the visual text is constituted in time by the continuous transformational process of perception, which generates movements in elements and changes in the interrelations of the energetic properties linked to these same elements. This process can only be described in the analysis by an intuition of the system which conditions it, as Hjelmslev further explains: “The process comes into existence by virtue of a system’s being present behind it, a system which governs and determines it in its possible development. A process is unimaginable—because it would be in an absolute and irrevocable sense inexplicable—without the system lying behind it” (39).
If we agree, further, with this linguist that “the characteristic proper to a process, and by consequence, to a text, is that it is submitted to the general rule of the order of positions” (89), we will propose that this characteristic, (which we assimilate to the topological notion of the order of succession) must be completed in the visual text by the determination of transformations that the process engenders in the elements themselves. If a minimal dimension of homogeneity is required for the elements of the visual language to enter into any relation, one must not conclude that the variables interrelations must necessarily lead to a homogeneity of the work itself as a totality. The function of spatial continuity must itself be revealed by analysis, the heterogeneity of visual elements remaining a component whose unification will always be only a provisional step in the analysis and one destined to yield to disjunction. As expressed by Kurt Lewin: “Homogeneity results from a unifying action of perception, as well as from the ‘data’ itself, and maximal heterogeneity is the inverse function of the unity of the group” (1941; 105).
For Hjelmslev, the idea of homogeneity is posited as one of those indefinable concepts that his linguistic system requires (besides the notions of “description, object, dependence” (29) or of “presence, necessity, condition, function, functive” (35). Semiotical analysis will have as a central objective the illumination of the possible meaning of the concepts of “homogeneity” and of “continuity” in the visual language.
We recognize fully, with Hjelmslev, the necessity of interpolating in the description the presence of certain “fonctifs” (or operators, latent magnitudes) that are inaccessible to knowledge through channels other than the process of analysis itself.
Therefore, the description which is the result of semiotical analysis will make use of dynamic and structural concepts, issuing not from the obvious surface properties but from the underlying structures.
To this end, and through the methodology instituted by Kurt Lewin, when he discerned those attitudes that opposed Galilean and Aristotelian thought-processes (1935), topological semiotics intends to use classificatory principles that are based on concrete objects observed situationally. This will not result in dualisms, but rather in dynamic seriations which serve to support an awareness that the intermediate stages are always active and present under the rarefied poles of perceptual reference (Kaufmann, 1968; 21-22).
It rejects as inadequate any intuition of classificatory concepts seen normatively or which grants both undue importance and a permanent character to the fact that an object belongs, at any given moment, to such or such a category. It refuses to consider as the only intelligible and determined objects those occurrences that are produced frequently, in a statistical sighting which denies the heterogeneity of the real to the profit of the aleatory existence of constant averages, substituted for the concrete situation. And, in contradistinction to linguistic semiology, it refuses to establish polar oppositional tables or dichotomies too abstract to fit or reach the complexity of the visual field itself: “As in physics, classification of phenomena and objects in opposite pairs and logical dichotomies disappears. It is replaced by a classification using concepts of series allowing for continued variations” (Lewin, 1935; 32).
This description, which has to be a classification, must necessarily be abstract; that is, conserve only certain dynamic elements disclosed by perceptual activity in view of elaborating a unified understanding of the diversity. But it must be concerned with the recognition of the most important structural functions that allow elements to interrelate in the totality of this environment that the visual work constitutes. To this end, it must be unceasingly open to unknown properties, those which have not yet been perceived because of the approximative and temporalized character of all operations of perception.
Before we continue, it appears necessary to stress the radical difference between semiotical analysis, as we have defined it, and a method which has been given the same name and which has been developed, particularly in France, in the course of recent years.
7.1.Typological Semiological Analysis
A method that is called “semiological analysis,” linked to computerized data analyses, has been used for many years, particularly in Europe, in researching the archeology of art, sculpture, painting, and ancient architecture, wherein it appears more immediately fruitful than in its applications to the domain of anthropology proper (Salomé-Lagrange, 1971).
However, in spite of its name, this form of analysis does appear to depend more on general typology than semiology, as a science of signs, because it is not founded on a specific theory of structures of the visual language. Moreover, the art objects which it treats are not defined as in any way differentiated from any other objects of human fabrication, except for the fact that they form part of the repertory of art history. From the verbal categories developed by that discipline, the objects are classified into groups, subgroups, and so forth. This simple classification is sufficient, apparently, for confirming or disproving earlier classifications and chronologies, definitions of types and styles, diffusion of models, and so on. That is to say, it is sufficient to verify the pertinence of the verbal discourse that has been applied to the same visual works by various authors.
According to Marie Salomé-Lagrange, this approach has as its target the verification of the validity of an already existing interpretation by confronting it with a list of pre-established distinctive traits:
This means, in particular, to try to find anew all the descriptive criteria used by the author to differentiate classes and to verify that they are not contradictory, aleatory or abitrary in relation to the ensemble of the description furnished by the author. To examine its significance is for us to know if it is compatible with external information: chronological or geographical distribution (1973; 16).
In the first stage, which consists in identifying distinctive, minimal and mutually exclusive traits belonging to subdivisions of diverse categories of verbal information, this analysis proceeds to an enumeration of occurrences of these distinctive traits in various classes of objects belonging to the repertoire of the history of art. At this point, it constitutes a form of typological analysis such as that defined by J. L. Chandon and S. Pinson: “It is a method of the analysis of facts which permits the grouping of objects, characterized by an ensemble of attributes or variables into classes not necessarily disjointed two by two. These classes must be, on one hand, as sparse as possible and, on the other hand, as homogenous as possible” (1981; 4).
This form of classification endeavors to establish classes of distinctive traits and associates them with other classes in order to verify external hypotheses, and not to develop any possible model of internal syntactic structures by which observed objects would be distinguished one from another. It responds to the following objectives:
What are the descriptive traits (or properties or attributes) which, being regularly or constantly associated, permit the distinguishing between types of objects? What types (or classes) are regularly associated in a way to constitute characteristic assemblages? What assemblages present a continuous and evolving sequence of traits and can be placed in relation with a chronological ordination? (Chandon, 1981; 26)
This type of analysis, as Salomé-Lagrange observed, requires a certain number of preliminary precautions in the choice of an homogeneous collection of objects of limited complexity, so that they can only be put into relation with a relatively short verbal text. This necessary reduction of variables and of classes, dictated by typological analysis, limits at a first level the application of this method to the analysis of visual language as such in a proper semiological manner.
Moreover, the classifications which serve typological analysis are based on strongly substantiated and individualized abstract conceptual schemas, for which it is relatively easy to adopt the basic analytical criteria of similarities and differences. In effect:
The method of typological analysis rests on the simultaneous taking into account of similarities and differences between objects. It is concerned with the regrouping of objects in such a way that elements of a group are “strongly similar” whereas elements belonging to different groups are “relatively dissimilar.” The multitude of typological analysis methods come from the fact that each of them gives a different sense to the terms “strongly similar” and “relatively dissimilar.” (Chandon, 1981; 4)
Following procedures analogous to those of graphic semiology—which itself cannot be classified as proper semiotics of visual signs—this type of analysis is founded on lexicalized notions, carriers of stable distinctive traits which must, like the objects to be described, be present in small numbers. Modeled on a substantialist and static conception of verbal linguistic process, this “semiology” can be adapted to the quantitative procedures of computer systems, but only after having excluded the consideration of the countless characteristics of visual reality.
It has been put to the test, at first manually, then by computerized treatment, with the matrix of the scalogram, assembling distinctive traits by proximity and remoteness. “The scalogramme is a configuration where cruxes of matrices are grouped in a more or less regular escalation in the diagonal zone” (Salomé-Lagrange, 1973; 24). This disposition allows the visual estimation of distances between classes in relation to three privileged positions: the two extremities and the center of the scalogram. The classes corresponding to extremities have no descriptive element in common. While allowing to see immediately separations between the series of classes, this matrix of the scalogram allows examination of “the distribution of descriptive elements in all classes, and in particular, to distinguish by sight the most discriminating associations of traits” (24-25). In addition to the scalogram, this analysis can use other types of matrices, like those of clusters, which define groups or classes of elements linked by graded coefficients of similitude (33).
But visual semiotics analysis does not consider fruitful this kind of typological approach presenting a yielding to the logic of identity and of the static substance in the description of its proper object of knowledge. On the other hand, the abundance and the variability of data in the visual language may require the instrumentation of the computer, as well as of the graphic representation of spatial curves that perception constructs in the visual fabric, but more under the form of analogous than digital symbols. But it does not seem, until artificial intelligence research is more developed in the visual field, that computerized systems can be of an immediate help for theoretical research in this field of semiotics (Saint-Martin, 1989).
It is not only by virtue of its basic connection with the analogous nature of perceptive movements that the semiotical analysis we propose finds premature the recent efforts in “formalizing visual language, in the sense of contemporary logics” (Gips, 1975). A clarification and simplification of the basic elements and operators in the visual language appear to us impossible as long as the theory of the grammar of this language has not been further explored, as has been the case for mathematical formalization or that effectuated on verbal grammar. In this regard, we share the view expressed by Kurt Lewin who, although he did not hesitate to introduce geometric and topological formalization with respect to certain zones or contexts where psychological phenomena are deployed, wrote: “However we would produce an empty formalism if we forgot that mathematization and formalization can operate only in relation with the maturity attained by the material of the object to be studied at a certain moment” (Lewin, 1951; 1).
7.2.The Ocular Circuits
In the case of painting, semiotical analysis accounts for the perception of the pictorial plane, being that energetic mass of coloremes integrated by the producer in the dynamic structure of the Basic Plane. It bears, therefore, on the organization of specific physical materials as they are perceptually constructed by human vision.
