The Semiotic Self

Sebeok, Thomas A. 1992. 'Tell Me, Where is Fancy Bred?': The Biosemiotic Self. In: Sebeok, Thomas A. and Jean Umiker-Sebeok (eds.), Biosemiotics: The Semiotic Web 1991. Berlin; New York: Mouton de Gruyter, 333-343.

Shifting sharply to a commonplace sphere of observation from the cascade of such lofty questions as Bassanio toyed with, note that the police officers in American cities, and perhaps elsewhere, when they arrest someone and invite him for a ride 'downtown' in the back seat of a police car, firmly press a palm on the handcuffed suspect's head when ushering him into the seat. This practice is so familiar that actors impersonating cops and robbers in a movie or a television show cooperatively simulate this very gesture, often, I hear, not knowingly or fully understanding why. Police officers explain their action as routine prevention procedure to make sure that the suspect doesn't contuse his head on the car frame while in custody and thereby later claim physical abuse.
To me, however, this ocmportment suggests an interesting and empirically quite accessible research problem: are human beings - or, for that matter, vertebrates in general (Hediger 1980: 44f.) - consistently aware of their body size, viz., their changing height? Given that we grow in stature at a relatively leisurely pace from childhood to adulthood, then tend to diminish somewhat as we become yet older, how are these reformations registered in consciousness and implemented so as to, for instance, know when to duck entering a car? (For that matter, how do drivers internalize their fairly accurate knowledge of the perimeters of their vehicles so as to avoid scraping surrounding objects?)
More generally, still, how are self-images established, maintained, and transmuted into performances? Sensory experiences may at times pose semiosic ambiguities, as in the following seemingly paltry example: A hole in one of my teeth, which feels mammoth when I poke my tongue into it, is a subjective symptom I may elect to complain about to my dentist. He lets me inspect it in a mirror, and I am surprised how trivially small the aperture - the objective sign - looks. The question is: which interpretation is 'true'? the one derived via the tactile modality or the one reported by the optical percept? (Cf. Sebeok 1986: 55.) (Sebeok 1992: 333-334)

In trying, since 1977, to come to terms with many more or less anomalous semiosic phenomena, only a sampling from which I can recount and illustrate here, I began to explore the notion of the semiotic self. Cited instances have to do with the somatic localization of such passions as love or such feelings as anxiety; the incorporation (as it were) of such faculties as one's own body-size; and the association of other private experiences, such as light-headedness, pains, twinges, nausea, hunger and thirst, 'funny feelings', or what Hungarians call, in a well-nigh untranslatable idiom, their közerzet (a generalized but amorphous state of good or ill health), with their respective outward manifestations of referents. The effects of wrongly parsed sign processes or their impairment, including long-persisting images of amputated extremities, constitute another profoundly enigmatic class of events, as in the eerie case of the one-armed Paul Wittgenstein's - Ludwig's brother's - amputated but lingering phantom right limb with its reportedly still virtuoso fingering technique for some new composition (Otten 1992: 45). (Sebeok 1992: 335-336)

Difficulties of this sort arise in part, I believe, from the fact that bodily sensations and the like, most saliently among them those connected with illness, are not amenable to verbal expression becaue they lack external referents; insistent intrusions though they may be into the routines of one's day or night, they can at best be denominated, for they resist unfolding into narratives, which are, by definition, always verbal. (Sebeok 1992: 336)

