Webpage:Language Brain and Evolution
also now see:Ascent of Intelligence and Pinker:The Stuff of Thought Language and the Mind

[Extracted from Nöth, Winfried ed. 1994. Sign Evolution in Nature and Culture, Part III Glottogenesis: Phylogeny, ontogeny, and actogeny, 255-268. Berlin: Mouton de Gruyter]

Language and the origin of semiosis

Robin Allott


Language is the type of semiosis which has been most closely examined and which has served as a model for considering other forms of semiosis. Semiotics has been based, certainly in the case of language, very much on the proposition of Saussure that the sign is arbitrary - a questionable idea (Holdcroft 1991) - and that the sign is conventional or social. If this fundamental idea of semiotics, and linguistics, is discarded, what does this do for semiotics, the 'science' of signs ? This paper seeks to trace out the implications for semiotics of a very different account from Saussure's of the origin, development and functioning of language, leaving it open whether one should conclude, in the light of this, that language does not constitute a paradigm or model for a general science of semiotics (and is not a typical or useful example of a semiotic system) or that language should be treated as the paradigm but a totally changed paradigm, so that the new view of language will require a restructuring of semiotics and lead to a much more biological and indeed neurological approach to the science of signs.

Perhaps the problem can be traced back to Locke. Locke’s remarks at the end of Book IV of the Essay (21.4) virtually amount to a specification of the program which Peirce proposed for himself but never completed. Locke's influence. though indirect, can also be seen in Saussure. A typically representationalist view of the sign was accepted by Saussure and underlies his account of the relation between language and the study of sign systems more generally. It was against this background that he outlined the specific character of language as a system of signs in which the only essential thing was the union of meaning and sound-images, comparable to "writing-systems, the signs of deaf-mutes, symbolic rites, polite formulas, military signals, etc." (Saussure [1917]: 16). Succeeding writers on semiotics have carried the argument to one or the other extreme. Kristeva (1969: 58) sees that for Saussure linguistics is only part of a general science of signs and proposes that linguistics in due course will merge in a universal science or, more colourfully, "La sémiotique se construira alors à partie du cadavre de la linguistique, d'une mort . . . à laquelle la linguistique se résignerait après avoir preparé le terrain à la sémiotique." At the other extreme, as already noted, Barthes asserts the unquestioned primacy of language from which all other sign systems must be derived.

The most sustained recent attack on the Saussurean account of the relation between language and semiotics is that by Holdcroft(1991: 156, 160):

[If one followed Saussure] the way to begin the study of language is to develop the theory of signs, to describe the principles that hold for all signs and sign systems, and then to derive a description of linguistic sign systems from the theory and the principles . . . However, . . . it is surely a questionable strategy [because] to begin with, it takes it for granted that different sign systems have something interesting in common, in spite of the fact that the examples given are very diverse . . . [More fundamentally,] . . . the conception of a sign that Saussure employs is . . . not a plausible candidate for one of signs in general.

What is one to conclude? To relate the structures and processes of language to those of other sign systems, the first step surely is examine what the structures and processes of language are. The science, or technique, that is meant to do this is linguistics. Linguistics as a science or craft has passed through many stages since Saussure's Cours was published. The stages have produced radically different accounts of the functioning of language; the Chomskyan system has been in process of nearly continuous change for more than 30 years. In either its earlier Transformational-Generative form or in its more recent Government-Binding form it is difficult to see how general coding and decoding principles, general sign system principles can be extracted which would be readily applicable to other aggregates or systems of signs as they have been identified by semioticians. And, of course, beyond and apart from Chomsky, there are many other competing and inconsistent accounts of language. How can one say that language is the semiotic prototype if there is absolutely no consensus about the proper way to characterize and analyze language?

Equal confusion reigns about the principles, characteristics, scope of any general semiotic science, and about the description and analysis of other sign systems identified as such by semioticians. There is uncertainty and disagreement at every level, from the lowest: the identification of the sign, to the highest concepts of underlying structure in sign systems. How can one say that language is an exemplar, a specimen of a sign system, subordinate rather than superordinate, if the criteria by which one identifies signs and the methodology for analysing sign systems are apparently completely lacking? A semiotic science has to be founded on recognition of commonalties between diverse sign systems, but there is no systematic account of what language shares with other sign systems.

