The hypothesis relates speech, gesture and perception. It treats phonetic symbolism as a manifestation of the natural foundation of language in the functioning of the human brain and body. The hypothesis is first presented as a series of assertions. This is followed by explanation of some of the terms used. Detailed development of one aspect of the hypothesis in the form of a systematic schema of the relation between speech-elements and gesture-elements is reserved for a separate chapter. Discussion is also reserved to a later chapter of some of the objections and difficulties that may occur to the reader since these can be considered more satisfactorily when both the broad outline and the detailed content of the hypothesis have been set out.

The main propositions which form the hypothesis are listed below. The underlining of some phrases refers forward to the explanation which follows later for the particular terms:

A. Associated with the speaking of every word is a specific invariant pattern of brain organisation. This is the pattern subserving the form and timing of the articulatory processes involved in the speaking of the word;

B. The pattern thus associated with the speaking of a word is not simply derived from the articulatory process. It is prior to the articulatory process and has a special relationship to the meaning of the word.

C. This special relationship between the pattern for a word and its meaning can have different forms depending on the category of word involved:

- the simplest case is for words referring to the human body, to parts of it or to actions referable to the human body. In this case, the pattern underlying the word is typically the product of the state of brain organisation that accompanies movement of the part of the body involved, indication of that part e.g. by pointing or more generally that accompanies awareness of that part of the body or of a specific body feeling;

- in this most straightforward case, the relation between the articulatory pattern for the word and the pattern of brain organisation for movement of the part of the body referred to exists because the brain is a single organ which operates integrally. A movement of a part of the body modifies the set of the rest of the body, including the articulatory organs and muscles;

- there is similarly usually a specific, non-arbitrary relation between words referring to acts of perception (hearing, seeing) and the particular percept which is the meaning of a particular word. So the hearing of a sound produces a pattern of brain organisation which is transposed into an articulatory process producing a word naming the particular sound;

- the relation of the word for a seen object and the character of the object is less clear (because a seen object may have a number of distinctive features) but there is normally a specific relation similar to that for the relation between a word and a sound to which it refers.

It is posited that the seeing of a particular object produces a specific brain pattern characteristic of the object (the pattern involved in recognition of the object) and that this pattern of visual perception in turn leads to a specific word - a specific articulatory sequence. The brain pattern associated with the particular perceived object takes the form of a modification of the optic component of the body-image: the body-image being a stable, general, internally-perceived pattern by which the individual locates and demarcates himself, internally and externally in the environment. It could be said that the specific word linked to a specific seen object is constituted by a modulation of the pattern in the brain constituting the body-image;

D. In any single language there must be broad consistency between the words forming the language i.e. there must be compatibility and sufficient distinctness between the words, This need for coherence between words in a language stems again from the integral operation of the brain - words modify and demarcate each other;

E. On the other hand, the hypothesis does not lead to the view that there should be a universally appropriate word for every distinct object or referent:

- the perceived object varies to a considerable extent according to time and place

- and in a given context the distinctive feature of a particular object may be different from the distinctive feature in another context;

- there is a non-negligible variation in physical make-up and specifically in the organs of articulation from race to race, Even small physical variations in the organs of articulation (or general physical structure) will involve differences in the associated motor patterns in the brain and so differences in the articulatory sequences associated with particular brain patterns i.e. a different word will result;

- where for the reasons given words for particular percepts differ from one language to another, there will necessarily be changes in other words because the brain operates integrally and changes are required in other words to maintain the overall compatibility and distinctness of the aggregate of words in a particular language;

- so differences in vocabulary between languages result partly from a process of 'seeding' (as in crystallisation). The patterns in the brain and in the articulatory process associated with a number of elementary words go towards determining the complete character of a language and lead to extensive divergence between words in different languages for the same objects, perceptions, actions, feelings;

F. Nevertheless different languages have the same natural foundation in the relation between the brain pattern subserving the speaking of a word and the pattern associated with a particular object, or action. Words in other languages associated with particular objects, perceptions, actions, feelings can be readily associated with those objects &c by speakers of a different language The ability to learn a foreign language is a result of this shared natural foundation and the observed phenomena of cross-linguistic phonetic symbolism find their explanation in this way;

