[Due to space constraints, the foregoing essay was be published in three instalments. Parts I and II can be found here and here.—Ed.]
Struggle between Symbol Equality and Logocentrism
The way out of the irrationality of imagery—understood as a united symbolic system—lies precisely opposite to physical objects, in reason. None of the philosophers mentioned earlier denied that the level of Kantian law, while being outside time and space itself, determines our capacity for apperception; but all art critics tried to solve the problem of artwork interpretation mostly by observation, merely referring to “something else” affecting their belief systems. Here, of all the interpretations of Kant (and there are so many that diametrically opposed ones exist on most issues), we will follow Sellars, for in his reading one can most clearly delineate a distrust of empirical observation of the object—and it is justified, as shown by the preceding parts of this text. We will take into account that Sellars’s focus was strictly that of scientific observation. We will also have to make some comparisons with the approach of Nelson Goodman as a philosopher of art, in order to find the distinction between current art and current science (both having productive eschatological views of themselves) which would become completely independent from their respective schemata.
In the introduction to Science and Metaphysics, Sellars acknowledges that Kant maintained an ambivalence in his approach to rational activity involving observation of objects. On the one hand, such activity requires the capacity for perception and, on the other, conceptualization—that is, understanding. This ambiguity gave rise to a school of philosophers of science whom Sellars grouped into the category of sensory data theorists—the truth of scientific observation for them being at the level of perception. Remaining in the approach to the external realm faithful to Kant, these theorists divided the undifferentiated stream of data into micro perceptions, and postulated the whole foundation of reason as a product made up of bricks of such micro perceptions.
As Sellars discovered, the viewpoint of these theorists does not withstand an attack made even after a cursory analysis. The empiricist theorist claims that one’s ability to perceive colors is innate, hence the perception of a red triangle (we do not know whether the schemata has any external referent) is primary and true; but this theorist also claims that one’s knowledge of the shape of the triangle is acquired, and thus the perception of the red triangle is both an innate sense and an acquired knowledge, which cannot be the case simultaneously.
The second, already brewing question, is whether the perception of a red triangle entails the existence of some physical object that meets both of these criteria. Again it turns out that, as the reader guesses, the observer may be hallucinating, or completely unaware of their surrounding reality due to, for example, poor lighting; and the question of what is behind the phrase “I perceive a red object” breaks down into several sub-questions: Does the theorist see a red object? Does it seem to the theorist that the object is red? Does it seem to the theorist that some red object exists in the direction of their gaze? As can be understood from this series of questions, the level of confidence in the truth of the judgment about the red object will vary greatly depending on which of these questions is answered in the affirmative. Sellars persists in making it difficult—for a scientist using methods of an art historian—to ask if we can say that the perception of a red object is itself red. To this he replies that even empiricists contemporary to him would not make such a claim, and would eventually arrive at the conclusion that it is only possible to unite these concepts discursively without making them uniform:
To ask how impressions fit together with electro-magnetic fields, for example, is to ask a mistaken question. It is to mix the framework of molar behaviour theory with the framework of the micro-theory of physical objects. The proper question is, rather, ‘What would correspond in a micro-theory of sentient organisms to molar concepts pertaining to impressions?’ And it is, I believe, in answer to this question that one would come upon the particulars which sense-datum theorists profess to find (by analysis) in the common sense universe of discourse. (italics in original)1
In an implicit argument with Charles Peirce’s “diagrammatic thought,” where logical procedures are necessarily corrected by visualization, Sellars claims that inside the human mind and in human practices—which he does not separate—there is no difference between symbols received through different sense channels. Knowledge in general produces a discourse that connects logical laws with the referents of speech in natural languages, ultimately structuring the very practice of speaking. If we look at the chapter Our Rylean Ancestors, we see that in science as an institution, in public speech in principle, and within the thinking of each individual, the same process of ordering utterances takes place. In science, it is called modeling: the application of mathematically computable laws to the life of objects familiar to the scientist. In everyday speech the same function, but with less rigor, is performed by speech episodes; and in the structure of the personality of the individual, it is so by what in everyday and in Cartesian speech is called “thought,” in Kant’s, “imagination,” and which Sellars renames to “inner episodes.” It is precisely at this level that the logical links with symbolic forms circulate both between different individuals and within each particular consciousness, with no structurally significant difference between the two streams. All that society produces is perceived by the individual, and all that the individual expresses becomes part of the general stream of symbols—yet Sellars insists that speech first arises intersubjectively, and only then is its structure used by the subject in inner episodes. The philosopher does not call this process thought. Not only because structural shifts caused in an individual by the structure of the flow of external symbols are not reasoning yet, but also in order to minimize what he branded as ghostly entities: hypotheses of an evil demon lying to the subject about the order of logical connections, present in some particular interpretations of the stage of doubt that Descartes went through until he got to “therefore exist.” For Sellars, so-called thoughts are internal commentaries on previous statements, meta-expressions:
For characteristic of thoughts is their intentionality, reference, or aboutness, and it is clear that semantical talk about the meaning or reference of verbal expressions has the same structure as mentalistic discourse concerning what thoughts are about.2
Simultaneously, at the beginning of Truth and Correspondence, Sellars takes an unmodified Platonist approach to the definition of thought as a “dialogue in the soul.” Such is done in a polemic vein, and he gets caught into the traditional notion of natural language as a set of aural and logical symbols interconnected in the process of human speech, while it is already clear that we at least do not use aural symbols in the thought process, according to Sellars. An issue arises: What do we do with artificial languages, prototypes of which are symbolic systems used in art, apparently not used in the thought process by a human being?
