Ecumenicalism in Semiotics

Thomas A. Sebeok

The Organism as a Text

In what was assuredly one of his most stupefying lectures, titled “Consciousness and language” (7.579-596), delivered about one hundred and ten years ago, Peirce raised the question: “What is man?” His decisive answer was that man is a symbol. Peirce elaborated and deepened his argument by comparing man with some other symbol, to wit, the verbal sign Six, and showed that, “remote and dissimilar as the word and the man appear, it is exceedingly difficult to state any essential difference between them...” (7.588). Elsewhere, Peirce claimed that “my language is the sum total of myself ” (5.314), a view tantamount to asserting that man is a string of signs, a process of communication (Fairbanks, 1976:20), or, in short, a text. Nor is man uniquely endowed with the property of life, “For every symbol is a living thing, in a very strict sense that is no mere figure of speech” (2.222). Furthermore, “the most marvellous faculty of humanity,” one which we share with all other organisms, is “that of procreation” (7.590). Yet we have even that capacity in common with signs: “Symbols grow. They come into being by development out of other signs. ... It is only out of symbols that a new symbol can grow.Omne symbolum de symbolo “ (2.302). Chomsky (1972:90-93)— whose way of envisioning the acquisition of knowledge of language vividly recalls (as he is well aware) some of the ideas of Peirce on the logic of abduction—reconstituted, with the linguist's fruitful reformulation of “the creative aspect of language use” (1972:6), this notion and turned it to challenging problems of modern linguistics. Incidentally, Peirce's splendid manner of argumentation was very far from an instance of the Pathetic Fallacy. Indeed, his view that signs have an inherent measure of self-responsibility was developed much further a century later by René Thom.

Thom, especially in his neoclassical essay on a theory of symbolism (1974, Ch. xi), which adheres closely to the Peircean semiotic model (cf. Sebeok 1976b), has authoritatively delineated the fundamental identity of the processes of biological reproduction and semiosis, bracketed in the frame of his powerful catastrophe theory, or a kind of mathematics applicable to phenomena that are neither orderly nor linear, and hence relevant to the so-called inexact sciences, of which biology and semiotics are representative, each in its own ways. In the course of the former, a parent organism or “signified” emits a descendant, thereby engendering a “signifier” which, in an eternal and universal flux, reengenders a “signified” each time that the sign is interpreted. A signifier turns into a signified given only sufficient time, say, the lapse of a generation. By virtue of this subtle fluctuation between two morphologies, bound by the simultaneous exigencies of reversibility and irreversibility, “la dynamique du symbolisme,” or, in Peircean terminology, semiosis, i.e., sign-action and sign-interpretation, “porte en elle . . . toutes les contradictions de la vision scientifique du monde, et qu’elle est l’image même de la vie” (Thom 1974:233; cf. Sebeok 1976a:69).

Two Traditions

The title for this summative essay was chosen less for its ecclesiastical connotations than to suggest that, in my view (Sebeok 1977a), the scope of semiotics encompasses the whole of the oikoumenē, the entirety of our planetary biosphere. The chronology of semiotic inquiry so far, viewed panoramically, exhibits an oscillation between two seemingly antithetical tendencies: in the major tradition (which I am tempted to christen a catholic heritage), semiosis takes its place as a normal occurrence in nature, of which, to be sure, language—that paramount known mode of terrestrial communication which is Lamarckian in style, that is, embodies a learning process that becomes part of the evolutionary legacy of the ensuing generations—forms an important if relatively recent component. This accords with the view of the Stoics who argued, in nuce, “that animals have ‘expressive’ reason, the power of speech, by which they communicate with each other by means of signs” (Philodemus 1941:144), although they also drew a sharp distinction between the aforementioned properties, shared by animals with men, in opposition to synthetic and combining reason, ascribed by them to men alone (cf. Simone 1972:6-7). It was, however, St. Augustine (354-430), one of the greatest thinkers of all times about sign functions, who, in his brief dialogue, De Magistro, first wrestled with the perennial question whether it is feasible for an organism to learn through signs at all and, if so, how.

The minor trend, which is parochially glottocentric, asserts, sometimes with sophistication but at other times with embarrassing naïveté, that linguistics serves as the model for the rest of semiotics—Saussure’s le patron général —because of the allegedly arbitrary and conventional character of the verbal sign. Much has been written within and about this tradition, which has both a conservative and a radical wing. The orthodox line was quite succinctly enunciated, in the weakest form, by Cassirer (1945:115): “Linguistics is a part of semiotics...,” but more emphatically so by Bloomfield (1939:55), who averred that “Linguistics is the chief contributor to semiotic,” then Weinreich (1968:164), who declared natural language to be “the semiotic phenomenon par excellence “ (cf. Sebeok 1976a: 11-12). The rallying cry of the ultrarevisionists was perhaps most enticingly proclaimed by Barthes (1964:81) in an often-repeated catchphrase: “... c’est la sémiologie qui est une partie de la linguistique.”

The distinction between what I have termed, although without elaboration here (but see Sebeok 1977a), the major and the minor traditions has lost its force, since it is amply clear that Darwinian and Lamarckian (i.e., sociogenetic) evolution coexist in subtle interaction in the human animal (Bateson 1968) (or, as Morris [1976:39] marveled about him in his delightful poem, “Animal Symbolicum,” “What a remarkable animal: physiological and symbolical./ Kin of every other animal, but symbol-haunted and dramatical”). Human evolution is thus not only a reconfirmation of the evolutionary processes which went on before man appeared on the scene, but continues as a dual semiotic consecution that can scarcely be uncoupled in practice: one track language-free (or zoosemiotic), the other language-sensitive (or anthroposemiotic) (Sebeok 1972:163). Semiosis must be recognized as a pervasive fact of nature as well as of culture. In such matters, then, I declare myself not only a Peircean but a (René) Thomist, yet one who stands ready to obey Karl Barth's injunction as formulated at the outset.