Where, then, is the 'semiotic self' located? Clearly, in the organism's milieu extérieur, on the level of an idiosyncratic phenomenal world, tantamount to J. von Uexküll's Umwelt (1973: 334-40) - a technical appelation I prefer to render as the 'model' of a species-specific segment of individual reality (Sebeok 1991a: chapter 5) - made up of exosemiosic processes of sign transmission. Miller, in a nice figure of speech, tells us that sensations happen 'in an isolated annexe called the self, and if that annexe is missing...the sensations float around in a sort of elsewhere' (1978: 20). This semiotic self, which of course enfolds and thus 'contains' in its milieu intérieur some body's immunocompetence, occupies, as it were, space/time in a sphere bounding the organism's integument, although the programs for the fabrication of subjective constructs of this sort are surely stored within the subjacent realms of its endosemiosic organs (semiotic aspects of pertinent boundary conditions were recently discussed in Hoffmeyer 1992). This semiotic self, furthermore, is composed of a repertoire of signs of a necessarily sequestered character; as J. von Uexküll - claiming that even a single cell has its Ich-Ton - remarked (1973: 68), 'bleiobt unser Ich notwnedig subjektiv'.
Peirce, in his canonical amplification of the classic definition of a representamen, wrote that a sign 'is something which stands to somebody for something in some respect or capacity' (c. 1897, CP 2.228). It is his addition precisely of the tag 'to somebody' which illuminates the semiotic self, and which doubtless engendered the Shakespearean notion of Peirce's 'glassy essence' (Singer 1984) that enswathes all living organisms within their private Umwelts in the manner of an impalpable, solid withal context-sensitively and environmentally supple (Sebeok 1989: 46) carapace, or, as I have previously dubbed it, a Hediger bubble (cf. Sebeok 1989: 45, 1991: 40). This invisible, malleable proxemic shell amounts to nothing less than what laymen call 'reality', to which all sign users and sign interpreters are knit by a formidable array of indexical representamina (Sebeok 1991a: chapter 13). (Sebeok 1992: 337-338)

Nature's indexicals are universally nonverbal, but in our glottocentric genus the former may increasingly, if always selectively, be enhanced - in a phylogenetic as well as in an ontogenetic sense - by verbal elements, including especially deictics, designators, reagents, metonyms, symptoms, clues, cues, synechdoches, and pars pro toto expressions. In Homo, nonverbal as well as verbal indexicals may be either vocal or nonvocal. Nonverbal vocal indexicals, such as groans and moans, are public signs of latent discomfort; and so are nonverbal nonvocal expressions, such as frowns or writhings. Signs of this kind are both promulgations of, for instance, pain, in contrast to exposures, usually out-of-self-control, such as yellow skin exhibited by a jaundiced person. (Sebeok 1992: 338-339)

The barriers which occlude and thereby separate each and every windowness monad from all the rest are such as to prevent any self from fully fathoming any other. Hurdles between Egos - unlike those between cells, minimal reproductive units that are surrounded by semipermeable membranes, allowing the passage of certain chemicals and thereby certain information - are insurmountable. We can of course, and regularly do, spin fantasies about, or 'image', the situation of an 'other', or even perhaps empathize with a fellow-human's or some pet's singular individuality; but our respectively impenetrable semiosic orbits must perpetually remain separated by a frigid intergalactic void: the self's perception of any other is composite, partial, and forever incomplete. We can approach the 'real' richness of the universe only by entertaining multiply contending, mutually complementary visions. I believe this is the quotidian implication of Niels Bohr's celebrated adage that it is 'wrong to think that the task of physics is to find out how nature is. Physics concerns what we can say about nature' (Pais 1991: 427). (Sebeok 1992: 339)

An act of interpretation is an act of as-sign-ment - that is, the elevation of an interpreted phenomenon to 'signhood'; indeed, this is what the word 'encoding' betokens. Interpretation is an autopoietic (i.e., actively self-maintaining) process, and one that operates, moreover, on the product of its own operations (Maturana and Varela 1987: 47-52, 253); that is, it is recursive, as both Peirce and the Uexkülls undeniably understood. Moreover, the elder of the latter family was the earliest to actually postulate a biological mechanism for the elucidation of the process (1973: chapter 5 and passim) - namely, the well-known 'functional cycle' (Funktionkreis), in the course of which a meaning is not merely con-sign-ed (Bedeutungerteilung), but also pragmatically verified (Bedeutungsverwertung). (Sebeok 1992: 340)

This paper reflects work in progress on the topic of the 'Semiotic Self'. Previous ponderings were reported on in two other papers, both now conveniently available in Sebeok 1991a [A Sign is Just a Sign]; especially chapters 3 and 4. (Sebeok 1992: 340)

Kull, Kalevi 1992. Evolution and Semiotics. In: Sebeok, Thomas A. and Jean Umiker-Sebeok (eds.), Biosemiotics: The Semiotic Web 1991. Berlin; New York: Mouton de Gruyter, 221-233.

