A new biological grounding for cognition and language

Abstract.

This paper explores the possibility that there is a small, tight-knit set of primitive factors that have a key role in the behaviour of biological systems in terms of their interaction with the environment. Collectively they can be regarded as a code that governs behaviour. Individually, they manifest in human beings as simple abstract facets of experience but they have not previously been recognised as the functional set that is proposed here. The factors are of two types: a set of parameters of the physical topology of entities in the world at a highly simplified level and a set of normative principles used by organisms to evaluate what they encounter in terms of their self-interest. The two types combine to potentiate the interaction of organisms and the environment. I call them ground factors. They were originally discovered in language where there is strong empirical evidence of their presence and necessary function as constituents of the meaning of words. There is direct evidence of their presence in human experience but their biological function remains speculative. The paper argues that there is a universal principle that determines the relationship between aspects of the external world and their representation within biological systems and that is responsible for interaction. This principle is similarity. For human beings similarity is between external entities and neural configurations across perceptual, affective and motor regions of the brain with which the ground factors are associated.  The way in which this system operates has evolved in several stages. If the nature and function of the ground factors is confirmed their place in our cognition and language, and in biological systems in general, may prove to have significant explanatory power in a number of fields.

Key words: biological systems; interaction with the environment; ground factors; word meaning; semantic primitives; gestalts; articulation; the principle of similarity.

1. Introduction.

            This paper explores the possibility that there is a set of fundamental factors, not previously identified, that are prerequisites for the existence and operation of organisms because they govern their interaction with the environment. These factors operate in a coordinated manner that is necessary for the viability of organisms. For organisms other than human beings they are necessary for their various forms of interaction with the environment. For human beings they are essential, as well, for cognition, categorization, memory and concept and word meaning. I have called them ground factors (zoemes in the other papers). The factors constitute, I propose, a structure on which the phylogenetic specifics of organisms’ detection and evaluation of environmental entities and conditions are founded. They are of two types, one sensory, for the detection of objects, and the other imparting normative value as a trigger for action. They were first identified in language as highly recurrent facets of word meaning that are iconically represented in the manner of the articulation of sounds to a degree that is highly significant statistically in at least some languages. The facets thus revealed (the ground factors) are arguably universal constituents of the meaning of words––semantic primitives of a new kind. More fundamentally they appear to constitute a necessary structure of human cognition and thought.

            The early parts of this paper explore the nature of the factors and their functions in primitive biological systems. Later I describe their function in human cognition, thought and language. This includes an endorsement of the idea (with a new rationale) that the basis of these faculties is the construction of similarities between brain and somatic states and the world. The factors inject new credibility into this position. I also derive a new model of word meaning in which the ground factors play a central part. The place of the factors in human cognition provides an opportunity for reviewing a number of aspects of linguistics, cognitive science and psychology, but this is not attempted in this paper.

            The ground factors are, I propose, operative in the life of any organism. For us they are dimensions of psychological space that are represented by simple, abstract, basic, mundane constituents of cognition and experience such as the sense of the particularity of entities, of large/small size, surface, activity, a favourable or unfavourable aspect, and of uncertainty. At a structural level they underlie every moment of our experience. They are necessary parameters and principles that occur in the interaction of organisms and the environment that have been stable throughout evolutionary time although they have become somewhat more complex in their operation in the latest stage of evolution. I call them ground factors because they are fundamental to the viability of biological systems and they provide a new grounding for human cognition. I have identified twenty-odd ground factors but this may not be the full set. There is nothing surprising in them individually, but as a set they seem to have an extraordinarily important function. I interpret them as a structural code or ‘alphabet’ of the process of interaction of living things and their environment. The factors are structural in the sense that they are the necessary framework for specific phylogenetic sensory and evaluative particulars.

            The ground factors first came to light in language in a way I will describe in section 6.3, but it soon became evident that they are a biological, rather than an intrinsically linguistic phenomenon. That is one reason why the paper begins their explanation in biology rather than in linguistics. The other reason is that there is a substantial stumbling block to the acceptance of their role in language that might be removed by understanding their role in biology first. The main aim of the paper is to establish the credibility of the ground factors in their operation alongside phenotypes, genetic systems and physical environments that are conducive to life, as co-essential prerequisites for the viability of biological systems. By the end of the paper I hope the nature and function of the factors will have strong credibility.

            The approach I have adopted in constructing the paper is to use several examples to trace the operation of the ground factors through a period of evolution from the Mesozoic era (although I believe they operated from the dawn of life), and then to the apes, to pre-linguistic hominids, and finally to Homo sapiens. As I work through these evolutionary stages the full significance of the factors should become clearer. Finally I will illustrate the ways the ground factors occur in language where they can be observed in action in the meanings of words. There is much that is speculative in the paper but throughout I am following the thread of the empirical evidence of the factors in language in order to try to work out their implications at several points in the evolution of life.

2. The ground factors.

2.1. What are the ground factors?

            But what exactly are these entities? They can be identified with a group of humdrum, abstract, very basic features of human experience and cognition. Each ground factor specifies a basic, universal, abstract facet of experience such as particularity, largeness/smallness, manifestness, uncertainty. As a group they seem to constitute the structure of human experience in that specific experiences are reliant on the framework of the factors. Without the ability to identify particularity we could not tell figure from ground. Our sensory experience would be an impossible blur. Without the ability to distinguish size our perception of the world would be hopelessly confused. If we were unable to be in a state of uncertainty or of satisfaction, reflecting our perception of states of the world, there would be a fatal lacuna in our make-up. The precise number of ground factors is uncertain (partly because there is some overlapping) but Table 1 is a good starting point. Despite possible gaps I believe the set of factors has a remarkable completeness and coherence as a structure of human experience that will become apparent as the paper progresses.

            The individual ground factors are very familiar in themselves as abstract features of quotidian experience, types of innumerable classes of tokens. They may seem too mundane to sustain a theory such as I am describing. My claim for their importance may seem as true but as pointless as asserting that oxygen, water and certain chemical elements are indispensible for life. The functions that are attributed to the factors suggest otherwise but in describing this phenomenon I am walking a tightrope with the risk of tumbling into triviality or the possible reward of finding some far-reaching explanations. In the meantime I must plead for some suspension of disbelief through the early parts of the paper. In due course the factors should speak for themselves.

  2.2 A thought experiment.

            I will conduct a brief thought experiment in order to develop an initial appreciation of the factors as they operate across biology. I will try to imagine the most basic set of factors that would be necessary to enable simple mono-cellular organisms such as amoebas, as well as others including ourselves, to live in interaction with their ecological systems.

            We will consider initially some factors that enable organisms to identify conditions and entities that impinge on them directly. These are parameters of the physical world that are identified by sensation and perception. Organisms need to be able to register such parameters in order to respond in ways that will secure their survival. Response is governed by innate activators or triggers for which the factors are structural prerequisites. The first factor is the material nature of entities that they encounter. An organism that bumps into something solid responds in some way because it is genetically programmed to do so, but its response is partly determined by the structural fact of the materiality of the object. The second factor is surface. Whether it is the surface on which they move about, or surfaces they come in contact with, this factor is primal for all organisms and is closely associated with the first. Next is particularity. Many entities that organisms encounter have a particulate form. They need to be able to distinguish figure from ground, providing the first step in identifying salient entities. Particulars come in large/small sizes relative to organisms’ own scale. Distinct physical configurations of particular entities may be pertinent to an organism. They may be characterised simply as particulars of a specific size, or as having a specific topology, perhaps a compressed, contracted form, or a convex surface (that may imply a biological entity), or as an extended, elongated form, or one that is characterised by physical intensity, that is angular or has prominent parts, or significant unusual movement. It is implicit that encountered entities are displayed or manifested to organisms in some way, generally in situations of openness or spatiality. Finally, tactility may distinguish other salient surface characteristics of encountered entities.

            The factors described above are accompanied by a second group that has an entirely different character that complements the first. This group is equally necessary for the viability of organisms. These are not parameters of the physical world. They are genetically attuned normative principles according to which organisms ‘evaluate’ entities and conditions in terms of their interests and survival and how they should respond. They enable organisms to ‘judge’ whether a sensed entity is favourable or unfavourable in terms of the organism’s welfare. They work, in combination with factors of the first type and specific inherited triggers in activating organisms’ responses to entities and conditions they meet. This last factor may be supplemented by other evaluative, qualitative factors such as recognition of the availability or abundance of something favourable. Another factor of an allied nature is the recognition of what I have called possession/self-interest, the ‘knowledge’ that an entity or conditions or a situation is well suited to the organism’s interests in some way, possibly in terms of ingestion or as a safe location. A further necessary evaluative factor is an ability to ascribe a degree of strength or value (fullness/emptiness) to any factor or combination of factors. In addition, any organism must encounter objects and situations that are not clear-cut, situations in which the factor uncertainty prevails, or a distinct sense of displacement from normal conditions or discomfort that requires some form of remedial response. In other circumstances organisms may experience what I have called a somatic affect, a specific form of somatic disposition that stipulates a particular response such as something analogous to fear or security.

            Of course we do not attribute affects or emotions to organisms or even to animals, but the factors of this second type have an affect-like result. For most organisms and animals these factors are genetically-determined aspects of response activators of triggers, but they operate as if they have experienced an affect. The basis of the response is a genetically (or behaviourally) based ‘knowledge’ that they should act in a certain way. This set of factors is implicated in determining the actions of organisms, the specifics of their interaction with the world around them. They provide the framework within which organisms can respond to what they encounter in ways conducive to survival.

            Table 1: The ground factors.

.......1.

Materiality
Particularity
Surface
Bodily roundness
Contraction
Extension
Display
Openness
Action
Largeness / smallness
Physical intensity
Intensity of energy

.......2.

Favourable / unfavourable
Somatic affect
Tactility
Abundance, generosity
Possession, self-interest
Uncertainty
Displacement
Fullness / emptiness
Existentiality
Community life world
Physical life-world

 

2.3 The ground factors as the structure of organismic interaction.

            Table 1 contains the set of ground factors that I have identified. As I have noted there is nothing unusual or surprising about these factors individually or as a group. What is of interest is the way they arguably provide, as a set, a single unchanging framework for the interaction of living things and their immediate environments in the interest of their survival. The ground factors are not made redundant by either their blatant obviousness or the specific innate activators of organism’s response that are associated with them. I have described them elsewhere as categories in the philosophical sense of that term. They are the ultimate classes, the highest genera of the prerequisites of interaction of organisms and their environments. Like most sets of categories, they are more notable for their architectural, than for their bricks-and-mortar character. The factors can be regarded as the elements of a bauplan for organisms’ behaviour.

2.4 Ground factors, gestalts and affectivities.

            The factors were derived from language in a way that I will outline later (when I will also explain the basis of the last three factors in the second column that appear to have no readily apparent rationale at this point). As I have noted, the ground factors and combinations of them cannot determine either identification or response on their own. They are the structure of tokens of perception and behaviour but they require a further element in order to be effective. This is the genetically prescribed (or learned) specific sensory registration that activates response (e.g. eat, mate, avoid, maintain the status quo). I call this the gestaltic form of percepts and sensations, or a gestalt. This is the actual, immediate form that occurs in sensation or perception in any modality. Gestalts are manifest (the building); the factors are implicit and concealed (the framework.) The gestalts are integral to any interaction of organism and environment. For a bacterium in a liquid medium, the gestaltic activator may be the detected specific concentration of a crucial chemical compound that determines the direction in which it will swim so as to remain in optimal conditions, for a toad it may be the position, size, shape and speed of a moving object that triggers its capture with its tongue, for a dog it may be a specific odour coming from a particular direction that determines where it will look for a potential meal. For us, this further element is, in part, the phenomenology that is the very fabric of our sensory consciousness.