2. The "science of signs" ?

Rather gloomy current assessments of semiology, or semiotics, of the prospects of the "science of signs" can be set against the hopes of the originators. I have already referred to Locke's prospectus for semiotike and Saussure's equally optimistic write-up of the then non-existent science of semiology. More recent statements of what semiotics might be or aim to be can be contrasted with the far-reaching ambitions of Peirce and Morris. Peirce saw semiotics as the key to human knowledge and, following Locke, as to be identified or at least to be closely linked to a new science of logic. His aspiration was "To erect a philosophical edifice that shall outlast the vicissitudes of time" (Peirce CP 1.1). The ambitions of semiotics have been large; the achievements remarkably small. Why should this be? One answer may be that there was unclarity at the very beginning in the concept of the sign, the atomic element in a hypothetical system. However, some scholars have foreshadowed a different and perhaps more promising line of attack for semiotics. Thom has expressed the view that semiotics would profit from closer contact with biology or ethology. In the set of replies to Koch’s valuable questionnaire, Pinxten suggested that the borders of semiotics are the neurobiological sciences where "absolutely amazing advances are being made today .. in neurophilosophy and the general model of connectionism" (Pinxten 1989: 34).

What can usefully be said about the fundamental problem in semiotics, the nature of the sign? Ever since Locke made the relation between signs (words) and ideas a central feature of human understanding, there has been debate and divergence about the nature of the sign. "Sign" is a word used in ordinary speech; if it is to be given a more precise meaning in semiotics, it is from the ordinary meanings of "sign" that the narrowing down must start. The range of meanings and applications of the word "sign" shows that one cannot use the word as though what it means is self-evident. Nor is there agreement about the classification of signs. Peirce’s three broad classes of sign: index, icon and symbol, are still in active use but, though he later went on elaborate his classification, the baroque, or Gaudiesque, proliferation of sign classes which he outlined has not been useful (any more than the rather similar elaboration by Morris).

There is no greater clarity about the "sign process" by which is meant the sequence of events or operations extending from the origination of the sign (typically by a human source), through its reception by the human destination and then on to its internal effects on the receiver. Peirce outlined the sequence in his account of the functioning for a Frenchman of the word/sign homme, quoted in the previous section. Peirce analyzed the process into three elements, object, sign and interpretant. This sequence of events can be described in mentalist or behaviorist terms. Morris translated the sequence into stimulus and response; characteristic sign-using behavior, on this view, is exemplified in our putting on a coat at the sight of rain clouds rather than in our reading a book.

There seem to be logical as well as psychological difficulties with the sign process as it has been described. Peirce said that nothing is a sign which is not interpreted, so on this view signs have no objective existence; whether a sign is a sign depends upon the receiver of the sign. Then the interpretant of the sign, which is what constitutes the sign as a sign, may be an idea or it may also be, or be described as, a habit, a behavioral disposition, a tendency to action. Yet the interpretant also may be a sign, or a more developed sign as Peirce described it, which within the receiver in its turn may give rise to or be associated which other interpretants, which also would be signs, in what could be, in Peirce's terms, part of an infinite series of signs. This mentalist account seems unclear, even bewildering. The behaviorist alternative has been discarded by most linguists, psychologists and neurologists.

3. Language origin and function: The motor theory


Language is taken to be the capacity of one individual to alter, through structured sound emission, the mental organization of another individual. In considering the origin of language, we should not look for a distinct, datable origin any more than we would look for a distinct, datable origin for the eye. Language is more than speech just as perception is more than the structure and functioning of the eye. In both cases we have also to be concerned with the neural organization underlying the functions of speech and visual perception. The theory is that language was constructed on the basis of a previously existing complex system, the neural motor system. The programs and procedures which evolved for the construction and execution of simple and sequential motor movements formed the basis of the programs and procedures going to form language.