The Distinctive Burden of Semiotics

Ecumenicalism in semiotics means, however, far more than a plea for attempting to capture global properties of sign systems in general and for unifying local variations that are criterial for heretofore unrelated information about the genetic (Marcus 1974), metabolic (in the sense used by Tomkins 1975), neural (Pribram 1971, Ch. 4, Rose 1973:100-110), intraspecific vs. interspecific (Sebeok 1972:79), nonverbal vs. verbal and nonvocal vs. vocal codes (Sebeok 1977b), beside a host of secondary modeling systems (Winner and Winner 1976:135-137). The term, which derives from the Greek oikos, designating “house,” seems to be appropriate because it once again calls attention to the holistic force of semiotics, which many leading contributors to the discipline have marked as its distinctive burden. This perspective was perhaps best articulated by Morris who, working initially in the overall frame of the International Encyclopedia of Unified Science (the program of which was unfolded at the 1935 International Congress for the Unity of Science, plans for which were laid the year before, interestingly enough for those familiar with the classical period of the Prague School, at Charles University), conceived of semiotic as an instrument of all the sciences. An acute historian of the semiotic endeavor since 1865 recently remarked about this vista that “The foundations of the theory of signs were the foundations for the unification of the sciences” as well (cf. Fisch, in press, #13). “Semiotic has a double relation to the sciences: it is both a science among the sciences and an instrument of the sciences,” Morris wrote in 1938 (1971:17-18), continuing: “The significance of semiotic as a science lies in the fact that it is a step in the unification of science, since it supplies the foundations for any special science of signs, such as linguistics, logic, mathematics, rhetoric, and. . . aesthetics. The concept of sign may prove to be of importance in the unification of the social, psychological, and humanistic sciences insofar as these are distinguished from the physical and biological sciences.” Morris later provided for a still higher level of unification with the two last-mentioned groups of disciplines, having shown that signs are simply the objects studied by these branches of knowledge in certain complex functional processes.

A closely similar view was sanguinely expounded, in 1943, by the Danish linguist Hjelmslev, whose glossematics professed to be grounded in Saussure’s Cours, but who was also comfortably familiar with the work of such logical empiricists as Carnap, who, in turn, acknowledged his indebtedness to Morris's semiotic. Hjelmslev (1953:69) echoed: “... it seems fruitful and necessary to establish a common point of view for a large number of disciplines, from the study of literature, art, and music, and general history, all the way to logistics and mathematics. . . . Each will be able to contribute in its own way to the general science of semiotics.. . . Thus new light might perhaps be cast on these disciplines, and they might be led to a critical self-examination. In this way, through a mutually fructifying collaboration, it should be possible to produce a general encyclopedia of sign structures.” French semiotic writings of the past decade seem often to have felt obligated to take a stance, whether pro or con, Hjelmslev’s amplification of Saussure’s outline of a prospective general science of signs; and Anglophone exegetes, like Whitfield (1969:258), who notes that glossematics views all sciences “as also being ... semiotics,” or like Lamb (cf. Lamb and Makkai 1976), are still endeavoring to reinterpret “the semiotics of culture” (Schwimmer, this book; Umiker-Sebeok 1977a) in various eclectic ways but seldom straying far from the broad if fragile framework erected by Hjelmslev. On the other hand, in his heroic but neglected attempt at a unification of the structure of human behavior, Pike (1967:63) has explicitly rejected both Saussure's theory of signs and Hjelmslev's version of it. Barthes, of course, as cited in the second epigraph, has persuasively and prestigiously rekindled the spark of coadunation.

The works of Morris, Hjelmslev, Barthes, and their numerous epigones on the holistic force of semiotics hardly exceed programmatic pronouncements. In truth, as Peirce came to recognize with increasing clarity after 1890, science is an ongoing quest that cannot be identified as a particular body of knowledge yet segments of which are delimitable in terms of particular social groupings of research scholars within a vaster community of investigators; as he wrote Lady Welby in 1908, “the only natural lines of demarcation between nearly related sciences are the divisions between the social groups of devotees of those sciences” (8.342). But Peirce also stressed that one of the first useful steps toward a science of semeiotic, “or the cenoscopic science of signs” (i.e., which does not depend upon new special observations), “must be the accurate definition, or logical analysis of the concepts of science” (8.343). Jakobson, in his masterful (though mistitled) essay, “Linguistics in relation to other sciences” (1971:655-696), has sketched out a way whereby the nomothetic sciences of man, and possibly some of the natural sciences alongside them as well, might be systematized to achieve a degree of logical filiation and hierarchical ordering exactly if linguistics— alias semiotics—is used as the point of departure for their tentative serial arrangement. However, Thorn's outline of a general theory seems to me to provide the first rigorously monistic model of the living being endowed with the zfacultas signatrix, and to offer one pure topological continuum in which causality and finality are combined. Assuredly it is no accident that his dynamical topology was influenced by Peirce on the one hand (Thom 1974:229-230), and von Uexkiill on the other (Thom 1975:xxiii, and passim).

