A model of behaviour

The object of this paper is to provide the basis of a model permitting the prediction of change in behavioural systems, and in particular human social systems, in the process of disintegration,

I shall make several assumptions:

• The first is a methodological one: a model is postulated, rather than induced from randomly accumulated data in accordance with empiricist theory. Its acceptability does not reside in the empirical verifiability of its postulates which is not only often difficult to achieve, but is also inconclusive as observations are themselves models postulated on the basis of the observer's general model or world-view. Acceptability will be taken as residing in the precision of the interpretations and predictions that the model gives rise to. (Appendix 1).
• The second assumption is that behavioural processes at all levels of organisations are sufficiently similar to be represented by the same basic model (see Appendix 2).
• The third is that behavioural processes are directive. (See Appendix 3).

Stability

The law of economy appears to be the basic law of behaviour. Things take the line of least resistance and move thereby to positions where free energy is reduced to a minimum i.e. equilibrium. Since positions of unstable equilibrium are, by their very nature, unlikely to be maintained for long without external or asystemic intervention, changes will be towards stable equilibrium or stability.

We can regard this teleonomic behaviour as passive in pre-life forms and increasingly active as life develops.

Stability is normally regarded as the ability of a system to return to its point of departure after a disturbance. A behavioural system as opposed to a man-made incomplete system cannot return to its exact point of departure, but to that which involves the minimum change compatible with the maintenance of a stable relationship with a changing environment and. hence with the internal stability of the corresponding supra-system (For this reason, C. H. Waddington suggests the use of the term homeorhesos [from the Greek: rhesus, to flow] rather than homeostasis to describe the negative feedback behaviour of behavioural systems as opposed to that of incomplete man-made systems. The term 'system' in this paper will be used to refer to the former.)

Systems can be more or less stable. The more stable, the smaller will be the disequilibria occurring between them and their respective environments, and the corresponding corrections. A system whose stability is increasing is referred to as 'damped'; one which is out of control as 'runaway'.

The behaviour pattern of a system can be represented by a series of oscillations corresponding to disequilibria and their corrections. Thus in a stable system oscillations are small and in an unstable one large. A damped system is one in which they are diminishing, while in a runaway system they are increasing.

At the moment the particularities of social behaviour and of policies to influence it, are judged purely in terms of their ability to achieve specific targets which are deemed desirable per se in terms of the society's social model or weltanschaung. Attempts to question the desirability of achieving these targets are regarded as unscientific, unobjective and falling within the category of value judgements. Once we accept that stability is the goal of behaviour, then we have at our disposal a precise criterion, an objective measuring rod for judging the desirability of behavioural trends regardless of the level of organisation at which they occur, and of their degree of generality.

Control

A system is defined is unit of behaviour, and is composed of a control mechanism plus that part of the environment that it controls.

The control mechanism is fundamentally the same at all levels of organisation. Relevant data are detected, transduced into the appropriate informational medium, and organised, or interpreted, in the light of the system's model, not of its environment but of its relationship with its environment. The responses mediated are those which appear the most adaptive in the light of the model. This adaptiveness must depend on the quality of the model. Relevant to this are the number of elements, their degree of organisation, and the time-lag between detection and the mediation of the appropriate response.

The role of the model in the determination of responses is more evident in the case of behaviour at a low level of organisation, for instance during protein synthesis; but it is equally important in a human society.

The latter's behaviour pattern can, in fact, be regarded as determined by its model, world-view or 'weltanschaung', interacting with its environment both internal and external. As will be shown later, institutionalised external controls, except superficially and in the short-term, are ineffective in counteracting behavioural tendencies associated with a given model or world-view.

Our society has developed a very singular world-view, according to which man is above the rest of nature rather than part of it and is thereby exempt from the laws applying to other living things. He possesses something called 'reason', and so long as he is provided with the necessary knowledge he will behave in a 'rational' way, which is interpreted as meaning that he will cease having too many children to prevent over-population, he will regard all men as his brothers - thereby, so it is believed, eliminating all human and social problems.

The material problems which are supposed to have afflicted man throughout his tenancy of his planet will also be solved by means of science, technology and industry which will enable him to eliminate such things as drudgery, poverty, ill-health, even death. Needless to say, this materialistic variant of paradise is quite unachievable. To do so would involve violating the basic laws of thermodynamics as well as many of those of biological, social and ecological organisation.

It is this model that is giving rise to our equally singular social behaviour pattern which is geared to the systematic substitution of man-made and man-controlled artefacts for the self-regulating processes of nature, i.e. to the methodical substitution of the 'technosphere' for the 'biosphere' - a process referred to as progress, but, needless to say, one that can only lead to the breakdown of society and the disruption of its life support systems.

Environmental Parameters

Evolution is a feedback process between a system and its environment. Seen slightly differently, the system evolves to fulfil a specific function within a larger system. The rate and extent of possible changes in the latter, as in all systems, must be limited by a number of different factors. For instance, in order to maintain its basic structure, the total rate of change must be limited by that of its slowest changing sub-system.

Hence environmental changes are only tolerable within certain limits. I shall use the term 'environmental parameters' to refer to the minimum and maximum values of the variables, in terms of which the supra-system is described, to which the system is capable of adaptive responses. Disequilibria caused by changes in which these values are exceeded are referred to as asystemic

Behavioural processes proceed from the general to the particular by differentiation (see Appendix 1). As the system differentiates, as a unicell, for instance, evolves into a complex metazoa, as or a foetus develops into an adult, so it becomes capable of dealing with ever more serious environmental challenges.

This means that these values will increase. One can draw a graph to illustrate this (see Figure 3).