A principal theme is the mosaic evolution of language, the fitting together of a whole array of elements, anatomical, neural and behavioral. Two important behavioral elements for language are imitation and the categorical perception of speech sound, both abilities found in some animals. Imitation of speech or other sound or bodily movement, involves a remarkable and complex linking of perception and motor organization. The capacity to discriminate categorically between human speech sounds in a way similar to that found in adult speech perception has, surprisingly, also been found in a variety of animals, notably in chinchillas, monkeys and in extremely young human infants. These and other behavioral prerequisites for language depend on the intimate involvement of the motor control system and their dependence on crossmodal processes. Development of the language capacity has resulted from the progressive establishment of new crossmodal or transfunctional neural linkages. Cerebral reorganization in the sense that the interconnectedness of different brain regions concerned with what are usually considered distinct functions has substantially increased. This extensive relation between language and the motor system is what one might reasonably expect, given the central role of the motor system in all behavior and the essentially motor character of speech production. The motor system forms the indispensable mediator between language and perception.

The essential additional hypothesis is that the motor system, prior to the development of language, was built up from a limited number of primitive elements -- units of motor action -- which could be formed into more extended motor programs. If this is so, then one can look for a direct correspondence between the primitive motor elements and the fundamental elements of spoken language. The processes of word-formation and syntactic rules for constructing word-sequences would then be derived from the neural rules governing the union of motor elements into simple and more complex actions. If language is in this way derived from the motor system, there is no reason to believe that any aspect of language – sound elements, words or syntactic structure – is necessarily arbitrary.

The relation between motor programming and speech programming can be examined at each level, the phonemic, the lexical and the syntactic. For phonemes, this leads to the idea of an invariant program for each phoneme, or "auditory targeting", a motor-alphabet underlying speech, related to the elementary motor-patterns underlying other forms of action. On the motor theory, the categorization of speech-sounds is derived from organization prior to language, and specifically from the categorization of motor programs used in constructing and executing all forms of bodily action. What the rhesus monkey, or the chinchilla, in their capacities for categorical perception of speech sounds, share with the young human infant is very similar skeletal and muscular organization. The specificity of the phoneme is the accidental result of the application of the different elementary motor subprograms to the muscles which went to form the articulatory system. The theory suggests that the phonemes of all languages are likely to be selected from a limited set of possible distinctive speech sounds.

The link between the motor system and the formation of words follows. The hierarchical structure of the motor system is built on the basis of a limited set of motor elements. These are combined in an unlimited number of ways (motor words). Words in speech are a read-out of neural structures in much the same way as actions or facial expressions. A word, as a neural structure, can be formed from the co-activation of the motor subprograms for phonemes which are then melded or shingled together to form a distinct neural program for the whole word. Experimental approaches with the creation of artificial words have suggested that there can be a lawful relation between speech-sounds and auditory or visual percepts. Research into sound-symbolism suggests that there is an isomorphism at the motor level between speech and the contents of perception. The object seen produces a motor pattern which is readily transferable as a motor program to the articulatory system and so becomes the associated word for the thing. The neuromuscular sequences which are the immediate motor programs underlying words are derived from the integration of the neural structures underlying perception in all its forms (visual, auditory, tactile etc.) and motor organization.

The phonemes which go to form any word manifest pre-existing motor programs for the general control of bodily activity, and if, in the simple case, the words relate to actions or familiar external objects and percepts, then inherent in each action represented by a word is its own overall motor program, its own combination of motor control elements. Motor programs are also of essential importance in perception, in vision (eye movements in scanning and in other adjustments) and in perception through touch. The motor theory suggests that the selection of the structure of phonemes for a particular object or action derives from the motor aspects of the action or perception to which the word refers.

There must, on the theory, be a close relation between the structuring of motor activity, motor syntax, and the organization of language, speech syntax. The motor theory suggests that syntax also has a physiological origin, an origin in cerebral organization for action, for motor control. If one treats syntax, at its simplest, as the putting together of motor-derived words, there is a clear analogy between the formation in speech of a sentence or a phrase (or more extended discourse) on the one hand and the formation of extended bodily movement, an action sequence, a complex neural motor routine on the other hand. The specifically syntactic aspects, i.e. ordering regularities, are in the motor theory derived from ordering regularities in the neural sub-structure for motor control and perception (ordering regularities in visual perception and in the execution of bodily actions). For each of the three main components in syntax (word classes, ordering regularities and function words), the relation to the motor theory may take the form of a relation directly with the organization of perception – the grammar of vision. Vision, like action, is substantially motor-based. Scanning of a visual scene by the eye is a serial process, just as spoken language is a serial process. The iconicity of syntax in terms of word order derives from this.

