The Message in the Bottle and Lost in the Cosmos

by Walker Percy

  The Structure of Symbolic Behavior

  It would not, perhaps, be inaccurate to say that American psychology, as well as other behavioral sciences, has settled on an eclectic behaviorism in which the cruder features of Watsonian psychology have been refined by the work of Tolman, Skinner, Hull, Mowrer, Dollard and Miller, Sears, and Angyall. In this view, also put forward at the pragmatic level of semiotic, the organism, whether human or subhuman, is regarded as an open system living in an environment and adapting to that environment through its response to elements which are called signs. A sign is defined as an element in the environment which, through congenital or acquired patterns of behavior, directs the organism to something else, this something else being understood either as some other element or simply as biologically relevant behavior. Thus, the scent of deer directs the tiger to the deer; the scent of the tiger directs the deer to flight. A good representation of this relation is the semiotic triangle, shown in Figure 5.*

  The relations between signs and interpreters and between interpreters and objects are of the nature of space-time transactions between an organism and its environment and can be studied by a natural science. The relation between sign and object, shown in Figure 5 as dotted, has been called an imputed, as opposed to a real, relation. But this imputed relation is ambiguous. Does it mean that naming is folly and not the fit subject of a natural science, or does it mean that it is a formal relation and open to study only by a formal science? But naming does happen. People give names to things as surely as rats find their way through mazes.

  The problem, it would seem, is how to give an account of symbolic behavior considered not in its formal aspects—as it would be considered by grammar, logic, and mathematics—but as a happening and, as such, open to a natural science.

  Although the semiotic triangle is a useful model of stimulus-response arcs and of learning behavior, the fact is that symbolic behavior is irreducibly tetradic in structure, as shown in Figure 6.*

  The second person is required as an element not merely in the genetic event of learning language but as the indispensable and enduring condition of all symbolic behavior. The very act of symbolic formulation, whether it be language, logic, art, or even thinking, is of its very nature a formulation for a someone else. Even Robinson Crusoe, writing in his journal after twenty years on the island, is nevertheless performing a through-and-through social and intersubjective act.†

  The new ensemble of elements and relations which comes into being does not replace but rather overlays the organismic interaction. People still interact with each other behavioristically as much as do dogs and bees, but they also enter into intersubjective relations and cointend objects through the vehicle of symbols. It is possible, and indeed preferable, to describe symbolic behavior in an operational language which omits reference to mental contents or even to “meanings.” “Ideas” are difficult to define operationally and even more difficult to bring into coherent relation with the observables of behavioral science. As for “meanings,” the word is itself so ambiguous that there is more to be lost than gained from its use. It seems least objectionable to say that in the particular communication event under consideration, an organism intends such and such a designatum by means of such and such a symbol.

  This approach still deals with elements and relations, just as does that of the neobehaviorist. A list of the elements and relations of the symbolic meaning-structure, and an example of their clinical application, follows.

  The intersubjective community. Whenever behavioral scientists are confronted with a concrete language event, appropriate questions are: What is the community? What is the status of the intersubjective bond? Who is included and who is excluded? Is the community I-you-you or I-you-not you (as it is sometimes when one goes to a very high-toned lecture: we are listening and understanding, and we are quite aware that those out in the street are not)? The community may vary from a face-to-face confrontation of two people and the various colorations of the I-you bond, to the scattered and numerically unlimited community of mass communication in which one person communicates with others through various media. In the latter case, still other questions become pertinent. What is the effect of the interposition of the medium between speaker and hearer? When the President says on television, “I am counting on you right there in your living room to make a sacrifice,” is the sentence received in the same way as it would be in a face-to-face encounter, or is it apt to constitute itself for the viewer as merely another item of “what one hears” on radio and television?

  It should be emphasized that this empirical approach does not require the settling or even the raising of the question of the ontological status of the intersubjective relation. The latter is introduced as a postulate which is valid to the extent that it unites random observations and opens productive avenues of inquiry.

  The object and the world. The notion of world here is not an epistemological construct, as it is in much of European phenomenology. I am not saying that the world is constituted by the Dasein or the transcendental ego. Nor do I say that a tree is exactly as it appears. I say only that if one makes an empirical study of sign-using animals and symbol-using animals, one can only conclude that the latter have a world and the former do not. Nor does such a notion require the entity “mind” in one and eliminate it in the other. It has only to do with the observable difference between sign behavior and symbolic behavior. A sign-using organism takes account only of those elements of its environment which are relevant biologically. A chick has been observed to take account of the shadow of a hen and the shadow of a hawk but not, I believe, of the shadow of a swallow. A two-year-old child, however, will not only ask for milk, as a good sign-using animal; he will also point to the swallow and ask what it is.

  A sign-using organism can be said to take account of those segments of its environment toward which, through the rewards and punishments of the learning process, it has acquired the appropriate responses. It cannot be meaningfully described as “knowing” anything else. But a symbol-using organism has a world. Once it knows the name of trees—what trees “are”—it must know the name of houses. The world is simply the totality of that which is formulated through symbols. It is both spatial and temporal. Once a native knows there is an earth, he must know what is under the earth. Once he knows what happened yesterday, he must know what happened in the beginning. Hence his cosmological and etio-logical myths.* Chickens have no myths.