            But gestalt, as I am using the term, differs in one important way from the purely sensory gestalts of psychology. These gestalts are not purely sensory. Affects or affect-like principles are essential features of them. Outside the human context these are, of course, ‘as if’ affects. No gestalt is without affectivity although it may simply be the affect lack of affect meaning take no interest, ignore in the case of perceived entities that are of no consequence to an organism. This accounts for most of what comes to an organism’s perceptual attention. It is the ‘as if’ affective content that determines the organism’s response or non-response to sensory stimuli.

            In brief, I propose that the full set of ground factors provides a structure on which biology operates from the behavioural perspective. They are the necessary, structural elements of situations in which interaction occurs. Stated more formally, the ground factors are the dimensions of the primal space of biological interaction. They can be viewed as taking the form of a matrix in which their two kinds provide the coordinates that are the framework for the phylogenetic constituents of interaction. The first set of factors is determined simply by the physical nature of the world at the level at which biology functions. The second set consists of ineluctable normative principles of biology that form the basis of any responsive mechanism. Any phylogenetic trigger must occupy one or more positions on the matrix. For example, for a bacterium sensing a specific concentration of a chemical compound that it needs to survive, the matrix positions physical intensity, manifestness, favourableness, abundance and self-interest would be operative. The appropriate selection of nodes, as phylogenetically determined, is the trigger for a specific action.

            It may still be difficult to avoid the belief that sense data alone are responsible for organism’s responses. The critical point is that there are two aspects of a response activator: an external source of energy or data that impinges on a sensory receptor and an aptitude to respond to the right datum. Both can operate only within the ground factor matrix by definition because the factors are inescapable parameters and principles of biology. Without this matrix life would be an impossible enterprise.

            In Table 2, which is a rearrangement of Table 1, I have given names to the two types, the alpha and beta factors. This table also incorporates a key feature of the ground factors––that each has a scalar or polar character that is not apparent from Table 1. This feature occurs in several factors in the earlier table. Here several factor have been linked on account of their polarity (e.g. surface and particularity) but it is evident that all the factors must be inherently polar. (This applies more clearly in language as will be demonstrated later). Thus display, manifestness incorporates concealment and an intermediate spectrum of clarity and obscurity.  Each factor incorporates its opposite and the graded territory in between. Polarity is a natural integral feature of the factors. The last two have important roles that I will explain later. A case can be made that the factors in the tables are the only ones involved in the structure of biological interaction although it is difficult to be certain that this is the case. I hope the rest of the paper will build confidence in this set.         

            Table 2: The polar ground factors.

Alpha factors

Materiality / non-materiality
Surface / particularity
Bodily roundness / lifelessness
Compression / extension
Physical intensity / featurelessness
Largeness / smallness
Display /concealment
Openness /limitation
Intensity of energy / rest
Action / inertness

Beta factors

Favourable / unfavourable
Somatic affect / tactility
Abundance / paucity
Possession, self-interest /loss, threat
Uncertainty / certainty
Displacement / comfort
Fullness / emptiness
Existential value
Social / physical lifeworlds

            In the following four sections I will traverse a large passage of evolutionary time in discussing some examples to illustrate the function of the ground factors in biology and some ways this function has been supplemented in line with evolving cognitive systems.

3. What is it like to be a tuatara?

3.1 What is it like to be a bat?

            My first example of the operation of the ground factors is the New Zealand tuatara, an extremely primitive lizard that is often referred to as a ‘living fossil’. The tuatara was chosen as a representative of a form of life that has (apparently) little in common with us. The title of the section is borrowed from a famous 1974 philosophical essay, What is it like to be a bat? by the American philosopher Thomas Nagel. On the basis that bats are ‘a fundamentally alien form of life’ Nagel concluded that we cannot know what it is like. The experiences of a bat ‘have….a specific subjective character which is beyond our ability to conceive’. He went on to state that ‘we cannot expect ever to accommodate in our language a detailed description…..of bat phenomenology’…and he further noted that ‘reflection on what it is like to be a bat seems to lead us….to the conclusion that there are facts that do not consist in the truth of propositions expressible in human language’. Nagel was partly right, but we now have something he was not aware of––the ground factors––that might help us to understand to a degree what it is like to be such an alien creature. We can begin to know from the inside what it is like to be a bat––or a tuatara. I will demonstrate later that the new element was, in fact, just under Nagel’s nose in the words he was writing.                                                                                

3.2 The tuatara.

            The tuatara is one of a very small number of species that have survived with minimal change from the Mesozoic era, the dinosaur period, of some 200 million years ago. It belongs to the order Sphenodontia and is related to other lizards, snakes, crocodiles and birds. Its relatives in the Mesozoic included the marine pleurosaurs. Tuataras grow to about 8o centimeters in length. They are normally observed stock-still, absorbing the warmth of the sun. Their most notable physical features are a spiny crest along the back, a hard grey skin, a unique form of dentition, and a parietal (third) eye that is of unknown function in the forehead. (It is visible only in hatchlings.) They have no external ear but can detect low-frequency sounds. Their body temperature is lower than that of any other reptile and their metabolism is very slow. They feed mainly on invertebrates such as beetles, crickets and spiders. Their growth rate is the slowest of any reptiles and they have an average lifespan of 60 to 100 years. These creatures’ brains are a typical reptile brain, far more primitive than those of mammals. Tuataras reproduce by eggs laid at intervals of three to five years. The reptile survives naturally only on a number of offshore islands and enclosed sanctuaries onshore, having been wiped out by introduced predators on the main islands. As survivors of the Mesozoic, the tuatara has lived through global extinction events such as that which wiped out other many creatures of that era. In more recent eras they have survived continental drift, volcanic eruptions, seismic catastrophes, tsunamis and drastic climate changes, but they would have been made extinct by introduced rats in the absence of off-shore island havens. The native Maoris of New Zealand regarded the tuatara as representatives of Whiro, a deity that was the personification of evil, darkness and death. Their presence was regarded as indicative of a zone of possible grave consequences. Maori women were known to have a tuatara tattooed on their lower bellies.

3.3 What it is like to be a tuatara.

            The tuatara is an alien species, perhaps more so than the bat. How can we possibly come to understand what-it-is-like to be one of them? Obviously we cannot know how spiders taste to them or even, or how nearby objects appear to them, or how they perceive human beings. We cannot know from the inside the nature of their experience of the world without anything like our conscious awareness of what is experienced and the experiencing self. In other words, we do not know what sheer biological interaction feels like for them. But I propose that the ground factors that govern organisms’ interaction with their surroundings provide an approximation of what-it-is-like. The ground factors can help us to gain a gleam of insight into tuatara being. The reason for this is that we share them with tuatara and other creatures and organisms. They have the same central role in the existence of these primitive reptiles as in human experience.

            Tuataras are often seen sitting on a rock soaking up the warmth of the sun and the heat of the rock as a means of conserving their metabolic batteries. In such situations a tuatara is responding to the materiality of the rock, the surface that it rests on, the openness or spatiality of the sunny situation, the abundance of the sun’s warmth, a consequent somatic affect and its occupation or possession of this favourable position. These ground factors come into play in influencing the reptile’s maintenance of this situation. As the sun declines and loses warmth two new factors come into play, uncertainty relating to this situation and eventually displacement or discomfort.  This combination induces the reptile to get itself into low gear and move (action), perhaps to fossick for food among pebbles and ground litter. There it recognises the particularity of pebbles and the extendedness of twigs, things that have no vital interest for it apart from the possibility of their concealing something edible, particularities with specific value. In recognising such things a range of ground factors come into play. These include the manifestness of the discovered beetle or spider as revealed to sight or hearing, its particularity that may be distinguished by the bodily roundness of a beetle or the physical intensity or angularity of the shape of a spider, the intensity of energy in the attempt to scuttle to safety. The capture and consumption of such a morsel is distinguished by tactility, abundance, possession, a somatic affect of taste and the favourable nature of the situation. After feeding, the tuatara may return to its burrow in a bank, a place marked by possession, ownership. The burrow, as it enters in characterised by contraction, compression of space, smallness, concealment and somatic affect in the form of a sense of security.

            These ground factors are intrinsic structural elements in the reptile’s interaction with its environment. The alpha factors enable it to locate objects around it while the normative beta factors provide the motivation for action through the values attached to objects of perception or situations either innately or through learning from previous experience. They are vital to its survival. But the situations I have described are abstract and incomplete. What is missing is the phenomenology of the tuatara’s sensory perception, the actual warmth of the sun, the physical shapes and spaces it sees, the sounds that penetrate to its concealed ear, the smells it constantly monitors as it fossicks. These are the sensory aspects of its gestaltic activators of response. The sensations and the affective factors are fundamental to its inherited make-up. They operate in distinct areas of the tuatara brain and neural and sensory systems that have been honed by millions of years of evolutionary process. The factors are necessary natural parameters and principles, instantiated in the reptilian brain, on which sensory gestalts are formed. As such the factors are the structure of the genetically driven interactive behavioural systems of the reptile.

3.4 ‘Knowing’ and similarity.

            This account is in conflict with the standard view of the basis of biological behaviour. The functioning of creatures such as the tuatara is normally described in terms of stimulus and response (S & R), with both being regarded as simple unstructured processes. A stimulus is received through one of the creature’s sensory systems and an innate autonomic response is triggered. I believe it is wrong to be so parsimonious. I propose that stimuli have a factorial structure as a matter of necessity. The factors are the structural means whereby organisms ‘know’ when and how to respond to events in their environments although the actual ‘when’ and ‘how’ of the organism’s action, is determined genetically. An organism must ‘know’, in some phylogenetically determined sense, the conditions or objects it encounters if it is to be able to respond in ways that favour its survival.

            Recall the act of reptilian perception described above. The tuatara sees a bug and captures and eats it. A perceptual gestalt triggers the reptile’s action. The ground factors seem to have no bearing on what happens. S & R appears to reign supreme. But this is not the case. The factors provide the very structure of the situation from the reptile’s perspective. It responds through an instinct that has been attuned by evolutionary process. Certain normative beta principles such as availability, positiveness and value underlie the evolutionary process that formed the response mechanism. They prevail at a structural level as necessary parameters of the tuatara-meets-bug situation. The reptile seems to respond solely to a perceptual simulacrum of the bug but the factors are the structure of the percept in the reptile’s autonomic neural system. One possibility is that the percept itself occurs in a purely factorial manner in the reptile’s rudimentary neural system in the form, perhaps of  particularity, small, convex, energetic (scuttling) configuration, rather than a well-formed simulacrum such as we experience when we see a bug. In this case the reptile’s vision and other senses are primarily alpha ground factor detectors. Either way, there is a similarity between what it perceives (registers neurally) and what is there within its line of sight. This is similarity in terms of both alpha factors’ correlation with external physical features and beta factors’ correlation of key principles that govern behaviour (such as a point on the favourable/unfavourable scale) and the object as viewed through the ‘tinted lens’ of the principle.

            Although similarity is redundant under the traditional S & R scenario, I believe the model that I am proposing reveals it in a very different light. Similarity, an isomorphism between a perceived external object and a configuration of the reptile’s neural system, is a basic mechanism that determines organism behaviour. Without similarity––no behaviour, no organism. But why should the mechanism for neural representation be similarity rather than an arbitrary abstract code of sensory signals that trigger response? Similarity, I propose, is simply a matter of necessity for biological systems. Sensory signals or data necessarily have an implicit structure that is derived from the topology of the word and its objects. The alpha factors provide a schematic way in which they can be represented. The beta factors have a similarly inescapable source in necessary principles that underlie interaction.

            The combination of these two types of factors is the mechanism that emerged at the dawn of life. In a monocellular organism the zoemic sensation and evaluation mechanism is represented in its genetic attunement to the world beyond its external membrane. This reflects the evolutionary history underlying the organism’s survival. The lessons of survival are packed into the genes. In the tuatara and other creatures with complex bodies and brains the mechanism is unpacked into body and brain in ways that facilitate more complex interactions with a high degree of reliability. The zoemes are represented in the architecture of the body-centred brain, in sensory, ‘affective’/evaluative and motor regions where zoemic configurations operate in a way that enables the creature to react in finely-tuned adaptive ways. In the case of a fossicking reptile that encounters a beetle, a neurally instantiated visual gestalt with an alpha factorial structure combines with the appropriate beta factors to effect registration and motor response through a neural network involving visual receptors and specific regions of the reptile brain. Here similarity occurs in the form of a visual gestalt that represents the beetle as seen by the reptile, that is, as coloured by the beta factors.