This description cannot be based on the spontaneous movements of the gaze across the canvas which has itself raised a problem, since these movements are potentially infinite and open to aleatory unfoldings leading to opposite conclusions (Marin, 1976; 113). As we have already noted, the ocular fixations only produce a problem when the centrations are not sufficiently numerous to allow for the greatest possible number of coloremes to be perceived and explored for their dynamisms and interactions, their differentiations and regroupings. Only when this does happen do the interactions produce temporary and transformable syntheses finally destined to be integrated with vaster and more comprehensive units.
The process by which centrations unfold does not rest, however, on a hypothesis concerning a supposed normal trajectory of perception, nor on the hypothesis of the existence of privileged zones of perception in the image, as proposed by C. M. Tardy and taken up by René Lindekens (1971; 56). Following some studies on the perception of images by given subjects, a relatively constant itinerary of the gaze has been proposed that would organize, in a stable way, the system of iconic signification. In particular, it has been argued that the eye has a tendency to move in the sense of the hands of a clock, and that it actually spends more time on the left half of the field, the inverse being true for left-handers. As for the succession of pauses, their points of application would seem to have a tendency to be concentrated on the same region of the image, almost necessarily to the left, and upwards in the projected image (if it is concerned with slides or film), whereas they would disperse on different shores in opaque photographs (Lindekens, 1971; 57). Obviously, these perceptual behaviors would not permit a true apprehension by the spectator of the totality of the pictorial field, of which certain regions are thus completely neglected. These behaviors correspond even less adequately to the multitude of centrations effected by the producer in the process of production, since he interrelates each minute region of the visual field to a multitude of others in a trajectory continuously recommencing in all possible directions.
These observations of Tardy would seem to corroborate the experiences of G. T. Buswell (1935) when he proceeded to record the movement of the eyes during the contemplation of a print by Hokusai; he collected a series of 70 consecutive points of fixation. But, as Molnar observed (1966; 140) the eight fixations of the eye that are produced during the first two seconds do not correspond to those elements susceptible of being associated with a possible interpretation. In fact, he commented about the group of fixations, “we can hardly count more than fifteen which fall on something.” These movements do not copy, therefore, the apparent structure of an image or of its “iconic” configurations, but correspond more to the structural locales linked to what we have called the Basic Plane. Arnheim (1954; 364) was categorical when he said that “there are very few links between the order and the direction of the fixations and the compositional structure of a work,” a structure established here by the traditional codes of the iconic reading.
Moreover, this succession in visual perception would seem to assume a different order and function, according to the linguistic medium considered: “The temporal order of our perception is not part of the composition when we look at a sculpture or a painting, whereas it is a part when we look at the dance” (Arnheim, 1954; 343). It appears that the method of appropriating those elements that constitute the visual text is not to be derived either from the usual behavior of perceivers or from a previous theory of the composition of a work, since this itself must be elaborated from the perceptual trajectory.
One must acknowledge, indeed, that the ocular paths are not neutral and that what has been perceived previously, from the left to the right, from the bottom to the top, from the center toward the periphery and so on, influences the characteristics of the elements which will be next perceived. The order of succession of perception is hardly indifferent with respect to the types of dynamisms and interrelationships that are retained in order to form the composition of a painting or a sculpture, at least in the rapid and partial apprehension with which a large number of perceivers are satisfied. Visual semiotics proposes that the composition, or rather the structure of the work, can be deduced only from a series of equilibria established between the elements. The energies and the regions they form, which are taken up and modified with a view to producing superior and more complex equilibria, can finally produce a state where the ensemble of movements and transformations produces a system which can then be offered provisionally as an adequate synthesis.
The semiotical description of the pictorial plane must thus offer a representation of the equilibrated system of transformations that structure the visual field, following sufficient centrations to explore the totality of the dynamic field. These multiple centrations have taken up, corrected, and modified the first experiences of tensions and visual movements, verifying all the possibilities of interdependence and interaction of the elements, and testing the attractions and the disjunctions and their effects on position, dimension, depth of various regions, and so on.
These circuits, which are extremely numerous and which ceaselessly construct new interrelations between the elements perceived in the first centrations and deconstructed in subsequent centrations, transforming the qualities and dynamisms of regroupings, do proceed according to certain rules. They do not lead to a type of experience where all would endlessly transform itself in all directions, as suggested by the entropic aesthetics of Umberto Eco in L’Oeuvre ouverte (1965). To the contrary, the very structure of the process of perception, which is not atomistic or associative but integrative, leads to the elaboration of global systems of equilibrated transformations proper to a given visual field. In Peircian terms, some “final interpretants” are periodically produced on the syntactic levels.
In effect, by the very mechanism of perception, a threshold of equilibrium of the systems of transformation of the elements is eventually offered to the perceiver with a certain quality of necessity. The field may conserve it at the time of subsequent centrations or it may be modified by the enrichment of further visual trajectories. Thus, the semiotical description always corresponds to a provisional halt in the process of perception, when the perceiver believes he/she has reached a more or less invariant equilibrium in the dynamisms of his perceptual movements and of those which animate the work.
Jean Piaget already explicated the function of this invariance which always represents a certain synthesis: “The fundamental property of this latter is not to be ‘left’ unchanged, as is often expressed, but rather to result from the composition of modifications themselves which must be constructed, whereas it did not appear at the heart of initial observations” (1980; 119).
The character of necessity of such an equilibrium, produced by a given perceiver, can appear to another observer as justified or unjustified. In the latter case, another type of system of transformations—more comprehensive and accounting for more in the dynamic of the pictorial work—should be furnished to replace that which is found to be inadequate.
If, by contrast, a certain necessity in the system offered is recognized, this does not imply in any way that this structure ascribed to the work is immanent in it and that it has now been disclosed to the profit of all, after having been hidden for some time. This structural necessity does not correspond to a hidden character of the work; it is only the outcome of a set of perceptual operations that alone can produce it. Genetic epistemology has examined this problem, one that can be resolved neither by some kind of innateness nor by an objective immanence of the constructions of the perceptual dialectic: “How to explain that the construction of new relations in the course of processes of equilibriation lead to results of which the internal necessity seems to imply that they were preformed or predetermined in anterior situations where the subject still did not perceive them or, more simply, did not yet become aware of them” (Piaget, 1980; 10).
The supergestalt that constitutes the integration of multiple stimuli in a field of perception is not found in the work through a simple search, but is constructed through continuous modifications of what seemed originally to be the first stimuli of a work. It exists only in and through the perception, while being this very text that semiotics must analyze in its actual constituents. However, if this semiotical description must aim at taking into account several structural levels, the parameters and the principles of analysis cannot remain the same when the levels of abstraction, which are necessary to the synthesis of awareness, become more and more elevated.
We would like to point out, finally, that the kind of ocular fixation required by semiotical analysis, if it is altogether a normal one, and does not require any “tour de force” of the eye, remains distinct from the habitual regard which we turn on things, with its more or less attentive and personal involvement. Inattentive centrations, in effect, even statistically numerous, which are not accompanied by sufficient motivation for the perceptual processes to be realized are practically equivalent to an absence of centrations or to a nonperception.
This type of gaze must, moreover, differ from that which regulates our daily activities, where, as J. J. Gibson remarked (1966), it is of primary importance to know or to interpret as quickly as possible the visual objects which are presented before us, that is, to reduce them to an object already known or to that which, in our previous experience, resembles it the most. This behavior, dictated by needs of survival, has become so automatic that it renders perceivers at times unable to perceive visual elements objectively. Thus, the majority of subjects before whom one projects on a screen an empty field in a weak luminosity, when asked “What have you seen there?” recognized a vast repertory of common objects with a precision all the stronger as the time of perception was prolonged (Frances, 1975; 225).
The sole requirement put forth to the semiotical gaze is to undertake an activity of perception vis-à-vis a visual text that one cannot presume to know before having read it. But the reading of a visual text, well before any possible repercussions of its semantic contents, constitutes a lived-experience which not only requires all the emotive and conceptual capacities of the perceiver, but also involves him in a relatively exacting process of personal transformation.
As is the case with the syntactic analysis of verbal language, the semiotical analysis of visual language is empirical, not only because it attempts to account for a reality which exists independently and outside of the perceiver, but also because it implies more extensively a sensory instrumentation. It depends, therefore, strictly on a theory of perception; as expressed by Paul Bouissac, any theory of the sign, explicitly or not, can only be linked to a theory of perception (1984; 8).
However, this necessarily experimental dimension does not permit its assimilation to advances made in experimental aesthetics. In this field, experimentation turns, above all, on the attempt to measure the reactions (surprise, interest, pleasure, or distaste) of perceivers of works. As a result of these reactions, the amount of originality or banality which a work can convey is assessed. But as such, the structure of the work remains an unknown.
Semiotical analysis aims, on the contrary, at a knowledge of the structures of organization of the works themselves, independently of the reactions, evaluations, or interpretations which the spectator adds, while imagining at times that he contributes thus to the semantic dimension. Syntactic analysis desires neutrality, in the sense that Molino wished it (1973), that is, to account for the phenomenon which any perceiver can observe who applies the given syntactic rules. It is only when this reading of the visual text is truly effected that semantic hypotheses can be intelligibly discussed, as can also the diverse evaluative reactions of the aesthetics.
7.3.Steps of the Semiotic Analysis
The study of a visual field which is offered as a phenomenon of language is necessarily started through a first approach submitted to peripheral vision, assessing the distance at which the work can be found, its dimension, and its general aspect. The desire to become acquainted with an object requires habitually the use of a greater angle of vision than that which macular and foveal visions can cover. The perceiver attempts, in the first place, to gather a variety of stimuli in a rapid scanning process, which institutes a certain number of links between points of aleatory centration. Foveal vision may choose to rest on these points but without attempting to inform itself in a particular way about these reduced zones which are directly accessible to it in its angle of vision. Macular vision will also contribute, at angles of 15 degrees, assembling particularly the general chromatic information offered by the field.