G. Where resemblances are observed between vocabulary in different languages, these are not necessarily an indication that the languages are related by descent or have a similar vocabulary as a result of diffusion. The resemblances may be the result of a natural appearance of similar words for similar perceptions by physically similar people in similar circumstances;

H. Between what are taken to be related languages, the picture of lexical relationships is a complex one. Divergences can develop from an originally common language as a result of the use of a few different words which lead to widespread compensatory modifications of the individual descendant languages in order to maintain the coherence and compatibility of the total vocabulary. Scattered or chance modifications at one or other point in the language, particularly for common objects or actions, will tend to lead to systematically related changes elsewhere in the language i.e. changes of the type observed in the Indo-European sound-shifts. Systematic differences between descendant languages are another manifestation of the integral functioning of the brain in the production of language;

I. The learning of words by a child is a natural process which is 'seeded' by the form of the elementary words that it first hears. The child is naturally programmed to develop a consistent language of some kind but the environment determines which language this should be. The child is essentially given clues as to the kind of language it should develop and the language is not simply learnt but unfolds in the child;

J. The final step in the hypothesis is that, if each word is the product of a specific patterning of the brain and has a relation to the whole bodily set reflected in the brain, then the speaking of the word involves a reflection of the specific patterning in every aspect of the state and activity of the body at the time the word is spoken. Facial expression, muscular tensions, bodily movement accompanying a word - and specifically the characteristic bodily movement observed as gesture - express the patterning inherent in the word;

- So natural gesture accompanying a word is homologous with the word spoken. The movement of the hands and arms in natural gesture is coherent with and derives from the patterning of the brain which constitutes the word and which is transformed into speech-sound through gesture of the vocal organs;

- Observation of natural gesture is thus a route for exploration of the brain patterning associated with individual words and for examination of the inter-relation of patterning between words which resemble each other in meaning or in sound;

- The speech-sound and the natural gesture are parallel expressions of a central brain-pattern associated with the meaning of a word.

Though often natural gesture is abbreviated or vestigial, in certain circumstances e.g. in emotional tension fully-developed gestures are observed and their structure can be related to the structure of individual words.

The following is a fuller explanation of some of the terms used above:

(i) SPECIFIC INVARIANT PATTERN OF BRAIN ORGANISATION

Every time we speak a word, the sound of the word is produced by a particular configuration and change over a specific time-interval of the articulatory organs -that is movements of the tongue, the vocal cords, the lips, the jaw, the chest muscles controlling expiration. As with any other voluntary bodily movement, the movements of the articulatory apparatus - the specification of the initial and final states - are controlled and precisely determined by nervous impulses transmitted to the muscles involved. This pattern of nervous impulses has to be finely co-ordinated in space and time; in principle the character of the movement of the muscles serving the articulatory apparatus is precisely specifiable in terms of the time at which each impulse is delivered, the duration of stimulation of the muscle, the spatial distribution of impulses between the various muscles, the temporal ordering of the nervous impulses between the muscles serving different articulatory organs. This pattern of nervous impulses, which produces a pattern of muscular change in the articulatory apparatus, has its origin in the central nervous system. To produce one specific sound (which requires an appropriate muscular pattern) there has to be a distinct central nervous pattern from that required to produce another specific sound, i.e. each speech-sound is associated with a prior distinct patterning in the central nervous system.

Though the neurologists are not yet in a position to specify the precise nature of the patterning of nervous impulses within the central nervous system for particular speech-sounds, the idea of such patterning as the basis for speech is reasonably well accepted. Lenneberg points out the essential identity of central nervous control of voluntary movement and speech-movements: "Motor co-ordination is driven by a rigid unalterable cycle of neurophysiological events inherent in the central nervous system. Language as any other type of behaviour is seen as a manifestation of intricate physiological processes .... The neuromuscular correlates of speech sounds are muscular contraction patterns among one and the same set of muscles .... Throughout the duration of individual speech sounds, muscles must be activated (or de-activated) at such rapid succession that a neuronal firing order must be assumed that functions with an accuracy of milliseconds. This can only be accomplished by automatisms consisting of intricate time-patterns .... It would be presumptuous to try to explain the nature of the innate events that control the operation of language. We may however assume that mechanisms are involved such as (1) the modulation of firing characteristics of nerve cells; (2) the triggering of temporal patterns in neuronal chains; (3) the modulation of oscillatory characteristics of endogenous activities; and (4) the production of spreading of disturbances."