In Figure 3, seriousness of environmental challenges is illustrated along the vertical axis, time along the horizontal axis; the curve shows the development of the ability to cope with environmental challenges with time, and the two dotted lines indicate the parameters, or limits within which systemic changes can occur. It is to be noted that the curve does not rise in linear fashion, but as a result of a series of jumps; this is because, during phylogeny, different critical points or levels of organisation are reached. When this happens, new behavioural principles enter into operation, permitting greater behavioural possibilities.

During ontogeny there are critical moments, such as birth when a child is suddenly removed from the highly ordered environment of the mother's womb and subjected to a considerably less ordered environment: that of a family unit. Puberty marks another important environmental change, as at this point in a stable society the child enters into the still less ordered environment of the community as a whole. It is significant that in most stable societies these critical moments in a child's life are marked with festivities often involving some traumatic experience (circumcision, for instance), which has a similar effect to shock treatment in helping destroy a now obsolete behaviour pattern and introducing those conditions most favouring the rapid inculcation of a new one.

In Figure 3, the periods lying between these critical points have been separated and labelled. The labels used, paleo-, meso-, and neo-, are purely tentative and the number of divisions arbitrary. It is possible to distinguish two different types of disequilibria: those in which the maximum and those in which the minimum values have been exceeded. The former situation can be referred to as deprivation, and the latter as saturation. Again the terms are tentative.

An example of the first situation would be a child subjected to insufficient motherly attention, i.e. being brought up in a family environment with insufficient order; an example of the latter situation would be a child suffering from excessive motherly attention. One can expect pathological manifestations caused by these two different types of disequilibria to be very different.

In addition one can distinguish between disequilibria occurring at the various stages of development. The earlier these occur, the more serious one must expect them to be, as by effecting the generalities of a behaviour pattern they must colour all the particularities in terms of which the former are differentiated. It is for this reason that children subjected to an unsatisfactory family environment in early youth will tend to be 'emotionally unstable'. They will be more likely to display pathological behavioural tendencies such as delinquency, drug addiction, etc., and will be very difficult indeed to educate - or, more precisely, socialise, whatever might be their apparent intellectual potential

It is therefore one of the conditions of stability that a system be able to develop from the very start in the appropriate environmental conditions. Responses to asystemic environmental changes will not only themselves be asystemic but, if occurring in the early part of development, will prevent the system from being able to respond systemically to any changes at all

It is suggested that this might provide a means for classifying disequilibria, and then corrections, systemic and asystemic. This is particularly necessary in a field such as psychology where the classifications used: psychoses, neuroses, etc., are non-functional, and mainly refer to symptoms. It should also prove useful in classifying social disequilibria.

Responses

A system can develop in two ways; firstly, it can increase its capacity to cope with environmental challenges and modify the environment in such a way that such challenges become both less severe and less likely to occur.

To a large extent it is by doing the former that it succeeds in achieving the latter. To increase its capacity to deal with environmental challenges, it must build up its model so as to improve its ability to interpret and predict environmental changes. At the same time it must increase its control over environment both in space and in time. This means expanding by destroying and assimilating systems otherwise organised. As a system does this, so, by the same token, it increases its capacity for dealing with environmental challenges and hence for further expansion. This process cannot go on indefinitely, and one must expect a hierarchy of negative feedback loops to be operative. In a social system some of these are likely to be of a cultural nature.

In the final instance, the law of economy provides the final such negative feedback: if we assume that responses occur in answer to a challenge present, predicted or imaginary, then expansion must ultimately reduce such challenges to a point where they are no longer sufficient to trigger off further expansion.

Entropy

Clearly, in the first stages of development when few predictions can be made regarding environmental change, the most adaptive organisation must be one of disorder or entropy. When the number of elements is maximal and order is minimal, the range of possible reactions to unpredictable environmental changes is maximised.

Variety

As the environment builds up, and with it one' s capacity for prediction, so more effective responses can be mediated towards different possible challenges. In such conditions a corresponding number of somewhat more complex reactions must be made possible - Variety can be said to be replacing Entropy. For reasons of economy, there must be a limit to the number of such challenges to which a system can respond adaptively. The higher the probability of the challenge that can be predicted, the more the system must be capable of reacting adaptively to it. Therefore, the greater a system's variety, the higher the improbability of the challenge to which it can adapt. What is taken for redundancy (animal populations, neuron populations, etc.) in a stable system is, in fact, variety.

Complexity

Normally, variety and complexity are used interchangeably. I prefer to distinguish between them. When it is possible to predict the occurrence of environmental challenges in a particular spatio-temporal pattern, the system gives rise to a correspondingly complex response. In this way it becomes specialised in dealing with a specific environmental situation - one that can be predicted as being highly probable. This gives rise to a damped system so long as unforeseen challenges can be prevented from occurring.

Centralisation

In communication theory, it is assumed that the recipient is interested in a message if it conveys sufficient information, i.e. if its improbability is sufficiently high. In a behavioural context this is not necessarily so, as a signal must also be important, i.e. relevant to the recipient's behaviour pattern or general to it. The more general it is, the greater must be that proportion of the system affected by it. That organisation favouring the detection of such signals and the mediation of the correct responses must display a high level of centralisation.

Compromise between the satisfaction of these requirements

Centralisation means reducing the variety of possible responses that can be mediated by sub-systems at a lower echelon of control. As a system becomes more complex so also is its variety reduced, since it is committing itself to a specific environment, thereby reducing the possible range of environmental changes to which it can react adaptively. Also, as a system becomes more complex, and hence more specialised, so is it likely to become less centralised, so that responses can be mediated as much as possible by the increasingly specialised sub-system, more intimately in touch with their respective equally specialised environmental situations.

It must follow that the response mediated by a system, and the organisation it will display, must be a compromise, that which in the light of its model can be predicted as likely to give rise to the most adaptive behaviour.

