Once the signals are received by a human being, information theory has nothing else to add and gives way to either semiology or semantics, since the question henceforth becomes one of signification—the kind of signification that is the subject of semantics and that is quite different from the banal signification that is the subject of information. On the other hand, it is precisely the existence of open works (that is to say, of the openness proper to works of art, the existence of messages which manifest themselves as sources of possible interpretations) that requires an extension of the notion of information.
It would be fairly simple to show that information theory was not conceived to explain the nature of the poetic message and that, therefore, it is not applicable to processes involving both the denotative and connotative aspects of language—so simple that everybody would immediately agree with the proposition. On the other hand, it is precisely because information theory cannot and should not be applied to aesthetic phenomena that numerous scholars have tried to apply it to the field of aesthetics; likewise, it is precisely because information theory is not applicable to processes of signification that some have tried to use it to explain linguistic phenomena.
Indeed, it is precisely because in their original usage the concepts pertaining to information theory have nothing to do with a work of art that, in this essay, I have tried to determine to what extent they can be applied to it. Of course, if they had been applicable to begin with, there would be no point in trying to find out whether they could be applied or not. On the other hand, the only reason I want to find out is that I think that, in the end, a work of art can be analyzed like any other form of communication. In other words, I believe that, ultimately, the mechanism that underlies a work of art (and this is what needs to be verified) must reveal the same behavior that characterizes the mechanisms of communication, including those types of behavior that involve the mere transmission, along one channel, of signals devoid of all connotative meaning, which can be received by a machine as instructions for a sequence of operations based on a preordained code capable of establishing a univocal correspondence between a given signal and a given mechanical or electronic behavior.
On the other hand, the objection would be insuperable if the following points were not now clear:
The application to aesthetics of concepts borrowed from information theory has not generated the idea of the open, polyvalent, ambiguous work of art. Rather, it is the ambiguity and polyvalence of every work of art that has induced some scholars to consider informational categories as particularly apt to explain the phenomenon.
- The application of informational categories to phenomena of communication has by now been endorsed by a number of scholars, from Jakobson, who applied the idea of integrated parallelism to linguistic phenomena, to Piaget and his followers, who have applied the concepts of information theory to perception, all the way to LeviStrauss, Lacan, the Russian semiologists, Max Bense, advocates of the Brazilian new criticism, and so on. Such a fertile interdisciplinary and international consensus cannot be seen as a mere fad or a daring extrapolation. What we are confronting here is a categorial apparatus that may provide the key to several doors.
- On the other hand, even if we were confronting mere analogical procedures or uncontrolled extrapolations, we would have to admit that knowledge often progresses thanks to an imagination that explores hypotheses and dares to take uncertain shortcuts. Too much rigor and an excess of honest caution can often deter one from venturing along paths that could well be dangerous but that could also lead to a plateau whence an entire new landscape would open up, with roads and highways that might have escaped a first, cursory topographic inspection.
- The categorial apparatus of information theory appears methodologically fruitful only when inserted in the context of a general semiotics (although researchers are only now beginning to realize this). Before rejecting informational notions, one must verify them in the light of a semiotic rereading.
Such a semiotic endeavor could not, of course, be encompassed in this essay. The objections I have tried to answer in this postscript were for the most part raised by Emilio Garroni, author of one of the few exhaustive and scientifically sound critiques of Opera aperta.25 And I do not pretend to have satisfactorily answered all his objections here. These comments are intended, in fact, to supply this essay, which still maintains its original structure despite numerous revisions, with a few answers to possible future objections. They are also designed to show how some of these answers were already implicit in the original argument, even though I did not make them explicit until stimulated by Garroni’s observations. It is thanks to these
III Infinmation and Psychological Transaction
I hope that this discussion has demonstrated how a mathematical study of information can provide the tools necessary for elucidating and analyzing aesthetic structures, and how it reflects a penchant for the «probable» and the «possible» that mathematics shares with the arts.
On the other hand, information theory evaluates quantity and not quality. The quantity of information concerns only the statistical probability of events, whereas its value can be measured only in terms of the interest we bring to it.26 The quality of information is related to its value. To determine the value an unforeseeable situation may have for us (unforeseeable but verifiable, whether it be a weather forecast or a poem by Petrarch or Eluard), and the nature of its singularity, we must consider both the structural fact in itself and the attention we have brought it. At this point, questions of information become questions of communication, and our attention must shift from the message itself, qua objective system of possible information, to the relationship between message and receiver—a relationship in which the receiver’s interpretation constitutes the effective value of the information.
The statistical analysis of the informative possibilities of a signal is, in fact, a syntactic analysis, in which the semantic and pragmatic dimensions play only a secondary role, the former to define in what cases and under which circumstances a given message may provide more information than another, and the latter to suggest what kind of behavior this information might entail.
But although the transmission of signs conceived according to a rigorous code, based on conventional values, can be explained without having to depend on the interpretive intervention of the receiver, the transmission of a sequence of signals with little redundancy and a high ratio of improbability demands that we take into consideration both the attitudes and the mental structures by which the receiver, of his own free will, selects a message and endows it with a probability that is certainly already there but only as one probability among many.
This, in turn, means that it may be necessary to add a psychological point of view to the structural analysis of certain communica tion phenomena—an operation that may seem to contradict the antipsychological tendency of the various formalist methodologies that have been applied to the study of language (from those of Husserl to those of the Russian Formalists). On the other hand, how could one examine the signifying possibilities of a given message without taking the receiver of the message into account? To consider the psychological aspect of the phenomenon merely means that we recognize that the message cannot have any meaning, at least formally speaking, unless it is interpreted in relation to a particular situation (a psychological situation that is also, by extension, historical, social, anthropological, etc.). 27
It is therefore necessary to consider the transactional rapport that is established, at both an intellectual and a perceptual level, between certain stimuli and the world of the receiver—a transactional rapport that constitutes the very processes of perception and reasoning. In the case at hand, this kind of analysis is more than a methodologically necessary stage: it confirms everything I have said up to now concerning the possibility of an «open» appreciation of a work of art. In fact, a basic theme of the most recent currents in psychology is that of the fundamental openness of every perceptual and intellectual process.
These perspectives are founded on a critique of Gestalt psychology, which maintains that perception is the apprehension of a configuration of stimuli, that is, of stimuli that already possess an objective organization—recognition more than apprehension, thanks to the fundamental isomorphism between the structures of the object and the psychophysical structures of the perceiving subject.28
Later, postGestalt schools have reacted against the metaphysical burdens of this psychological theory, and have described the cognitive experience as an experience that occurs in stages, as a process that, far from exhausting the possibilities of the object, highlights those aspects of it that lend themselves to an interaction with the dispositions of the subject.29
American transactional psychology, an outgrowth of Dewey’s naturalism (and other French currents, of which more later), maintains that although perception is not the reception of physical stimuli, as described by classical associationism, it nevertheless represents a relationship in which my memories, my unconscious persuasions, and the culture I have assimilated (in other words, my acquired experience) fuse with