These multiple traversals recognize, mainly through gestaltian laws, a certain number of regroupings and disjunctions between the visual variables, their focal or peripheral distribution, certain major vectorialities, chromatic characteristics, and so on. This first scanning represents indeed the type of perceptual attempt with which the majority of perceivers of visual works are usually contented. It involves a certain number of perceptual reactions which will remain attached to later perceptive trajectories to a variable extent. However, this level of scanning, more or less strengthened or prolonged, must be considered as presemiotical, as it segments the field by blurred mechanisms of peripheral vision in regions manipulate by analysis.
We will call semiotical analysis a process which prolongs this first contact in a specific way and at two different levels. The first level, of a more phonological than syntactic nature, studies the nature of the basic elements of visual language, the coloremes. We call this study exploratory or colorematic analysis. The second level, properly syntactic, studies the way in which the elements regroup themselves to form functional ensembles in a given spatial globality, according to the operators defined by the syntax of visual language.
This form of analysis is derived from the very structure of visual grammar which, like verbal grammar, is made up of two sections: one dealing with the properties of the constitutive elements of visual language and the other with the syntactic laws which specify their interrelations and sequences in possible statements. The fundamental difference between the two grammars is that the laws of perception are an integral part of the syntactic structure because they alone contribute to the actual or possible modes of interrelation of elements in a field. At more fundamental levels, concrete relations exist only between given coloremes, regrouped by topological and gestaltian rapports, at the heart of a previous energetic structure, that of the Basic Plane or the Virtual Cube.
The semiotical analysis necessarily develops, therefore, at different structural levels. It can be represented and unified through a certain number of grids which are superimposed according to larger and larger degrees of abstraction and of synthesis in relation to the concrete characteristics of coloremes. But from the permanence and importance of foveal vision, it follows that even when the analysis is related to the more synthetic levels of regroupings of coloremes, the ocular centrations are endlessly reengaged in the perception of coloremes. Even if these have already been made the object of experimentation at the exploratory level, any new context in which they are perceived can modify the nature of their visual variables and their proper dynamisms, while multiplying their interrelations with more distant regions in the visual field. These relations are always effected, however, through basic syntactic rules. But let us first deal with the presyntactical level of analysis which describes the organization and function of the basic elements of visual language.
7.3.1.Analysis of coloremes
The colorematic (or coloremic) analysis describes the aggregates of the visual variables which correspond in the visual field to that which is perceived by the foveal centrations. After having analyzed the very components of each of the perceived coloremes, this analysis describes the transformations which a coloreme undergoes by its interrelations with the other coloremes of its immediate entourage through macular centrations. The analysis proceeds thus at a first regroupings of coloremes through the topological relations which establish the first perceptual construction and structure the energetic exchanges between coloremes. This exploration of the “building blocks” of the spatial construction must theoretically be carried out on the totality of the field offered by the visual work. Nothing can let us presume that the producer has used the same visual variables in the same way in two different places, and no region of the field can be defined as empty or denuded of any specific energetic information, resulting from the particular reverberation of luminous rays on opaque matter.
The colorematic analysis has as its aim the recognition of the dynamism and the tensions specific to the first unit of perception, before they are inserted in a more elaborated syntactic structure. It determines, therefore, the characteristics of the visual variables used in this basic unit which is the coloreme, such as their mode of integration, their potentiality of expansion, their vectoriality, their vibratory energy, and so on.
On the subjective plane of the production of a percept, this analysis has a primordial function: it permits the beginning awareness of visual events correlative to perception and of the energetic experiences they cause in the organism of the perceiver as well as in the visual field. By this awareness, we mean the act of retaining in short-term memory, for several seconds, the energetic effect produced by these visual tensions perceived in the coloremes, in opposition to the process of ordinary visual perception, which is largely unaware and nonreflexive concerning the material causalities to which it is linked.
As noted previously, any centration in the visual field puts especially into action the energy specific to foveal vision, the one which is richest, most precise, and particularly suitable to reacting to visual energies present in the region which it establishes, within an angular opening of two or three degrees. If foveal vision alone can on this register and at a medium distance, see with precision minute visual elements, this does not imply that foveal vision ceases or cannot be used when less distinct organizations of visual variables are offered to it. Suitable to perceiving all variables present in its point of centration, color as well as textures, vectors and forms, the trajectories of the fovea are necessarily surrounded by elements perceived by macular vision; but these two types of perceived masses do not possess the same type of differentiation and precision, if perceived by the first or second type of vision. Further, foveal vision and macular vision cannot entirely exclude from the visual field the effect of perception by peripheral vision. However, the normal functioning of the eye easily permits attention to characteristics of percepts produced by foveal/macular or peripheral vision.
The visual variables which constitute the coloreme, (or the correlative region of the centration) necessarily differ from one coloreme to another even if it is only by the different effects produced by the variation in their position. This means, among other things, that the intensity and energy of a coloreme will be modified by its later insertion in the proper syntactic networks. But as such, the coloremes can be the object of the semiotical description, accounting for the nature of the visual variables that constitute them, their organization and function, their possibility of expansion, their vibratory energy, potentiality of junction or of disjunction, and so on. Structured as a topological mass, the coloreme will offer in addition different characteristics in its central and peripheral layers, in the structure of their boundaries, and so forth.
This analysis of the visual variables does not have as its aim a purely physical or taxonomic description of the characteristics of elements. The simple enumeration or accounting of the modes of appearances of the visual variables (color, texture, dimension, form, etc.), juxtaposed in a neutral inventory, without a priori notions concerning their functional possibilities of liaisons/disjunctions, could not deliver any information about the discourse that they serve to construct. Regrouped in the unit of the coloreme, these visual variables are already inserted in a minimal syntactic structure, that of a topological region capable of entering into energetic relation with other units. The visual variables are therefore immediately perceived at the level of the tensions that they produce inside the coloreme and secondly at the level of those which arise when they are reconnected or regrouped in an ensemble of coloremes.
We have already noted that even when more abstract structures of organization have been elaborated upon in order to take into account the interactions between ensembles of coloremes, the inevitable perceptual return to the work will always be at the level of the coloremes, as a full experience, replete with concrete energy, where the pregnancy or the real pertinence of larger and more abstract structures and their capacity to transport the more profound tensions issued from the work, may be either verified or contested.
Semiotical analysis, therefore, must linger at this stage of experimentation with the visual dynamisms which constitute the coloremes, which indeed appear as a microcosm of the particular syntactic structures used by each artistic producer. The internal/external structure of coloremes perceived in a work by Cézanne, Kandinsky or Mondrian, will be very different from one another and from the basic units that may be perceived in Matisse or Pollock. Without a doubt, this study is the sole instrument which can permit us to expand our knowledge, so rudimentary to this date, of the energetic structures of visual language.
In order to regulate this pursuit, and to permit communication between researchers, we propose a procedure which can serve to identify the perceptive constants and the first interrelations in the heterogeneity of visual percepts.
7.3.2.Systems of partition
The observations of colorematic analysis will be recorded in a theoretic grid of partition placed upon the visual work. This grid intends to provide only external parameters, situating the description, and not to implement a real division of the field or of the visual or pictorial process. The visual field can only be known in its globality or any of its subdivisions, following an integrative accumulation of perceptive centrations, that is, at the end of a syntactic analysis of the work.
Not only does the organization of this preliminary grid not attempt to mimic a possible hypothesis about the composition of a visual work, but its most immediate objective is the deconstruction of the perceptive interpretations that usually arise from the recognition of such or such aggregates of visual variables when recalling previously known or imagined objects.
Rather than being founded on these hypotheses of objects and on the verbal discourses that they make possible, the perception of the visual text must attempt to produce the experience itself of coloremes constructed by the producer to precise energetic ends. Any new tension, in effect, of the visual variables forming the coloremes involves a significant transformation in the discourse of the work.
It must be noted that theoretically this grid of partition must be used in a confrontation with the visual work itself, and not in relation to a reproduction of any kind. Any reproduction modifies, in an essential way, all the visual variables that the analysis of coloremes proposes to study in a given work. Even though certain printed color reproductions of the best possible quality can serve as a memory aid once the work has been observed in its concrete and physical reality, it is necessary to prohibit the use of slides which by their transparency misrepresent, in an irremediable way, the memory of percepts obtained through the actual experience of the work. If for reasons of semiotic training, reproductions of visual works are used, it should be well understood that information gained thereby cannot be attributed to the original work, but only to its deformed “translation.”
The use of grids of partition is common in typological types of analysis, the choice of a partition of more or less compressed meshes being left to the discretion of the user, provided, of course, that it leads to fruitful results. This decision is always arbitrary if one recalls, for example, that the partitions which can be established between a group of about 15 objects to be analyzed add up to 1.4 million possibilities. However, in practice, it has been verified that the analyses carried out on ensembles of 30 to 350 objects favor the production of nine groups or fewer, the average being six (Chandon, 1981; 163). The grid used in the colorematic analysis has the particularity of determining, by itself, the number of objects which will indeed be observed, that is, the number of coloremes which will be examined and the number of surrounding coloremes with which they will be put into relation.
The only type of grid of partition which appears to us to be necessarily excluded is that in which certain coordinates could be identified with the most important internal structure of the Basic Plane, being the cruciform, horizontal and vertical axes, in order not to obscure, in taxonomizing and underscoring them, the dynamics specific to these fundamental axes in the pictorial or sculptural production.
Thus the exploratory analysis can use grids formed by ensembles of vertical/horizontal coordinates in a square or rectangular format, forming nine compartments, or again, ensembles of coordinates producing 15, 20 or 25 compartments (see figure XVI).
We propose the use of a grid of partition of 25 compartments identified from left to right and from top to bottom by the letters of the alphabet. Each of these compartments will, in turn, be divided into five zones identified by the figures from 1 to 4 according to the hourly sense (in quarters of an hour) in order to be terminated by a return to the center marked by the figure 5. These zones can therefore be described as: Al, A2, A3, A4, and A5, as in figure XVII.