The kinship of semiotically based programs, such as those mentioned, to the movement known as General System Theory (Bertalanffy 1968), or GST, is seldom underlined yet their common denominator is rather obvious. As a “natural philosophy,” both these variants of a single metatheory can be traced back to Leibniz (ibid., 11; Dascal 1972) and his program for a mathesis universalis, and it was certainly not a coincidence that the founder of GST first presented his ideas, in 1937, in Morris's semiotic seminar at the University of Chicago (Bertalanffy 1968:90). Although the two different sorts of mobilization toward synthesis continued to develop and flourish independently one from the other, their proponents have not infrequently crossed paths here and there; for example, both Morris and I were occasional participants in discussion groups of scientists assembled, from the most various traditional disciplines, in the 1950s, “for the purpose of approaching a unified theory of behavior,” or, as was sometimes grandiosely claimed, human nature (e.g., Grinker 1956:v). The San Francisco psychiatrist Jurgen Ruesch was an animating spirit in this striving for integration, drawing “from theories that derive from social anthropology, sociology, philology, linguistics, and related disciplines,” all of which, he explicitly acknowledged much later, “are concerned with semiotics” (1972:11-12; cf. 127). Recently, the two main preoccupations of system theorists appear to have settled on action, or the movement of matter-energy over space, and communication, or the “change of information from one state to another or its movement from one point to another over space” (Miller 1975:346), but the two naturally flow together since information is always borne on a marker (id. 1976:299). The distinction immediately calls to mind Peirce's alterity of dyadic or dynamical action vs. triadic or sign-action (5.473), which, again, brings us back to his “doctrine of the essential nature and fundamental varieties of possible semiosis” (5.488). If the writings of Peirce had been available to von Uexkiill, who also inspired von Bertalanffy (1968:288) (although the latter’s avowed knowledge of semiotics seems to have been confined to some writings of Cassirer), GST would very likely have taken a radically different turn. The approach of von Bertalanffy consisted of treating sets of related events collectively as systems manifesting functions and properties of the specific level of the whole. This presumptively enables the recognition of isomorphics across ascending levels of organization—variously dubbed, e.g., org (Gerard 1957), holon (Koestler 1967:341), integron (Jacob 1974:299-324), or, processually, shred out (Miller 1976:226-227)—from which can be developed general principles or sometimes fundamental laws that operate commonly at all levels of organization in contrast with those which are unique to each. Semiotic ideas pervade GST, but are seldom anchored in a comprehensive doctrine of signs. For instance, Miller (1976:227) writes: “... the evolutionary shred out of the channel and net subsystem is from slow, inefficient, chemical transmission by diffusion at the cell level up to increasingly rapid and cost-effective symbolic linguistic transmission over complicated networks at the higher levels of living systems.” However, as critics have justly pointed out—among other strictures (Phillips 1976:48-67)—GST suffers from an aura of vagueness over what is meant to be encompassed within it. The missing ingredient, the addition of which, I suggest, is most likely to dispel the fog, can only be supplied by way of a judicious linkage with a semiotic which must be firmly based upon the empirical facts of current biological thought as foreshadowed by Peirce, developed by Morris, independently furthered, in his own lambent fashion, by von Uexkiill, and recently restated in the language of Thorn's catastrophe theory (which, pace its detractors in some American mathematical circles [Kolata 1977], has already vastly enlarged the horizons of semiotic inquiry [Sebeok 1976b]). I emphasize the historical and logical associations between holism and semiotics because these are nowadays often overlooked: for instance, although Phillips devotes a section of his exploratory essay to the holistic connotations of structuralism (1976, Ch. 6), including its embodiment in Prague-style linguistics (pp. 83-84, 88), he ignores the more patently tenable and much more fundamental connections with semiotics. Precisely how semiotics can function as “an organon” (Morris 1971:67) of all the sciences, and the wide humanistic implications of the assumption that semiotics “provides a basis for understanding the main forms of human activity and their interrelationship” (ibid., 69), since all such activities and relations are mediated exclusively by signs, pose a host of further questions that need to be widely as well as urgently debated since, among other consequences, their satisfactory reformulation might well lead to badly needed improvement of the chaotic curricula which still inform students of the human sciences worldwide.

Communication and Manipulation

The Tampa Colloquium coincided with the heyday of an era in international relations epitomized by the fashionable slogan détente, which an interim President of the United States later tried to declare taboo. In the global pursuit of semiotics, we have no need for such artifices of labelling to characterize a relaxation of tensions. Wherever our workshops happen to be located—from Bloomington to Moscow, as the newest introductory manual would have it in the perhaps exaggerated heading of one of its principal chapters (Voigt 1977, Ch. 8)—the semiotic network amicably binds us together, to our immense profit, with only sporadic intrusions of that stain of darkness in the fabric that both mars the otherwise seamless web and invites ridicule upon the preposterous advocates of the intrusion of politics into our affairs (e.g., Sebeok 1976a: 166). I do not mean to suggest that the practitioner of semiotics is more enlightened than the practitioner of politics, but only that the two trades are right to be uneasy with each other. Margaret Mead (1972:286) expressed the distinction between cooperative and manipulative behavior (Sebeok 1976a:80) with great delicacy and her customary vatical flair when she said that in inaugurating the new science of semiotics it “is of the utmost importance that continuing work. . . should take place not in a context of power,” but that “we must affirm the acquisition of this greater systematic insight as part of man's scientific knowledge on behalf of man. Otherwise we may expect to encounter continual blocking resistance, misunderstanding, and downright fear and hostility both from outsiders and from members of disciplines included in the new science. ...”

Indeed, the sign expert can supply the policy maker with pointers on how to sharpen his tools designed to exploit resources in certain very important areas of human activity, such as strategic conflict, say, in the transnational arena. Game theory, in particular that subclass to which two-actor zerosum games belong, clearly articulates, in yet another kind of holistic construct (Bertalanffy 1968:90), communication with manipulation, or, if you will excuse the reiteration, sign-action with dynamical action. The book of Jervis (1970), although held to the descriptive level and without the imposition of formal game models, serves as one outstanding illustration of a creative projection of semiotic theory and techniques to the level of diplomacy in international relations. Beyond this sort of service, entailing a heightened level of awareness about games in general and about the pervasively institutionalized human con-games (Goffman 1952) in particular, an insistent exertion of the semiotic point of view may merely embarrass politician and administrator alike. It is perfectly all right for our masters to be wary of what we know, do, and might yet uncover, but is it too much to ask for their trust?