Another paradigm means another language, with other values for the description of already known things. It means another world picture. If the semiotic paradigm were to be adopted in biology, then all biological phenomena would be viewed differently, from the semiotic viewpoint. (Kull 1992: 221)

The main mechanism of recognition according to some preexisting model is thus a consequence of the ability to reproduce. Recognition must proceed according to certian constraints in order to work properly; it is a result of development through selection and reproduction of the most suitable forms. Recognition seems to indicate selective retention of certain elements of experience, and extinction of other, irrelevant elements. Recognition is, to use Heschl's (1990) term, a result of cognitive gain. (Kull 1992: 222-223)

It has been noticed by many biologists that in the case of uniparental organisms - i.e., many prokaryotes, or those eukaryotes which have lost the capacity for sexual reproduction - clear typical species are usually absent (Grant 1985). These are the so-called 'difficult groups' for taxonomists, like Hieracium or Alchemilla among plants, in which huge numbers of microspecies have been described with no real hope for their identification by other investigators. When biparental reproduction is lost, the clear boundaries of the species will usually be lost as well. (Kull 1992: 226)

The same logical mechanisms that creates species in the world of organisms is also responsible for creating words in language. First, note that words are reproducible entities: every time we pronounce a word, we actually reproduce it. Physically, we are able to pronounce an almost continuous variety and unlimited number of different sound sequences, and the same is true for all speakers. But if some of these sound combinations have nay meaning to other speakers, they may be reused (= reproduced), whereas the sounds which are not recognized will be reused less frequently. In this way, only the utterances which can be understood (recognized) by other speakers are likely to be remembered and to take part in the development of speech. The almost continuous variability of sound sequences in a child's utterances is eventually split into words, each word having a certain variability in pronounciation but remaining compatible with the same words as pronounced by other people. Between the ranges of variability of similar words there are hiatuses - i.e., the intermediate forms which can be pronounced by the speaker, but which are not usually used because they would not be identified (recognized) by other people as specific words. Different people always pronounce the same words in slightly different ways, but the range of these differences is regulated by recognition capabilities and the closeness of similar words. The requirement to recognize the words limits their variability to a certain range; it keeps the pronounciation exact. This situation is exactly analogous to speciation and stabilizing selection. The intermediate forms should be as rare between the biological species as they are between the words of a language. (Kull 1992: 228-229)

Classification of Interaction
Let us summarize briefly the general conditions required in order for semiotic relations and qualities to arise. We will compare the relationships between systems dependent on their ability to reproduce. Dividing systems into reproducing (R) and non-reproducing (N), we get the following three types of fitting interactions:
- N-N: interaction between inorganic systems. This may lead to mechanical congregation of similar systems, if they fit to each other, as in the case of crystal growth. No semiosis.
- R-N: for example, the relationship between organism and abiotic environment. The organisms that fit to their environment better will reproduce more; as a result, the distribution of organisms changes toward better fitness. Using the equivalent expression, organisms are adapting; in other words, organisms recognize their environment. This leads to the primary semiotic relationships, in which the environment will hold meaning for the organisms.
- R-R: reciprocal recognition (= reciprocal adaption) between reproducing systems will create internally compatible groups of limited variability. This makes possible the origin and development of language.
The R-N interaction is the basic one in the mechanism of natural selection, investigated by the Darwinistic theory of evolution. The R-R interaction is considered to be the basic one by the recognition concept of species, which gives an alternative explanation of the origin of species. I believe that the adoption of a semiotic paradigm as the paradigm of general biology is a good basis for the systematization of theoretical biology. (Kull 1992: 230)

Csányi, Vilmos 1992. The Brain's Models and Communication. In: Sebeok, Thomas A. and Jean Umiker-Sebeok (eds.), Biosemiotics: The Semiotic Web 1991. Berlin; New York: Mouton de Gruyter, 27-43.