            This discussion of similarity is significant because I will propose that the principle of similarity between a neural representation and what it represents has a continuous role through the evolutionary story––even, controversially, into human thought and language. This is a much-debated philosophical issue.

3.5 Reptilian categorisation.

            Tuatara cognition of the world through the ground factors and sensory and perceptual data provides the basis for perceptual categorisation, the ability to distinguish, in variable sensory data, specific categories that are important for the survival of the species––a scuttling bug, a basking temperature range. Categorisation involves the matching of a percept with an appropriate exemplar or prototype in memory. Membership of categories is governed by similarity within specific bounds. Categorisation has an adaptive function in imposing some order on perception of the variable physical environment and specifying precise adaptive responses. I propose that a specific degree of commonality of appropriate ground factors across a range of variant gestalts is crucial to categorisation. The full potential value of categorisation lies well beyond the tuatara in the ability of some higher animals to accumulate additional knowledge about a categorised entity that is available inferentially when the entity is perceived. This emerges much further down the evolutionary track. Tuataras have negligible inferential knowledge of particular physical entities, but bare categorisation is crucial to their survival.

3.6 Biological meaning.

            Another way to express this situation is to say that the beetle percept has ‘meaning’ for the reptile. This meaning is clearly in a ‘proto’ form.  I have called it biological meaning because I hold that this kind of meaning is common across all biology. I define it as simply ‘knowing’ how to respond to perceptual stimuli in a way that is consistent with a creature’s survival. A bug scuttling over the ground in front of a tuatara has, I propose, a ‘meaning’ for the reptile, a very specific meaning that triggers its autonomic response. This is not a matter of S & R, but a consequence of a biologically meaningful situation that involves in its construction several ground factors such as those spelt out above, plus the appropriate gestalt. Tuatara meaning is primitive, but in the way it is constructed from components to form a proto-representation it has surprising parallels with the human meaning of concepts and words that, I will propose, are also configurations of ground factors and gestalts. (See section 6.4). Even tuatara meaning has the componential form that is often assumed for the meaning of human concepts and words. (They consist of components rather than being of a unitary, atomic form.)

            There is a further way in which reptile ‘meaning’ is similar to ours. For decades many linguists, psychologists and cognitive scientists have assumed that the meaning of our words and concepts exists in the form of highly abstract representational symbols that are manipulated by the mind or brain using syntactic rules in a similar fashion to the computation of a digital computer. The process is cerebral. The body and its sensori-motor system is considered to have no direct role in the formation of meaningful words and statements. This view has been challenged by a number of authorities in recent times. (For example Lakoff and Johnson (1999), Johnson, (2007), Damasio (1999), Barsalou (1999, 2007), Prinz (2002)). They hold that lexical meaning is essentially embodied. It does not occur as abstract symbols like those in a computer. It is not written in some highly abstract mental language. Meaning is intimately associated with the body. Consistent with this view, tuatara meaning is instantiated in firings in specific body-related regions of the reptilian brain, that is, in sensory, motor and affective regions.

3.7 Some conclusions.

            This discussion points to a number of conclusions that I will build on in subsequent sections.

            1. The ground factors that have been identified in human language appear to have a prominent and necessary function in the interaction of tuataras and their environment.

            2. The factors are the essential structure of the sensory images/gestalts associated with interaction. They constitute a matrix in which sensations can be phylogenetically formed or ontogenetically developed and employed.

            3. Sensory images and their associated ground factors are manifested in perceptual categorisation. Tuatara categorisation is limited to an adaptive broadening of its differential response to tokens of types of entities perceived.

            4.  Sensory perception in both the form of gestaltic representations of external entities and their structure of ground factor configurations operates on the principle of similarity of neural states (ground factor configurations and gestaltic form) within the organism and states of the world beyond its peripheries.

            5. For the tuatara, the factors are necessarily embodied. There is no alternative here, no mental level to invoke.

            6. The factors normally occur, not singly as sensory triggers are assumed to do in many cases, but in tight configurations that are integrated in gestalts. These are proto-representations that are componential rather than atomic entities.

            7. The ground factors can be interpreted as forming, in association with gestalts, units of biological meaning, with this term defined as knowing how to respond. Although a reptile does not possess anything like the cognition of human beings, it can be said to ‘know’ biologically in a way that is essential for its adaptive interaction with the environment.

            Summing these conclusions we can surmise that the matrix metaphor can be taken to mean that the tuatara brain literally operates as a kind of matrix that governs its interaction with its environment. Neurons across sensori-motor and affective regions of its brain are continuously being activated in ever-changing distributed combinations of the units of the factorial code in response to messages from its senses, its limbic region (the source of affectivities) and its motor system. This conclusion will resonate through the rest of the paper. The conclusions I draw from the discussion in this section are that biological meaning is essentially embodied, componential and founded on world-neural system similitude and that tuatarameaning’ may well be a paradigm of meaning in the human situation.

4. What is it like to be an ape?

The ape family emerged some 140 million years after the tuatara. This is reflected in the far greater complexity of their brains, neural systems, phenotypes and behaviours. With their partly arboreal habits, vulnerability to predators, dependence on food sources that were often unreliable, and their high degree of sociality, apes had developed a cognitive ability, intelligence and complexity of behaviour and perhaps the beginning of self-awareness, that are far in advance of those of reptiles. There is no need to replicate the earlier illustration of the operation of the ground factors in the behaviour of apes. As they are intelligent social creatures and because their bodily form and their behaviour have close parallels with ours, we need only a slight effort of imagination to appreciate that the ground factors in Table 2 have an almost palpable presence in the ape world. The size, architecture and power of the ape brain are crucial features in distinguishing ape from tuatara in terms of the ways they interact with the world. Apes have a far greater ability to categorise, to cognise full visual scenarios and to associate them with similar scenarios from the past using values associated with earlier experiences. These abilities have major consequences for the species’ survival in hostile environments. At this point I want to compare the way the ground factors operate in the ape and tuatara contexts.

            1. The ground factors have the same fundamental role in apes’ interaction with the world as they have for tuataras in that they operate as the structure of perception and the activation of response.

            2. As a consequence of its greater neural complexity ape phenomenology and cognition are vastly richer than those of tuataras. The sensory and cognitive constructions that are built on the ground factors are much more varied and numerous but the factors remain the necessary framework for any experience. For example, an ape seeing another reaching for and grasping food has a rich visual and affective experience that is built on a configuration of factors relating to the situation: action, materiality, display, openness (the visual setting), extension (the reaching), intensity of energy, contraction, tactility, possession (the act of grasping), self-interest, abundance and fullness of value (with these an ape empathetically recognises the other’s satisfaction and feels its own lack of it).

            3. Ape perceptual categorisation is structured by the ground factors and is supplemented by much fuller visual gestalts than those of the tuatara. It is also supplemented by a limited ability to accumulated information about individual categories that is useful in determining response. This much more advanced categorisation is a precursor to conceptualisation.

            4. Similarity continues to be the principle for the formation of perceptual categories and their representational states in the ape neural system. Sensory gestalts with their factorial structure are the basis of such similarity. The much more sophisticated perceptual system of the ape allows more veridical forms of similarity.

            5. As with the tuatara, the factors are intrinsically embodied in ape cognition. The alpha factors operate through visual, tactile, kinaesthetic and other senses and the beta factors work through deep affective value regions of the brain. The factorial matrix is ever-present as the basis of sensory gestalts and as motivators of action.

            6. The factors operate in configurations that are associated with phenomenological and affective gestalts as in the example above. An ape grooming another ape has an experience that can be interpreted as quite a complex combination of sociality, generosity, favourableness, fullness, tactility, intensity of energy, contraction (focussed attention) and others, together with a variety of sensory gestalts. Componentiality prevails in the perception of both single objects and complex scenes.

            7. The ground factors, whether occurring singly or in combination, operate as the structure of expressions of biological meaning, knowing how to respond to whatever the ape interacts with. The sight of a fellow ape reaching for food represents a complex meaning for the viewer. There is an act of empathic simulation, more than just the registration of what is happening. The biological meaning lies in the simulation. This passive kind of meaning does not exist with the tuatara. The ape has a more open, curious orientation to its surrounds, especially its fellow apes. The world has more meaning content. The observational kind of knowing seems to anticipate the meaning that human beings derive from passively observing what happens around them. But apes have active lives and biological meaning with its affective motivators is the driver of ape behaviour. In ape-world, as with the tuatara, the factors operate hand-in-glove with first-person perceptual experience.

            If there is ape thought of some kind it has an immediate character. Its subject is what is under perception and some limited inferences that can be drawn from it. The formation of concepts that can be utilized in the absence of their objects is still several million years down the evolutionary track. Apes cannot engage in prior cogitation and planning for situations that they are not already involved in. Nothing that might be termed a conscious mental activity prevails. Ape interaction is predominantly autonomic but the stage has been set for this situation to change.

            We can form a reasonable idea of what it is like to be an ape.

5. What was it like to be a pre-linguistic hominid?

         The next evolutionary stage I have selected in a smaller leap forward is a notional late hominid species that preceded Homo sapiens. It would probably have lived around 150–200,000 year ago, been close to sapiens in appearance and brain size, lived in small bands, engaged in group hunting of large animals, used fire and made fine stone tools. These were creatures with few natural defences living in a very hostile environment but they possessed some powerful new abilities. I assume they had near-human cognition but only minimal language, gestural or vocal, if any.

 5.1 Key cognitive attributes.

            The main new attribute I am assuming in pre-linguistic hominids (PLHs) was what Edelman and Tonini (2000) call primary consciousness which they saw as having four key components. The first is perceptual categorisation, the ability to recognise complex variable patterns, sometimes spanning several sensory modalities, that apes and possibly some other animals already possessed to some degree. The second is the ability to form concepts which are defined as ‘the ability to combine different perceptual categorisations related to a scene or an object and to construct a ‘universal’ reflecting the abstraction of some common features across a variety of such percepts’. (2000, p.104) A key feature of concepts is their decouplement from the physical presence of what they denote. The third requirement is memory that is sufficient to enable substantial recall of percepts, categorisations, and concepts. The last is value and value systems that ‘affect the dynamics of individual memories which, in turn, are established or not, depending on positive or negative value responses’ (p.105). E & T assert that ‘value, emotional responses and salience provide strong constraints on the establishment of conceptual, category-based memories’ (ibid.). They conclude that ‘with such means in place, an animal would be able to build a remembered present––a scene that adaptively links immediate and imagined contingencies to that animal’s previous history of value-driven behaviour’. This model of primary consciousness appears to be a useful assessment of the cognitive situation that would have prevailed in the PLH era.

            The emergence of primary consciousness was of monumental significance. In particular, conceptualisation would have been a highly adaptive tool for interacting with the environment. This ability to consolidate past and present experience opened new frontiers. Concepts connect with others in associative networks that can provide more effective ways for dealing with events. They also give access to inferences that are one of the motors of human thinking.

            PLH’s decoupled concepts would have incorporated altered perceptual gestalts, as our own do. These would seem to have been schematic greatly simplified images or simulations that retain their full factorial structure. As in previous eras, percepts as configured in the neural system in gestaltic and factorial form, would have been produced on the principle of similarity. The neural representation of sensations and percepts would have had a palpable resemblance to their objects on both counts. On the other hand, PLH concepts can be conceived as mental states that were formed in the same way and in the same regions of the brain. Major differences from perceptual gestalts would have been their decouplement from the presence of a perceived entity, the schematisation of gestaltic form compared with the rich phenomenality of percepts, and a consequent tendency for the ground factors to be more prominent. These decoupled amalgams of ground factors and sensory information must have had profound consequences by, for instance, assisting in exploring inferences, combining separate concepts, combining concepts with current percepts, and allowing beliefs, desires and feelings to influence and modify concept. With such abilities in operation thinking, albeit of a purely solipsistic kind, would have dawned. Pre-mankind had taken a large step forward.