The Contribution of Adnormal, Denormal, and Abnormal Semiotics

Most of the papers delivered at our conference can be pigeonholed as multifarious examples of work touching upon the semiotics of culture (Winner and Winner 1976, Umiker-Sebeok 1977a), broadly defined, in the sense of imperative language involvement. None dealt with the necessarily language-independent semiotics of nature (Sebeok 1977a), a marked/unmarked opposition the many consequences of which I do not want to pursue here. Instead, I should like to allude to another set of emergent distinctions that I think must rapidly be assimilated to a unitary pansemiotic (cf. Sebeok 1977b, Section 7). Sign processes in everyday life, or the semiotics of the normal, need to be closely reexamined in the light of their ontogenetic formation, or the semiotics of the adnormal; their dissolution in the course of a human life, or the semiotics of the denormal; and their modifications when caused by injury or disease, or the semiotics of the abnormal. In brief, two polar oppositions intersect here: one between the semiotics of ripening childhood and the semiotics of retrograde second childhood; the other, which Kleinpaul tried to make much of (1888:103-111) in a pertinent frame, between the semiotics of the “normalen . . . Formen der Gesundheit” and the semiotics of “Krankheiten.”

The study of adnormal semiotics is shared with neurology and several branches of psychology; of denormal semiotics with social gerontology; and of abnormal semiotics with different medical specialties. As an example of distinguished consolidating work rooted in adnormal semiotics, let me single out McNeill’s (1975) demonstration of the intimate connection between the production of speech by adults and the representation of conceptual structures on a sensory-motor level. His apt expression “semiotic extension” makes double reference to the basis of contact between “normal” adult conceptualization with the development of speech production in children and spontaneous gesticulations during speech that, “far from being mere embellishments,” can also “be seen to arise directly from the operation of the speech mechanisms themselves” (ibid., 373), viewed, that is, as an external dynamic trace of the internal speech program. It is good news that McNeill is finishing a book on his fruitful notion of “semiotic extension,” a kind of meaning relationship, and especially encouraging that for his underlying theory of signs he builds upon the strongest yet available—Peirce’s.

The development of human communicative competence, including the unfolding of man’s repertory of prelinguistic signs, “those which occur in the child’s behavior before it speaks, or which later, even in the adult, are independent of language signs,” according to the fruitful formulation of Morris (1964:58), involves a gradual increase in both nonverbal and verbal specializations and specificities that Eric Lenneberg has together termed “differentiation” (Lenneberg and Lenneberg 1975:1:32)—a label fittingly chosen, for it evokes, to my bent quite accurately, the proper embryological homology, or, rather, continuity. It has not yet been sufficiently realized that this amply documented process of semiotic differentiation in children not only perdures through adulthood but that it is an essential feature of senescence. While the aging and the aged shed, bit by bit, semiotic competencies “normal” adults take for granted, they also tend to acquire a new-sprung stock of message-types as a part of their adaptive reaction to the impact of their changing environment with which they must all learn to cope. Since it is one of the most dramatic and significant facts of demography that the number of people characterizable as elderly—i.e., 65 years of age and older—is increasing faster than that in any other age group, and that this population trend is likely to continue in the foreseeable future, the study of caducity and communication (Oyer and Oyer 1976) is becoming an ever more urgent research concern of our decade, just as adolescence and communication was—as it still is—a pressing preoccupation throughout this quarter of a century. The heterogeneous age-related alterations in the patterns of communication—which are more likely than not complicated by a host of other variables, notably by health-related impediments—in 11% of our population, or about 21 million Americans alone, a figure which is predicted to rise, for this country, to about 20% by the turn of the century, constitutes an unprecedented opportunity for the intensive semiotic quest whose time has come. The signs of old age, seen as adaptive strategies which are reasonable in particular semiotic environments, were studied by Stafford in a pioneering thesis (1977). He showed that many such signs are routinely “misread” in the community where he did his fieldwork as indices of physiological deterioration. He ascribes the embarrassment, alienation, and, eventually, the mental anguish of old people to the consistent misinterpretations of their messages on the part of their young interlocutors and interactants, or age mates, who don’t share with them what I am here provisionally proposing to call their denormal code, resulting, according to Stafford, from a socially imposed decrease in responsibility coupled with a corresponding increase in dependence. “Old age,” thus conceived, is, in fact, a system of signification, to which Peirce's notion of infinite semiosis is eminently applicable and which is fecundly analyzable within his trichotomous scheme of sign relations. In Stafford's favorite example, “repetitiousness” in the senescent is for them an index of old age, which, however, is regularly mistaken by others for an unwitting symptom of a physiological deficit, leading him to formulate the insightful maxim that “one man's index is another man’s rheme,” or, as I would prefer to phrase it in the vocabulary of cryptanalytic transformations, “one generation’s cleartext is another generation’s encicode.” Nothing short of the armamentarium of cryptanalysis must be applied to appreciate the bracketing phases of man’s full life cycle, from adnormal to denormal, perceived, at least within our own culture (cf. Umiker-Sebeok 1977b), as an essentially specular process.

