October 23, 2017

A model of behaviour (short version)

This paper outlines Goldsmith’s general theory (or model) of behaviour, which provides the basis for predicting change in behavioural systems, and in particular human social systems, in the process of disintegration.

A longer version of this article can be found here.

Note: click on the figures to enlarge them.

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 homeorhesis (from the Greek: rhesos, 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 (see Fig. 1).

Fig. 1. Stability

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 “weltanschauung”. 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.


A system is defined as a 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 organized, 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 (See Fig. 2).

Fig. 2. A self-regulating system

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, worldview, or weltanschauung, interacting with its environment both internal and external. Institutionalised external controls, except superficially and in the short-term, are ineffective in counter-acting behavioural tendencies associated with a given model or world view.

Back to top

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 subsystem.

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 hand 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. As the system differentiates, as a uni-cell, for instance, evolves into a complex metazoa, or as a foetus develops into an adult, so it becomes capable of dealing with ever more serious environmental challenges. This means that the values will increase. One can draw a graph to illustrate this (See Fig. 3).

Fig. 3. Environmental parameters

In Fig. 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 by 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 Fig. 3, the periods lying between these critical points have been separated and labelled. The labels used, paleo-, 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 affecting 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.

Back to top


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.

Back to top


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.

Back to top


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.

Back to top


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.

Back to top


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 low. 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.

Back to top

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.

Back to top

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.

Back to top

Alien Sub-Systems

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 an 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.

Back to top

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 cause 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 less 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 function 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 are 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.

Fig. 4. Dealing with surplus energy and random information

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.

Back to top

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 they 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 to 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 worldview.

Back to top

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 Boyden refers to such reactions as pseudo-adaptations. They tend towards an equilibrium position, but 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 response are mediated at a higher level of complexity, i.e. at the level of a human society, is an interesting one.

Fig. 5. A model of a modern society

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 reinterpret in such a way as to render them tolerable. What is true of individuals is also true of society. A society will tend to interpret asystemic situations so as to 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 worldview is almost certain to be interpreted in such a way as to ensure their reconciliation.

Back to top

The Runaway Social System

Unfortunately the system’s instability will increase by positive feedback. This means that the asystemic corrections required 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 4 billion acres of desert have been created in the last 70 years.

Fig. 6. A model of science

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 system’s instability will increase by inevitably be caught up in this process, and introduce further positive feedback into the system. The process continues until the disequilibria are so serious that no expedients are available to prevent collapse.

  • Twitter
  • Facebook
  • Digg
  • Reddit
  • StumbleUpon
  • Diaspora
  • Identi.ca
  • email
  • Add to favorites
Back to top