Disruption of the System

A system breaks down when the self-regulatory mechanism essential for ensuring adaptation ceases to be operative. In such conditions, there is no means of maintaining the level of variety, complexity and centralisation that would enable it to meet environmental challenges. The system, no longer under control, becomes progressively less stable until it collapses.

What is likely to cause disruption of this sort? Geophysical changes can bring about serious upheavals. They can lead to ecological invasions, by alien sub-systems. Since these were not developed to fulfil specific functions within the system, it is likely that they would not have developed the capacity for ritualising their behaviour, thereby limiting its impact on the new environment. Also, it is likely that the latter would not provide the necessary controls for keeping their population in check, enabling them to proliferate and destroy the system.

Therefore, it is not surprising to find systems at all levels of organisation equipped with rejection mechanisms to exclude elements alien to it. Whether one likes it or not, such mechanisms are operative at the level of a human society, so long as it remains capable of self-regulation and hence of adaptation.

When such mechanisms break down, the introduction of alien sub-systems in any quantity could lead to increase in randomness or in a reduction in order. This can be counteracted by incorporating these new sub-systems into the system's basic structure, which can be done at different levels of organisation. Immigrants can be assimilated at the level of the individual, assuming that their cultural pattern can first be broken down, or foreign groups can be incorporated in a cultural symbiotic relationship (as has occurred with the spread of Hindu civilisation). However, this can only occur in specific cultural conditions.

Surplus Energy

We who live in a society that equates progress with increasing energy consumption should consider that there is an optimum amount of energy required for operation of any system. Plants only exploit a minute fraction of available solar energy, not because they are inefficient, but because, if they were to photosynthesize more, the nutrients in the soil would be exhausted and the environment would cease to be capable of supporting them.

It is significant that stable societies appear to exploit various strategies for channelling surplus energy into those uses that will result in the minimum supra-systemic disruption. I shall refer to this as the 'ritualisation of economic behaviour'. It means providing a maximum outlet for surplus energy with the minimum use of natural resources in such a way as to cancel the minimum social disruption.

Thus in many societies we find a large proportion of the society's resources being channelled into feasts and other forms of ostentatious spending. The best known example being the Potlatch of the Kwakiutl and other Indians of the American North West coast. The production of subtle and highly contrived human artefacts, and indeed artistic activities in general, can also be regarded as ritualisations. Artisans spending their life in carving a cathedral door will use considerably fewer resources and cause correspondingly less pollution than if they were employed in an automated ball-bearing factory.

Also the finished products of their activity do not interfere with the optimum functioning of their society, as do the utilitarian consumer products manufactured by modern industry.

It must be remembered that the family, a sine qua non of a stable society, is an economic as well as a biological unit, and it is difficult to see it surviving in a society in which food, clothing and other basic requirements normally produced at a family level are manufactured by some vast company and available in the local supermarket. Nor can it survive when the functions normally fulfilled by the father have been usurped by the government's social services: (education, welfare, health, etc.).

The same is true of the small community whose survival is menaced by the ever greater centralisation of industrial activity required to fully exploit the so-called economics of scale which manifest themselves as production becomes increasingly capital intensive. In such conditions a small community is either deprived of its livelihood or forced to commit itself to the production of a specific commodity which must seriously affect its basic social structure as well as rendering it particularly vulnerable to changes likely to alter the demand for the commodity in question.

On the other hand economic activity geared to the production of non-utilitarian commodities, such as works of art have no adverse effect on social structures. If anything they are likely to reinforce them by causing the society to concentrate on those religious and artistic activities that distinguish the society from its neighbours, and that provides it with the pride that will lead them to divert ever more energy and ingenuity towards its preservation.

We are thus led to the paradoxical conclusion that to preserve stability and thereby ensure the survival of a social system, energy should be channelled from the production of utilitarian products to that of non-utilitarian ones.

Random Information

As already mentioned, learning is very similar to other behavioural processes. The generalities of the information are contained in the rudimentary model present when the process starts, and are differentiated step by step through interaction with each new environmental situation. Information must be introduced in a specific order and from a specific source. This is clear if one considers that in the case of a social animal the learning process occurring during ontogeny is designed to enable a child to fulfil its functions as a member of its family and community.

Education is, in fact, nothing more than socialisation. In a feedback system the information enabling the system to react to a particular part of its social environment must be obtained through contact with that specific part of the environment, rather than from some arbitrary source.

At all levels of organisation interference with the learning process must cause serious systemic disruption. Thus when the information contained in the genes or the nucleus of a cell is modified by radiation or chemical action, the model ceases to represent its environment adequately, interpretations of environmental signals will be wrong and the responses mediated unadaptive.

If any of the feedback loops linking the systems to its environment are severed, if behaviour becomes 'institutionalised' as in the case of modern government, then behaviour can no longer be influenced by environmental requirements and from the point of view of the supra-system it must become random.

The introduction of random information into the system must have a similar effect. Unfortunately, as our society "progresses" so its inhabitants tend to be bombarded with ever greater quantities of it. Obvious sources are television personalities, newspapers, and, unfortunately, one must include to an ever greater degree our educational system which is increasingly institutionalised and centralised, and hence ever less capable of fulfilling its basic function: that of providing that information which will enable people to fulfil their functions as members of their (now defunct) family and community.

A system will only tend detect and interpret signals that are relevant to its behaviour pattern, i.e. will affect the value of the variables used. This means that random information which is irrelevant to a man's behaviour pattern, is likely to be filtered out. However, as the environment changes, and the signals become more relevant, they are more likely to be detected.

It is more difficult to filter out random data in childhood. A child's brain is not designed to encounter random data, normally excluded as much as possible from the protective family environment. Also, in the child's case it will do more damage by affecting the generalities of the learning process which will colour the subsequently developed particularities of his world-view.