In their contribution to this book on some “Unexpected Semiotic Implications of Medical Inquiry,” Shands and Meltzer discuss a fascinating array of semiotic observations characteristic of certain unskilled laborers following industrial accidents. At the same time, they emphasize that the “application of semiotic principles to the field of medicine is in its infancy, although it has long been clear that the physician’s stock in trade is interpretation.” This protracted nonage of clinical semiotics is perplexing when one considers the Hippocratic roots of the doctrine of signs as a whole (Sebeok 1976a: 125-126, 181-182, and 1977a), and particularly the explicit contributions of the immensely influential Galen (130-201 a.d.). It was Galen who classified semiotics as one of the six principal branches of medicine, and the strength of Galenism, according to Temkin’s dependable account (1973:179), “reposed in no small measure in its having provided medical categories ... for relating the individual to health and disease, [including] semeiology (the science of signs).” Galen subdivided the field into three abiding parts, inspection or diagnosis, cognition or anamnesis, and providence or prognosis. He regarded everything unnatural occurring in the body as a symptom, and a syndrome as an aggregation of symptoms (athroisma ton symptomaton), clearly recognizing that while both of the former directly reflect clinical observation, the formulation of a diagnosis requires causal thinking.

It is my layman’s impression that semiotic ideas have, naturally enough, penetrated most pervasively into that area of medicine which is devoted to disturbances in signing function consequent to selective ravages of brain injury—the condition Steinthal termed Asemie in 1881, and Hamilton (later copied by Jackson) asemasia in 1878, but which is, of course, more commonly known, in specific reference to a variety of abnormal language functions, by the general label aphasia. But since all medicine, from the Corpus Hippocraticum to the present day, is in essence hermeneutic, the pertinence of semiotics, in a deep sense, to a universal theory of disease is not surprising. Note what one leading figure in cancer research told an interviewer: “I think the most important diseases amount to a problem of bad communication. . . . Cells and tissues respond to misinformation or misinterpretation from the environment and they do so in a certain way. They overreact.. ..” (Thomas 1976:114).

Children diagnosed as suffering from infantile autism, the descriptions of which are most often negatively stated as consisting of a minimal behavioral repertoire and almost no control at all by the social environment save, perhaps, for some elementary avoidance responses, offer one kind of a challenge to semiotic analysis, because this is a form of malady which is typically characterized by disinterest in interpersonal relationships (even within the family), let alone in looser forms of association. It is often asserted that the autistic child is indifferent to faces. One such boy was pictured as mute, and what activity there was consisted almost entirely of simple, stereotyped maneuvers, “such as moving his foot back and forth, flipping a shoe lace, rubbing a spot on the floor or flipping small objects pivoted between the fingers for many hours at a time” (Ferster 1964:317). Usually, the presence or absence of a potential interactant is reported to have hardly any effect on the patient’s behavior, but it has occasionally proved feasible “to achieve an increasing level of social interaction and even considerable amounts of productive speech” (ibid., 326), as in the case in point by the interposition of a machine, i.e., by means of a form of operant conditioning. It is not at all clear whether infantile autism, variable as it is, is due to some genetic defect, personality disorder, or brain damage perhaps suffered at birth, or to a combination of such and still other “psychosomatic” conditions intuitively sensed to coexist in different measure (cf. Ingram, in Lenneberg and Lenneberg 1975:2:243).

Kanner’s syndrome, or infantile autism in its classical form (ibid., Creak et al., 2:242), is recognized by an array of diagnostic features each of which represents some impoverished sign function, beginning with an underlying inability for sustained message coupling between source and destination, or, to phrase this in more positive fashion, its fundamental defining criterion limits the linkage between the child and any adult to mere symbiosis that approaches a zero degree of normal semiosis: while the child is indifferent to his caretaker’s attention, the latter’s frequency of responses decreases proportionately with the amount of stimulus aversiveness and may ultimately become extinguished. The autistic child is also apparently unaware of his personal identity, and may exhibit this anomic state, among other symptoms, by a confusion of personal pronouns. He will often be echolalic in the vocal mode and correspondingly imitative in the visual mode, as well as produce significantly more stereotyped utterances and contextually inappropriate remarks than a developmentally aphasie child. Autistic children are attested to “differ even in their cries from normal children and from other subnormal children” (Cromer 1974:251). Adding to these the bizarre postures and ritualistic mannerisms, such as spinning themselves or objects, it is small wonder that the Tinbergens—as “animal ethologists who have for years specialised on social encounters in higher animals, which are of course nonverbal” 1972:11)—have focused upon autism as a prime target for their ethological approach, noting “the considerable potential” residing in this method for the investigation of nonverbal communication in man and, of course, “in general.” Since, in my view, the biological study of comportment and the semiotic study of communication and signification are largely overlapping disciplines (Sebeok 1976a, Ch. 5, and id. 1977a), I perceive here one common meeting ground where both methodologies could prove illuminating, although we have contributed, so far, next to nothing to either theory or rehabilitation. These brief remarks are meant to be read chiefly as a plea for the reabsorption of clinical semiotics into the comprehensive doctrine of signs, along with the signs of growing up and the signs of growing old.

Pavlov’ s Mice: Three Concluding Lessons

Before bringing these summary observations to an end, I would like to address the issue of overarching principles guiding semiotic inquiry. I shall do so briefly because I deal with it in substantially more detail in a separate publication soon to appear (Sebeok 1978). Also, I prefer to approach the question here obliquely, introducing it by way of an anecdote from the history of physiology in the 1920s.