4. Motor theory and other semiotic systems

From the summary of the motor theory in the previous section, one can pick out a number of distinct aspects which may be helpful in looking at other semiotic systems.

Animal (vertebrate) precursor features. In other semiotic systems one ought to consider what the evolutionary history might have been and whether anything in the system has parallels in animal organization and communication.

Mental organization change. The role of language is to produce changes in the mind (brain) of the hearer which structurally resemble those in the mind (brain) of the speaker. This presumably would be a feature of all semiotic systems where there is a human source and destination. It may even have relevance where the communication is mediated in space or time e.g. processes similar to secondary forms, such as written language.

Neural basis. All semiotic systems and "messages" ultimately will be represented by changes in neural patterning: changes in synaptic strengths, establishment of new excitatory or inhibitory connections, dendritic growth (as hypothesized in PDP or connectionist approaches to neural modelling)

Crossmodal processes. The role of any semiotic system may be to produce specific action from perception, whether the perception is of the behavior of a human being or of changes otherwise in the external environment. This will depend on the existence of crossmodal links between different brain functions. What these may be can be examined in relation to particular sign systems.

Motor system centrality. All behavior in evolutionary terms is a matter of motor primacy, and one would expect the motor primacy to be reflected in the construction and functioning of other semiotic systems.

Primitive motor elements (programs). Because of the limitations of the human (and other) brains -- there cannot be pre-formed neural programs to provide for every possible sentence or every possible action -- and the demands of an unpredictable environment, semiotic systems have to be open-ended. The parsimonious way is to rely on a limited set of primitive elements which can be combined in many ways to meet the needs of any situation. Other sign systems may use the primitive motor elements underlying language or other primitive elements.

Categorical perception. Production and perception of natural sign systems must have evolved together for them to operate as communication. In perception, there must be provision for extracting primitive elements or combinations of the primitive elements reliably from the incoming "message".

Non-arbitrariness. Insofar as the motor elements are the product of evolution of neural organization, language rests ultimately on a non-arbitrary basis. It is for examination how far other systems have a (perhaps covert) non-arbitrary base.

Hierarchical pattern of system. Motor control is a hierarchical process, with relevant parameters being fed in at the appropriate level. This is a necessary organizational economy and one can look for a comparable hierarchic structure in other sign systems. The primitive elements in any sign system must be capable of being melded together to form higher-level structures, to provide a "lexicon" to match the multiplicity of objects or actions in the field to which the sign system applies.

Derivation of second-level patterns from the structure of perception. In the case of language, the particular structure of words is derived from the structure of the perception or action to which the words refer. Whether something like this happens in other sign systems can be examined.

Extended sequences. In the case of language, syntax is based on motor syntax. Motor syntax may also be a source for the organization of extended sequences in other sign systems, or if not one can examine exactly what rules govern these sequences. Besides forms relating to actions or percepts, there will also be schemes for organizing sequences, both in terms of order and through functional devices (function words in language).

Diversity . An obvious feature of language is geographic and community diversity. Can one identify similar diversity in other systems? The existence of geographic or community diversity strongly suggests an original population-genetic basis, manifested physiologically or neurologically, affecting preferences in the choice of the sign set by the sign-using group.

Genetic/cultural development. Sign systems are dependent on the existence of shared abilities for production and reception of the signs between members of the community. The current forms of a language are the product of a mixed genetic/cultural development. In other sign systems where geographical or other diversity is found, one would expect a similar mixed genetic/cultural pattern of development of the system.

In a very tentative way, one can comment on how far these aspects of the motor theory of language may apply to other forms of communication, to other sign systems. Given the primacy of cerebral motor control in the theory, the most likely group of sign systems to which the theory may be extended are those with a prominent motor element. This includes:

Gesture. Gesture can take various forms. The most familiar is gesture directly accompanying speech. McNeill has in many papers and books defended the thesis that such gesture in some way appears to be a direct derivative of the speech-process, illustrating the concepts with which the utterance deals. Gesture of this kind certainly could be analyzed very much along the lines of the motor theory.