These five zones correspond to the localizations of centrations effected by one or more observers in a way that permits a comparison between the results of perceptual processes of various individuals. A tighter grid—one of finer interstices—or one of wider character could establish an equal or greater number of loci of fixations. Broadly speaking, the number and the disposition of five internal zones seems a bare minimum for the study of any given compartment.
In each of these zones, the analysis describes the aggregate of the visual variables correlative to a foveal vision centration. Theoretically, this analysis describes 125 coloremes (there being five zones multiplied by 25 compartments) regularly dispersed across the pictorial plane. The parameters of description correspond to the ensemble of the visual variables, since they all are always implicated in any place of the centration of the gaze on the visual field. However, as the coloremes are not truly isolated by the grid of partition or the zones division, the description will include their relations with those coloremes which immediately surround them. This will involve the description of topological relations inside and on the borders of each compartment, the state of their boundaries, and so on.
Thus the description of coloreme A1 requires the consideration of its dynamic relations with A4, A5, A2, and Bl. Similarly, for the others:
A2 → Al, A5, A3, B4, Bl, and B3.
A3 → A5, A4, A2, B3, B4, and Fl.
A5 → Al, A2, A3, and A4.
Bl → Al, B2, B5, B4, and CI.
Ml → M4, M5, M2, LI, G3, H3, 13, and Nl.
and so forth.
One must note, moreover, that these compartments are not endowed with characteristics of equal energies since they are structurally diversified by their placement in the Basic Plane. Thus, the compartments A, E, U, and Y are animated by the potentialities of the energies of the corners; C, H, M, R, and W are sustained by the central vertical axiality, and K, L, M, N, and O by the horizontal axiality; A, G, M, S, and Y are integrated in the disharmonic diagonal; and U, Q, M, I, and E in the harmonic diagonal.
In the same way, the peripheral compartments envelop the central compartment M, and the compartments G, H, I, N, S, R, Q, and L are both enveloping and enveloped in relation to the others. However, it is at the level of subsequent syntactic regrouping of compartments that their interior energies will be assessed in relation to the energetic matrix of the Basic Plane, and not at the level of exploratory analysis. It will become quickly apparent, furthermore, that each compartment of the grid contributes the added energy of diagonal and cruciform axialities, given the gestaltian quality of peripheral decoupage which is attributed to them.
In effect, analysis of coloremes already requires a form of syntactic operation, in the sense that their interrelations are always established at the plane of the dynamic of the visual movements which constantly transform them. One cannot conceive of the visual variables or their integration in a coloreme as discrete units, previously cut out and identified in the internal/external experience, to be submitted to operators which would regroup them from the exterior without modifying their identity, as proposed by the logical schema: “X = F (a, b, c,),” where “F” represents a similar function acting on each variable.
If one compares this procedure of colorematic analysis with that used in formal grammar, one might observe that the analyst in the present case is not free to use only one or the other of visual variables in his description of visual language, but rather that he must use them all in modalities which are always, however, different and which cannot be determined a priori. In the context of a system of axiomatic description, one might say that the alphabet will always be constituted by the finite ensemble of six variables (texture (T), color/tonality (CT), dimension (D), vectoriality (V), implantation (I) and form (F)), but in instances of occurrence where each is always different from what it was on a previous occasion.
The coloremes are also described as symbols composed of certain types of visual variables capable of forming ensembles according to certain so-called syntactic rules.
Thus, we would have:
C1 : Ta, CTb, De, Vd, Ie, Ff . . .
C2 : Ta’, CTb’, De’, Vd’, Ie’ Ef’ . . .
C3 : Ta”, CTb”, De”, Vd”, Ie”, Ef” . . .
One must point out that the topological relation of succession which identifies the order in which an ensemble of coloremes can be inscribed cannot be reduced to the function of concatenation, which is not applicable as such in the visual language, whereas it is fundamental in verbal, logical, or mathematical language. In effect, in a visual field, the simple relation of concatenation cannot exist, neither between the visual variables nor between two coloremes, in the sense of a simple adjunction or juxtaposition. The variables or coloremes cannot be added to one another without being mutually transformed. Thus, color, for example, is transformed according to its dimension or its texture, if the contours are different, as much as by the interrelations with close or farther positioned coloremes.
The syntactic rules are therefore “operators of integration” at the level of the coloreme as well as at the level of regrouping of the coloremes. They can be described in a nonterminal vocabulary as categorical components of grammar, whereas coloremes form a part of a terminal vocabulary accompanied by markers specifying certain of their characteristics, their immediate or distant environment, their position in the Basic Plane, and so on.
This analysis will ideally recover the entirety of the perceived visual field, step by step, in the heterogeneity of visual movements effectively produced by the visual variables used by the producer of the work. A list of distinctive traits, which can be observed among the visual variables grouped in a coloreme, as well as of the interrelations of adjacent coleremes, is presented in Appendix III.
In summary, the colorematic analysis takes account, for each centration, of the constituting elements of a coloreme. It will describe, therefore, the visual variables constituting the perceived coloreme, the types of liaisons or of interrelations of multiple variables in a coloreme, and the interrelations of these coloremes with adjacent coloremes. In addition, the first perceptive syntheses, especially those of the topological species, describe the predominant vectorialities in each of the zones.
If the analysis of coloremes can take account of the basic dynamic structure of the energetic elements that a given producer privileges over others in the construction of his particular representation, it cannot serve to demonstrate how these basic units are spatially brought together in a particular structure. The ensemble of these coloremes is regrouped in significant spatial sequences through syntactic rules which govern the infrastructure of the medium, as well as the colorematic interrelations at more and more dialectic and synthetic levels.
7.4.Syntactic Analysis
Syntactic analysis accounts for the application and the particular functioning of syntactic rules in a determinate visual text. In a manner analogous to the numerous syntactic rules of grammar of verbal language, the syntactic rules of visual grammar, which we have examined in Chapter Three, are juxtaposed in order to regularize and specify the functional potentialities of the liaisons/disjunctions of colorematic groups, to form specific spatial statements.
Owing to the state of ignorance we are still mired in today concerning the syntactic structures of visual language, more than two millennia after a constant practice of this discourse by all human societies, topological semiotics will establish steps in the apprehension/analysis of this discourse. The base units of visual language, of the phonological type, are not able to furnish the hypotheses of regroupment on the syntactic plane. Visual semiotics proposes as a preliminary hypothesis of segmentation necessary for any analytic approach, the segmentation of the visual field by scanning of the peripheral vision, in order to facilitate the observation of the syntactic mechanisms between large aggregates of coloremes.
The various intuitive modes, which have served until now to describe the enunciatory function of visual language, have involved extremely disparate syntactic consequences. Until now the most common model simply applied the operator of concatenation, typical of verbal language, to previously constructed iconic fragments in an aleatory order. Most often, this construction remains insensitive to stylistic differences in the process of iconization, capable of modifying the syntactic functions of elements. Moreover, this form of iconic segmentation seems to offer two defects. First, being indissociable from verbal lexicological units which are a guarantee of the completion of the process of iconization, the iconic units always present themselves already as a plane of content. This procedure does not maintain the separation between the plane of expression and the plane of content which appeared essential to Hjelmslev in any analysis of linguistic function. Second, this process conceals and rejects from the plane of expression the noniconizable zones of the visual field which form at times the largest part of it. Finally, it cannot integrate into the verbal meanings the effect of the various visual treatments to which these images have been submitted.
Instead of this operator of concatention, which rules in a linear and irreversible way the digital relations between discrete elements, such as numbers or morphematic units of verbal language, we propose the introduction of operators which can take account of the spatial, analogical and continuous function operating in visual language.
In recognizing the dynamic character of the variables constituting the material plane of visual expression, we propose a syntactic analysis which can account for the differentiated networks of energetic groupings produced by the topological and gestaltian perceptual operators. These groupings will be constituted by aggregates of informal visual variables as well as by iconizable fragments, interconnected not in some linear order but rather in the three dimensions of space itself.
However, in referring to the scanning of the peripheral vision to obtain a first reading—a first segmentation of the visual field—one has to be aware that this perceptual route takes into account only a restricted number of the elements of the visual field, leaving a larger number of other sections not perceived. In this scanning, the eye establishes general junctions/disjunctions outside of the dynamism of the chromatic elements, inaccessible to peripheral vision which is endowed with rods and not with cones. This preliminary segmentation serves as a scaffold for the analytical work and must be replaced, at the conclusion of the syntactic analysis, by a more adequate segmentation of the visual field.
This scanning, which is effected in a back and forth fashion from left to right and from bottom to top on the visual field, recognizes larger or smaller aggregates as units. They are quite similar to those which Rorschach described as global responses (G), responses of great detail (GD), or responses of little detail (Dd). It may again regroup in a much larger ensemble a multiplicity of small regions, which more easily lend themselves to a more enveloping form or gestalt. It is of relatively little importance that this decoupage is made in one given direction or another, provided that it takes account of the totality of the field. This decoupage serves only as an armature for a description of relations which are established between various regions of the field. These relations themselves, by syntactic analysis, will provide the appropriate model of adequate segmentation that semiotical perception must produce in the visual field.
To this primary peripheric level of reading, syntactic analysis adjoins the perceptual possibilities of foveal/macular vision in order to recognize the effective differentiations supported by all regions and subregions, being individualized through characteristics of their visual variables, even if these figures are presented under irregular or blurred forms, as bad gestalts or nonlexicalizable figures. Thus, the decoupage by peripheral vision of large previously seen regions permits the establishment of a mode of description and nomenclature of characteristics and relations linked to this potentiality of the process of perception, but does not prejudge anything about the syntactic functions of these regions before they have been submitted to a syntactic analysis.
Less suitable in reacting to chromatic energies, peripheral vision, however, maintains a primordial function in perception, since, in being opened to a larger angle on the visual field, it is the sole instrument of analysis and of grouping of regions of large dimensions. It never acts alone but always concurrently with foveal/macular vision, whereas these latter can, in a process of intensification of attention in the centration of the gaze, operate on the smaller regions of the visual field.