The episode, which is well authenticated (Gruenberg 1929:326-327, Zirkle 1958:1476, Razran 1959:916-917), centers upon Ivan Pavlov, who has justly been called “the 20th century’s empiricist par excellence” (Razran 1959:916). It concerns experiments he caused to be conducted to demonstrate the transmission of conditioned reflexes in mice from one generation to the next. The experiments were actually carried out by an obscure research assistant of his, named Studentsov, whose published data showed 298, 114, 29, 11, and 6 conditioning trials for five successive generations of mice. Pavlov, who had “shared the Lamarckian predisposition, common to Russian bio-scientists” (Razran, ibid.) with his country’s intelligentsia, accepted these inheritance data initially or, at any rate, failed to repudiate the explanatory hypothesis even after he obtained no evidence to support it. Not until 1929, in an informal statement made during the Thirteenth International Physiological Congress, in Boston, did Pavlov offer an alternative interpretation for the amazing data: he explained that “in checking up these experiments it was found that the apparent improvement in the ability to learn, on the part of successive generations of mice, was really due to an improvement in the ability to teach, on the part of the experimenter!” (Gruenberg 1929:327, fn. 1.).

To me, this fascinating little story suggests three morals of particular interest to our community of semioticians, the first of which is both the most general and, at least at first blush, the most banal: that even the greatest scientist can err. What is important for us is to distinguish between certain major sources of error, and, notably, to recognize self-deception. Self- deception is akin to the extremely widespread effect known as self-fulfilling prophecy (Rosenthal and Jacobson 1968, Part 1), but will perhaps be best recognized by scientists by an emblematic tag known as the “Clever Hans phenomenon.” Pavlov's gullibility was due to his wish, consonant with the ambience of his times, to believe that acquired characters were inherited, just as, a few decades earlier, an unfortunate German high school mathematics teacher, Herr von Osten, very badly wanted his horse, Hans, to perform feats of arithmetic, spell and read, solve problems of musical harmony, and answer personal questions. “Pavlov, of course, was an honest scientist” (Zirkle 1958), and there is no evidence that von Osten ever had fraudulent intent; both simply fell victim to their passionate expectations.

Fig. 1. Mr. von Osten with Clever Hans

Self-deception, then, must be sharply separated from deception of others, or deliberate fraud.^1. There are, alas, not a few cases of academic double-dealing, some British and U.S. perpetrators of which were exposed in recent issues of Nature and Science. I have followed with special attention the failed efforts of the 1960s to engage in complex communication with dolphins (Sebeok 1972:53-60), and parallel endeavors to unconvincingly (Limber 1977) attribute the use of human language to chimps and gorillas in the 1970s. Without impugning in any way the good faith of most investigators of the intellectual capacities of marine mammals and primates, I see the ever-present specter of Clever Hans, except in those instances where chicanery cannot be ruled out. After all, despite the innocence of von Osten, most of Hans’s performing ancestors and descendants were operated for profit in confidence games, as for instance the famous “oat-eater” Morocco, “the dancing horse” mentioned by Shakespeare (in Love’s Labour s Lost I.ii.51), exhibited by an Elizabethan rogue named John Banks, whose methods were exposed, in 1612, by one Samuel Rid in semiotic terms that are amazingly fresh: note, Rid wrote, in The Art of Juggling, that the horse can do nothing “but his master must first know, and then his master knowing, the horse is ruled by him by signs” (in Kinney 1973:290). It would surely have saved Oskar Pfungst (1965) a lot of time and effort had he known the case history of Morocco, a detailed account of which had been published (cf. Halliwell-Phillipps 1879), but which chanced to be reckoned in the field of literary exegesis instead of experimental psychology. Various other tricks enabling animals, horses as well as dogs, cats, pigs, and many kinds of birds, to say nothing of fleas, to perform such marvels of misdirection—as by a technique widely employed by mediums but known to only those in the trade as “pencil reading”— pour épater le bourgeois and, of course, for a modicum of profit, are illuminatingly and amusingly bared by the fine magician Christopher (1970:39-54).

The second lesson is a more palpably semiotic one, the essence of which was perfectly captured by the leader of the Hans-Commission, Carl Stumpf, and his brilliant investigator Pfungst, when they spoke of “looking for, in the horse, what should have been sought in the man” (cited by Rosenthal, in Pfungst 1965:xxx). As we saw, it took Pavlov a while to shift his attention from the mice to his overly zealous assistant, back from the event being scrutinized to the experimenter, from message destination to message source. The role played by the observing subject in physics is fairly well understood and respected: since the properties of the observer, especially with regard to magnitude, are usually very different from those of the object observed, the subject is likely to remain relatively outside the system that he is studying and hence one may assume that his report will be reasonably unbiased. In the life sciences, however, and a fortiori in the communication sciences of man, the characteristics of the observer are likely to be similar to those of the observed and must be supposed to influence the processes under scrutiny. The subject himself will likewise be seriously affected by the stream of ongoing events (cf. Ruesch 1972:54-55). In linguistic fieldwork, for example, hardly anybody has fully appreciated this, or, if so, conducted himself accordingly. A sophisticated exception was Leonard Bloomfield, who, according to Voegelin (1960:204), “preferred being corrected, when he made an error as a child-like speaker of Menomini, to asking a direct question on how do you say so-and-so to a bilingual Menomini—for fear of obtaining a false analogy. During the three summers Bloomfield and I recorded Ojibwa texts together at Linguistic Institutes, he never once asked a ‘how do you say’ question. ...”

Fig. 2. John Banks with Morocco, the wonderful horse of an. 1595. In this woodcut, the steed is stomping out the numbers on a pair of rolled dice.