American Sign Language and other sign languages(BSL). These have been extensively researched and analyzed by Stokoe, Bellugi and others to demonstrate their rightful claim to be treated as languages though not direct derivatives of spoken language. These sign languages contain iconic elements as well as conventionalized elements; they have morphological and syntactic structure though understandably different from the morphology and syntax of spoken language. Again an obvious candidate for treatment in terms of the motor theory.

Facial expression. Ekman for many years has undertaken research which shows the universal and natural character of facial expression and the scope for analysis of facial expressions in terms of primitive elements of muscular action which combine to form complex expressions. Neurological research, for example by Ojemann and Mateer, has demonstrated a close cortical relation between speech areas and areas controlling facial expression.

Postural expression. All bodily actions, including articulation and gesture, can be seen as modifications of bodily posture. Posture is not something static but a dynamic, constantly modified state. The motor elements which form part of the motor theory of language are modifications of posture which result in particular bodily movements.

All these motor-based sources of expression are important elements in making possible non-verbal communication and the empathetic perception by one individual of another's intentions and emotional state. Less obviously, the motor theory or aspects of it may be relevant also for:

Music. The performance of music is a highly motoric activity and perceived music is powerful in producing responsive movement in the listener. In language also there are aspects with some relation to music, tone and pitch, rhythm, the melodic line of the sentence. In song, language and music join to produce their effect on the listener. Macdonald Critchley (1977) and others have written perceptively about music and the brain. It would be worth considering whether features of the motor theory might be helpful in analysing musical production and musical perception.

Dance. There have been attempts to prepare an alphabet of choreography in terms of primitive elements from which more elaborate dance-patterns are constructed. Dance is a motor-response to music, whether the dance is classical ballet or disco-dancing.

Insofar as the motor theory proposes and requires a neurological/ physiological matching between motor action and perception (particularly visual perception), there is some prospect that features of the motor theory might be relevant even for:

Figurative Art. Drawing and painting are motor activities and one could treat both the creation of art and the response to it in motor terms. It is interesting that the categorical perception of colour in many ways resembles the categorical perception of speech-sounds.

Alphabetic systems or hieroglyphs. These are specific examples of visual patterning. Kohler's MALUMA/TAKETE experiment shows the possibility that visual patterns such as these may reliably be linked to word-patterns.

Obviously a much closer examination has to be undertaken of each of the sign systems before one could reach firm conclusions about the relevance of the motor theory or of aspects of the motor theory. Nevertheless, the motor basis of the sign process looks a good candidate to serve as a unifying concept in semiotics.

5. Semiosis, language and perception

The confusion about what constitutes a sign makes one wonder whether there is something fundamentally wrong. Is the whole idea of the atomic sign a mistake? Are not "signs" only meaningful in the context of the total organization of the individual's "knowledge-structure" and of the total relevant environment as perceived by the individual – in much the same way as the meaning of a word depends upon the whole context in which it is spoken? The key questions then would be: How is the sign marked off, distinguished from the totality, that is, how does the subject decide where to focus his attention, whether in regarding the external world or regarding his own interior world ? What makes for the salience of the sign? How does the "sign" modify the subject's internal mental [ neural ] organization, that is, the mode of interpretation of the sign? This brings one very close to issues about the nature and process of perception.

A sign plays a role in perception. The sign must, in some sense, be perceived before it can function as a sign by being matched with its hypothetical interpretant (a pre-existing idea or concept, tendency to action or whatever an interpretant is supposed to be). Perception thus is essential to, and prior to, the sign process. Peirce and Morris touched very briefly on this: Morris simply noted "the difficult question as to whether or not perception is to be regarded as a sign process"(Morris 1946: 34) but Peirce (CP 4.424) more categorically said: "The immediate object of all knowledge and all thought is, in the last analysis, the Percept".(1960 IV:424) Thus, semiosis in some sense is perception. From survey of current views of psychologists and neurologists on the nature of perception, one can draw the following points:

1. Perception cannot readily be analyzed sequentially from the object, through the sensory channel to the 'mental' pattern of the concept, and interpretation of the concept. The perceiver brings a prior structure to each act of perception and the act of perception is directed (as an aspect of attention), not simply a mechanical pick up of whatever information there happens to be.