The perceptual mechanisms of peripheral vision provide two types of syntactic information on the visual field: 1) they determine the superregions, formed by an aggregate of coleremes which are amalgamated or differentiated through the laws of junction/disjunction of the gestaltian theory and which are capable of entering into relations with other superregions through the same mechanisms; and 2) they put into play or shed light on the energetic network constituted by the pictorial plane in its relation with the Basic Plane.
In the schema shown in figure XVIII, one could regroup these various rules, of which the boundaries not only are not airtight but also continuously transport the energies of each level towards the others:
The central zone could be called “more connected” to the objective aspect of the visual field, and the other zones more linked to the subjective dimensions of perception. But one can only take account of visual language while conceiving it as a product of a continuous interaction of all these structural levels. Also, any description of the functioning of the language at one of these levels results from an artificial distinction made to serve the ends of the analysis.
The syntactic analysis of visual language, which we call operative analysis, since it describes operations, but which can only know them in effectuating them, establishes the nature of the interrelations existing between those large, more or less autonomous regions, recognized by peripheral vision upon approaching of the work and their eventual subdivision into smaller regions.
It uses the ensemble of syntactic laws, and its topological and gestaltian operators, to disclose networks of liaisons/disjunctions between agglomerates of coloremes constituted by peripheral vision into totalities. These present themselves as series of regions reunited/separated from one another on two fundamental levels:
(1) In a bidimensional expansion, that is, groupings/separations deployed in height and in width;
(2) In a tridimensional expansion, that is, at various distances in depth through a variety of factors such as boundaries, vectorialities, chromatic contrasts, relations with the Basic Plane, effects of perspectives, and so on.
Before commenting further on these, we present in summary the fundamental steps of a syntactic semiological analysis that are required to account for the particular dynamisms of any field of visual representation:
I. Differentiation of the visual field into regions by peripheral vision.
II. Description of regions according to their preponderant visual variables by foveal/macular vision: chromatic poles and recognition of phenomena of interaction of colors, dimension, boundaries/forms, texture, vectoriality, bidimensional implantation. The visual variable of implantation in the plane of depth requires an important ensemble of perceptual trajectories, establishing: 1) a putting-in relation of regions by scanning; 2) the analysis of the type of vision solicited from the eye; 3) the points of view, sightings, and openings of the angles of vision used by the producer; and 4) the spatializing potentialities of the diverse systems of perspectives used.
III. Recognition of topological liaisons of each region with each of the others.
IV. Establishment of gestaltian interrelations between each region and recognition of gestaltian regroupings endowed with an iconic function; pressure of the ‘good form’, figure/ground productions, and so on.
V. Recognition of the main subregions in each region and description of their internal/external structures.
VI. Insertion of each of the regions and of certain of their subregions in the infrastructure of the Basic Plane determining a particular energization of the spatial curve of the pictorial plane. This description establishes: 1) actual differential energies; 2) potential and virtual differential energies; and 3) the production of specific continuous/ discontinuous spatial models.
VII. A regrouping of the visual field according to regions of stronger or weaker energies, accentuating or nonaccentuating their topological proximity or distance in depth.
VIII. In this energetic segmentation, recognition of dominant events (or statements) from their subordinates, of liaisons between them and with more separate regions.
IX. Recognition among regions and segmented subregions of the arrangements of visual variables linked to the structural spatialization of various organic spaces.
The syntactic analysis of a visual representation is no less detailed and complex than verbal syntactic analysis, requiring moreover the intervention of a less familiar conceptual instrumentation, inasmuch as it refers to dynamic processes in constant interaction.
The reader will find in Appendix IV a group of abridged notations facilitating the recording of visual variables, topological liaisons, and so on.
7.4.1.Topological liaisons
While assisting the integration of visual variables in a coloreme and intercolorematic relations, topological relations will also be active in the apprehension of relations between larger aggregates of coloremes constituting regions circumscribed by peripheral vision and their interrelations.
One must not confuse the relations of neighboring, separation, or envelopment, with the topographical or geographic notions of adjacent juxtaposition or of distancing in bidimensionality. Instituted in the subjective dimension of the percept, they establish more fundamental links between the dynamic elements transformed by forces with which they enter into relation. The neighboring relation effects a strong bringing together of percepts offering tensions of attraction of equal intensities. In the same way, separation, which recognizes and valorizes the energetic variations in contiguous or foreign groups, maintains those regions in a juxtaposition, permitting their oppositions to remain active. Far from denoting an absence of relations between two elements, the perceptual separation is the locus of an awareness of reciprocal dynamic interrelations in depth.
Similarly the relation of envelopment, which plays a fundamental and paradigmatic role, since it represents the fundamental structure of the Basic Plane as the reservoir of energies which constitute the linguistic fiction in visual materials, becomes dialectic at all levels of analysis. It defines the structural schema of the coloreme as the relations between a central layer and the diversely surrounding peripheral layers. It also determines the fundamental relations between regions, their subregions and superregions, by means of continuous or discontinuous relations in the encasing, the embedding, and so on.
7.4.2.Infrastructure of the Basic Plane
We have described the infrastructure of the Basic Plane as: 1) The system of limits defining the spatial matrix of the pictorial or sculptural visual work and the foundation of their discursive autonomy; and 2) as the reservoir of energies permitting a spatial fiction to expand into the three dimensions and not remain in the bidimensionality of the material pictorial support.
The syntactic analysis of the visual work must, therefore, describe the way in which the pictorial plane, as produced by the artist, interrelates, that is, does or does not actualize the actual and potential energies of this infrastructure. In other words, the aggregates of the visual variables forming the pictorial plane will see their proper energies increased tenfold, transformed according to their type of implantation in the original matrix.
The syntactic structure depends, in the first place, on the dialectic and energetic exchange produced between regroupings of coloremes and the centers of energy of the Basic Plane, producing diverse spatial curves in the pictorial plane. These diversely spatialized regions are interrelated in view of the production of functional, organic, or practical spaces. They are amplified at the levels of increasing complexity calling for extremely subtle and mobile spatial curves. They may also coalesce into illusory spaces, usually more static, since these depend less upon the specific dynamics of reflected light as a privileged constitutive material and rely more upon collateral information.
Let us take as an example the grid of partition of 25 compartments described previously (see figure XIX), in which we have observed that the positioning of different compartments is equivalent to varied energetic levels.
The formative sides of the Basic Plane are indicated by the letters identifying the compartments, at their moment of origin and termination, being AE, AU, UY, and EY (see figure XX).
The corners identified as FAB, DEJ, PUV, and XYT constitute those regions which possess the maximum of actual energy, which decreases in the central section of the sides, more remote from the place where the angles meet.
By reverberation, the maximum potential energy will reside in the peripheral compartments, which are A, B, C, D, E, J, O, T, Y, X, W, V, U, P, K, and F, with a similar attenuation in the central compartments of each side. In the same way, the compartments of the central and intermediary region, which are M and G, H, I, N, S, R, Q, and L, possess a virtual linear energy of axial, diagonal, and cruciform coordinates which cross them.
The energetic system of the Basic Plane could be illustrated by a hypothetical quantification where the total of the actual energies of the corners and formative sides and the total of potential and virtual energies in the internal mass of the Basic Plane would be equivalent.
In other words, the actual energies of the angles are equivalent to the potential energies of diagonals; the energies of each of the sides are equivalent to the virtual energy of the central zone where the diagonal and cruciform axes meet:
Actual energy of angles in A, E, U, Y = 160
Potential energy generated diagonally = 160
Moreover, the actual linear energy (ALE) of the formative sides, is identified from A to E by the quantities 15, 10, 10, 10, 15 = 60 ALE.
It is equilibrated by the virtual energy of cruciform vectors calculated at half of the energies of mass: VLE = 60.
The equilibrium between the actual, potential and virtual charges involves a distribution of energies, different according to certain zones, producing three particular regions, and identified as follows:
(1) The compartment M regroups the virtual linear energies: 40.
(2) This region, formed by compartments G, H, I, N, S, R, Q, and L, possesses a minimal potential mass energy (20 for each compartment).
(3) The peripheral zone is more diversified, since the corner compartments possess a potential energy equal to 40, the intermediate compartments: 30, and the middle compartments: 20. These latter compartments will obtain an accrued charge, if we describe them as points of anchorage of virtual energy of the cruciform axes and of the internal diamond area (figure XIX).
The energetic elements which constitute the visual variables used in the constitution of the pictorial plane will be all the more powerful depending on whether or not they are conjoined with this specific reservoir of energy by an ensemble of mechanisms that syntactic analysis will describe.
In a general way, any actual reiteration of the vectors of the infrastructure of the Basic Plane, by the coloremes of the pictorial plane, multiplies their energies and their function in the whole of the work. Moreover, any activation of segments of the elements of the Basic Plane reinforces in a proportionate way the energy of the localized visual variables. This activation can be realized by a reiteration of sections of the matrix, by a lightly shifting rendering or by colorematic regroupings which punctuate their trajectory, producing a virtual vector of the pictorial plane upon the emplacement of a virtual vector in the Basic Plane.
The mechanisms which reactivate the energy of the corners and of the formative sides include the production in the pictorial plane of open forms in these regions, in strongly saturated chromaticities, or their repetition in proxemic shifting in what art theoreticians have called ‘deductive structures’ (Fried, 1966; 405).
The maximal peripheral energies are significantly attenuated by the inscription of coloremes offering tonalities, that is, an incorporation of black/white in the chromas, or by the inscription of a closed form which would establish the peripheral zone as a more or less distant background in relation to this figure.
The analysis must also account for the energetic transformation which results from the reiteration on the pictorial plane of the vertical and horizontal axes of the Basic Plane, or of their shifting duplication in the pictorial mass. The intensification of the central zone, in particular, in an attempt to focalize, often is accompanied by a decrease of the peripheral forces in the pictorial plane.