An area of special concern to us all is the art of healing, to wit, the workings of the physician-patient dyad, and the so-called placebo effect, which means the therapeutic sequel consequent to the patient's belief in the efficacy of the treatment. This is a semiotic problem in its most crystalline form: since a placebo is, by definition, a pharmacologically inert substance that the doctor administers to a patient to relieve his distress when the use of an active medication isn't called for, “its beneficial effects must lie in its symbolic power. The most likely supposition is that it gains its potency through being a tangible symbol of the physician’s role as a healer” (Frank 1961:66). We learn from Shapiro’s riveting sketch that, throughout the centuries from shaman to physician, this very coupling—this bond between the healer and the troubled—“comprised all that the doctor had to offer the patient” (1960:113-114); and he quotes other authorities who affirm that the history of medicine is a history of “dynamic power” of the nexus between doctor and patient (cf. once again Peirce’s dynamical action and Thorn’s “dynamique du symbolisme”). The placebo works best with those patients who have favorable expectations from medicine and, “in general, accept and respond to symbols of healing” (Frank 1961:70), but “the fact that one of the best educated major religious groups in the United States is able to deny the rational efficacy of any treatment or medicine and to assign all treatment benefits to faith” (Shapiro 1960:114) strongly underlines the purity of the semiotic import of the phenomenon, where faith itself acquires the essence of a symbol and thus quite admirably meets Peirce’s criterion of “an imputed quality” (1.558), such that the two constituent sides of the sign, the signifier and the signified, are bonded one to the other in “a relation which consists in the fact that the mind associates the sign with its object” (1.372) regardless of any factual connection. In passing, it should be noted that in medical history as well as contemporary practice one may observe a gamut of procedures in the administration of placebos from out-and-out quackery to their sensitively controlled use by responsible physicians. What all have in common is the confidence, the enthusiasm, and the ritual on the part of both interactants, and it is therefore undoubtedly true “that the ability to respond favorably to a placebo is not so much a sign of excessive gullibility, as one of easy acceptance of others in their socially defined roles” (Frank 1961:70). In other words, placebo-responsiveness assumes, so to speak, an indexical function in respect to the ability of patients to trust their fellow man as embodied by their physician or, to a lesser degree, his anointed surrogates (e.g., nurses).

The more or less synonymous sets of dyadic ties I have variously referred to in this section as source and destination of messages, observer and observed, subject and object, physician and patient, man and animal (horse or mouse, porpoise or primate), could be multiplied ad nauseam to encompass further particularized couplings, notably of parent and child. To my taste, the most suggestive pair of labels comes from the argot of the criminal world engagingly explicated by Maurer (1949; see also Goffman 1952): operator and mark. This consummate metaphor works something like this, to give only one example from an area of application that ought to pique the curiosity of all who would heed Pavlov’s second lesson.

From time to time, claims are made that there are dogs that actually talk. There are records in the Académie Française of one such animal that, according to no less an authority than Leibniz, “had a vocabulary of thirty words, which was put to effective use when he wanted something specific to eat or drink” (Christopher 1970:51). The question is: how is such a con perpetrated, even on smart apples like Leibniz? Briefly (cynical data after Johnson 1912), a dog is conditioned, in the customary manner, to bark in response to certain verbal cues. Don, a seven-year-old German setter, who was reported by numerous observers to answer questions if food were held before him, replied to “Was heisst du?” (“Don.”), “Was hast du?” (“Hunger.”), and to other such queries, and was alleged moreover to answer categorical questions by “Ja” or “Nein” Don’s behavior was thoroughly investigated by the same Oskar Pfungst who, in 1907, had unveiled the secret of Hans, and he concluded (1) that the dog does not use words with any consciousness of their meaning to the hearer, and (2) that he is not using words learned by imitation. Instead, he gave this comparatively simple explanation: the dog produced vocal sounds which induce “illusions in the hearer. . . . The uncritical do not make the effort to discriminate between what is actually given in perception and what is merely associated imagery, which otherwise gives to the perception a meaning wholly unwarranted; and they habitually ignore the important part which suggestion always plays in ordinary situations.” In other words, “we may expect the majority of animal lovers to continue to read their own mental processes into the behavior of their pets. Nor need we be astonished if even scientists of a certain class continue at intervals to proclaim that they have completely demonstrated the presence in lower animals of ‘intelligent imitation’ and of other extremely complicated mental processes....” This is also precisely what happens in the circus, as Bouissac has emphasized when he referred to inferences drawn from the humanization of animals: “The most efficient training. . . evokes a behavior from the animal that, within the constructed situation, subtly creates the impression that the animal has humanlike motivations, emotions, and reasoning” (1976:118), and when he drew attention to a characteristic feature of every circus act, namely, that it “constructs its own situation, its own immediate context....” “The circus act must be completed,” he adds (in this volume), “by the semantic operations which are performed in the process and account for the meaningfulness that such circus acts convey to their audience.”^2. The underlying principle, that we construct a context-based advance hypothesis about what the message will be to tune in our perceptual equipment to favor certain interpretations of the input and reject others, was elegantly demonstrated by Bruce over twenty years ago (1956), although how such contextual constraints actually work is still unknown. In sum, I read the second lesson of the Pavlov yarn as a strong injunction for us to look in the men and women (Lilly [1967] and Margaret Howe, Bernadette Chauvin-Muckensturm [1974], the Gardners, Premack, Rumbaugh, Francine Patterson, and their fellow trainers) for what has hitherto been assiduously sought for in Peter Dolphin, the Greater Spotted Woodpecker (claimed “to represent the phenomenon of man-animal communication analogous to that found in monkeys by the Gardners and Premack”), Washoe, Sara, Lana, Koko, and the rest.^3. This is not to deny the inherent interest of studies in animal intelligence. Quite the contrary; but I would like such studies additionally to take the crucial effects of man’s participation into full account and, as in the Lana case, even the extremely subtle secondary effects of the computer used as the operator (Rumbaugh 1977, see esp. pp. 159 and 161; cf. Sebeok 1978). Incidentally, I concur with Hediger (1974:28-29) totally that the view held by some scientists that all contact between observer and observed must strictly cease in experimental (and, by extension, clinical) situations is basically untenable. There are two corollaries to this skeptical stance: first, that the source that generated the message has got to be at least as thoroughly investigated as the destination that is to interpret it; and second, that the properties of the channel(s) interconnecting the two organisms be probed with equal diligence—which brings us to the third implication of the Pavlov story.