The definition of the term 'model' has changed considerably; nowadays, under strong ingluence from the systems sciences, we generally use the following one: a 'model' is always a simpler system in which the components and their interactions are isomorphic to the components of a more complex system, and some of their interactions. Model building, therefore, is always a simplification and a special identification between two different systems, one of which is the model and the other the system being modeled. We make use of the model by operating it, and based on its operations, we predict the behavior of the system being modeled (Csányi 1989). (Csányi 1992: 27)

Even the simplest of nervous systems is characterized by a triple partition. Separate receptor neurons interact with stimuli, motor neurons create or mediate behavioral instructions, and interneurons between receptors and motor neurons serve as a reference for the activity of the three as a whole. The actual state of the network of interneurons, or their past activity, influences the choice of the actual behavioral instruction and the extent of response, if response occurs at all. With the concept as the basic functional unit of the brain, the orgnaization of any kind of behavioral act can be described. The phenomena of sensitization and habituation in simpler animals are good examples. In the course of sensitization, the excited state of receptor cells (key), transmitted to the interneurons (referential structure), not only evokes an immediate reaction (action) within the nervous system, but also produces a lasting state of excitement in the interneurons, which is reflected in their referential role. These cells react to the reappearance of the stimulus with a more immediate, more direct, and more effective reactions. The mechanisms of habituation is essentially similar. These three functional components of behavior are retained even by higher nervous systems, but the complexity of the parts is increased enormously and functional overlaps are developed. (Csányi 1992: 28-29)

During the evolution of linguistic ability in humans, a fundamentally different mechanism of model-making emerged. In the act of naming something, a key (the word) arises which has only a very loose connection with percepts and actions. The 'word-referential structure-action' segmented units could be combined not only through grammatical and logical rules. This resulted in the very complex superstructures of conceptual thought. Human thinking is a rule-driven wandering on the surface of concept superstructures. A linguistic concept may be regarded as an utterance of human thought. It can also elicit actions, but primary experience is no longer a prerequisite for these actions, as it was in the case of other animals. Mental superstructures built up from linguistic concepts also reflect experiences, and therefore may also be regarded as models of the outer world.
The animal's brain, if it belongs to a long-living higher species, is able to construct complex concept-superstructures from individual experiences. However, because of the very nature of animal concept units (Key-Referential Structure-Action), these superstructures are bound, exclusively and finally, to outer reality. They are only the representations of the external environment, good or better, but nothing more.
In the evolution of man, the symbolic information content of the brain's models plays the most important role. Animals are capable of thinking in their own ways, but according to our knowledge, only man uses descriptions in thinking. Descriptions have a double function. First, they are representations, models of the outer reality. On the other hand, they are active entities in the human brain, and can interact with other concepts of the brain through logical and linguistic rules. Their 'meaning' is connected to this second function. Meaning is an active property of a description, and is tied to the whole conceptual system by which the description is made. Linguistic concepts can be detached from reality. If the perceptual keys are transformed into words, the referential structures may evoke actions that are themselves also words, words that may be spoken or written and might become keys again. This feature of linguistic concepts contributes to the creation of a self-generating system of concepts that are only occasionally influenced by reality.
The development of linguistic concepts - the ability to form conceptual thought in man - has led to the emergence of a genuinely new brain system. Abstract thinking creates a system of self-organizing concept-superstructures which are not only primitive models of reality but autonomous entities; their internal structure and dynamics pertain not only to the outer environment, but to the relation between the emerging new system and reality. Self, imagination, fantasy independent of experiences - their connections and the relations among them are the most important features of this self-created world that we call mind. With the help of his/her mind, not only is the individual able to react appropriately to changes in the immediate environment, but it can view itself as part of the environment or as acting object, analyze the relation between itself and reality in a wide range of the space/time continuum, and project its own position into the past or the future. The mind can create a world of fantazy wherein self plays a relatively subordinate role, but rigorous rules exist concerning the dynamics of other abstract entities, such as the world of mathematics. Religions are created worlds in which everything in which everything revolves around the self without reflecting any constraints of reality. (Csányi 1992: 29-31)