The Growth of Instability

Once changes occur beyond the system's environmental parameters, it is no longer possible to deal systemically with its challenges. Some of these the system will not even have the means of detecting. Thus we are not provided with mechanisms for detecting the 3,000 or so chemical additives which are systematically introduced into our food, nor, for that matter, the pollutants, such as heavy metals, pesticide residues, and radio-isotopes which also find their way into it, and into the water we drink and the air we breathe. We thus have no means of behaving adaptively towards them.

Even when we are capable of detecting the presence of asystemic elements, the tendency is to mistake them for outwardly similar elements of which we have some phylogenetic and ontogenetic experience. In this way dangerous pollutants such as Strontium 90, for instance, are introduced into our life-processes. Stephen Boydon refers to such reactions as pseudo-adaptations. They tend towards an equilibrium position - one in which stability is reduced rather than increased.

This must be so because they are designed to satisfy a single supra-systemic requirement, rather than to provide the optimum compromise between the various supra-systemic requirements, thereby leading to increased stability. They must, by their very nature, create further disequilibria, each giving rise to further pseudo-adaptations, leading to further disequilibria, etc. - hence causing the system to proceed ever more rapidly towards inevitable breakdown.

I prefer to refer to such reactions as asystemic. Practically all the behaviour of our industrial society falls within this category.

The mechanism whereby asystemic responses are mediated at a higher level of complexity, i.e. at the level of a human society is an interesting one.

It is known that perception is not an objective means of acquiring information. The perceiver tends to see what he expects to see, i.e. what in the terms of his particular model he can predict is likely to be there. Extremely unpleasant situations whose correct interpretation would be intolerable to him, i.e. would lead to the breakdown of his mental equilibrium, or personal control system, he will tend to re-interpret in such a way as to render them tolerable. What is true of individuals is also true society. A society will tend to interpret asystemic situations so as render them acceptable. Thus any information whose correct interpretation would lead one to cast doubt on the basic tenets underlying the society's model or world-view is almost certain to be interpreted in such a way as to ensure their reconciliation.

A schizophrenic behaves in the same way when he persuades himself that he is Napoleon or Julius Caesar in order to render tolerable an otherwise intolerable situation. People like Professor Zuckerman, Mr. John Maddox etc., who stubbornly refuse to face the realities of the present environmental crisis are behaving in like manner.

In the words of Professor Forrester, once the environment has changed sufficiently it becomes 'counterintuitive', i.e. normal human intuition fails to provide satisfactory interpretations of it, then responses must be 'counterproductive'.

The runaway social system

Unfortunately the system's instability will increase by positive feedback. This means that the asystemic corrections required to restore, however, precariously, the environmental conditions in which the system can function must become of an ever more radical nature, causing ever greater disruption, and hence still further increasing instability. Thus in our industrial society, measures required to bridge the widening gap between population and food supply are becoming ever more desperate and ever more destructive of the soil's food-producing capacity. It is no coincidence that four billion acres of desert have been created in the last 70 years.

As measures become increasingly desperate, so they tend to become ever more dependent on advanced technology and capital-intensive, highly centralised industry, putting an ever greater stress on natural resources and generating an ever greater amount of pollution. Human activity will also become increasingly utilitarian to the point where ritualised behaviour will tend to be regarded as anti-social and unethical in that it does not contribute towards providing the temporary relief of increasing human misery.

Unfortunately, the society's principal institutions must inevitably be caught up in this process.

Government

The behaviour of government is predictable with reasonable accuracy. Like all institutions, it is primarily interested in self-perpetuation. This means that first of all it must obtain votes, which in turn means pandering to public opinion i.e. to the basic values underlying the society's world-view. As the society disintegrates so will these values come to reflect more and more the requirements of the alienated individual rather than those of the family and the community; so will they become more trivial and short-term; and so will their satisfaction further reduce the stability of the society as a whole.

In order to perpetuate itself the government also requires finance. This it can obtain by encouraging industry and taking all those measures designed to increase the gross national product. Unfortunately this simply means Introducing ever more surplus energy into the supra-system and ever more asystemic controls, which, as we have seen, can only increase instability.

It is illusory to suppose that any government will subject such considerations to that of ensuring the long-term stability and hence the survival of the society it has been called upon to direct.

Industry

Industry is also interested in self-perpetuation and requires finance for this purpose. It depends on continually increasing the gross national product, hence on introducing more surplus energy into the system, and corresponding asystemic controls. The restoration of self-regulating controls is unlikely to be encouraged since these involve the replacement of energy consuming asystemic controls by energy-conservative systemic ones. In fact it must, in the long run mean a considerable reduction in the GNP, and hence in industrial activities.

Science

It may be surprising that science has not contributed towards reversing the present runaway situation. Unfortunately scientists are also people, and as such are imbued with the same set of values as the rest of society. It is nevertheless surprising that they should have accepted the world view of our industrial society 'hook, line and sinker'.

Indeed, rather than serve as the critics of our technological society and offer us some protection against its worst abuses, they have been as involved in it, as instrumental to it, as the technologists and industrialists who have exploited their 'discoveries'. Functionally speaking, they are its priests. It is they who have formulated the world-view that provides its rationale, and they have couched it in the most up-to-date 'scientific' terminology, and supported it with a wealth of empirical data, which confer on its principal tenets a degree of indubitability seldom enjoyed by religious dogmas.

What is more, this priesthood is backed by massive government subsidies and its prestige and influence is as great as that of the most firmly established of conventional religions. Like other priesthoods, it has reserved for itself the sole right to dispense the mana, or vital force whose accumulation in terms of the current weltanschaung is a measure of one's power over nature. In our society this power is called 'scientific knowledge'.