Similarly Teuber discussing the neurological basis of speech says "We cannot begin to understand the neural basis of speech unless we begin to comprehend how spontaneous and selective movements can be initiated in the central nervous system, how they can be confined to particular configurations of neurones, and how they can be stopped in order to permit the organisation, in time, of varying motor patterns that follow upon each other". Penfield speaks about the acquisition by children of what he calls word-formation units, the image of how to speak a word "really a pattern of the motor complex required to produce the word".

(ii) INTEGRAL OPERATION OF THE BRAIN

This concept is important to the presentation of the hypothesis. It is the basis for saying that there is a relation in the brain between the pattern controlling the speaking of a word and the pattern controlling a particular voluntary movement to which the word refers, for the assertion that in any given language there is an obligation for coherence and compatibility between the words forming the language, for the proposition that in closely related languages wide differences in vocabulary can originate in and spread from changes in a few elementary words (including changes on the pattern of Grimm's law) and finally and most important for the proposition that there is a close relation or identity between central patterns of nervous organisation associated with a particular word and the central pattern responsible for natural gesture accompanying a word.

At first sight, the idea of integral operation of the brain seems a natural one which is reflected in the unity of consciousness and the unity of behaviour of the individual. On the other hand, the concept of integral action must not be confused with the older idea of holism i.e. that the brain functions as an unanalysable whole with no part having distinct functions from any other part. There is no conflict between assuming that the brain operates integrally (in the same way as any large organisation might operate as a unity) and at the same time recognising that the brain has significant internal structure and specialisation of function. For example, the cerebellum in serving the total muscular co-ordination of the body performs a function which is different from those of the rest of the brain. There is the other misapprehension which also needs to be avoided that some functions such as speech are in some sense located only in a particular part of the brain viz: Broca's area and can be separated from the rest of brain function so that the brain would be a collection of distinct organs serving speech, perception, movement and so on.

The proposition here is that speech, perception, movement, hearing are not controlled by water-tight compartments of the brain but by areas of the brain which connect with each other, interact with each other and form part of a larger continuous organisation. On this view, it would be natural to assume that perception would affect speech, speech would affect movement, movement would affect perception and so on for other distinct functions.

Synaesthesia as observed is some demonstration of the likelihood of this view. So also are experiments which show interference between the process of hearing and the process of speaking (e.g. where speech is played back with a time-delay to the ear of the speaking subject). Brain in an article commenting on views of Head and Holmes discusses recognition of phonemes and words in terms of physiological phoneme-schemas and word-schemes (rather similar to the central pattern underlying words already discussed) and he comments: "A word-schema must possess links with the physiological bases of perception and thought".

Lenneberg provides a clear statement of the view here put forward: "In the brain ... there are no independent parts or autonomous accessories. In vertebrates, and probably many higher invertebrates, the entire brain is a functionally integrated system with constant, spontaneous and inherent activities that involve all healthy structure ... Any modification on the brain is a modification on the entire brain. Teuber more specifically says in referring to the question of cross-modal and supramodal learning: "If it were generally true that the different senses are hermetically sealed off against each other, it would be extremely difficult to understand the function of those large clusters of cortical and sub-cortical neurones which seem to respond to more than one kind of sensory input ... It would seem essential that there be some central mechanism for transcending the division between the different senses, for identifying an object felt with an object seen, and both with the object we can name; there should be some form of cross-modal processing resulting in supra-modal, rather than sensory categories, extracted from or imposed upon experience.

Finally Penfield commented on the relation between speech and gesture: "With severe aphasia at any time, the individual's ability to convey meaning by gesture of head or hand is lost. He may use the muscles of neck and hand for other purposes but he cannot nod assent in place of the word 'yes' nor shake his head in place of the lost word 'no'. It must be assumed then that the characteristic gestures employed by anyone to convey meaning while speaking or instead of speaking, must have neuronal units in the speech mechanism. This applies to meaningful gestures as it does to writing".