5.2 Pre-linguistic thought.

            But what would have been the medium of this thought? In the absence of a spoken language to think in, does it imply a language of thought (LOT) along the lines of what Fodor  (1984) has asserted as operating in human beings?

            The ground factors could have provided the basic elements of the medium of thinking. If PLH had the concept GO-TO-THE-RIVER-TO-CATCH-FISH it would seem possible that the primary form of the thought could have arisen in two ways. One possibility, is a linear sequence of word-like formations composed of ground factors and gestalts, a language-like mode of thinking without a natural language. The other possibility is a single composite model-like construction formed from the relevant configurations of ground factors such as action; self-interest, abundance, positiveness, fullness (intention to find food); extension, surface, display, intensity of energy (the river); action, intensity of energy, contraction, possession (catching), bodily roundedness, extendedness, intensity of energy, abundance (fish) together with appropriate gestalts. This implies the construction of situational macro-gestalts with the ground factors perhaps also assuming a combinatorial function. Such a construction could be replicated in several variations to explore alternative courses of action.

            The emergence of particulate, simple concepts, however, would have involved a radical change in the form of meaning. I have characterised meaning up to this point in evolution as purely biological, knowing how to respond. However concepts are not normally immediate triggers of response but rather tokens in the calculation of whether or not and how to respond. They have a semantic rather than a purely biological form. They imply a self-contained meaning. At the same time they retain their biological features of being founded on the ground factors and ultimately being responsible for action. In this way the early concepts were precursors of language. For the PLH biological meaning, knowing how to respond, had become, through decoupling, a much more effective instrument in its interaction with the world. The power of biological meaning had continued to expand in parallel with more complex phenotypes with more flexible forms of behaviour. This language of thought, with concepts composed of ground factors and gestalts, is very different from Fodor’s. In particular its concept tokens are not innate but are based on the eons-old ineluctable ‘alphabetical’ structure of the innate factors.

            These conceptual processes continued to be embodied in the sense that concepts were distributed across body-related and affect- or value-related (alpha and beta) regions of the brain. PLH concepts were brain states, patterns of neuronal activity that were primed for action by the beta factors. Sensory, kinesthetic, motor, proprioceptive and affective regions of the brain participated in the formation of simple and complex, word- and model-like concepts. A neural matrix structured along ground factor lines was operative in PLH thought.

5.3 Similarity again.

            It may be useful to explore a little further the proposal that the concepts of PLH and human beings are founded on the similarity of their brain states and the entities in the world to which they referred. They were similar in two ways. First, the compressed sensory gestaltic aspect of concepts would have occurred in the same somatic regions of the brain as the manifestly resemblance-based percepts that they represented. They would accordingly have exhibited similarity in the same way as in percepts apart from their more economical form in concepts. The meaning of BISON was largely composed from the bison gestaltic image in an abstracted form. Second, the ground factor structure of the concepts, including the normative beta factors, is identical to that of the corresponding perceptual experiences. The concept BISON included the alpha factors particularity, largeness, display and bodily roundness and the beta abundance and positiveness reflecting its potential as a source of food. The concept GIVE with its rich beta content (possession, abundance/generosity, positiveness) has a neural isomorphism with the act as it is normally experienced. Ground factor configurations are characterised by similarity with their referents because they consist of structural elements of referents as experienced at first hand. The proportion of similarity attributable to these two constituents of concept meaning (sensory gestalt and factors) is highly variable for reasons that I will outline in section 6.5, but the similarity principle is very firmly based in conceptualisation.

            This principle has been widely rejected as the basis for human conceptualisation, but in this evolutionary context I believe its credibility is enhanced. One of the reasons for its rejection has been the failure to identify how brain states could resemble things in the world. How could they represent elephant or give or laugh? This failure stems in large part from an antipathy on the part of many authorities (most notably Pylyshyn) to the idea that images (or gestalts), as opposed to linguistic descriptions, form any part of concept or word meaning. This view is beginning to change under the influence of a number of thinkers as I have noted. I believe the ground factors and the model of concept meaning in which they have a key part adds strong support to this position. This issue will be discussed further in the next section.

5.4 The beginnings of language?

            In the absence of an ability to communicate by language, the power of this medium of conceptual thought to generate new patterns of thinking would have been limited. It is, however, difficult to conceive hominids living in cooperative groups and in possession of this faculty, not using it to communicate, at least to a limited degree. Two things point towards the possible emergence of language in gestural or vocal form fairly soon after the emergence of conceptual thought. First, although the thought of these hominids was a private affair their lives were not solipsistic. They lived in family groups and small clans. Much of their experience was shared. Percepts would have often been the subject of joint attention. As well as common sensory experiences, they would have shared to a large degree the affectivities and values that were associated with things, events and behaviour. These would have formed the basis of an empathy that is likely to have been stronger than that which occurs in apes. The presence of empathy based on the common possession of the ground factors and conceptual structures would seem likely to have generated a potent urge to communicate, especially in situations of joint endeavour or competition. Second, the experiences of all individuals were structured by the ground factors and similar concepts were formed from similar factorial configurations. The cognitive lives of PLH individuals ran along more or less parallel paths.

            In fact, PLH possessed something that may have been the natural bridge from this medium of thought to language. The ground factors have a feature that may have facilitated the opening foray into language, the magic moment of the first words or gestures that enabled hominids to begin to communicate. This is their embodiment, their psychosomatic expression, which invests them with a unique property: they are, to various degrees, eminently suitable for gestural representations of their embodied content. One somatic thing leads to another. Most of the factors, for example, particularity, surface, contraction, smallness, largeness, abundance, possession, uncertainty, intensity, display, lend themselves to being conveyed by iconic (imitative) hand or bodily gestural signs for concepts in which they are prominent. They can also be conveyed by iconic oral gestures, forms of articulation that mime factorial meaning, the essential structures of human experience that are the essential structure of conceptual and lexical meanings. A common example is the use of open vowels that involve an enlarged space in the mouth in words that have largeness as a prominent part of their meaning, like the vowels in far or huge as discussed later.

            A small set of vocal icons based on single ground factors could not make a lexicon, but the factors can be combined, as we have seen, to represent concept meanings with the aid of associated gestalts. I am suggesting that the first language, perhaps just a small set of terms for key entities and actions, was highly iconic in that the articulation of word sounds often imitated facets of word meaning thus highlighting salient qualities of meaning. Schematic gestalts would have added a specifically imagistic dimension. Once the language habit became established, and the lexicon expanded, the form of the vocal (or hand) gestures would have become conventionalized to a degree in ways that diluted the iconic factorial content but there is ample evidence that this content continues to play a prominent part in language. There is, as we will see in the next section, a pervasive pattern of factorial iconicity in modern languages. This supports the possibility that the ground factors may have facilitated the birth of language.

            We can conclude that there must have been a strong continuity in several respects in the nature of cognition from PLHs to Homo sapiens. All of this presupposes slow evolutionary developments in the architecture of the brain that must account for the difficulty of the birth of this faculty that is evidenced by the apparently very slow pace of cultural development until well into the sapiens era.

6. What it is like to be human.

            In this section I want to relate the ground factors to key features of human experience and cognition and to draw out some implications. This is a large task because it engages with some difficult and unresolved issues. I can only treat them briefly and superficially here but I hope some fairly cogent conclusions will emerge.

6.1 But what exactly is it like?

            We know very well what it is like from the viewpoint of our self-awareness. It is vivid in every moment of our conscious existence. Our phenomenology is rich and clear and self-reflective. But in another way we are almost as much in the dark as in speculating on what it was like to be a pre-linguistic hominid. We still lack much definitive knowledge about the nature of our mental processes, the very things that most distinguish us from animals, and how they operate. For example, the nature of our thought and the medium in which meaning operates are still deep puzzles. This paper suggests that we have often viewed these matters in the wrong way. We have been bemused by our privileged possession of minds. We have not taken sufficient account of continuity in the evolution of these faculties and we have, until quite recently, largely ignored the inevitability that our physical bodies and the ways in which they interact with the world around us are fundamental to the understanding of conceptualisation, thought and language.

            This section explores how the ground factors that pervade the whole field of biology may help us to understand such aspects of what it is like to be human. I am proposing that they are something of a missing link. They seem able to loosen some tight knots in our understanding of things ‘mental’. There is evidence that the factors continue to carry the same function in the semantics of human thought and language and in the structure of human experience as they did in perception and interaction in previous eras. They continue to provide the structure of our interactions with the world.

            But first we need to clarify what differentiates us in terms of cognitive ability from the pre-linguistic hominid that I have postulated. I follow Edelman and Tonini (2,000) in singling out higher-order consciousness and language as hallmarks of sapiens. Higher-order consciousness is built on top of lower-order consciousness in human beings. The lower-order type prevails in young infants and the higher order emerges as the child develops a clearer awareness of the complexities of the world and masters language. The chief features of the new faculty are an awareness of the self that is much stronger than the biological, autonomic self-awareness of animals, infants and PLH, the ability to use language, greatly enhanced memory, a refinement of phenomenological experience, and a much heightened conceptual ability and capacity for deliberative thought. These features interact with one another to produce a new level of cognition and enormously enhanced new ways to interact with the world and other human beings. Language is central to higher-order consciousness.

            I speculated in the previous section about one possible mechanism that may have facilitated the emergence of language in the transition from PLH to sapiens but numerous interlocking neurological, physiological and social elements and influences were at play. The ground factors appear to have provided a continuing, unchanging, fundamental element––an unbreakable thread, in the emergence of language, higher order consciousness and thought. I have proposed that the ground factors together with gestalt schemas constituted the emerging semantics of the pre-linguistic thought of earlier hominids that was superceding the age-old biological meaning, knowing-how-to-respond. But the situation with Homo sapiens is further changed. Knowing how to respond is transformed into a deliberative, flexible and dynamic form of thought that can generate, particularly with the aid of linguistic exchange, complex concepts and scenarios. Responding to what we encounter in the rich episodes of daily living in busy and phenomenologically rich cultural settings is a much more attenuated affair than it was for PLH; but it is still the ultimate function of our cognition.

6.2 The ground factors in human cognition.

            We have seen that the decoupled cognition of PLH must have consisted of two components, sensory and affective data that is correlated with the alpha and beta ground factors and the complementary sensory features that form gestalt schemas. The combination brought together object categorisation and value attribution in emerging thought. At the neural level sensorimotor and affective areas of the brain were involved in simultaneous cognitive activities. I have used the metaphor of a matrix to describe the way the two kinds of ground factors form the structure of the cognition of percepts and categories in animals, concepts in PLHs and human beings and our word meanings (see section 6.4). The matrix links external entities and events with regions of the brain in ways that enable it to construct meaning. There has been much puzzlement about how physical brains, using chemistry and electricity, could construct cognition in the form of configurations that refer to entities in the world in ways that are meaningful to us. The ground factors may help to resolve some of the difficulties associated with knowing, referring, representing and meaning. There are two keys to this process, the embodiment of meaning and the similarity of concepts in the brain and their external referents. The puzzle of the nature of concept and word meanings is how these entities that have the form of some kind of brain state can refer and bear meaning. Philosophers of language have theorised at length on these questions. Description theory holds that these meanings are essentially in the form of verbal descriptions and definitions. They are dependent on language. Causal theories propose that meanings derive from specific causes that initiate, inform and validate meanings. Each meaning has a history that explains how it was formed, with an original baptismal event. Amodal theories claim that sensory data is held in a non-sensory abstract form in the brain for interpretation as representations. There is a further and older approach that was thoroughly discredited for decades but that is beginning to find favour again. This is that word and concept meanings are based on a similarity between mental states and the entities they refer to. But how can mental states, events of chemistry and electricity, resemble external things, events and situations in their concrete actuality? The similarity principle has been severely criticised as, on the one hand, so general and unconstrained as to be ineffective, and on the other as being pictorial or depictive in a way that is incompatible with the structure of the brain. I hope to steer through these extremes.