The reaffirmation of the infrastructure of the Basic Plane by the zones constructed by the pictorial plane attempts to integrate them in the spatial coordinates of the Basic Plane, that is, in the distance of depth proper to this topological mass. But this expression of neighboring can be thwarted by other characteristics which these zones possess and which will be explained at another stage.
7.4.3.Gestaltian regroupings/disjunctions
The analysis of regroupings/disjunctions between the elements of visual language constitutes primarily an analysis of visual movements in a specific field. However, contrary to what a certain interpretation of Gestaltian theory tends to hold, these visual movements are not uniquely and principally those establishing stable, simple, and regular forms (closed forms) from visual elements which lend themselves more or less easily to this end.
The application of gestaltian mechanisms in syntactic analysis concerns, first of all, the constitution of the regroupings of regions or superregions through factors of junction/disjunction which gestaltian theory has revealed and which we have treated earlier in this text. If the gestaltian factors tend to produce certain effects of regroupings of the coloremes in a unified form or of groups of coloremes in larger aggregates, they also contribute in an equal manner to the phenomena of separation, distancing, differentiation, distantiation, and isolation, which structure in a very fundamental way the mass of coloremes in the pictorial plane.
Syntactic analysis will proceed systematically in the examination of factors which bring together or separate elements in the three dimensions, through a functional application of gestaltian laws to the pictorial plane. The analysis will afterward establish those correlations which arise between regions through the factors of proximity, similarity, and complementarity.
The factor of proximity between two elements can stress distance as well as heterogeneity, since perception, which endeavors to bring together contiguous regions, will be hindered by the disparities between some of the visual variables, the structures of their boundaries, or other factors. In the case of an energetic homogeneity, the factor of proximity will lead to an effect of neighboring. But in the case of heterogeneity, perception will refer to more complex hypotheses resulting from other syntactic levels in order to diminish the separation between these elements.
The factor of similarity produces visual movements between elements separated by more or less important intervals. Syntactic analysis takes into account the links which are established between elements which are similar in color, texture, boundary, orientation, angularity, vectoriality, or form. By definition, this similarity established between one or more visual variables does not recover the whole ensemble of visual variables, thus creating particular tensions between regions which are brought closer by a certain similarity, but which are differentiated, separated, and diverted by other factors. Indeed, the similarities perceived between coloremes or groups of coloremes are not the properties of objective elements which would be attached to them as distinctive traits. They are the result rather of subjective gestaltian approximations, resulting from relations established by the operations of the perceiver on the basis of multiple visual trajectories in the field. These are applied to specific aspects of the visual variables, while deliberately ignoring some of their other dissimilarities (figure XXI).
Insofar as a similarity of properties, movements or functions has been established, one can conclude an operation of abstraction or distancing in relation to the group of effective movements of coloremes or a regrouping of the same—a temporary cessation of perception of visual movements in certain regions to the benefit of the establishment of liaisons in others. However, a resumption of the trajectory of perception can lend to the elements or regions previously determined as similar a dissimilarity or heterogeneity and thus fracture the regrouping effected on the basis of similarity. This newly recognized dissimilarity will be the nucleus of another type of dynamic interrelation between these coloremes and the other regions now considered, leading to different regroupings and positionings in the three dimensions.
This perceptual and dialectical process does not carry any contradiction, but rather points to a simple coherence at the heart of the perceptual process, open to heterogeneity of the real and which recognizes that the establishment of relations between elements can only be the result of a perceptual, subjective, and perspectivist operation. Without the recognition of heterogeneity, no possibility of analysis and of regrouping is possible. This is recognized by typological analysis itself, when it “assumes that the population of objects is always heterogeneous, that is, that groups exist” (Chandon, 1981; 176). Certainly, only processes of abstraction, which retain only certain particulars of the real, permit the establishment of homogeneities setting up the very notion of ensembles or of sets among heterogeneous elements.
The effects of regrouping or of neighboring, produced by the factor of similarity, must be observed from all the visual variables, of which chromaticity and tonality appear particularly dynamic. However, the visual movements between coloremes carried out through gestaltian regulation will be again transformed, in the case of color/tonality by the laws of interaction of colors which possess their own structure.
The interrelation between regions which resemble each other by their characteristic of ‘closure’ usually entails the visual movement productive of a figure on a ground, just as regions do which seek an immediate recognition through a hypothesis of objects of external reality to which they can seemingly refer. But it is necessary to remember that syntactic analysis must proceed in the placing into relation of all the elements of the visual field. It is extremely rare that closed forms are more numerous in a given visual field and their syntactic function can only be defined by way of the structure of their variables and of their interrelations with other open regions in the visual field.
Moreover, the normal functioning of the eye, according to the three types of vision (including foveal, macular and peripheral), constantly gives rise to the appropriation of individual coloremes specific to foveal/macular vision which can interfere with the gestaltian process of the completion, that is, of the assimilation by peripheral vision of variables pushed toward their ‘good form’, in spite of their specific reality. The foveal/macular centration may systematically reinstitute the irregularity or opening of a form which one has closed or regularized, the discontinuity of texture, an opposite vectoriality in the ensemble, a more faithful chroma of a color which has been pulled toward its chromatic pole, and so on.
These perceptual contradictions, the sources of incessant movements and tensions, lead to the necessity for syntheses. By virtue of such syntheses, the homogenizing gestaltian productions can lose their organizing force to the benefit of more complex spatializations than that, for instance, of the simple mechanism of a form on a ground. If certain characteristics of the visual variables correspond very closely to the pressure for ‘good form’, the perceptual process at times attempts vainly to regroup them into more flexible and open structures.
7.4.4.Regroupings by interaction of colors
The junctions/disjunctions between visual areas by way of the laws of interaction of colors are, above all, the product of the foveal/macular centration. If peripheral vision is particularly sensitive to variations in tonalities and to the movements of mass which it institutes, foveal/macular vision reacts vividly also to clear and dark aspects of chromas which are subjected to the same tonal laws.
The chromatic contrasts termed ‘simultaneous’ or ‘successive’ are of a different order and contribute at an autonomous level to the regroupings/disjunctions of regions in the pictorial plane by laws altogether different from those of gestaltian theory. As we have already described, the phenomenon of interaction of colors creates very particular visual movements which literally transform the material chromas placed on the pictorial plane. For example, by a centration effected by the fovea, to which chromatic percepts of the macula gather, the eye projects a percept of a complementary color to a given region, which will bring it closer or distance it from its surroundings.
A chromatic transformation of a region, through interactions of color, simultaneously modifies all the visual variables of this region and, consequently, the relations of this region with adjacent or more distant regions. Through their mobility and the instability of their effect, the interactions of color constitute extremely powerful, yet at the same time subtle, unstable, and unexpected visual movements. They permit the regrouping or the opposition of regions (while subtracting or multiplying their reciprocal chromas) and the joining of distant regions which are constituted as complementary. They introduce into distant regions forms, colors, vectorialities, and so on, which reiterate totally, partially, or inversely those which have been perceived elsewhere, in relation or not with the energies of the Basic Plane.
7.4.5.Modalities of perspectives
The modalities of perspectives belong properly to the syntactic structure of the field of visual representation. While they constitute instances of subjectivity in defining the point of view taken by the producer on the field of representation—as well as the experimental network of percepts that he puts into play in his work—they constitute ineluctable infrastructures of the pictorial plane. Moreover, they are always concretized in a specific treatment of visual variables and of their modes of regrouping, thus constituting observable elements in the visual work.
As explained in chapter 5, perspectives make explicit, above all else, the distance that the producer takes vis-à-vis his field of representation, the specific sighting of his point of view, and the distance that he puts between the objects which constitute his representation. These parameters are at first established by the recognition of the type of vision required or sought by the various aggregates of the visual variables on the whole surface of the pictorial plane.
The characteristics of the regroupings of coloremes give us data, in effect, according to their internal precision, the treatment of textures, the saturation or luminosity of chromas, the distinctiveness of boundaries, the contrasts of dimension, and so on, on the representation envisaged by the producer and on the type of discourse that he holds at a near, far, or medium distance. They equally reveal whether the percepts sought belong to proxemic spaces, indebted to a particular sensory experience with respect to spaces at a medium or a far distance.
Grammatically, the chosen distances, completed by sightings, angles, or different points of view taken on these fields, are established by programs of organization of the coloremes which cannot be modified without the given perspective being partially or completely transformed to the benefit of another.
However, certain regions of the pictorial plane are frequently treated according to a code of a certain perspective, whereas other regions are treated altogether differently, indicating in effect a transformation of the point of view, of the distance, and of the type of perceptual referents in play. This is why syntactic analysis of perspectives must not presume, on the faith of the perception of certain regions, that the ensemble of the pictorial plane belongs to the same perspectivist modality.
It is extremely important, to the contrary, that the analysis systematically examine the characteristics of diverse regions of the pictorial plane and recognize among the perceptual fields the type of perspectives that are constructed.
In so doing, the analysis establishes the types of spatialities that may be relatively disjointed from one another, both by the position in depth and by the dynamic of the treatment of the visual variables.
If the various proxemic perspectives, constituted by the very mechanisms of perception in the space constructed by the Basic Plane, present among them greater potentialities of regroupings—although they can offer numerous and divergent points of view and perceptual spaces—the illusory perspectives at a medium or a far distance present stronger disjunctions. To the maximum extent, the areas treated according to modes of perspectives entirely heterogeneous in relation to those which organize other regions appear like gaps, ruptures of spatial tissue, or wedges, which no longer possess the elasticity needed to be incorporated into a wave which appears continuous.
In particular, the illusory perspectives often propose irreconcilable points of view to perception because they require a virtual ubiquity of the producer or the recourse to sightings which can only belong to several producers and not to the one that produces the discourse, since he cannot occupy at the same time two or more positions in space. These structures depend upon an abstract or imaginary trajectory which is not linked to the elaboration of a concrete space, but rather to allusions to the hypotheses or logical deductions derived from other discourses than that of visual language. Similarly to what happens in graphic language, they no longer function according to visual language dynamics and its proper creative potentialities. They are debased to illustrating themes already manufactured by other languages (Saint-Martin, 1989).