This is defined in a very subtle way. It only refers to data accumulated as a result of experimentation. Information deduced from basic principles does not qualify unless it can be "tested" empirically in the artificial conditions of a laboratory. Thus it is obvious that wide-spectrum chemical pesticides cannot possibly work, as they accumulate up food chains and thereby do more damage to the predators than to the target species that they control. In the same way it is quite evident that efforts to eradicate infectious diseases by waging chemical warfare against their vectors must be counterproductive, since one is thereby substituting a precarious, highly simplified, externally controlled, and hence very unstable device for a much more complex set of highly stable, self-regulating controls. However, such information is not regarded as constituting "scientific knowledge" because it is not backed by sufficient experimental data. Needless to say, this can only be acquired by trying out these iniquitous devices, thereby providing our technologists and industrialists with the green light.

The outcome

It is easy to predict the outcome of the runaway process our society finds itself in. Disequilibria will become increasingly difficult to correct. Eventually the situation will be such that no expedient will be available to prevent total collapse.

The question we must ask is whether or not there is a means of reversing the process. In order to do so, we must seek to understand it in greater detail. It is hoped that this paper provides some of the material required to permit the building of a model of the process involved. It then remains to determine a detailed programme of change designed to restore some semblance of stability. A first attempt to do this (A Blueprint for Survival) was published in January 1972 in The Ecologist.

The Unified Science Institute has been set up to examine its implications, and monitor it in the light of the projected model.

Appendix 1: Facts and hypotheses - a false dichotomy

Most people - including many scientists - assume that there is a difference of kind between a 'fact' and a 'hypothesis'! A fact, it is usually considered, is something that has been established 'empirically', i.e. by observation; a 'hypothesis' is just a hunch that remains to be verified empirically.

This is a false dichotomy, and is only made by those who are ignorant of the nature of perception or observation.

The latter is not simply a mechanical process like taking a photograph, as empiricists would have us believe. It is an organisational one. Data is detected, 'transducted' or translated into the informational medium of the brain and organised in that pattern of information that the brain contains.

It is only when this has occurred that the data constitutes information; and this information is not a fact but rather a hypothesis based on the interpretation of the data in the light of our particular model of the system. Information is organised in our brain to form a model of our relationship with our environment, i.e. of the system of which the two are part.

However, each interpretation must vary in accordance with the model used, and since, as a result of our different characteristics and different experience, we have all built up different models of our relationship with our environment, so our models, and hence our interpretations, will be different. It is for this reason that people see, hear and smell different things and that perception is so subjective.

The information obtained by means of observation in terms of which we verify our hypotheses is itself but another hypothesis, from which it must follow that there can be no such things as facts as distinct from hypotheses.

The main reason why this is not more apparent is that for day-to-day requirements, the subjective probability of the hypotheses postulated on the basis of our perception is so high, that for all practical purposes, they can be regarded as certain.

As a result, we have an innate tendency towards what might be called 'perceptive realism' - a form of subjectivism in which we assume the reality of our perceptions, and which provides the psychological basis of the empiricist fallacy.

Thus if I see a dog sitting on the green and eating a bone, I feel I have established a fact. In reality I have postulated a hypothesis. It might for instance, be a jackal, or a wolf, or a mechanical contrivance dressed up as a dog. Similarly, what I have interpreted as being a bone might be something quite different - a stone for instance, or a porcelain figurine, or a piece of wood.

However, I shall almost certainly treat these suggestions with the scorn they deserve, for the simple reason that my original hypothesis fits in extremely well with my model, and therefore has high subjective probability, whereas the alternative hypotheses simply do not.

The 'hypothetical' nature of perception becomes much clearer in the case of models with a lower degree of probability. Thus, to return to our dog; once I am satisfied that he is eating a bone; I might postulate a number of further hypotheses.

For instance, I might identify it as John Smith's dachshund eating a bone. I might go further and make certain assumptions about the origin of the bone. Since I know that John Smith's family is away on holiday and that the dog is being looked after by John Smith's mother-in-law, who is notoriously mean, and even more notoriously indifferent to dogs, I would be pretty certain that the bone had been given to the dog by the local butcher.

I could complicate my model still further by guessing what sort of bone it was; at what time in the morning the dog visited the butcher; the expression on the butcher's face when he gave the bone to the dog, etc. Even the staunchest empiricist would admit that these final details were in the nature of hypotheses, and that they did not constitute 'empirical knowledge', as did the original facts alluded to.

Empiricists would thus establish a sort of dualism between valid knowledge obtained by perception and not so valid hypotheses. If so, however, where then is the frontier to be drawn between these two categories of information? Thus, if it cannot be said that I saw John Smith's dachshund eating a veal bone given to it by a smiling butcher at 11am, since much of this information is assumed or deducted, can I say that I saw John Smith's dachshund eating a bone given to it by a butcher?

If not, can I say that I saw John Smith's dachshund eating a bone? If I must limit myself to saying that I saw a dog eating a bone, why should the frontier be drawn at this point rather than any other? The answer is that I did not 'see' any of these things, if 'seeing' refers to an objective mechanical process such as detecting.

What I saw was a mass of lights and shadows which I then proceeded to interpret, in terms of my systemic model, by postulating the most probable hypothesis, whose generalities I feel certain of; i.e. have very high subjective probability, and whose successive particularities I am increasingly less certain of; i.e. have increasingly lower subjective probability.

One is thereby forced to the conclusion that all knowledge simply consists of hypotheses with greater or lesser probability or what is the same thing ... that there is no dichotomy between facts, however well they may be verified empirically, and mere hypotheses.

Appendix 2: The development of the ecosphere as a single process

The Hellenic philosophers assumed that the world could be explained in terms of one all-embracing theory. They built general models that may appear naive today, but many of which were probably, at different moments in time, the best that could be built with the available knowledge. Since then, the tendency has been for science to split into ever more specialised fields, each using its own method and terminology. Only in the last decade has a movement arisen to link the various fields into one general science. The principle involved, however, is still little understood, and the necessity for such a general science is only recognised by a minority of enlightened people.