            I have described how the embodiment of cognitive/interactive activities stretches back through the whole of evolution. I propose that embodiment is the key to the similarity of meaning (as represented neurally) and the referent. The ground factors are embodied not just in that they occur in the physical brain, but also in their enactment through specific regions of the brain that control the body’s interactions with the environment. Sensory, motor, kinaesthetic, proprioceptive and affect-generating regions are central to cognition. When I think of laughter a highly schematic but distinctive gestalt of the action, appearance and sound of laughing occurs in my brain. Key structural elements of the gestalt are the ground factors largeness/loudness, display in open space, intensity of energy, physical intensity, positiveness, communality, humour (associated with extension), and physical intimacy (associated with surface). These same factors form the structure of laughter when I hear and see it occurring or when I hear or read the word laugh or remember laughter. They also form the structure of the similitude, the link between the brain and the world. In the absence of actual laughter this configuration of ground factors distributed through the brain in association with the sensory gestalt can be experienced as a simulacrum of physical laughter. The gestalt may contain such sensory features as memories of the distinctive sound, the image of a laughing face, the feeling of laughter erupting. Both factorial structure and sensory features occur in parts of the brain that are connected to parts of the body that are associated with laughter. This is the manner in which I propose similarity operates. We read our knowledge of the world in our rehearsed bodily awareness.

            Each of the ground factors in this example occurs as the activation of a specific region of the brain. The meaning of the word, to the extent that I have described it (contextual and associative elements may supplement this meaning), can be attributed to a number of neural activities occurring in concert. The ground factors help us to understand the neural construction of the meaning and its derivation from sensory data in a way that has not previously been available. The alpha factors provide a coded input of data that assists the brain in identifying such external entities with the necessary support of the normative element of the beta factors and the necessary input of the sensory features of the full gestalt.

            The view that similarity is fundamental to representation, categorization, reference and meaning remains controversial but I will not engage with these discussions in any detail here. Important opponents of similarity include Goodman (1968), Cummins (1989), Pylyshyn (1984, 2007), and Fodor (1984) while Kosslyn (1980, 1994), Goldstone (1994), Barsalou (1999, 2007) Prinz (2002) and Edelman (2001) have supported its role in several allied fields. Similarity has been dismissed for at least three reasons: first, because there is held to be no effective means whereby similarity between entities in the world and states of the brain can be generated; second, because it is said to fail to produce the meanings of abstract terms; and third, because bare similarity is unconstrained and so an acceptable means of specifying its parameters has not been apparent. A further factor that has contributed to antipathy towards similarity as a basis for mental representation is its apparent incompatibility with the belief of many of the authorities that computation, using abstract symbols as conducted by computers, is the most appropriate model of representation and the manipulation of representations.

            I believe the embodiment of the ground factors, together with the gestalt schema, neutralizes the first objection. In effect they provide a medium in the brain in which meaning is generated and directly cognized. Meaning (at least of core words) is an amalgam of coordinated simultaneous brain states that are associated with felt states of the body. The nature of these is such that they have enormous versatility in representing states of affairs in the world. I will illustrate this point later. With regard to the second objection, I will demonstrate in section 6.4 and section 7 that abstract meaning can be handled satisfactorily under this system. Ground factors combined with sensory/affective gestalts are surprisingly effective in describing abstract terms. The third issue, the ineffectiveness of bare similarity without constraints that limit the parameters of similarity to a manageable level, is invalidated by the performance of the ground factors combined with perceptual gestalts as parameters of similarity. The delineation of meaning by sensory gestalts and their associated ground factor configurations provide adequate constraints. Finally, the computer model of thinking, becomes close to untenable, I believe, if similarity, embodiment and the gestaltic and factorial aspects of meaning are accepted. This does not invalidate the idea that the brain computes word meanings and other outputs but their computation is not the digital computation of the abstract symbol kind.

            Critics of similarity have sometimes objected that a homunculus in the brain is a necessary assumption behind similarity theories because perceptual data needs an interpreting agent in the brain. This has been a stumbling block because such an agent would seem to require its own internal agent, with a consequent endless regression. The model I am advocating requires no homunculus. When I hear laughter regions of my brain, acting in concert, form a physiological configuration that simulates the concept, scenario or word meaning. It is registered in my sensory, motor, kinaesthetic and affective and perceptual systems. The alpha factors are a direct coding of key physical characteristics while the gestalt schema(s) and the beta factors join to form a recognised event. My brain state refers to the event via the similarity metric that I have described. It has a direct meaning that I am able to enrich by associating it with memories of a similar factorial and gestaltic character. I have a matching prototype in my ‘mind’. It has a natural iconic, non-linguistic form and I can give it a name (which also has a partly iconic form). A PLH would probably have engaged in this process except for the last step. An ape in the wild could hear the laughter only as background noise.

            The concept laugh has a well-defined content. How did that content happen to come together in my mind? This has been a source of controversy. I do not believe, although some authorities (particularly Fodor) do, that the concept is innate and unlearned. I propose, however, that the ‘alphabet’ of meaning, the ground factors from which it is partly composed, has these characteristics. It is innate and this is the basis of its alphabetical character. From very early childhood I heard human sounds of similar peculiar qualities in similar contexts. I registered the relevant ground factors and the spectrum of gestalts associated with these events and came to recognise laughter as a discrete, variable, recurring, easily recognisable event. I had formed a concept and held it in long-term memory in its ground factor/gestaltic form. In due course I heard the word laugh associated with the event. The concept came to have a name. It seemed a fitting name somehow. (Perhaps I intuitively recognised its phonological iconicity (as described in section 6.4). If the name I heard had been yitch or bunt it might have been a little harder to learn because, unlike the great majority of core words, it would have had negligible phonological iconicity. I would soon have learnt it but I might always have had a niggling feeling that it was not quite right).

6.3 Sound/meaning iconicity in language.

            This paper might have begun here. As noted earlier, I first identified the ground factors in the meanings of words––as semantic constituents that are manifested in a previously unrecognised similarity between the articulation of individual sounds of many core words in English and facets of their meaning.  This was a chance discovery that began with the identification of a small cluster of words with this characteristic in one section of the dictionary that lead me to explore further with a surprising result––a series of correspondences involving phonemes of similar articulatory form that transpired to be a compact set of semantic primitives. It was also a rather awkward discovery because the best authorities had long ruled out the possibility of such an entity. One of the most fundamental tenets of linguistics for the last century has been the Saussurean arbitrariness of the sign, the idea that meaning is not attributable to sounds in their occurrence in words. This principle has appeared to be very soundly based for two reasons: because there was no obvious pattern of association of sounds and meanings apart from onomatapoiea and sound symbolism, which are not very significant in the full lexicon, and because any wide-spread bonding of sound to elements of meaning would severely compromise the freedom of language to provide the multiplicity of words that are necessary to meet the ever-increasing demand for words for new concepts.

            The unquestioning universal acceptance of this principle has lead linguists to overlook a very pervasive and rather obvious pattern of sound-meaning association in English, where I first identified it, and in at least some other European languages, and in the only non-Indo-European languages that I have examined (Finnish and Maori). I have concluded, with considerable confidence, that there is likely to be, at least a tendency towards this feature in all languages for reasons that will become clear. (This will generally, of course, involve sounds other than those in English, but I hypothesise a pervasive articulatory iconicity that may use quite different sounds for much the same component of word meaning in other languages).

            To describe the feature more fully, it consists of a similarity between a facet of the meaning of a word and the manner in which a sound, often but by no means necessarily the first in a word, is articulated. This is a case of sound miming meaning. There is one example that has been widely acknowledged without any recognition of its being part of an extensive pattern. It shows how things as different as the way sounds are formed and elements of word meaning can be similar in a significant way. The example is the frequent occurrence of an open vowel, one that uses a relatively large space between tongue and palate, in words that denote largeness, while close vowels in which this space is constricted, frequently occur in words that express smallness in various ways. Examples to illustrate in English are legion. (There are contrary instances but they occur less frequently.) The surprise, in the light of the arbitrariness tenet, is that a similar association can be frequently identified in English for virtually all the main sounds (or clusters of sounds with similar forms of articulation e.g. /k/ and /ch/, /t/ and /th/), although the articulatory basis of the relationship is obscure in some instances. I have called these iconic associations of sounds and facets of word meaning semes and this feature of language, sematicity.

            The reason for my confidence in the wide occurrence of sematicity in the world’s languages is that the meaning elements of the semes, the sematic meanings, are in fact, the ground factors. As structural elements of human cognition the factors are operative in the meanings of words and concepts. I have described them, in the human context, as categories of experience, using that term in the philosophical sense as necessary and sufficient ultimate classes, or the highest genera of human experience. They have the scope and power, on a more mundane level, of the categories of Aristotle and Kant. It was only after I gained familiarity with this new feature of language that I came to recognise its apparent place in biology that I have been describing. Given the functions of the ground factors in human experience it is hardly surprising that they are prominent in word meanings. This prominence is not limited to sematicity as I will explain shortly.

            The semes have been described in more detail in other papers but I need to outline their nature and characteristics here. They are set out in alphabetical order in Table 3 which shows the associations of specific letters/sounds and sematic meanings/ground factors as identified in the English language. These associations were identified by a simple process of distinguishing common facets of meaning in many of the most frequently used words in the sections of the dictionary with their common initial sound. In some sections these common facets almost fall out on a perusal of the core words. (I define core words as the proportionately very small group in each section that are central to everyday experience, are widely used in all forms of discourse and that normally have meanings that can be largely delineated by gestalts combined with ground factors. Interpretations of the core will vary but not to a degree that affects the outcome). In other sections they require a closer examination of the data. Anyone can repeat this method. It is not sensitive to the number of words selected as long as they are confined to the most commonly used words as illustrated in Annex 2. The results are somewhat more clear-cut in sections such as j, l, n, q, r, v, y and z than in others. Robust results are obtained from the other sections but they are sometimes more complex and oblique.          

Table 3: The ‘alphabet’ of the English semes.

a: action; largeness
b: bodily roundness; natural being
c: the community life-world
d: positivity / negativity
e: the physical life-world
f: surface; superficiality
g: abundance; availability; generosity
h: acquisition; possession; self-interest
i:  smallness; interiority
j: energy; humour
k: physical intensity
l: manifestness; display
m: materiality; magnitude

n: contraction, reduction; negation
o: openness; spatiality
p: particularity, specificity
q: uncertainty
r: intensity of energy
s: somatic affect, the body
t: tactility
u: displacement; negation
v: fullness / emptiness
w: existential value
x: strangeness?
y: extension; stretching
z: energy / emptiness

                        See Annex 1 for fuller descriptions.       

            An easy test of the results (using the word lists in Annex 2) is to apply the meaning associated with one sound say, r, to words in another section such as the l section. It will be very clear that the r meaning is a poor fit there. An obvious objection to the validity of this phenomenon is that the large majority of words in any section of the dictionary do not display any sematicity. This may seem a fatal flaw but it is not and the reason is spelt out in Section 6.6. Sematicity is largely confined to words that are the most strongly connected to everyday bodily experience, words that are among the oldest in the language and that relate to our immersion in the physical and social environments. A further objection that might seem to invalidate sematicity is that it is easy to identify the meaning attributed to one sound to numerous words that do not contain that sound. This is perfectly true but it does not affect the conclusion because the sematic meanings, as ground factors, have a much larger role in lexical meaning than that evident in sematicity. The meaning of many core words is spelt out by the ground factors to a significant degree. Sematicity is the tip of the iceberg of the role of the factors in language.