(iii) BODY-IMAGE

The body-image is introduced in the presentation of the hypothesis as a means of explaining the relation between the form of a word referring to a particular visual percept and the meaning of the word, the referent. The body-image is a stable internal perception of the form and state of the body. It contains optic components as well as somaesthetic: that is, we have a pattern of brain organisation related to the pattern of muscular tension, the postural system of the body, awareness of position and weight as well as the figure of the body. The body-image, and more specifically the postural image of the body, operates as the basis of reference by which we determine the position of individual limbs, the extent and direction of particular movements. There are closely related compensatory systems in the functioning of vision which allow perception of movement from a stable point of view.

Schilder explains: "The image of the human body means the picture of our own body which we form in our mind, that is to say the way in which the body appears to ourselves - there is the immediate experience that there is a unity of the body. We call it a schema of our body or bodily schema or, following Head who emphasises the importance of the knowledge of the position of the body, postural model of the body. The body schema is the tri-dimensional image everybody has about himself .... The postural model of the body, the knowledge of the limbs and of their relation to each other, is necessary for the start of every movement ... The perception of our own body is not very different from the perceptions of any other outside objects. Localisation is built up by optic and kinaesthetic impressions, by bringing the single impression into connection with the postural model of the body."

This explains in a preliminary way what is meant by body-image. The hypothesis goes on to suggest that particular acts of visual perception are related to the body-image in a way parallel to that in which particular movements of the body are related to the postural image. And there is the further idea implicit that since the postural image is a part of the total body-image, modification of the optic element by visual perception can be accompanied by modification of the postural state i.e. movement of the body in gesture or otherwise. This was put more clearly by Lashley who observed: "close interrelationships of visual and kinaesthetic space suggest that the perceptual processes in vision may be far more dependent upon interaction with the postural-kinaesthetic system than is ordinarily assumed ... It is not impossible that all the spatial characteristics of vision .. are dependent upon integration with the postural system. It may be that we shall have to seek the source of visual percepts in the integration of these two systems ... "

Unavoidably, the expression of what is involved becomes more difficult because one is dealing with the most complex part of brain function - the recognition of pattern and the inter-relationship of patterning drawn from different sources. Lashley emphasised the relation between the stimulus, say an event of visual perception, and the existing state: "Input is never into a quiescent or static system but always into a system which is already excited and organised. In the intact organism, behaviour is the result of interaction of this background of excitation with input from the designated stimulus .... The spatial properties of the visual stimulus are translated by integration at a series of levels into modifications of the general pattern of postural tonus".

In rather vaguer and less precisely neurological terms, Werner and Kaplan applied this sort of approach to the relation between words and their meanings:

"Objects are given form, structure and meaning through inner-dynamic schematising activity which shapes and inter-twines the sensory, postural, affective and imaginal components of the organismic state ... The organismic view that the translations of non-sonic properties into sound-patterns are based on similarities deriving from the primordial unity of the senses ... From this global unity of sound and bodily movement ... the development of the specific modalities ... of motor and of vocal reference occurs ... Phonic properties may 'synaesthetically' represent shapes ... (On looking at a particular object) the material phonemically-unique sequence (contained in the word referring to the object) is articulated into a production whose expressive features parallel those ingredient in the percept".

It may be helpful in the light of what is said above on this complex subject to set out more simply the chain of ideas which is involved in the hypothesis under examination:

(1) the visual perception of a particular object produces a particular pattern in the brain corresponding to it;

(2) the brain at the time when this pattern is produced is not in a neutral state but in a dynamic state which results from the representation in the brain of the body-image, including the postural image i.e. in mental terms our current awareness of our own state;

(3) the pattern produced in the brain by the perception of the particular object interacts with the patterning already existing in the brain i.e. it interacts with the body-image and the postural image;

(4) changes in the body-image and the postural image produce changes in bodily behaviour. The activation of the articulatory organs to produce a particular word is one consequence of the modification of the body-image and the postural image (which is the expression of total muscular set);

(5) but expression in observable change in behaviour of the change in the brain-patterning following perception of a particular object is not limited to the making of a sound - a word. It can also be observed in changes in the positions of the limbs (or in facial expression) and in gesture;

(6) So - seeing a particular object can produce in parallel a word uniquely referring to the object and a gesture associated with the object and with the word.