It is worthwhile comparing animal and human model building processes by way of an example. It is well known that in the cooperative groups of certain higher mammals such as the wolf, individuals use hunting tactics in which they are very attentive to each other's actions. Each pack member positions itself in such a way that the appearance of a fellow member at the right moment and the right place is clearly supposed. This significant form of cooperation can exist because of the high similarity of the brain's models in each wolf. Individuals can identify with fellow members and can predict pack members' next actions - that is, they have appropriate models of the behavior of others. This kind of cooperation occurs without the exchange of plans, intentions, or thoughts. Wolves are capable of informing each other only about certain parameters of their internal state, such as hunting or aggressive mood.
Not only can the members of a human hunting group predict the behavior of their fellow members, they can divide common tasks among themselves - they can make plans and then assign to each member a particular role by means of linguistic communication prior to the action (Eibl-Eibesfeldt 1982). In this way, concepts existing in individual brains become parts of a higher collective structure, which then determines the goal and the precise ways of achieving it. We call this higher structure of individual an idea. Every idea is in itself a superstructure, a part of a social super-model - that is, a system built up from concepts that are, physically and organizationally, components of the brain models of individual members of a given human group.
The concepts that comprise an idea are not selected at random; rather, they form a functionally organized set that makes the performances purposeful and possible. It is not important for each member of the hunting group to know everything about the tasks or roles of the others. It is enough for the leader to know the main program, but even for him it is unnecessary to learn the finer details of the tasks assigned to particular individuals. Ideas can be organized hierarchically, and the whole is only available in the group as a whole. Individual concepts existing in the brains of the group members can be functionally combined only by a specific self-organization of ideas. Only the ideas that contain those and only those concepts suitable to achieve the given goal can act and accomplish something.
Besides linguistic competence, it is the 'rule-following behavior' of our species that makes ideas possible. Most of the concepts of an idea are simple behavioral rules, which are the elements of collective action. Language itself can be described by a series of elementary ryle-following behavior actions. (Csányi 1992: 31-32)

Man-made objects are always expressing ideas; that is, they can also be translated as systems of organized rules of behavior. For instance, consider how many rules are followed when we use an instrument as simple as a key. I have to take the key with me when I leave, and I must have it with me when I return. I have to hold it in a given way if I want to open the door. Different rules apply to whether or not I will leave the key in the lock. The notion of the key, ownership, the lock, the fitting together, etc. are also formulated in complex ideas. But the making of the object itself - for example, the key mentioned above - can be described as a series of fixed rules. In the course of preparing the casting mould and the metal, and during the casting and the rest of the work, the worker must obey well-defined rules; the object of use is a result of all these steps. Obeying the rules is the most essential biological feature of man. Our nearest relative, the chimpanzee, can be taught to do many things. With enough training, he may even be taught the maneuvers and operations used by people living a simple country life. But if a few hundred chimpanzees were taught in this way and placed in an empty village, the social life characteristic of human communities would never develop, as the colony of chimpanzees is unable to obey complex systems of rules. If hungry, individuals will acquire food at any price, and they also satisfy their sexual desire immediately and forcefully. A human may die of starvation without attempting to touch the food in the supermarket if he has no monmey to buy it; this cannot happen with an animal. For man, obedience to the rules is more important than anything else. Even if we do break a set of rules, we generally do it on the basis of a system of rules that we consider to be more important than the one we disobey. (Csányi 1992: 33)

Human-to-human communication is a very special case. In the transformation of referential information to description and then back to referential information, we see the exchange of components that are parts of the same shared super-model, and which are functional in both individual model systems. The correspondence will therefore be extremely high, due to the great similarity of the brain models of the individuals. This permits an effective exchange of nonreferential information which, because it is encoded into linguistic terms that are easily understood by the receiver, can immediately be transformed into referential information. (Csányi 1992: 37)

From the study of primates and apes we know that individuals in groups constantly observe the activity of other members and try to predict future actions of important individuals and use their social skills to manipulate others (Byrne and Whiten 1988). Since they are capable of only Type I communication, and since the correspondence among such brains is low, they can only predict certain parameters of the internal state (motivation, for example), occasionally intention (aggression, submission, etc.), and, in rare cases, changes in the environment (danger). (Csányi 1992: 38)