            Several surprises emerged as the scope of sematicity in English unfolded. First, that robust common facets of meaning were readily identifiable in most sections of the dictionary and in the others they were found with only a little more effort. Second, in most sections it was not difficult to find a credible association between the meaning and the articulation of the sound. But the greatest surprise was that the full set of sematic meanings was not just a loose collection like the large group of phonaesthetic meanings in English, but seemed to exhibit strong coherence and systematicity. They had the appearance of being, not only an alphabet of meaning, but also an alphabet of the structure of human experience and possibly of biological interaction in ways this paper seeks to argue.

            The main features of sematicity can be summarised as follows:

            1. The descriptors I have chosen should be regarded as indicative rather than definitive at this stage. They are set out more fully in Annex 1.

            2. The descriptors are generalisations from the common sematic meanings of the core words in a section of the dictionary. Their manifestation in any sematic word meaning may diverge somewhat from a literal interpretation of the descriptor but still must be consistent with it. The sematic meanings are sometimes used metaphorically.

            3. Semes operate most noticeably in the common, everyday words that represent human experience at its most elementary level, the core of the lexicon. They also occur in other words, but to a diminishing extent as words become more specialised, artificially constructed or exotic. Although semes are most readily identified in initial sounds, they can occur in any position.

            4.  Most of the semes can be identified in English in a remarkably high proportion of the core words beginning with their sounds/letters, between 50 and 80 percent on my calculation in most sections of the dictionary.

            5. The bond between meanings and sounds is the form of the articulation of the sound that is generally iconic or imitative of the sematic meaning to various degrees––one is motivated by the other. The articulatory motivation of most of the individual semes is described in the Annex.      

            6. Most of the sematic meanings have a polar and scalar character that often involves two separate semes. This polarity is represented in Table 2.

            7. A key feature of the sematic meanings is their somatic (bodily) character. I use this term to contrast the sematic meanings with the apparently cerebral nature of lexical semantics. This somatic character is a consequence of both the articulatory aspect of semes and the ‘felt’ nature of the sematic meanings with their basis in body-related areas of the brain. Most of them relate either to sensually cognised facets of the external world, to physical activity or to affective responses (‘gut reactions’) to entities. The sematic meanings are profoundly somatic!

            8. Although the sematic meanings tend to bond with sounds as described above they also occur very widely in word meanings without such bonding. In fact I propose that the core of all word meanings is constituted to various degrees (some to only a very small degree) from sematic meanings without the linkage with sound in most cases. This does not detract from the demonstrable significance of sematicity.         

            Two examples of semes will need to suffice here in supporting the validity of sematicity in conjunction with the word lists in Annex 2. The n section has a very robust seme. It’s meaning is relatively straightforward. It is contraction, compression, reduction and negation. The first three descriptors are indicative of a single meaning that no single word expresses satisfactorily while the fourth is an extension of this meaning. Some commonly used sematic words in this section are: nail, naked, name, narrow, navel, near, necessary, neck, need, nest, never, new, night, nipple, no, nod, nose, nothing, now, nub, numb, nut. The presence of the sematic meaning is not difficult to detect in most of these words. We can identify the physical compression in the elongation and sharpness of nail, the exclusivity of a name, new and now, the topological reduction of narrow and neck, the reduction of naked compared with the clothed state, the contraction of navel and near, the compactness of nest, the concentrated form of a nipple or a nut; an essential generalised reduction in necessary and need; a diminution of light in night and feeling in numb. The few non-sematic core words are nephew, net, niece, noise and north. The dominance of the seme in the core words of the section is impressive. There is no pole in this seme, but another seme, y, carries the natural polar meaning, extension, stretching that is expressed fairly clearly as in the tiny core yard, yawn, year, yell, yesterday, yet, yoke (n.), yonder and, less transparently in you and young. The only non-sematic core words are yellow and yolk.

            The core words I have singled out make up a very small proportion of the words in these sections of the dictionary. I have noted that sematicity is most evident in the core group. Any section of the full lexicon contains a great variety of words of different kinds––derivative, specialist, exotic, technical and scientific words, made-up words, those introduced from other languages and so on. Many of these have not been subject to the normal processes of word formation within the English phonological system. They cannot, accordingly, be expected to conform to patterns of association of sounds and meanings that have developed in English and its precursors. Similar situations probably prevail in other languages.

6.4 The ground factors in word meaning.

            In the final feature of sematicity described above I noted that the sematic meanings (the ground factors) have a much wider role in word meaning than their iconic representation in sematicity. Many word meanings consist to a significant degree of ground factors that have no sematic connection with their sounds. This is a consequence of the vast scope of their role in our cognitive systems and the non-feasibility of constructing word meaning extensively on a predominantly iconic basis. A significant portion of the meanings of many words can be formulated as a configuration of the ground factors using, simply for convenience of identification, the letters associated with the relevant semes in English to represent the appropriate ground factors (while recognising that sematicity is not present in the great majority of cases). In addition a symbol is needed to represent the necessary gestalt schema in any word meaning, the image-like or ‘felt’ intuition that the phonological form of a known word evokes. This is the larger element of word meaning for which, I propose, the ground factors provide the structure. The gestalts and their configurations of ground factors are the central elements of the concentric model of word meaning I will outline in section 6.6.

            Several conventions are followed in the analysis below: 1. A simple asterisk indicates the presence of the gestalt schema. I do not attempt to analyze the gestaltic content here. 2. The factors are set out within square brackets. 3. The initial two to four factors have a taxonomic function that defines the word’s broad field within the vast scope of referentiality. The following round-bracketed entries spell out, more or less, the structure of the gestalt. 4. The bracketed factors are ordered in terms of their approximate prominence in the word’s meaning. Their semantic effect is cumulative. 5. These factors operate additively to generate the structure of the core (non-taxonomic) meaning with any sub-bracketed entries operating as composite sub-elements. 6. Polarities apply, but where they are not indicated the first meaning of Table 3 applies. 7. Numerical subscripts indicate that a specific secondary meaning described in Annex 1 applies. Where there is no subscript the initial meaning applies.

            The analyses of word meanings that follow are very different from those that have been undertaken in recent decades. In particular they do not use the descriptive features, described linguistically, that most authorities have employed to describe word meaning. The formulas depict the non-linguistic nucleus of word meaning. They represent the initial instantaneous non-linguistic intuition of meaning that we experience immediately when we hear or see words. This is supplemented, as context requires, by linguistic elements of meaning. Word meaning that was restricted to the core would be of limited utility in speech and written discourse. It is the supplementary third zone that gives the lexicon much of its infinite scope and richness.

            The unfamiliarity of this method of analysing word meaning may make it strange at first but this feeling should be reduced as the palpable presence of the factors is recognised and the full gestaltic core is built. I need to emphasize that the formulas I provide are far from definitive. The whole structure I am describing in this paper needs to be reviewed before the method can be used with greater confidence. In examining these formulas it may be useful to bear in mind that the factors represent the contents of specific regions of the brain that are sensitised to the registration of specific facets of the external world. Experienced sequentially we can build up the structure of the meaning. Experienced simultaneously they provide a proportion of the full meaning that requires the gestalt to achieve a well rounded fullness.

            As two initial working examples I will use long (as an adjective) and laugh (as a verb), the meaning of which I analysed above. The meaning of the first can be written as follows:

Long (a.)*[e, m2, y, ((a2, y), (l, o), v1, w). In plain English this is equivalent to: operative in the external life-world; implies materiality and extendedness which, in association with the preceding factor, indicates an adjective; (these are taxonomic functions that specify the broad area of cognitive space in which the concept lies); large in extended terms (a succinct definition in itself); displayed in open space (further description that relates to the typical visibility associated with length); fullness (this is a versatile function that here simply amplifies the preceding description); existentiality  (this is a non-specific function that indicates a basic-level concept, one that is intimately associated with human experience).

Laugh (v.) – *[c, m2, a, (p, (a2, l, o), (k, r, j), (f4, s), v1, w)] or, operative in the community life-world, implies materiality, an action (taxonomic functions); particulate, large (loud) and displayed in space; physically and energetically intense and associated with humour; implying social intimacy and somatic affect; fullness, existentiality.

            The factors with their structural nature delineate meaning quite fully for these words, more so than is often the case, but the gestalt, the flesh on the skeleton, contributes decisively to the meaning. The gestalt for long is both visual and tactile (as in the gestural accompaniment to a fisherman’s tale) and for laugh it has elements that are auditory, visual, visceral and affective. Finally, here is a small selection of words to illustrate both the strengths and limitations of the ground factors in spelling out word meaning.

      now – *[c/e, m8, (p, n, i, l, o, v1, w)]

      wrong – *[c, m8, y, (u, s, d2, l, v1, w)]

      have – *[c/e, m2, a, ((p,p), h4, h1, g1, n, t, s, d1, v2, w)]

      appreciate – *[c, m8, a, ((p,p), h4, g, s, l, d1, v2)]

      real – *[e/c, m2, y, (t, q2, s, l, v2, w)]

      believe – *[c, m8, a, ((p,p), (h4, (i4, l), q2, s,  d1, v1, w)]

      cow – *[e, m, (p, a2, b, g)]

      cupboard – *[c, m, (p, (o, i2, l3))]

      pink, blue, green – *[e, m, y, (f, l)]

      carburettor – *[e, m, (p, (i2, f3))

      psychosis – *[c, m8, (h4, u, d2)]

      gibberillin – •[e, m, (g, k, v1)]

      lasagna – *[c, m, (g, d1)]

            I will comment briefly on each. 1. As none of the factors relate directly to time the precise nature of the gestalt is hard to ascertain but it seems to have a visceral component relating to a need to attend to the demands of the present moment. Manifestness refers to that quality of the present compared with the murkiness of the past and future. The o represents the openness and immediacy of the present moment. 2 The bracketed descriptive factors are indicative of a strong visceral gestalt. 3. The bracketed section, self-interest/possession, availability, exclusivity, contact, felt affect, positiveness, fullness, existentiality describes the semantic content of this important word remarkably fully. The (p,p) component is a device to show that this is a two-argument (actor, patient––a possessor and something possesed) verb. The gestalt has a kinesthetic element as with grasp, but weaker, or hold. 4. This and the next two words demonstrate that abstract meanings can be quite effectively represented by the ground factors (see also Section 7). The gestalts associated with these words are affective and visceral. 5. The descriptive section is effective in outlining the meaning: a mental state in which tactility, definitiveness, somatic affect and display are prominent. 6. Another effective formula. The sub-bracketed component represents the paradigmatic test of touching. 7 to 9. These illustrate the weakness in the factors’ ability to delineate everyday objects and qualities with strong visual schemas. The formulae are highly ambiguous and the onus of meaning is on the distinct and robust visual gestalts. 9. This is typical of formulas for sensory qualia that have strong gestalts and weak formulae. 10 to 13 demonstrate the typical failure of the factors to provide an effective structure for the meaning of specialist words that often lack a clear gestalt but compensate with verbal descriptions that are often essential for distinct meanings in this type of word (see Section 6.5). The effectiveness of the schemas is dependent on familiarity with the referent in such words.

            These examples are a tiny fraction of the words I have analysed in this way but they support the following conclusions of the wider analysis: 1. The ground factor formulas are often remarkably effective in providing a robust structure for the meaning of many of the core words of the lexicon. 2. They are surprisingly effective in many common words with abstract meanings. 3. The factors on their own fall well short of depicting the meaning of words for everyday highly visual entities such as animals, furniture and sensory qualia, all of which have strong compensatory gestalts. 4. They also fail to provide more than scant structures for the meanings of most specialized, technical, scientific, cultural and introduced terms. These are heavily dependent on linguistic descriptions. Nonetheless, the ground factors with their timeless evolutionary lineage are ultimately the essential constituents of these descriptions through core words that constitute the descriptions.