The different depths, established by the different modes of perspectives, can be described by means of a schema termed a ‘grid of depths’, which has been illustrated earlier (figures XIVe and XIVd). Representing a hypothetical profile of the pictorial plane under consideration, it establishes a series of numeric notations which account for an estimation of the various levels of depth where the observed regions seem to be situated.
Whereas the proxemic depths, which are registered inside the topological mass of the Basic Plane, are considered as ranging inside a maximal distance of one meter in relation to the position of the producer, the medium or far distances are evaluated in relation to the metric measurements used by the external world to which they allude. The distances of depth, therefore, may be registered from an intimate distance of several centimeters to a far distance leading to a point of infinity, interpreted as being situated at more than three hundred meters. The intermediate distances will be defined proportionally at the interior of these limits. It is important to point out again that the depths termed ‘indefinite’ belong by virtue of certain characteristics to a remote visual distance but are registered by other traits in the elasticity of the topological depth.
Moreover, in the representations known as ‘nonfigurative’, which do not make reference to metrically measurable remote distances but which are registered in the topological depths, different parameters will be developed where the various positionings in optic depth will be identified on a scale of one to ten. The dialectic interrelation in proxemic depths is as varied as that of intermediate and far perceptions of distances.
7.4.6.Spatial continuity
The production of more and more synthetic relations between heterogeneous elements produced by the visual variables tends toward the construction of a continuity. This term refers to a liaison or functional integration so close and proximal in nature that the field can be constituted as a unified space in spite of its internal, partial, and different types of spatial organizations.
A space (or a spatiality, a spatialization) is the product of a perceptual interconnection of elements in an unbroken level (or undulation) of depth. The inherent elastic potentiality of the topological regions is the paradigm of any aptitude to spatialization, since other geometrical types of regions may tend to break and generate different orders of levels of depth devoid of continuity.
This continuity can only be the result of an active interrelation between perceptive functions acting on previous disjunctions between the elements of the visual work. The coexistence of discontinuous elements in the visual field or the pictorial plane does not imply as such a continuity but, as its very name indicates, a simple coexistence or juxtaposition of elements, whose unification is undertaken according to hypotheses of spatialization capable of creating strong enough liaisons between them to constitute a unified space. In the latter, the function of the various regions is not negated, excluded, or concealed to the profit of others, but rather affirmed and assumed by the sensibility.
Without unification of the visual field in a given work, no awareness and knowledge of a visual discourse can be achieved. There only results a simple confrontation between partial elements, diversely heterogeneous and at worst, a confrontation leading to a chaos of unrelated percepts. Even the conscious intentionality of a speaker to construct a discourse full of discontinuities or heterogeneities, on the planes both of expression and content, can only be interpreted through a hypothesis which recovers and embraces these very disjunctions—a hypothesis that, in effect, unifies them in the appropriate dialectic continuity. Otherwise, these elements would not find a concrete ground for relations but would call for abstract interrelations produced by criteria external to visual language.
This function of continuity constructed by perception can never be interpreted as an objective trait, in the sense that the pictorial plane would produce it per se, or that the perception would succeed in homogenizing most of the dissimilarities between visual variables. As we have already explained, even a monochrome field without any variations in textures offers a specific spatial disjunction between its peripheral energy, strongly accentuated by the activation of the formative sides and corners and the nonstructured central regions, and only animated by chromatic aftereffects produced by perception. The static effect of similarity between its material constituents has to find ways of integration with the perceptual mobility, so as to be felt as parts of a unified field. If not analyzable in terms of organization, it cannot offer a visual counterweight to the actual energies of the periphery.
The description of regroupings/disjunctions, of vectorial tensions, of positioning of regions in topological or remote illusory depths, leads to the perceptual evaluation of a spatial curve specific to a given pictorial plane which represents the succession of the different levels of depth of the different regroupings of regions. When some regions cannot be perceptually integrated in a continued undulating spatial curve, they are qualified as “problematic.” They play an important role in the syntactic and semantic structures of the work, both by their specificity and autonomy, and by the need they create to summon up for their interpretation super-syntheses, which may be more logical and dialectical than spatial.
This semiotical description can be realized by a variety of means, which make use of numeric functions or graphic representations. In effect, any curve of various surfaces can be represented by a numeric function. Through a division into regular units, the vectorial movement which animates the surface toward laterality and depth can be described by a sequence of numbers in which the position indicates the chosen parameter (figure XXII).
This spatial curve can be explained by the addition to the numbers of an alphabetic notation of ten elements (a, b, c, d, e . . .) in order to demarcate in the ensemble of a work the levels of depth occupied by the regions, from those closest to the perceiver to those farthest away (figure XXIII)
Other graphs based on numeric functions can be developed in order to represent regions of a stronger neighboring or disjunctions, of stronger chromatic activities, of contrasts in perspectives, or else to show the interrelations between the potential and virtual structure of the Basic Plane in relation to that of the pictorial plane.
Other types of graphs are developed for the characteristics common to a certain number of compartments, determining a boundary around these regions in which the different spatial curve is analyzed in relation to those which are produced at the exterior. In the same way, in these descriptions, textural and tonal indices can identify the places where particularly intense energetic nodes polarize the principal vectorialities (figures XXV and XXVI).
These and other graphs, which can be developed in response to various needs, present essentially the undulation proper to a pictorial plane—but in a static way. They must be augmented or completed by other types of schemas symbolizing the vibratory movements which at times cause the pictorial plane to oscillate in a respiratory rotation or alternation regularly activated by the energetic equilibria of the visual work.
To evoke the highly developed energetic phenomena produced by certain contemporary visual works, it is necessary to borrow those vibratory schemas which the physical sciences use in their attempt to represent the experience of movement and organization of nonlinguistic matter. We allude to trajectories and rotations of the ‘spin’, vibrations of gaseous clouds, catastrophic curves or photonic fluxes, as well as to classical movements which animate, for instance, the nucleus of the atom, that is a global vibration of mass which alternately expands and contracts, which changes its shape by spontaneous reorganization of its energetic nodes and which oscillates irregularly in a lateral or vertical movement by a redistribution of its tensions and vectors (Bertch, 1983). Whether it concerns limited regions in a pictorial plane or the ensemble of its visual movements, it seems altogether impossible that nonspatial schemas, such as words, can in some way approximate the events which regularize the syntactic rules of visual language.
Thus, through the modes of relation between coloremes, regions, and various perspectives, the ensemble of these descriptions permits us to recognize and distinguish the various perceptual spaces which correspond to the specificity of these regroupings developed by the producer, offered to the eye, even if they correspond to spatial experiences relevant to different sensory fields. In this case, the so-called visual field of the pictorial plane, which offers the apparatus for the elaboration of different perceptual spaces within visible space, is used with a view to construct a representation of the organic spaces constituted by stimuli whose modes of interrelation and integration differ to a great degree from those which can regroup visual stimuli.
In a general fashion, the spaces organized in the zones close to the body itself may be said to refer to sensori-motor, emotive/conceptual experiences and interrelate tactile, thermic, postural, buccal, kinesthetic, and auditory spaces. The spaces constructed at a medium distance refer to kinesthetic, auditive, and visual experiences. The spaces evoked at a great distance refer to auditive experiences but especially to visual ones. All these spatial experiences are intermingled in representation as they are in reality.
The majority of the more traditional pictorial works in the occident are constructed by the disjunction/juxtaposition of tactile and postural spaces evoked at the first plane, kinesthetic spaces and their volumetric connotations at the intermediate plane, visual spaces mostly at the third plane.
These disjunctions between the perceptual spaces involve the construction of a spatial curve which encompasses the various depths and which will appear more or less chaotic or disjointed, according to the very potentiality of the ensemble to be reintegrated at a more synthetic level by certain of its structural characteristics.
In opposition to certain procedures of typological, probabilistic, or statistical analyses, semiotical analysis cannot brush aside, as a nonelement of the structure of the work, any of those regions which seem to be disjunctive, heterogenous, idiosyncratic, or not linked to a possible global perceptual space that can be constructed in a given pictorial plane. These oppositions/disjunctions form the very semiotic material of the visual discourse inasmuch as it is a cognitive and performative discourse, aiming at the production of a unified spatiality out of the very heterogeneous character of experience itself.
The purpose of semiotical analysis, by means of an actual experimentation with the tensions/contrasts/ruptures affecting the elements of the discourse, will be the construction of a spatialized representation permitting their perceptual integration. These mechanisms of integration, of which the guidelines are furnished by the particular treatment accorded by a producer to the visual variables, are based on the emotive and conceptual tensions of the perceiver himself. Actualized in various experiences of the work, this should lead to the elaboration of a unified construction and offer integration of the internal discontinuities. For want of this necessary leap in the production of a spatial model of the representation, the visual work itself remains nonunified, relatively chaotic, in other words, not known and not understood by the perceiver.
7.4.7.Syntactic articulations
The aim of syntactic analysis is the recognition of the specific articulation of the spatialities within a visual text based on positions in the third dimension. It must produce a final segmentation of the pictorial plane corresponding to the diversity of its spatialized regions; their components, functions, and intensities; their reciprocal interrelations; and the possible isolations of some of them. This representation establishes a few more or less autonomous or interdependent “propositions” within the whole that they articulate. Given the energetic character of these “propositions” whose sum will correspond to the performative statements realized in a visual representation, they would be better named by the term “events.”
This final segmentation can present analogies with the division of the field already operated on by peripheral vision, since depth is a perceptual construction based on points of reference in the visual field. But the synthetic segmentation will diverge essentially in that, instead of isolating some larger superregions, it will distinguish among them an energetic hierarchy which discloses their liaisons and specific interrelations.
The last steps in the syntactic analysis of a visual representation consist in a perceptual assessment of the differences in depth between regions. This perusal will lead also to the recognition of the most important subdivisions of these large regions. These subregions can be as numerous as their characteristics are influential in the internal or external liaisons between regions for a given perceiver.