(iv) VARIATION OF THE PERCEIVED OBJECT

If there was a direct and invariable relation between the perception of an object and the word formed to refer to the object, then one would expect that in all languages the same words would be used for the same objects. But this is obviously not the case. There are a number of reasons why there should be different words in different languages (as outlined in the presentation of the hypothesis) but one important reason is that between languages, between nations and between cultures, the words which are usually translated to mean the same object do not in fact refer to the same object -or a different feature of an object is chosen as distinctive in one culture from the distinctive feature chosen in another culture.

Taking a familiar object such as a CHAIR, in every dictionary of foreign languages into English, one will find foreign words which are translated by the English CHAIR. But one must examine the word CHAIR as referring not to an object defined by its function i.e. something to sit on, but an object which is taken in by a specific act of perception. If there is a relation between the percept CHAIR and the word CHAIR as Werner and Kaplan argue, then the form of the word depends on the form of the percept. If one collected together from different countries the perceived objects for which words translated by CHAIR in English are used, there obviously would be extremely wide differences in the physical structure and appearance. In some cultures, a Chair is an exotic piece of furniture - or a ceremonial object. The Chair of STP. Peter is not necessarily much like the common English idea of a chair. A chair to a Japanese is a strange European structure.

If the percepts, which are classified by the dictionaries as chairs, vary widely in form then the pattern in the brain on perceiving them will also vary widely - and it would not be at all surprising if in different languages the words associated with the percept CHAIR should also vary widely.

This may seem to be labouring the obvious - but it is a temptation to say that it should be easy to construct a list of words which have a clear and uniform reference in different languages - natural objects - trees, flowers, animals, the wind and the sea. But trees for example obviously differ widely between localities. The trees familiar in Australia or America are not necessarily at all in shape or appearance like trees familiar in Europe. The word TREE in Europe refers to a collection of objects which resemble each other within certain limits - the concept has so to say a centre of gravity, an average quality of treeness located at a particular point. In other countries and other continents, the average shape and appearance of trees is likely to be more or less far removed from the average European concept of tree and accordingly foreign words for TREE will differ from English or European words if the form of the word varies in parallel with the form of the tree.

This kind of analysis can be extended indefinitely. The problem becomes to find concepts or percepts which remain constant or nearly constant between widely separated countries and races. It is only on the basis of such constant concepts and percepts that one can usefully examine what sort of relationship exists between the objects referred to and the structure of the sounds - of the words - used to refer to them in different languages.

(v) NON-NEGLIGIBLE VARIATION IN PHYSICAL MAKE-UP

The sounds which go to form words are produced by changes in the shape and position of the articulatory organs and interaction of the specific pattern of the articulatory organs with the flow of air moving from the lungs as a result of changes in the chest-muscles controlling expiration. It is obvious that differences in physical make-up between those speaking the same language can result in very wide differences in the character of speech. At the most obvious, the difference between men's voices and women's voices is the result of differences in the dimensions of the vocal cords; but there are other differences beyond this. Malformations of the articulatory organs can produce generalised differences in speech, lisping, thick speech and so on It is more difficult for some people to pronounce particular sounds than for speakers of the language generally. There may be physiological differences underlying dialectal differences.

Now if one assumes that the brain-pattern for the speech-sequence - the word - used to refer to a particular percept or action is related to the particular act of perception or particular other action, the brain pattern can only take effect through the structure of the articulatory organs as they exist in the individual. A person whose tongue is paralysed will produce a different sound from one who speaks normally, even though in both cases the idea of the word to be produced is the same. So if between races there are significant differences in the structure or shape of the articulatory organs, one would expect this to lead to differences in the sound sequences produced - in the words following on from similar brain-patterns related to the meanings of the words to be used. Beyond this it is conceivable that in addition to variations in physical conformation of the speech-organs between races, there may also be genetically-determined differences in the central nervous structures - the accepted observation is that even between persons of the same race, each brain differs substantially from the next in fine structure and functional organisation. It would not be surprising if between widely separated races, there were systematic differences in the brain structures subserving speech and other functions.