6.5 A model of word meaning.

            On the basis of extensive analysis of word meanings using this methodology I have developed a three-part model of a concentric structure of word (and concept) meaning that explains the pattern described in the last paragraph. At the centre are the configurations of ground factors that provide a broad taxonomy and a structure of what is denoted by many words. This is the nucleus of word meaning. For some words the factorial configuration alone provides quite a full depiction of meaning (for example in wrong, have, believe, real) but this is not normally the case.

            In the surrounding second zone is the phenomenological aspect of the gestalt schema that fills out and disambiguates meaning. This aspect can involve any phenomenal modality, and quite often a mixture of modalities, as well as affectivities. The factorial nucleus has no life or viable function on its own. The factors are integral to the schema as its necessary structure. These two parts constitute the semantic core of word meaning. The core is what we normally experience instantly when we see or hear a word.

            The outer concentric zone can take a range of forms that represent supplementary ways to construct meaning where the first two zones are insufficient, as they are for many words. These include verbal descriptions of key features of the referent, definitions, linkages to other words and concepts, items of personal experience, inferences and encyclopaedic information. These constitute a subsidiary element of the meaning of most core words but they are indispensible for the very large and ever-expanding category of words described in 7 to 13 in the list above. Without this specific supplementary knowledge, that is generally expressed in words, these terms have insufficient semantic content for the non-specialist to understand them. Specialists with intimate knowledge of the referents, however, may form effective gestalts for some such terms. In any modern lexicon a large proportion of the words’ meaning are heavily dependent on this supplement.

            The first two zones are non-linguistic. The factorial structure has an embodied character as discussed earlier, but its intrinsic nature, as described in the earlier sections, is that of innate biological parameters. In ontological terms it is a set of necessary conditions that govern the behaviour of biological systems through its representation at a structural level in sensation and perception (the alpha factors) and in inherited activators of response to environmental condition and events (the beta factors). The factors are necessary constituents of sensory gestalts in which they serve a structural function. Gestalts are essentially perceptual, sensorimotor and affective in various proportions.

            This zone vastly expands the scope for the meanings of words. It is, however, hierarchically dependent on the core language with its rich factorial and gestaltic content. Words in this large and ever-expanding part of the lexicon generally have meagre factorial nuclei and gestaltic cores, and are generally almost entirely dependent on words from the core of the lexicon for their semantic structure in so far as it consists of verbal features and information. Ultimately the ground factors are the nucleus of the whole vast empire of language.        

            The last three sub-sections have provided evidence that the ground factors have a foundational role in word and concept meaning and in the whole field of language. This has been built on their eons-old role in the interaction of organism and world. There is a further case that I cannot argue here that the ground factors are fundamental to human thought, that they can be regarded as the ‘alphabet’ of a language of thought that has the universality of Fodor’s Mentalese but has some very different features.

7. Similarity, some conclusions.

            I have proposed that similarity is a mechanism that pervades biology in its provision of the basis for the behaviour of organisms, their interaction with the environment. Organisms mirror, darkly but adequately, the ever-changing kaleidoscope of selected perceived features of the world in order to determine how to respond to conditions, entities and events they encounter. This is biological meaning in operation. In sections 5 and 6 I briefly discussed a role for similarity in PLH concepts and human language and thought.

            But what exactly is similarity in this context. I have described it in several forms in other contexts. First, as a global element of organismic interaction (perception and response) where it incorporates necessary biological parameters (the beta factors) that are integrated into inherited mechanisms of evaluation and response; second, as iconic perceptual and categorial representation that is the result of a process of neural simulation in which the factors are integral features of neural representation; and third, in PLH and human concepts, as simulated, highly schematic, isomorphic neural representations that are decoupled from direct experience. It is in this third form that, I propose, similarity operates in word and concept meaning. The ground factor nucleus, gestalts and third zone semantic mechanisms participate in highly variable degrees in the simulation of the meanings of words. Meanings incorporate similarity at several levels: factorial nuclei of neural representations mirror those of perceptions; mental gestalts reflect perceptual gestalts in a highly schematic manner; and the third zone linguistic elements parallel those that can be involved in perception and understanding. This is a model of semantics at the lexical level.

            But how, more definitively, can we establish that similarity is the founding principle of word and concept meaning given the enormous edifice of the modern lexicon where it is far from evident? The answer was foreshadowed in the last two sections. It is my contention that the operation of similarity can be discerned without undue difficulty in much of the core lexicon, and beyond, with the aid of the ground factors and recognition of the role of gestalts in the construction of word meanings. I have tried to give credibility to these two key constituents of word meaning through the evolutionary story that I have presented. I believe the ground factors and gestalts are readily identifiable in word and concept meanings once their existence is recognised. But this requires the abandonment of existing models of word meaning such as those that use language-based constituents of core meaning like the linguistic features that make up prototypes and exemplars or descriptions, those that assume that meaning is formed from amodal symbols divorced from perception, and Fodor’s atomic constitution of meaning.

            Here are a few further examples of factorial analysis of word meanings to illustrate this point: see – *[c/e, m2, a, ((p,p), s3, l, o, v1, w)], look – *[c/e, m2, a, ((p,p) r2, s, l, o, v1, w)], listen - *[c/e, m8, a, (s3, i4,  l, o, v1, w)], touch - *[c, m2, a, ((p,p), s3, t1, l, v1, w)], lift, *[c, m2, a, ((p,p), t, r, u, l, o, w)], push – *[c/e, m2, a, ((p,p), r, f, t1, u, w)], enter - *[c/e, m2, a, ((p, p), r2, (i2, o, n), w)], wear - *[c, m2, a, ((p, p), b1, o4, v1, w)], forget - *[c, m8, a, ((p,p), (h4, i4), u, l3, g4, d2, v1 w)]. I believe such examples together with those in section 6 and those below provide substantial validation of the principle of similarity as well as of the ground factors themselves and their semantic function. If the ground factors were unreliable, accidental entities with no significant role in cognition it is inconceivable that they would operate so deeply in the delineation of the structure of word and concept meaning.

            One of the major developments of human cognition was the evolution of abstract concepts. There is no consensus as to how abstractness emerged from concrete forms of cognition but the model of meaning that I am outlining suggests a possible route.  It may seem surprising that the factorial and gestaltic components of the present model, with their somatic nature, are quite effective in representing the meaning of abstract terms. The basis of this capacity lies partly in the ability of the somatic quality of the factors to delineate facets of the meaning of abstract terms as effectively as those of concrete terms. This reflects something that is commonly overlooked. This is that abstract terms do not reside in purely cerebral regions of the mind. Their constituents are both of a beta and an alpha factor character, with the first often interpreted metaphorically. This interpretation flows from the appropriateness of the alpha factors, as expressed somatically, to form elements of the meaning of abstract terms as the examples in the next paragraph demonstrate. The other element is the somatic quality of the gestalts associated with most core abstract terms and many others.

            Abstract words are generally regarded as problematic in terms of understanding the nature of their meaning. Barsalou (1999) describes them as ‘a classical challenge to perceptual theories of knowledge’. To illustrate the capacity of the type of analysis I have been describing here are a few examples from many that I have examined: abstract - *[c, m8, y, (t1, v5, i2, n, l, p)], meaning - *[c, m8, ((h4, i4), p, g, n, l, q2, v2, w)], cognition - [c, m8, ((h4, i4), t3, q2, l, v2)], truth - *[c, m8, (f, t, l, q2, s, v2, w)], zero - *[e, m8, (p, n2, v5)], quality - *[e, m8, (v2, l, s, q2, d1)], quantity - *[e, m2, (v6, a2/i1, q2)], hope (v.) - *[c, m8, a, ((h4, i4, s), (g2, v5), (v6, d1, l), w)], funny - *[c, m2, y, (h4, j4, s, l, d1, v2)]. The results of this analysis may be highly surprising. Similarity is very effective in identifying the structure of abstract meanings.

            An alternative theory of abstract meaning is an element of Lawrence Barsalous’s well-known perceptual symbol system (1999). While the system itself has much in common with the treatment of word meaning in this paper I believe Barsalou’s position on abstract meaning is unwieldy and untenable. He rejects, as I do, the use of metaphor to illumine abstract meanings as proposed by Lakoff (1999) and others. Barsalou also holds that it is difficult to attribute abstract meaning using perceptual symbols directly. In essence he proposes that the meaning of abstract terms cannot be analysed into more primitive terms. Instead he advocates an ‘alternative strategy’, the use of a complex mechanism involving the framing of abstract concepts in illustrative situations that utilise perceptual symbols and introspective events in order to understand abstract concepts. Even in describing examples of his wider system he is unable to specify the semantic content of his symbols beyond the fact that it is sensory in nature. The ground factors present a radically different picture by providing a crucial feature of such content for concrete terms and, as the previous paragraph demonstrates, for abstract terms. The recognition of the role of gestalts and third zone constituents fills out the picture. The model described in this paper meets the ‘classical challenge’ in a direct manner.

8. Conclusion.

            If the hypotheses that I have constructed from the empirical evidence for the operation of the ground factors in English are valid it follows that this previously unrecognised but easily identifiable group of semantic and biological primitives occupies a central position in language and cognition and possibly in the whole biological enterprise. In the latter case it also follows that human cognition, language and thought has a strong continuity with elements of organismic interaction with the environment that extend very far back into evolutionary time. The consequences of recognition of the ground factors seem likely to be numerous and applicable across many fields of linguistics, psychology, cognitive science, biology and the philosophy of language and mind, but it would be premature to explore them here. As I have noted, the research and thinking behind this paper has had no significant peer review due to my lack of relevant academic credentials combined with the ‘left field’ nature of my findings and hypotheses. One of the main aims of this paper is to seek to initiate that process. Further papers are planned on specific implications of the ground factors.

References.

Barsalou, L. W. 1999. Perceptual Symbol Systems, in Behavioural and Brain Sciences 22, 577–660

Barsalou, L. W. 2007. Cognitive and Neural contributions to Understanding the Conceptual System. Current Directions in Psychological Science.

Barsalou, L. W. 2007. Grounded Cognition. Annual Review of Psychology, Volume 59, 2008.

Cummings, R. 1989. Meaning and Representation. MIT Press.

Edelman, G. M and Tonini, G. 2000. Consciousness: How Matter Becomes Imagination. Penguin Books, London.

Edelman, S. Representation is Representation of Similarities. Brain and Behavioral Sciences. 2001.

Fodor, J. 1975. The Language of Thought: A Philosophical Study of Cognitive Science. Hassocks: Harvester Press.

Goldstone, R. L. 1994. The Role of Similarity in Categorization: Providing a Groundwork. Cognition, 52 125 – 157.

Goodman, N, 1972. Seven strictures against similarity. In Problems and Projects. Bobbs Merrill.

von Humboldt, W. 1998. On Language: The Diversity of Human Language and its influence on the Mental Development of Mankind. Trans. Heath, P. Cambridge University Press.

Hinton, L., Nichols, J. and Ohala, L. 1994. Sound Symbolism. Cambridge University Press.

Johnson, M. 2007. The Meaning of the Body. University of Chicago Press.

Kosslyn, S. M. 1980. Image and Mind . Cambridge, MA: MIT Press.

Kosslyn, S. M. 1994. Image and Brain: The Resolution of the Imagery Debate. Cambridge MA: MIT Press.

Lakoff G. and Johnson M. 1999. Philosophy in the Flesh: The Embodiment of Mind In western Philosophy.  Basic Books.

Lloyd, T. 2003. Fossils of Language. Steele Roberts. (New Zealand).

Lloyd, T. Forthcoming. The Iconic Alphabet of Experience, Thought and Language.

Lloyd, T. Publication pending. The Alphabet of Meaning.

Nagel, T. 1974. What is it Like to be a Bat? Philosophical Review 83. Pp.435 – 450.

Nanny, M and Fischer, O. (Eds.) 1999. Form Miming Meaning. John Benjamins.

Prinz, J. 2002. Furnishing the Mind: Concepts and Their Perceptual basis. MIT Press.