The issue of the various ways in which symbols are expressed, and of how it is possible for a subject to distinguish between symbols, when those are, in fact, external objects among any others, constitutes a strong and recurrent theme in criticism of Sellars’s method. Most such critics would either offer to return to more naturalistic, Helmholtzian ideas of how mind functions, with a neat explanation of contour-rendering by eye movement (and no explanation of a further destiny of this contour), or take a radical pragmatist stance, scrapping the idea of any truth in symbolic comprehension/processing/expression altogether.
The most elaborate criticisms, going beyond dismissals, were collected by Sellars’s disciple Jay Rosenberg, in his Wilfrid Sellars: Fusing the Image.3 The most perplexing concept in Sellars’s writings for this particular text is that of language-entry. It apparently replaces perception in his system, but not taking into account a) the fact that language is primarily collective, while perception is at least believed to be individual and b) the fact that internal episodes are not language per se, precisely due to the fact that they use no visual or aural symbols and are not communicable, at least with today’s technical capabilities. While not giving an answer to these objections revolving around fragile personal boundaries, Sellars said that if it is possible to structure the thought of a subject with symbols, these play a different role in the life of subjects than the objects of nature and shall be treated differently. The scope of his approach would be more limited if we treat artworks as natural phenomena than if we refer to them as symbolic systems. Still, both approaches are possible. We can thus apply any scientific theory which can explain anything of interest about artworks and reveal instances when appearances deceive us, as Marxist theorists do, but we can also treat artworks as the statements that they are.
As we will find later in Goodman, non-linguistic symbols have the same potential for structuring perceptions as linguistic ones. Their ocularity is not a separate problem, but the thing that has distinguished art statements so far has been an abundance of eccentric “frames of reference” in it. Whereas in natural languages we could meet a person who would say “the sun never moves” but still agree on other issues with that person—because this expression marks only one frame of reference distinct from our own—in artworks everything can be put in frames of reference unusual for a natural language speaker.
We shall nuance this approach with a remark that such individualistic approach to referencing is particular to Western, twentieth century art, while symbols remain highly conventionalized in all other known canons of depiction. At the same time, we cannot call art statements necessarily primitive, syntactically speaking. The symbolic systems of art vary in complexity. They can be found at the level of primitive syntax, what Sellars called “jumblese:” juxtapositions of symbols with a hierarchy based on their position in a statement; or in another version of a simpler system, a map language indicating the connections between points. Moving up, they’re also found to be working at almost the level of scientific languages—if and when artists intend their works to be explanatory tools, although not necessarily in cases of scientific modeling, which can be used as well.
At such a level, symbolic systems get interconnected with propositions about symbolic units called facts, statements that something is the case—making the whole expressions if not as complex, almost as complex as those seen in science. Taking into account flexibility in symbolic system use, art could introduce artificial languages of any complexity, providing those with learning interfaces. The biggest trouble is that such systems are not immediately within reach for a human organism and will remain on the stage of sketches if no agent with a default visual channel of communication appears. We would call them sketches until any such system alone is able to produce an infinite range of meanings. As for a perceived semantic vagueness of artworks (included by Cassirer in his reference to the language of poetry), Goodman would answer that the symbols are as precise as could be for their frames of reference, otherwise statements would lose coherence.
However, if we look more closely at Goodman’s texts in order to depart from logocentrism even further in our argument, we will see that his description of the structure of thought suffers the same fallacy as Peirce’s. For example, in the first chapter of The Languages of Art, Goodman said that a symbol that represents an object is not necessarily similar to our perception of the object itself. As to the extent to which he thinks that perceptions can in any way be codified, his concept of similarity gains somewhat more clarity in another work, Ways of Worldmaking. There, Goodman distinguishes between representation as the equivalence of an object in a symbolic system and exemplification as the reflection of a number of perceived qualities of an object in the exchange of information between people. As an example of representation, he cites paintings by authors of different eras, made in different visual canons, but, judging by their titles, devoted to the same objects. As an instance of exemplification, Goodman suggests fabric samples in a sewing goods store, which accurately convey the characteristics of the product: all, except those that are expedient to report mathematically (such as width). In both this and another case of exemplification—the demonstration of physical exercise by a trainer in a gym—we will see that we are ultimately trying to get a grasp of the objects in their pre-symbolic form, not at all with a symbolic function.