Owing to our tendency towards subjective classification, we recognise that certain events among which a connection can be made within our immediate experience can be regarded as forming one process, while, on the other hand, we refuse to admit that this can be the case with events whose connecting bond lies outside our experience.

Thus, we are willing to admit that the development of a foetus into an adult is a single process, and that it is difficult to examine, separately and in isolation, any of its particular stages apart from the process as a whole. On the other hand, we are less ready to regard evolution in this way.

We still imply that radical frontiers exist between life at different levels of complexity, in spite of the fact that they are part of the same evolutionary process. Yet, it can be demonstrated that no such frontiers obtain. When Kohler synthesized urea, the barrier between the 'organic' and the 'inorganic' was suddenly shattered, as did that between the 'animate' and 'inanimate' when the virus was found to manifest certain conditions associated with life on being confronted with a source of protein, and at other periods to display the normal behaviour pattern of a crystal. Again, it has been demonstrated repeatedly that no barrier exists separating men from the simpler animals. He is more 'intelligent'; and that is about all that can be said. (see "Towards a Unified Science" The Ecologist Vol. 1 No. 7.)

If this is so, we should be able to establish laws of development applying to the process as a whole, i.e., it should be possible to build a general behavioural model that will be applicable to behaviour to all levels of complexity.

It is only when one attempts to do this that we realise the necessity for such a model.

Thus, it can be shown that any principles that appear to apply in the initial and more general stages of development, must also apply at the later, and more particular ones. The best illustration of this is the applicability of the laws of physics not only to inanimate objects, but also to the most sophisticated organisms, such as a human being.

Thus, if you drop a rock and a university Professor from the top of a tower, in both cases the way they fall will be predictable in terms of the same physical laws. They will both obey the law of gravity, for instance. This essential principle I refer to as the accumulation principle. It follows from the fact that behaviour proceeds by the accretion of successive strata, each one of which will constitute a differentiation of the preceding one.

The accumulation principle is apparent, also, from the following consideration. In its development from the simple to the complex, matter passes through certain critical stages, where the possibilities of a particular type of organisation are exhausted and further advance can only be achieved by the development of a new type.

Thus, an atom can be developed only up to a certain point. This point will vary with different types of atoms, some of which, such as the tungsten atom, are relatively large.

Beyond this critical point, however, development can occur only by associating several atoms together to form a molecule. As soon as the latter stage is reached, the constituent atoms undergo a considerable change, in that a radical division of labour occurs, in accordance with the law of economy.

To explain their behaviour now requires the introduction of new principles. These, however, do not replace those required to explain the behaviour of the atoms before their association; rather they complement them. An accumulation has occurred.

The same thing happens when we pass to the next level of complexity, the cell, which is made up of associated and differentiated molecules, and so on. In each case, as we proceed to a higher level of complexity, there must be an increase in the number of disciplines required to explain behaviour. The sociologist who deals with behaviour at the highest level should thus understand behaviour at all the preceding ones: for a society is made up of men, made up of organs and tissues, in turn made up of cells, in turn made up of molecules, atomic particles, etc.

The accumulation principle is also apparent from yet another consideration. The genetic instructions transmitted from one generation to the next are not determined by, the experience of the previous generation. If they were, modern science would not condemn so radically the notion of the inheritance of acquired characteristics. On the contrary, that part of the instructions that can be ascribed to the experience of the preceding generation is but a minute fraction of the total instructions contained in the genetic material.

The latter in fact, will reflect the experience of the unit of phylogeny taken as a whole, i.e. of the species to which the system belongs, taken four-dimensionally. From that it must follow that if we are to understand the process of phylogeny, it is the latter that must be taken as behaving, and not one of its differentiated parts.

The same principle is infinitely easier to understand in the case of onto-genetic development. Each step in the embryological process is not regarded as separate. The embryo as a whole is taken as the unit of behaviour.

Thus, the accumulation principle makes it clear that to understand a process one must not only take into account the unit of behaviour that appears to be directly involved, but the vast four-dimensional system of which it is an integral part, from which it derives its general instructions, and of which it constitutes but a differentiated part.

For this reason, sociologists, who attempt to explain behaviour without reference to the preceding stages of development, are like neuro-physiologists who seek to understand the development of the cerebral cortex in a child without reference to the midbrain, the brain-stem, and the other parts of the nervous system. The study of processes, which are but part of much larger processes, in an artificial vacuum can give rise only to the most superficial understanding.

Another principle of development that emerges from such an approach can be referred to as the sequential principle, or the principle of succession as it is known in ecology. All behaviour is made up of a sequence of steps. These steps must occur in the right order. If one step in the sequence does not occur, the sequence can proceed no further. In addition, the environmental situation to which they constitute adaptive reactions, and to which each one is therefore linked, must also occur in exactly the right order.

Thus, if a given step does not occur at the 'right time', it will not occur at all, or will occur imperfectly. Once more, embryology furnishes us with a very clear illustration of this principle.

Behavioural reactions, though they may occur spontaneously, are also 'triggered off', by corresponding environmental situations. The latter are said to act as 'stimuli'. The less discriminating the system concerned, the more specific will be the stimulus required to determine a given reaction.

Discriminatory ability is low in an embryological system, where the cytoplasm constitutes a very highly ordered environment. In such a situation, environmental situation 'A' triggers off reaction 'a', which in turn gives rise to a modified environment, 'B', which in turn triggers off specific reaction 'b', etc. It is evident that in these conditions any departure from the correct sequence of environmental situations and of behavioural reactions, will prevent the total process from occurring.