Pylyshyn, Z. 1984. Computation and Cognition: Toward a Foundation for Cognitive Science. MIT Press.

Pylyshyn, Z. 2007. Things and Places. How the Mind Connects with the World. MIT Press.

Rosch, E. and Lloyd B.B. (eds.) 1978. Cognition and Categorization. Erlbaum.

Saussure, F. de 1995. Course in General Linguistics. Bally, C. and Sechahaye, A. (Eds.) New York Philosophical Library.

Sloman, S. A. and Rips, L. J. (Eds.) 1998. Similarity and Symbols in Human Thinking. MIT Press.

Annex 1. A full description of the ground factors.

The alpha factors.

m         Materiality  •  1. Materiality; 2. associated with the material world;
............3. extendedness,  horizontality; 4. continuity; 5. magnitude,
............multiplicity; 6. measurement; 7. manipulation of materials;
............(8. immateriality, the mental).

p          Particularity  •  1. Particularity, specificity, compactness; 2. exclusivity;
............and the following character of things: 3. small and compact;
............4. slim and vertical; 5. long, thin and pointed; 6. flat and
............bounded; 7. seriality.

f           Surface  •  1. Surface; 2. flatness; 3. the manipulation of surfaces; 4. intimacy;
............(5. insubstantiality, unreliability; 6. excessively detailed).

b          Bodily roundness, natural  being  •  1. The fullness of the human body;
............2. of a rounded character; 3. natural being; 4. parameters of natural
............and human being; (5. violence, destruction).

n          Contraction •  1. Contraction, compression, constraint, focus, reduction,
............subtraction, exclusivity; 2. negation; (3. dispersion).

y, j       Extension  •  1. Extension, stretching; 2. continuous, ongoing; 
............3. personal and objective energy;  4. humour.

l           Display  •  1. Display, manifestness, conspicuous visibility; 2. illumination;
............(3. concealment, obscurity).

o          Openness   •  1. Spatial openness; 2. personal openness; 3. possibility,
............opportunity, (4. concealment, confinement, absence).

a          Action, largeness   •   1. Action; 2. largeness; enlargement; 3. locational parameters.

i           Smallness, interiority      1. Smallness; 2. interiority; 3. essence;
............4. having an mental or abstract quality.

r           Intensity of energy   •   1.  Intensity of energy; 2. activity, movement;
............3. salience, (4. inactivity, rest).

k          Physical intensity  •   1. Intensity in action or physical quality; 2. the prominence
............of parts of things, angularity; 3. the complexity, intricacy and refinement of things.

The beta factors.

t          Tactility  •  1. Touch, tactility, contact; 2. sensation; 3. pointing, deicity. 4. uprightness.

s          Body / mind  •  1. Somatic affects; 2. the body, bodily actions; 3. sensations;
............4. bodily excretions and secretions.

d          Positiveness / negativeness   •  1. Affectively bright and dark aspects of the world,
............happy/sad, good/ bad, favourable/unfavourable etc.

v, z       Fullness / emptiness   •  1. Fullness of explicit value (as expressed by previous zoemes);
............2. fullness of intrinsic value such as  positiveness, salience, forcefulness, presence,
............mindfulness, high visibility; 3. physical fullness; (4. lack of explicit value; 5. emptiness,
............insubstantiality, lack, absence, negativeness; 6. capacity to contain).

g          Abundance, generosity • 1. Abundance, availability; 2. generosity, prodigality, givenness;
............3. productivity, effectiveness; (4. deprivation; 5. seeking, grasping).

h          Possession,  self-interest  •  1. Possession, acquisition, satisfaction; 2. want, desire;
............3. self-interest; 4. subjectivity, personhood; (5. dispossession, detriment).

q          Uncertainty  • 1. Uncertainty; indeterminacy; (2. definitiveness, authority).

u          Displacement • 1. Displacement, alienation, strangeness, dissonance, discomfort; 2. negation;
............3. disparagement, (4. rightness, comfort, satisfaction, consonance, order and utility).

w          Existential  value  • 1. The existential  significance and value of the everyday.

c, e       Lifeworld  • 1. The communal (c.) and 2. physical lifeworlds (e)

For Annex 2 see the second annex to The Sound of Meaning.

         This annex provides further information on the semes and ground factors associated with six sections of the English dictionary. (The full set is provided in Lloyd (publication pending)). The annex is intended as an aid to understanding how sematicity was identified and how it operates. The examples of sematic words include all the basic-level words (in bold) in the sections as I have identified them and an admixture of other words selected to illustrate the scope of sematic meanings. The examples of non-sematic words are confined to basic-level words. Words with polar meanings are marked with an asterisk. There is scope for disagreement on what words exhibit sematicity because judgements are subjective and sematicity is quite often obscure and sometimes decisions are marginal. I do not expect readers to agree with all my examples but the sematicity of the great majority should be fairly obvious. For each seme I will provide, for convenience, the fuller description of the sematic meaning from Annex 1 (with polar meanings in brackets) before postulating the articulatory motivation of each seme and setting out the examples. I also make some comments of a general nature on the individual factors. The breakdown into a range of variants of the headline meaning is based on my experience in analyzing English words. They represent distinct facets of the headline as they emerge in this language. They should be regarded as provisional pending the input of other interested parties. The analysis of words in other languages may also require modifications.

Alpha factors.


l         Display • 1. Display, manifestness, conspicuous visibility; 2. illumination;
..........(3. concealment, obscurity).

Articulation.
The main phonological features of /l/ are its alveolar position, its voice and its laterality (which refers to the spreading of the lips relative to other consonants). In articulatory terms it is distinguished by its characteristic release of the following vowel, its ejection into the world by the release of the tongue from the alveolar ridge, putting the meaning of the word on display, as it were.

Examples.
label, labyrinth*, lack*, laconic*, lady, lament, lamp, land, language, large, laugh, laud, launch, law, lazy*, lead (v.), leaf, leap, learn, leave, lecherous, leer, lend, less*, let, letter, libation, library, lick, lie (v., position), lie (v., n. falsehood)*, life, lift, light (n., a.), like (v., adv.), limbo*, line, linger, lip, list, listen, litany, literal, little, live (v., a.), livery, load, lock*, locution, look, loose, lord, lose*, lot, loud, lout, love, low, lucid, luck, luminous, lure, lust, lyric. Non-sematic: lamb, lame, lean (v.), left (a.), leg, lemon, lion, liver, lung.

            This is a paradigm of sematicity. In word meanings it is often paired with o.


o        Openness  • 1. Spatial openness; 2. personal openness; 3. possibility,
..........opportunity; (4. concealment, absence).        

Articulation.
This is described as a mid-back, rounded or labial vowel. The motivating feature seems to be the size and orb-like shape of the oral cavity.

Examples.
oath, obey, obfuscate*, object (n.), oblige, obliterate*, observe, obstruct*, obtain, obvious, occasion, occult*, occupy, occur, odour, of, off, offer, often, oh, omit*, on, once, one, only, opaque*, open, opinion, opportunity, oppress*, option, or, oral, order, ordinary, organ, organise, orgasm, origin, ornament, other, ostensible, ought, our, out, over, overlook, owe, own. Non-sematic: odd, oil, old.

            Oath, obey, oblige, ought, owe, own represent interesting metaphorical extensions of the core sematic meaning. This seme has much in common with l and often has a natural pairing with it in the meaning of words to indicate, often pleonastically, display in space (sometimes an interior space as in learn and  listen).


r         Intensity of energy  • 1.  Intensity of energy; 2. activity, movement;
..........3. salience, (4. inactivity).                               

Articulation.
/R/ is described as a voiced post-alveolar consonant. The relevant iconic features of its articulation are the tense lateral curvature of the tongue and the presence of voice.

Examples.
rabid, race (v.), ragged, rage, raid, rain*, raise, rally, ramble*, rampant, rancour, ransack, rant, rape, rapid, rare*, ravage, rave, raw, re-, reach, real, reap, rear (v.), reason, rebel, reckless, recline*, rectify, recur, refer, reflect*, refuse, regress, reinforce, reject, rejoice, relate, relax*, relieve, rely*, remain*, remember, remote*, remove, renown, repeat, repel, repent, reply, report, repose*, representrequire,rescue, research, resemble, resent, reside*, rest*, resurrect, retire*, retreat*, return, reveal, revel, revenge, revere, rich, rid, ride, ridicule, right, rigid, rise, risk, riot, rip, roam*, roar, rob, rocket, root, rot, rough, round, rouse, rout, row (v., n.), rub, ruck, rude, ruin, rumbustious, rumour*, run, rush

Nonsematic: race (n. =  ethnicity), rather, red, responsible, river, road, roof, room, rope, rose.     

This section is notable for the easily observable scale of its sematicity from pole to pole––from rage to rest. There are many words with the re- prefix in the section. With its meanings backwards and repetition, many words with the prefix carry a mild strain of sematicity that reflects the effort of repetition and reversal.


Beta factors.


g        Abundance, generosity  • 1. Abundance, availability; 2.  generosity, givenness,
..........care; 3. productivity, effectiveness ; (4. deprivation; 5. seeking, grasping).

Articulation.
This is a voiced velar stop. Its position of formation is further back in the mouth than any consonant apart from /h/. This sound, together with /h/ and velar /c/ form a small group that is characterised semantically by their predominant reference to the communal and personal as opposed to the external and impersonal which is often associated with more frontal consonants.

Examples.
gain, game, gap*, garden, garland, garment, garner, gasp*, gather, gay, gear, genealogy, general, generous, generate, genuine, geography, get*, giant, gift, girl, give, glad, glamour, gleam, glee, globe, gloom*, glory, glutton, gnosis, go, goal, god, gold, good, gorgeous, gormandise*, govern, grab*, grace, grail, grand, grandeur, grant, grasp*, grass, gratify, gravity, great, greed*, greet, gregarious, grief*, grip*, gross, ground, group, guess, guest, guide, gulf, gush, gust, gusto, gut, guzzle*. Non-sematic: gas, gate, glide, grey, grind, groan, gun.`

            This and the following sematic meaning, and all the beta meanings, have a powerful biological rationale in that they are associated with biological motivators and triggers of action. Although both these have poles of their own there is a deep polarity between the two as exemplified in giving and having. The sematicity of go is marginal.


h        Possession,    self-interest  • 1. Possession, acquisition; 2. want, desire,
..........appetite; 3. self-interest; 4. subjectivity, personhood;
..........(5. dispossession, loss, detriment, hostility, threat).

Articulation.
The sound is described as a voiceless glottal fricative, pure breath, almost. It is formed with a small heave of the chest, the deepest articulatory action of any sound in English. This may be the motivational element.

Examples.
habit, habitat, hall, hand, handle, happen, happy, harem, harm*, harvest, hate*, have, haven, head, heal, health, heap, hear, heart, hearth, hedonist, heaven, heir, hell*, hello, help, her, here, heredity, hero, hibernate, hide (v.), him, hire, history, hither, hoard, hobby, hold, holiday, holy, home, honour, -hood, hope, hospitality, host, house, huddle, hug, human, humble, humour, hunt, hurt*, husband. Non-sematic: hair, half, hammer, hang, hard, hat, heavy, high, hill, hit, hollow, honey, hook, hop, horn, horse, hot, hour, how, hundred, hurry.

            The non-sematic group is unusually large but this does not detract from the robust sematicity.


q        Uncertainty  • 1. Uncertainty, indeterminacy; (3. definitiveness, authority).

Articulation.
This letter occurs in English only in association with u. The motivational factor is difficult to determine.

Examples.
quagmire, quaint, quake, quality*, qualm, quantity*, quark, quarrel, quasi-, queer, quench*, quest, question, quibble, quiet, quintessence*, quirk, quite*, quiz, quorum*, quota*, quote*. Non-sematic: queen, quick.

            This is another important beta factor with its function of signaling uncertainty that needs to be resolved. Its pole, which anomalously carries a negative sign, has a significant affirming function.

A new grounding – August 2009.