The fact that in some areas of life people have not invented symbols for the physical qualities of objects they need in practical activities does not mean that these qualities are contained in the subject’s mind in an unchanged form or even that they are communicable: the cloth sample and the exercise example lose much of their functionality when fixed in photo and video form. In Languages of Art, Goodman stipulates that processes similar to exemplification also take place in art. For example, he states that a caricature may be more like the person it depicts than a photograph. This passage raises doubts as to whether his concept of “likeness” describes a homogeneous series of phenomena and has scientific rigor. In the first two examples, we have seen the objects themselves presented in such conditions that they can communicate the desired qualities to the tailor’s client, while in the case of photography and caricature these are two symbolic systems that can more or less reflect one’s ideas about the portrayed person. Trying to find a special place for visual symbols in the subject’s internal structure, we get caught up in all the contradictions of this approach and are left inside two disconnected and non-communicating systems of perception, which seemed to be the dead end from which we had just gotten out of. We have no choice but: To abandon this idea altogether and take the risky step of equating different kinds of symbols in all aspects, except for the path these symbols take from the understanding of the subject to sense organs; and to state that until symbols remain in consciousness, they are operated as identical and interchangeable discursive units if conceptually equal.
Still, Goodman’s texts will also bring us to a conclusion that, against Sellars, only historically has natural language been the most refined channel of information transmission. Goodman cites traffic lights as an example of an extremely narrowly defined visual symbol. We could also speculate that traffic lights with about ten combinations could already become a base for a modern language, and an imaginary philosopher would say that we think in rounded colorful blinks, and dismiss any shapeless or square-blink theories. Consider the scientific discovery that language-use troubles in subjects lead to grave dimensional-perception aberrations. Such discovery may be attributed to a high degree of refinement of natural languages, as compared with other communication systems, rather than to something particular to this channel of information transmission.
In this sense, whereas all theorists mentioned in the text before Sellars have approached artifacts of art as natural phenomena in a broad sense, we will treat statements in art in accordance with their purpose: to shape our perceptual systems. Although due to a variety of constraints these belong to unfinished symbolic systems which are impossible to use unmodified in inner episodes, they picture the world in a manner highly unusual for a natural language speaker, and, as a result, we do not quite understand the message. Using this method, we will find out that art is as much an explanatory mechanism as is science; but rather than what is the case, it explains what ought to, or might be the case; a mechanism having no authority of science, readily disclosing its epistemic eccentricities.
If we assume that art and science will merge one day because they may use any symbolic system deemed as most suitable and have a single goal of life improvement, and if bodies able to complete visual systems appear in our world, we can already see a big trouble in their co-development from the current state of affairs. Whereas science “measures all things” and forms a normative appearance of states of affairs in public perception with a humble aspiration of mending things in the world, of removing the spell of objects traumatizing knowledge to oblivion, many non-normative appearances of art can be produced alongside—free from the institutional confines of science and its responsibility, more symbolically flexible, and, although this is barely possible now, eventually they will compete with the pictures of science in explicative power, because the distance between “might be” and “is” is not a constant.
Some will object that the author of this text elevates quasi-intellectual fakes and conspiracy theories, all of which do exist in art today, to the level of advanced knowledge. Still, no symbolic system is immune to refinement, and past a certain point, we will not apply such terms to most artworks—terms reserved for systems of knowledge of a finite range of development. One day, science and its arch-enemy—one it is not even considering to be a minor problem now, wasting its discursive attacks on silly folk beliefs: two major symbolic systems having an opposite attitude towards the fact that the only certain thing about them is that they form an appearance (science despising, and art boasting it)—will meet in a final reconciliation of accounts.
The petit-bourgeois drama of contemporary art (not a tragedy, as we will believe) is that it misuses its symbolic system sketches. The visual art that we see, or rather, that we discuss, has not even connected itself to public discourse. The legitimation of new artworks (though, returning to the introduction, often produced with a different intent) is based on their assumed stylistic similarity to older artifacts recognized as artworks, not on any discursive principle. Any claims that art is engaged in the popularization of knowledge obtained in other discourses, break against an argument that artworks themselves do not contribute anything to these discourses, only drawing attention to them at best. But a profound change, which would turn art workers into public speech participants with statements in various media, cannot use edutainment as its strategy either—this is too small of a purpose for a project with no limits ahead. In order to beat science on its premises, the artist must themself be a participant of the research process and use the languages of art in those situations where their own ideas about what exists lay the foundation for weird deliberations about anything other than that.
The presence of a separate and falsely autonomous entertainment sector, where all the things we are discussing take place, remains an obstacle for this to be seen anytime soon. Another problem is that it will become completely unnecessary for there to be separate people, writers and managers, that are nowadays trying to attach artistic expressions to discourse by using all sorts of advertisement tricks. These expressions will finally be acknowledged as statements, with necessary recoding procedures included by an artist or a caring community. The occupation of an art critic, already vanishing, will be confined to the vocation of historian. But this would be a small sacrifice for a better world, that may one day grow tired of its modern mythology and set expectations for art that it can meet, if we still have time for this to happen.
NOTES