This sequential principle is also apparent in everyday behaviour. If a man is hungry, he goes to the kitchen to make a sandwich. He cannot possibly perform the steps in reverse, i.e., eat the sandwich before he has made it, and before he has gone to the kitchen to collect the ingredients. The correct sequence of steps must be observed.

Similarly, in the development of an ecosystem, or of the ecosphere as a whole, the steps must occur in the right order. An ecosystem cannot support carnivores until it has first given rise to herbivores, and the latter cannot possibly come into being unless the requisite vegetation has first appeared. Only a fixed sequence of events, from which but slight deviations can be tolerated, can account for the development of the highly complex biosphere of which we are part.

This principle once more confirms the need for a general behavioural model. There is every reason to believe that this principle must apply to all behavioural processes. A third principle of behaviour is worth considering. In embryology, Van Baer's law states that development is from the general to the particular, from which it must follow that the earlier an interference occurs the greater the damage it will do. The reason for this is that development occurs by differentiation.

Van Baer's law can be shown to apply equally well to everyday behaviour.

When a man decides to eat a sandwich, a general instruction is issued by that particular centre in the brain that mediates eating behaviour. This message is differentiated at more and more particular strata, at each of which the instructions are adapted to specific environmental requirements. Similarly, when a General issues an order at Army HQ, the instructions will be differentiated at each echelon, i.e. at divisional HQ, brigade HQ, battalion HQ, company HQ, platoon HQ. etc., and further adapted to local systemic requirements.

It is also evident that as we pass from the amoeba, whose single cell fulfils all those functions that are necessary to the maintenance of life, such as the seizing of prey, its digestion, the excretion of waste matter, respiration, reproduction, locomotion, etc, to the complex multi-cellular organism into which it eventually evolves, these same functions are fulfilled in an infinitely more differentiated manner.

Specialised mechanisms have developed, perfectly adapted to fulfilling functions that were previously fulfilled in a more general way by a single cell. The same is also true as the artisanal workshop evolves into the large commercial enterprise, or a tribal society into a large centralised kingdom.

If those processes, occurring at a particular level of development, are but a differentiation of the more general processes occurring at the previous level, it is impossible to understand the former without reference to the latter. Once more, we find ourselves faced with the necessity for a general behavioural model in order to understand any of the differentiated parts of the process ensuring the development of the total eco-system.

Appendix 3: The directivity of behaviour

Science consists of organising data or putting 'cybernismic' order into the environment. Things that appear unrelated and haphazard are arranged in such a way as to appear orderly. The environment is four-dimensional, or, more precisely, it can best be represented by a four-dimensional model. Thus it is not three-dimensional things into which order must be put, but four-dimensional processes. To put order into the latter involves knowing in what direction they are moving. If one cannot do this, they remain unrelated and haphazard, i.e. disorderly.

All behavioural processes must therefore be taken as being 'directive' - a term coined by Russell in 1938. [1] I prefer this term to the term 'purposive', which in fact means the same thing. Unfortunately, when we talk of somebody's purpose, we are not thinking of his role within some general system, but rather of his 'conscious' motivation.

If man's behaviour is determined by a mysterious force called the 'free-will', then 'purpose' refers to the direction in which the exercise of 'free-will' is leading him, and in terms of which his behaviour can be explained. It must follow that since animals other than men are supposed to be governed by 'blind instinct', they are not capable of exercising 'free-will', and thus of displaying 'purposive' behaviour.

Even if we use the word 'purpose' functionally, its old metaphysical connotation tends to linger. If we use it, for instance, in connection with the behaviour of such lowly animals as sea-urchins or fiddler-crabs, subconsciously we cannot help but imagine these humble creatures consulting their little 'wills' before deciding 'freely' which zooplankton to have for tea. As this is not the image I wish to convey, it is easier to abandon the term 'purposive' in favour of one with no such undesirable connotations.

To deny directivity is in fact to deny that processes can be the object of scientific study. In spite of this, empiricists obstinately persist in so doing. This is partly because they tend to regard three-dimensional things and one-dimensional processes apart, as though they were self-sufficient units.

It is not currently realised that these units are nothing more than anthropocentric abstractions, units of our thought-processes and of our language but not of the world they represent. There are no such things as dogs that do not eat and drink and reproduce, except as photographs, pictures, concepts and words; nor are there such processes as eating, drinking, breathing and reproducing taken apart from the organisms involved.

To deny directivity is to deny that cybernismic order can be put into dynamic processes, and hence that they can be subjected to scientific examination, and, since all the constituents of the world display different degrees of dynamism, that science itself is in fact possible.

The evidence of directivity is so overwhelming at all levels of complexity, that its denial seems inconceivable.

Dc Beer writes [2] :

"The structure of an animal shows a number of exquisitely delicate adjustments: the splinters inside a bone are situated exactly where they are required to withstand the pressure to which the bone is subjected; the fibres of the tendon lie accurately along the line of strain between the muscle and the bone to which it is attached; centres of nerve cells in the brain are situated close to the ends of the nerve fibres, from, which they habitually receive impulses, and when in phylogeny there is a change in the nerve fibres from which any given nerve-centre habitually receives its impulses, the nerve-centre is found to be situated near its new source of stimulation."

Bierens de Haan [3] writes:

" ... that the weaving of the web by the spider is purposeful for the catching of insects, and the collecting and storing of caterpillars by the wasp purposeful for the nourishing of its future larvae, are facts that are so self-evident that it is not necessary further to elucidate them."

The evidence that is occasionally mustered to oppose the notion of directivity consists of examples of the behaviour of systems, ostensibly contrary to their personal interests, but that, if examined more closely, are seen to be in the interest of the more general system of which they are part. Indeed, if the sub-system is regarded in vacuo, its behaviour may not appear directive. If it is regarded, as it should be, as a differentiated part of a larger and longer-term system, its directivity then becomes apparent.