However, for the time being, the question is how far there are significant differences between races in the organs involved in speech. These organs are the tongue, the lips, the teeth, the palate or roof of the mouth, the nose, the glottis, the vocal cords, the lungs and chest muscles. It is obvious that there are wide differences between races in facial conformation. So the lips of Africans have a different conformation from those of Europeans or Chinese; the noses of Chinese and Japanese are different from those of Africans or Europeans. The bodily characteristics of the different races vary widely and it would not be surprising, indeed it is to be expected, that there should be differences in the internal conformation of the speech organs, in the shape of the palate, in the dimensions of the tongue, possibly even in the detailed musculature of the different organs. At the minimum, differences of these kinds would be likely to lead to preferences for use of a different range of sounds by the different races - but the hypothesis presented goes further than this. It suggests that even if the process of perception is the same between the different races, differences in the structure and functioning of the speech-organs between races would mean that the same central brain pattern associated with a percept would have a different translation in terms of the articulatory sequence flowing from it i.e. quite naturally and inevitably different races would tend to produce different words for the same objects

To make somewhat more concrete the ideas here, it may be useful to recall that speech is produced by a precise and complex control of sequences and extents of muscular contractions altering the shape and position of the vocal organs. Lenneberg says: "We cannot state exactly the number of muscles that are necessary for speech and that are active during speech. But if we consider that ordinarily the muscles of the thoracic and abdominal walls, the neck and face, the larynx, the pharynx, and the oral cavity are all properly co-ordinated during the act of speaking, it becomes obvious that over 100 muscles must be controlled centrally. ... the rate at which individual muscular events occur (contraction or relaxation) (throughout the speech apparatus) is of an order of magnitude of several hundred events every second".

Given this order of complexity in the relation between central nervous control and the articulatory system, it would be surprising if there were not systematic racial genetic differences in the articulatory system of an importance at least equal to the obvious differences in the appearance of the different races. Very small changes in mode of articulation even in a single language are required to change one sound into another e.g. P into B and differences between races in articulatory organisation could be expected to produce substantial differences in the speech-sounds used.

Brosnahan studied the role of genetic factors in the development of sound systems and this led him to consider racial differences in the vocal apparatus. He recalled that it is traditional in the textbooks to emphasise the basic similarity in the vocal apparatuses of all human beings: "In the stressing of this fact of similarity, however, the no less observable fact of differences in the vocal apparatuses has tended to be neglected There is basic similarity but by no means identity of form and structure". He quotes from the literature on differences between races. So differences in lip shape also include differences in the distinct muscles available for controlling lip-shape; one muscle which is found in 75 to 80% of Europeans and 80 to 100% of Chinese and Malays is not found at all in 80% of Australian aborigines or 40% of Africans.

The differences in the tongue between races are even more striking. One study found that the average length of tongue varied widely between Japanese and Negroes; the extreme range found was from 55 millimetres in Japanese to 123 millimetres in Negroes, i.e. the tongues of the negroes were more than twice as long as those of the Japanese. Such a great difference, given the importance of the tongue in articulation, is bound to lead to considerable differences in the articulatory processes of the two races, and so one would not be surprised to find great differences between the words used for similar objects.

In relation to the larynx, large differences are found as regards the possession of certain muscles - 86% of Germans have them, 80% of Japanese do not The form of the crico-thyroid muscle which influences the tension of the vocal cords shows large racial differences; the form of the muscle which is possessed by 82% of Hottentots is not found at all amongst Europeans and only amongst 8% of Japanese. Brosnahan goes on to refer to the neurological evidence for individual genetic variation of structures in the cortex from which the muscular activities in speech appear to be controlled -and one might draw the further deduction that if differences with a genetic base between individuals are discernible, then it is very likely that there should also be broad patterns of difference between races in the brain structures controlling articulation - from which again differences between languages would flow.

The conclusion reached is that to explain differences between languages, there are ample physiological sources of difference which, coupled with the differences between the percepts named in different languages, can go far to explain the existence of distinct languages without relying on the traditional explanation of difference, that is the development of different conventions of language in different societies. In brief, there is ample natural variation available to explain the multiplicity of languages