  Nor does the symbol refer to its object in the same mode as the sign does. True, one can use the word mean analogically and say that thunder means rain to the chicken and that the symbol water means water to Helen Keller. But the symbol does something the sign fails to do: It sets the object at a distance and in a public zone, where it is beheld intersubjectively by the community of symbol users. As Langer put it, say James to a dog, and as a good sign-using animal he will go look for James. Say James to you, and if you know a James, you will ask, “What about him?”

  The genesis of symbolic behavior, considered both ontogenetically and phylogenetically, is an all-or-none change, involving a symbolic threshold. As Sapir observed, there are no primitive languages. Every known language is an essentially perfect means of expression and communication among those who use it. As Helen Keller put it, once she knew what water “was,” she had to know what everything else was. The greatest difference between the environment (Umwelt) of a sign-using organism and the world (Welt) of the speaking organism is that there are gaps in the former but none in the latter.* The nonspeaking organism only notices what is relevant biologically; the speaking organism disposes of the entire horizon symbolically. Gaps that cannot be closed by perception and reason are closed by magic and myth. The primitive has names for edible and noxious plants; but he also has a name for all the others: “bush.” He also “knows” what lies beyond the horizon, what is under the earth, and where he came from.

  The distinction between Welt and Umwelt has been made before. Buber characterizes man as t
he creature who has a world and sets it at a distance, beyond the operation of his drives and needs. But, insightful as such an observation may be, it is of doubtful value to the behavioral sciences until it can be grounded in a coherent theory of symbolic behavior.

  The being-in-the-world. Here again, my element is different from the Dasein of the existentialists, akin as the latter is to the transcendental subject of Kant and Husserl. It is no more than a working concept arrived at through the necessity of giving an account of the organism who participates in symbolic behavior. The organism who speaks has a world and consequently has the task of living in the world. It is simply inadequate to describe him in the organismic terms of adjustment, adaptation, needs, drives, reinforcement, inhibition, and so on. A psychiatric patient is, to be sure, an organism in an environment. He is also a creature who is informed by his culture. But he is something more. He is an organism who may not forgo the choice of how he is going to live in his world, for the forgoing is itself a kind of choice by default. It becomes pertinent to ask in what mode he inserts himself in the world. May has suggested that it sometimes seems more appropriate to ask a patient Where are you? rather than How are you? Certainly, becoming aware of the threshold of symbolic behavior makes one very curious about modes of existence: How does the person go about living in his world?

  The intentional or quasi identity between the symbol and that which is symbolized. The mysterious “unreal but imputed” relation between the symbol and its designatum, the “wrong” identification of word and thing which the Polish semanticists condemn, never really fitted into a behavioristic theory of meaning. How did it come about that responding organisms imputed an unreal semantical relation between signs and things? How does an organism behave perversely by making a semantic identification at the “wrong” level of abstraction? What kind of organon is the unified science of signs when symbolic behavior is recognized as such by the formal sciences but disqualified by the natural sciences?

  Once it becomes clear that what is to be studied is not sentence forms but particular language events, it also becomes clear that the subject of investigation in this instance is not the sentence itself but the mode in which it is asserted. The sentence can be studied only by a formal science such as grammar or logic, but a sentence event is open to a rich empirical phenomenology that is wholly unprovided by what passes currently as semantics. Nor can a neobehavioristic psychology make sense of assertory behavior; it can only grasp a sequence of space-time events which it attempts to correlate by constant functions. But assertion—the giving of a name to a thing, this is water, or the declaring of a state of affairs, the water is cold—is not a sequence. It is a pairing or identification of word and thing, class and thing, thing and attribute, and so on. Stimulus and response events are studied by a quantitative science. But the quasi identification events of symbolic behavior can be grasped only by a qualitative phenomenology. This qualitative scale must take account not only of true-or-false-or-nonsense statements (water is cold, water is dry, water is upside down), but also of various modes of magic identification. It does not suffice, for example, to say that the assertion of a Bororo tribesman of Brazil, “I am a parakeet,” is false or nonsense. Nor is it adequate to say that it is false scientifically but true mythically.* It is necessary to understand the particular mode of identification of a particular language-event.†

  Sentences exhibiting the same syntactic and semantic structure may be asserted in wholly different modes of identification. For example, the sentence “My son John has become a roentgenologist” has the logical form of the assertion of class membership, a ? A. It has this form regardless of the particular language event in which it is asserted. The sentence can be asserted in more than one mode, however. Thus, if a psychiatrist should hear his patient utter the above sentence, he may very well understand, knowing her as he does, that she is asserting a magic mode of class membership. Her son John has gone off to a scientific place where he has undergone a mystical transformation and emerged as a roentgenologist. Another patient may assert the same sentence and be quite clearly understood to mean that her son has acquired a skill which it is convenient to speak of as a class membership.*

  The action sentence “John treats patients with X-rays” may also be asserted as a transparent vehicle intending a nonmagic action not utterly different from everyday actions of pushing, pulling, hitting, shooting, and so forth. Or it may be asserted magically: John makes a scientific pass with his paraphernalia and his ray, and the patient is cured.