Fig. 3. Blondie with Clever Daisy. © King Features Syndicate, Inc., 1977. Reprinted by permission.

No one, it seems, knows how Studentsov unintentionally instructed his mice. Rosenthal, as a part of his ingenious research on the generality of the effects of the experimenters’ expectancy, employed rats, but had no idea how the animals came to differentially perform as they did: “... we cannot be sure of the sense modality by which the experimenter’s expectancy is communicated to the subject” (1966:165), although these rodents are known to be sensitive to visual, auditory, olfactory, and tactile cues (for recent work on the surprisingly complex social interactions of rats, and especially the functions of their ultrasonic calls, cf. Lore and Flannelly 1977). Rosenthal thinks, however, that the specific cues by which an experimenter communicates with his animal subjects “probably varies with the type of animal, the type of experiment, and perhaps even the type of experimenter. With Clever Hans as subject, the cues were primarily visual, but auditory cues were also helpful. This seemed also to be true when the subjects were dogs rather than one unusual horse” (ibid., 177).

Detailed analysis of the carryings on in a laboratory shows that there are “sighs, groans, finger tapping, guttural exclamations or muted whispers” (Pilisuk et al. 1976:515) which are seldom discerned, not to mention “illicit communication” through all the other modalities that goes, if perceived, unreported. Pilisuk and his collaborators found that such signs “did, in fact, have a significant effect upon the level of cooperative behavior,” and advocated further studies of “behavior that takes place ‘on the sly’ since human interaction seems to be full of just such activity” (ibid. 522).

The most prescient of animal psychologists, Hediger, rightly insists that a wide-ranging, intensive program be launched to examine all known and potential channels which are or might be sign-carriers, and thus to determine what is really happening in the interaction between organisms. To this plea, he immediately adds, “Of considerable help in this endeavour is a new branch of science known as semiotics: the study of all possible signals in technical as well as biological and psychological fields” (1974:29). He then enumerates a few situations where reciprocal understanding between man and animal has been achieved via media of communication that are out of the ordinary, stressing not only that hitherto unknown channels are at work but “that familiar ones may already play a role even if outside our threshold of perception” (ibid. 34). The Hess pupil response (which I recently discussed in Sebeok 1977b, Section 5) furnishes one nice illustration of this kind of a sign-process; the equally important eyebrow flash, “discovered” and described by Eibl-Eibesfeldt (1972:299-301), another. The pupil movement and the eyebrow flash are both extremely common if mostly out-of-awareness signs to which we respond strongly in defined contexts; they are used globally by men, except where culture overrides nature (as in Japan, where the flash is considered indecent).

A much more arcane illustration, from the underground literature of semiotics, is furnished by the case history of Eugen de Rubini, the famous Moravian “muscle reader,” who came to San Francisco shortly after World War I. This young man had the uncanny ability to read the “thoughts” of others from the patterns of their muscle tensions. Had he not been carefully investigated by three of the most important contemporary psychologists, eminently including Edward C. Tolman, his feats might today be considered fictional. The very careful experiments they conducted were reported in the Psychological Review by Stratton, under the intriguing heading, “The control of another person by obscure signs” (1921). Stratton and his colleagues adopted Pfungst’s paradigm for Clever Hans, but they never did succeed in pinning down the precise cues employed by de Rubini, who modestly laid no claim to mind reading. He was judged to depend “to some appreciable extent, even though subconsciously, upon visual cues... of a highly elusive kind” (ibid., 309-310). Tremors of the floor, faint sounds of feet, of movements of arms and clothing, together with those made by changes in breathing, were diminished signs, but not excluded under the laboratory conditions imposed. “The experimenters each and all assumed that [de Rubini’s] successes depended upon sensory cues of some sort, and not upon immediate influence of mind upon mind” (ibid., 310). This admirably cautious paper refrains from extrapolation beyond the evidence. Stratton believed that his subject “caught, in the very periphery of his visual field” (ibid., 309), postures or motions which assisted him, but couldn’t prove so. It is instructive to contrast this with the plethora of early 1960s reports of observations of sight through the face (“dermooptical phenomena”): blindfolded subjects were said to read texts without using their eyes. Observations were accumulated, although no possible mechanism ever came to mind, until fraud was discovered. The last moral of Pavlov, then, is a simultaneous call for action, along lines set forth by Hediger, but foreshadowed by Jakob von Uexkiill (Sebeok 1977a), yet tempered by a need for unceasing vigilance in the face of imposture. Clearly, we are all prisoners of our senses, which are imperfectly understood at best, and the weaknesses of which are exploited by con-artists, academic or laic, witting or unwitting.^4.

This book opened with Wells’s “fundamental message. .. of caution” for semiotics. I do not disagree with his wise bidding that, “whatever claims we make, we must be prepared to support,” even though the theme of my closing chapter may have implied this, for I have taken the opposite tack here. My musings touched on many subjects which had in common what I have chosen to call ecumenicalism in semiotics—the pursuit of boundaries and of ways of transcending them. I believe that in semiotics, as in so many other areas of intellectual endeavor, the imaginative and sometimes intuitive search for invariance must go hand in hand with the empirical scrutiny of variation.