Thus, for instance, it is argued that during the mating season, the male stickleback undergoes colour changes that render him conspicuous and hence more vulnerable to predators. [4] It has been shown that the object of the colour change is to attract the attention of females.

That the stickleback has enemies who have learned to take advantage of this conspicuousness (as the predator's behaviour is also directive) is only to be expected and does not detract from the directive nature of its colour change for breeding purposes. The latter remains adaptive so long as the breeding advantages to be derived, from it outweigh its disadvantages for the purposes of phylogeny.

An infinite number of examples of the same principle can be cited, thus:

• Certain fish learn to tolerate smaller fish that enter their mouths and clean their teeth. This is known as 'cleaning symbiosis'. However, predators have 'learned' to imitate these cleaners, and have grown to look exactly like them. They are consequently tolerated by the larger fish, a fact they take advantage of by taking an occasional bite at their unsuspecting hosts. [5]
• In many species of ants, specialised workers have evolved to look after the larvae. Certain cuckoo-like parasitic beetles, incapable of looking after their own larvae, lay their eggs in the ants' nests. These later hatch into larvae that are indistinguishable from the ants' and which, after having been carefully looked after by the workers, hatch into predator beetles that gradually take over the Colony. [6] [8]

These are but two of an infinite number of examples of parasites that take advantage of certain features of a host's behaviour pattern. Does this mean that these features are not directive? Undoubtedly not. It is clear that cleaning symbiosis is very useful to the host; it is also clear that looking after the larvae is a necessary function within an ant colony and is directive to the survival of the young. The fact that, for these functions to occur successfully, a number of individual members of the species will fall prey to parasites is no argument against their usefulness.

Such behaviour only appears non-directive if we regard the individual in vacuo, i.e., apart from the family or the community of which he is part, which we know to be impossible.

Again, it is pointed out that the fierce competition obtaining in certain animal societies for the possession of the choicest female or of the most desirable territory is not conducive to the survival of the individual. Indeed, in such competitive societies as those of the baboons or fur-seals, casualties often can run quite high, especially under conditions of overcrowding. [7] But such behaviour can only be interpreted as contributing to the selection of the fittest individuals and thus to the adaptation of the species as a whole to the challenges of its environment.

It is also occasionally pointed out that in certain species the individual at one or more stages during its life-cycle, is subjected to so many environmental challenges that its chances of survival are in fact minute. This is especially the case with certain parasites. Miriam Rothschild and Teresa Clay write:

" ... the eggs of the grouse roundworm lie scattered all over Scotland, but millions and millions of their young, which hatch out and wriggle up the sprigs of heather around them, perish because that particular plant is never eaten by a grouse. Similarly vast numbers of immature ticks cling hopefully to blades of grass, waiting for the millionth chance which will bring an animal brushing through the vegetation within reach of their waving forelegs.

Owing to the difficulty of finding a host-a difficulty which is superimposed on the more familiar hazards of life- the mortality among most parasites is enormous. A vast number of eggs or larvae have to be produced in order that the species can survive at all. Consequently, a characteristic feature of most parasites is a relatively enormous development of the reproductive organs, which frequently come to dominate the body. Intestinal worms produce eggs by the million and even brood-parasites like the cuckoo, lay four to five times as many eggs as their hosts. The difficulty of host-finding can often be estimated by the number of eggs laid." [8]

Surely nothing could be more directive than this automatic regulation of the number of eggs laid in accordance with the number required to produce the optimum number of adults. Once again, directivity is apparent if one realises that the unit of analysis must be the larger unit-in this case the species as a whole, four-dimensionally--and not the individual.

Other arguments against directivity are based on the disadvantages to individual survival of the so-called inflexibility of instinctive behaviour. Thus Hingston [9] tells of a clubionide spider in Central India. These spiders live in grassy meadows. They are the same colour as the grass and are capable of lying in a particular position that enables them to blend perfectly with their background. When threatened, their instinct is to remain perfectly immobile and thus hope to pass unnoticed. Hingston found that, in such circumstances, there was no way to make them move, neither by pushing them with a straw, by sticking a pin into them, nor even by cutting off one of their legs. They would inevitably remain quite immobile.

Canis azarae, the pampas fox, apparently behaves in a similar way. Now, can one say that such behaviour is not directive? Undoubtedly not, statistically; therefore, from the point of view of the species, it must constitute the reaction most conducive to survival.

A further example is the phenomenon of blinking. The human eyelid closes to prevent a foreign particle from entering the eye. The performance of this task suffers from the same shortcomings as does the behaviour of the famous insectivorous plant, the Dionaea fly-trap. Neither system can distinguish between the various foreign particles, most of which are harmful, but some of which could conceivably be beneficial, such as the medicinal drops which an occulists may wish to insert into a diseased eye.

Does this detract from the usefulness of the blinking function? The answer is no. The experience of phylogeny has established that, statistically, blinking, like digestion and the circulation of the blood, is best mediated at a low neurological level. The possibility that a foreign particle entering the eye might be beneficial is so remote that it is best not taken into account. The cost of doing so, in terms of an increase in the size of those cerebral mechanisms required for increasing discrimination, would just not be worthwhile.

Indeed, in spite of the inflated view we may have of human intelligence, it is probable that if this 'faculty' were allowed to govern all those elaborate processes necessary to sustain life, which are at present mediated by lower centres in our brain and spinal cord, the result would undoubtedly be a serious increase in inefficiency.

Blinking may appear indiscriminatory, but this lack of discrimination is a low price to pay for the advantages of automatism and for the protection it enjoys from the ravages of 'intelligent' behaviour that is at present wreaking such irreparable damage to the less well-protected parts of our biosphere.

References