  The connotations of words themselves, apart from assertory behavior, undergo a characteristic semantic evolution which can be understood only by a science proper of symbolic behavior, for it is the particular word event which is studied and not the “semantic rule” by which it is applied to its designatum. The scale ranges from the almost miraculous discovering power of the word-vehicle as a metaphor in the hands of the poet, to its sclerosis through usage and familiarity until it becomes a semantic husk serving rather to conceal than to disclose what it designates. When Shakespeare compares winter trees with

  Bare ruined choirs where late the sweet birds sang†

  the words come as fresh as creation from the symbolizer and serve to discover for the reader what he too saw but did not know he saw. But when, in everyday conversation, I tell you, “Last summer I went abroad and had some interesting experiences and saw some historical sites,” the words act as biscuit cutters carving up memory into the weariest shapes of everyday usage.*

  THE SYMBOLIC STRUCTURE OF A THERAPIST-PATIENT COMMUNICATION EVENT

  To determine how the generic structure of symbolic behavior is relevant to the therapist-patient relation as an instance thereof, I shall consider briefly a hypothetical language event.

  Patient: “Here is a dream which may be of some interest to you. Since you are an analyst, I am sure you will agree it has psychiatric implications.”

  Therapist: “Sounds interesting.”

  Patient: “In this dream I was walking down a strange street. A sexy-looking woman standing behind a Dutch door beckoned to me. I hesitated for a second, then against my better judgment, I went into the house.”

  Therapist: “Horrendous! [Pronounced heartily with a j: horrenjus!]” †

  In the study of a spoken language event, a written transcription is, of course, wholly unacceptable.‡ All phonetics and vocal modifiers are omitted. Even a tap recording is inadequate since it does not transmit gestures. In my comment on this exchange, moreover, I will say nothing of such strictly linguistic analyses as might be made of phonemes, morphemes, and grammar. Nor shall I say anything about the “content” of the exchange—for example, the dream and its “meaning”—important though this may be in the patient’s dynamics. But if one does not consider the linguistics and content of the language event, what else remains to be said about it? What remains is nothing else but the particular structure of the symbolic behavior, of which the symbolic tetrad is the generic type (see Figure 2). The assumption that all that is going on is an interaction between organisms deprives the investigator of the means of taking account of the molar event of communication, leaving him only with the alternative of fitting as best he can the qualitative traits of interpersonal behavior into the Procrustean bed of a response psychology. But once the generic character of symbolic behavior is recognized, then the modes of intersubjectivity, “world,” “being-in-a-world,” and assertory identity are seen as particular expressions of the fundamental possibilities allowed by the structure of interpersonal process—just as drives, needs, reinforcement and extinction, stimulus, response, are the fundamental categories of organismic interaction.

  The mode of assertory identity. It may very well be that some of the assertory behavior in this example is magical. The patient is an educated layman, the sort who takes pride in being well informed in scientific matters, especially psychiatry, and in his use of psychiatric jargon. He quite consciously uses “analyst” rather than “psychoanalyst.�
�� One often notices in psychiatric interviews a kind of pseudo reversal of the roles of scientist and layman. The patient often uses such phrases as “Oedipus complex” (he would never say “inferiority complex,” since it passed long ago into everyday usage, passing, moreover, as a semantic husk of very questionable value), “sibling rivalry,” “aggressions,” and so forth, while the therapist is careful to steer clear of them, partly because he does not wish to use a technical phrase the patient would not understand, but perhaps even more because he is intuitively aware of the magic abuses to which expertise is peculiarly susceptible.* The patient in question may have, by reason of this very knowledgeability about psychiatry, fallen prey to a magic mode of identification. The clause Since you are an analyst very likely asserts a mystical transformation by which an ordinary human being is transfigured and informed by the resplendent scientific symbols “psychiatrist” and “psychoanalyst” and finally by the shorthand expression used among the elite, “analyst.”

  The world of the therapist and his being-in-the-world. Insofar as he is a scientist, the therapist has assumed the posture of objectivity. As a consequence of what might be called the Thalesian revolution, men have learned, beginning at about the time of the Ionian philosophers and the Vedantists of the epic period,† to strike a theoretical posture toward the world which would enable them to discover the underlying principles and causes by which particular things and events can be understood. The scientist is not in his world in the same way, as, say, a member of a cosmological culture like the Bororo tribesmen, nor as a wanderer between cultures like Abraham, nor even as his fellow culture members, the businessman and the streetcar conductor. Insofar as he practices his science, he stands, in Buber’s phrase, “over against” his world as knower and manipulator of that which can be known and manipulated. The scientist may so be characterized without pejoration—indeed if he were in his world in any other way, he could hardly be a scientist. Yet as a psychiatrist, a “participant observer,” he must also re-enter the world in some mode or other as a person who is friendly and sympathetic, or anyhow appears so, to his patient.