2. Perception is not intermittent but continuous, producing continuous change in the perceiver. The prior structure which the perceiver brings to each act of perception is an elaborate neural structure which in some way constitutes a model of the expected environment as well as a model of the individual himself, body and mind. Each new perception modifies the total model.

3. The creation of these models is closely linked, perhaps directly dependent on the representation in the brain of the perceiver's own body. The body image serves as an organizing construct from which all other perception derives. The structure of oneself is the means by which we order our experience of the world (the structure of oneself including the permanent awareness of oneself as body plus accumulated experience, and attitudes).

4. Perception is associated with action in many different ways. The act of perception is itself a motor performance, requiring direction of the head, the eyes, even the bodily posture. And perception is closely involved with the determination of patterns of action following perception. Not all perception leads immediately to action: reading a text may modify the reader's "ideas", produce neural changes, without resulting in any overt motor action. But, in evolutionary terms, perception developed to serve the need for action, for response to the environment, and the structuring of perception and of bodily action are likely to be very similar, complementary and interlocked.

5. Both the structures of action and the structures of perception are something that we have inherited, that evolved long before Homo sapiens. It is likely that our organization for perception and for action are the same as or very similar to those of other primates, and probably much further down the evolutionary scale. Signification for us and for animals must be a very similar process.

What bearing has this view of perception on semiotics, the science of signs, communication? What distinguishes perception from communication? Here we encounter problems with the vagueness of terms: if we identify communication with all sign processes, messages from the environment, then perception and communication become identical. The cloud communicates information about rain to us. This, however, does not seem a very helpful use of "communication". The narrowing down of the concept seems to require that we should call communication only the transfer of meaning from one animate creature to another. Our awareness of the other extends not only to awareness of them as a particular body but also of aspects of their self, perceived through language and through interpretation of the indicia of the other's attitudes, structure, emotions, character etc. If this is so then the concept of the "sign" takes on a quite different property: it will be because one human and another, one animal and another, share similar behavioral organization, perceptual organization, language organization, ultimately neural organization, that an action of one can serve as a sign conveying meaning to the other. The meaning of "sign" then has to be deepened to recognize it as the product of a complex shared structure.

Perhaps we should seek to construct a theory of signification (the extraction of meaning) rather than a theory of signs. We should not conceive of the neural processes as successive processes of coding and decoding but rather as transformations which result from propagation of the input through a complex system (the total network of neurons). Rather than serial processes, something much more like the ideas in parallel distributed processing will be relevant. Connectionism leads away from the atomic sign, the atomic idea, the atomic concept, to the complex global modifiable network, producing expectancies which control perception, that is, how changes in the environment are processed from moment to moment.


Allott, Robin 1989 The Motor Theory of Language Origin. Lewes: Book Guild.

Holdcroft, David 1991 Saussure: Signs, System and Arbitrariness. Cambridge: Cambridge University Press.

Koch, Walter A. (ed.) 1989 Culture and Semiotics. Bochum: Brockmeyer.

Kristeva, Julia 1969 Semeiotike: Recherches pour une sémanalyse. Paris: Seuil.

Locke, John 1706 [1972] An essay concerning human understanding. (5th edition.) [1972] 2 vols. London: Dent

Morris, Charles 1946 Signs, Language and Behavior. New York: Prentice-Hall.

Peirce, Charles S. 1931-1958 Collected Papers of Charles Sanders Peirce. 8 vols. Eds. Charles Hartshorne, Paul Weiss and Arthur W. Burks. Cambridge, Mass.: Harvard University Press.

Peirce, Charles S. 1982 Writings. Vol. I (1857-1865). Ed. Max H. Fisch et al. Bloomington: Indiana University Press.

Pinxten, Rik 1989 "Culture and Semiotics", in: Walter A. Koch (ed.), 34-35.

Saussure, Ferdinand de. 1916 [1966] Course in General Linguistics. [Trans. Wade Baskin.] New York: McGraw-Hill.