The tensions arising from the dissimilarities between the visual variables of boundaries or contours will lead to the analysis of the passages between the regions, establishing the quality of the frontiers (firm, blurred, degraded), their gestaltian quality (regularity, diffuseness, cohesion, figure/ground disparities, openness/closure). It is mainly through the visual variable of closed forms that regions or subregions lend themselves to the process of “iconization” through which they are interpreted as similar to the mnemonic percepts of objects belonging to external reality. On the other hand, chromaticity may be more active in the “recognition” of open forms (sky, earth, sea, forest, etc.). This iconization which is usually realized differently by several perceivers would require that a definitive list of iconic elements assigned to a visual representation be corroborated by an adequate group of observers (Saint-Martin, 1987, 1988).
While the syntactical function in the visual text establishes the spatial interrelationship between elements and their perceptual correlatives, the iconic regions, always linked to a verbal label, install thereby a “verbalized” level in the text whose semantic correlations are different from that of a spatialized language.
The analysis of the modes of insertion of the regions into the Basic Plane infrastructure will reveal higher levels of energetic intensities in parts or in the totality of the regions. The accrued intensities of regions anchored in various ways into this energetic system result in their movement toward the front in a more proxemic depth. Phenemona linked with the interaction of colors, mainly in relation to the simultaneous and successive contrasts of color and tonalities, will contribute to strong movement toward the front. The after-effects appear indeed like chromatic films hovering above the “surfaces” of the regions and intermingling with them, diminishing the distance between them and the perceiver that may be built from the other visual variables.
Disparities between the visual constituents will produce specific tensions between regions, when some are at the same time attracted and pushed away by different visual variables. These separative and distancing effects will first define the implantation in the two-dimensional plane, which is not equivalent to a topographical localization of regions in the field, but to the nature of the links which tie, contract, or expand them, contributing to the rhythmic expansion of space.
These disparities between regions will produce even more forcefully variations in the implantation in the depth or third dimension. But to these factors, localization in depth will add a certain amount of “deictic” information, referring to the producer and establishing his position in relation to his spatialized field of representation. This will be recognizable through: a) the type of vision according to which the visual variables are produced on the pictorial plane, the products of foveal vision being closer to the producer than those of macular or peripheral vision (This analysis is better achieved through the use of the grid over the whole field.); b) the overall distance from himself that the producer establishes for the ensemble of the representation; c) the angle of the position of his whole body in relation with the field (above, under, oblique, etc.); d) the angularities of specific sightings or directions of the gaze on the same regions; and e) added information is given about the producer through the various distances in depth he establishes between the objects themselves positioned in the field of representation.
These variations in depths, added to the recognition of specific traits in the visual variables, reveal the unified or contrasted use of modalities of perspectives, both expressive of a “syntactic meaning.” The intimate relation between syntax and meaning was demonstrated by Tesniere in his analysis of the structural syntax of verbal language when he concludes that “the structural expresses the semantic,” or more precisely: “To structural connections are superimposed semantic connections” (1976; 42). This is even more important in visual language as this form of language provides two levels of signifiers, one which can be termed “factual” or representing direct production of junctions/disjunctions or topological relations and another which remains linked with the verbal or iconological forms of representation. These two levels which carry different emotional and semantic contents are basic to syntactical constructs (Saint-Martin; 1987).
Graphs can be produced of the in-depth distribution of the regions and subregions with two entries: their effective distribution from left to right in the field and their estimated perceptual depths. Such a linear graph illustrates at once the leaps in depth which may exist between contiguous regions on the pictorial plane, some so abrupt as to break the spatial continuity and appear as problematic disjunctions in a discontinuous milieu (see figure XXVIII).
The energy and the interrelations of the various regions, multiplied by the intensification provided by the Basic Plane potential and actual energies, will finally define the different levels of events in the field, leading to the proper syntactic segmentation and articulation of the parts in the spatial text. They will allow for the recognition of the rhythmic oscillations and vibrations in the field, leading to definition of its main perceptual referents.
In using the term ‘text’ to designate a visual work, such as painting, photography, or sculpture, etc. analogous to its use in the verbal domain, we wish to retain two characteristics. The first is the fact that a text possesses a certain structural autonomy by virtue of which the parts are understood as a function of the whole, though we do not pretend to determine what can be truly defined as an autonomous text in the verbal domain. In addition, we would like to evoke the synthetic and pluralistic connotation of the term, in opposition to a comparison of the visual work with a single propositional phrase as those which compose verbal prose.
On a belief in the possibility of the translation of visual language into verbal language, Umberto Eco already observed that the simple description of a silhouette of a horse drawn on a sheet of paper requires the cooperation of a large number of verbal statements (1978; 104). In this regard, the description of a more complex visual work would correspond to a very large number of statements indebted more to a textual grammar than to a propositional grammar.
But in all respects, it is more fitting to speak here of a grammer of the field, since no subensembles of the visual work can correspond to the structure of the verbal statements composing a text. These are commonly described as formed by the sequence of syntagms of noun, verb, and predicate (NS—VS—PS) at the interior of syntactic paradigms, such as gender, temporal modalities, and so on.
Unfortunately, the analysis of the visual field cannot find support in a general textual grammar, in fact nonexistent to this day, except for some narrative schemas linked to a theory of action, relatively foreign to the activity of representation put into play in artistic visual expression.
Despite several recent works, such as those of Ray Jackendoff (1983; 170-175) showing a growing concern with the cognitive and perceptual functions of language, most verbal grammars are founded on a radical omission of the spatial dimension in human experience to the benefit of temporal succession, in accord with the most tenacious tendencies of occidental philosophy (Frank; 1986).
To counteract this assimilation of dynamically interrelated regions in the visual field to verbal statements, and although both languages serve the same basic functions of mental representation, we call “events” the agglomerates of regions endowed with a specific and contrasting energy which “autonomizes” them as dynamic nodes in the field and positions them at sufficiently distinct levels on the scale of depths.
Inasmuch as these regions/events represent very strong condensations of energy and appear to occupy disparate positions, implying bifurcations and leaps in the field, they would not contradict the proposition of Rene Thom (1972) to the effect that “any event is a catastrophe.” They could eventually be describable by a grammar of catastrophes as we have already suggested for the poetic text (Saint-Martin, 1985b).
These visual events are presented, in effect, as complex, qualitative reorientations of energy, but always suitable to be interrelated by various mechanisms of envelopment, emboxing, optic attraction, to produce the rhythms of topological successions. In addition, they present, on an internal plane, subevents more or less integrated or discontinuous in their containing region, in active interrelation with other events of the field through their particular environmental position. Moreover, certain radical discontinuities in the three-dimensional liaison, often explained by systems of heterogeneous perspectives, institute problematic energetic nodes, forming antagonistic poles to the discourse of the surrounding regions.
In its progressive modeling of structural elements of the visual field, visual semiotics will be careful to avoid the obstacle that has constituted, for numerous verbal grammars, a mode of graphic representation using a discontinuous neutral trait to represent the relations between its principal terms. This graphic artifice, as demonstrated by J. Petitot-Cocorda (1985; 141), most frequently leads to a substantiation of terms, a fixed reification of the relation, and their subsequent assimilation by the logic of the identity. As suggested already by Kurt Lewin, it appears more heuristic to call on notions of topological and “ensemblist” wholes to represent, at more abstract levels, the dynamic liaisons between elements in transformation.
We propose, therefore, to substitute for the straight lines used in the ‘stemma’ schema proposed by Tesnière (1976; 162), the curves in helix, like corkscrews, capable of recalling the continuous and reciprocal interaction between regions (figure XXIX).
This form of graphic representation shows, moreover, that we generalize to all the interrelations between regions of the visual field, the “rubber-band” effect observed by Arnheim in many visual representations (1982; 240). It is the term he applied to the magnetic inter-tension which links the peripheral regions of a work to its focal region. This elastic tension exists, according to topological semiology, between all regions of the field, even those defined as separate or foreign with respect to certain of their characteristics, because the perceptive tension itself always generates movements bringing them closer or moving farther off in the three dimensions. In particular, this elastic tension is characteristic of the pulsating link between the regions of the pictorial plane which redouble or reiterate the structures of the Basic Plane and its energies.
The visual events can be represented by oblong shaped figures, of variable dimensions according to their energetic impact, on which will be positioned, by intersection, other events which are subordinated to them by encasing or embedding. The liaisons offered by the pictorial plane itself between the principal or subordinate events resulting from superimpositions, optic links, effects of shadows, and so on, will be represented by dotted curves.
One could schematize the syntactic structure of a hypothetical visual work in the following manner, by regrouping the primary peripheral regions forming an event under a capital letter (EA, EB, EC . . .), and the subevents by these initials and an added capital letter.
In this illustration, the principal events EA and ED are interconnected by a strong disjunction. The regions EB and EC are the subordinates of EA by encasing and embedding. The subevent SEE links EA to EB; SEF links EA, EC and ED; SEG links EC to ED. While describing factual energetic liaisons, these systems of separation and dependency, quite different from the early peripheral segmentation of the field by peripheral vision, will be analyzed as to their perceptual space organizations, providing a basis for interpretation.
The magnitude of an actual application of syntactic analysis of a visual representation, which is a quite lengthy procedure, prevents its publication in this book. The reader may refer to a sample in another publication (Saint-Martin; 1986).
We are also entrusting to a future work the development of the semantic system of topological visual semiotics which constitutes its aim and last term. Any attempt at the interpretation of visual language can only rest on the analysis of the syntactic data it offers. But semantic hypotheses have been alluded to in the very study of syntactic theories. This is inevitable since any grammar consists of the set of linguistic methods by which meaning is conveyed by the intermediary of particular sensory materials and codes.
We use cookies to analyze our traffic. Please decide if you are willing to accept cookies from our website. You can change this setting anytime in Privacy Settings.