Published in The Ecologist Vol. 5 No. 2, February 1975.
We live in an Age of Faith, not in God but in Science. If most of us are still capable of facing the mounting problems of the world today with relative. equanimity, this is because we believe that Science will provide us with the means of solving them: half a millennium ago we would have expected God to do so.
Scientists are functionally the priests of our industrial society. It is only they who are capable of mobilising, for our purposes, the limitless powers of Science, of acting thereby as the intermediaries in our relationship with this new and formidable deity.
It is not surprising that their writings are imbued with an aura of sanctity previously reserved for the holy texts of the established religions. If a proposition is classified as ‘scientific’, then it must be true, indeed incontestable. If, on the other hand, something is branded as ‘unscientific’ then it must be the work of a charlatan. This has provided the Scientific Establishment with the power to prevent any undesired deviation from scientific orthodoxy. In the same way, the Catholic Establishment of the Middle Ages would excommunicate a heretic whose teachings constituted a challenge to their authority.
Indeed, one finds among the annals of the Scientific World some which are strangely reminiscent of mediaeval witch-hunts. Consider, for instance the response of the Scientific Establishment to the publication of Limits to Growth. It was branded as unscientific by both Nature and Science, the world’s two most prestigious scientific journals.
In Britain the inquisition was led by Lord Zuckerman, once Chief Scientist to the British Government. It is easy to see how he exploited the terms ‘scientific’ and ‘unscientific’ to discredit this very important work in the following outbursts in a speech delivered in Stockholm during the 1972 United Nations Conference on the Human Environment:
“Our newspapers urged on by a plethora of pseudoscientific books, articles and speeches are filled with items which warn us that irreversible damage is being done to our physical environment . . . I have referred to a book Limits to Growth which has been hailed . . . mainly by the scientifically uninitiated as a scientific statement about man’s environmental problems . . . for my part I have no hesitation in saying that I am among those professional students of environmental problems who dismiss the book as unscientific nonsense” 
What Is ‘Science’?
In view of this, it is clearly important that one should know just what ‘Science’ is, and precisely how one determines what constitutes a ‘scientific’ proposition.
‘Science’ does not appear to have ever been adequately defined. In general, it seems to involve the accumulation of knowledge. But what is knowledge? Here we encounter a major snag: to answer this question we must leave what is generally regarded as the realm of ‘exact science’, and enter that of Epistemology or the Theory of Knowledge.
For scientists to regard Epistemology as being outside the scope of Science is to renounce the responsibility for examining the assumptions on which their work is based, for determining in fact, to what extent it is justified.
This task is delegated to people who, working outside the field of Science, know very little about it, and who like most specialists today, tend to regard their field of study as largely autonomous, as something that can be studied in isolation from everything else. As a result one finds little in current epistemological writings which can serve to provide a theoretical basis for Modern Science – a lamentable situation. As Einstein wrote “Epistemology without contact with science becomes an empty scheme. Science without epistemology – in so far as it is thinkable at all – is primitive and muddled.”Back to top
What is Knowledge?
Knowledge is clearly some sort of information. To qualify as knowledge, however, this information must display certain characteristics. According to Ayer, who appears, to be one of the principal spokesmen for the modern school of Empiricism, it must be true, we must know it to be true, and for the right reasons. This implies, above all, that knowledge is conscious information of some sort. This is presumably the only type of information which can be studied empirically. Also, it is by basing one’s behaviour exclusively on such information that one is regarded as acting ‘rationally’.
If epistemologists knew a little about such subjects as Cybernetics, Ethology and Psychology, they would realise that conscious information plays by no means a determining role in the behaviour of even the most sophisticated members of the species ‘Homo sapiens‘.
To understand the use of conscious information without reference to that of unconscious information is simply not possible. In fact to understand the use of information in the brain is difficult without examining it as part of a general theory of information, which must mean examining the way, it is built up and made use of by systems at all levels of organisation. Such a study would reveal that information in the brain is built up and used in very much the same way as it is in a gene-pool or a fertilized egg and that there is only one way of organising and using information among natural systems.Back to top
The reason for this is that information is only built up for one purpose, and that is to constitute a model of the relationship of the system, of which it is part, with its particular environment. Information is, in fact, of no value by itself as a basis for behaviour. To identify a technological device as being a nuclear power station, for instance is of no value if one has not previously built up a model of the relationship between a nuclear power station, the biosphere of which we are part and the rest of the technosphere of which it is part. Only in this way can one understand what are its implications and hence how we should react towards these diabolical contrivances. It is a serious illusion to suppose that the mere fact of attaching a label to something provides information about it.
If information is only built up for a single purpose, this is also true of the model of which it is part. A model is only built up for the purpose of serving as a basis for the control of a system’s behaviour towards its environment.
This whole notion of control is largely ignored by epistemologists as well as many scientists who have implicitly adopted the empiricist position. There is a good reason for this. If a system is controlled, it must mean that it is goal-directed or purposive, for what else can control mean but to keep something on its correct course? And how can it be kept on its correct course if it doesn’t have one? The goal, needless to say cannot be pin-pointed in space-time. It is simply that course along which discontinuities and their corrections are reduced to a minimum. By taking such a course a system is capable of maintaining its basic structure in the face of environmental challenges, i.e., of remaining stable.
It is also by taking such a course that free energy is reduced to a minimum over a long period. In this way the system remains in four-dimensional equilibrium with its environment. This principle of directiveness is irreconcilable with empiricist philosophy, since it cannot be induced on the basis of observation, i.e., according to what empiricists regard as the only legitimate way of building up knowledge.
Also, it justifies a methodology for building up knowledge which is in competition with induction. I refer to deduction from the general principle cited. Thus one could postulate that to maintain its stability in specific environmental conditions a system must be able to achieve a given set of sub-goals. A specific behavioural act could therefore be explained in terms of its contribution to the achievement of a sub-goal, and judged in accordance with its ability to do so. This is in fact the cybernetic as opposed to the reductionist approach.
To reject the directivity principle, however, is to reject the very principle of organising information, and hence, among other things, the possibility of Science. The reason for this is very simple. Information is built up out of data, the raw materials of information. Data, as we shall see are interpreted in the light of a system’s model of its relationship with its environment, and then constitutes information. This means putting ‘mental order’, I suppose one might call it, into what might previously have appeared to be random data. This is only adaptive if this order corresponds to something, i.e., if it reflects an ordered situation. Since behaviour is, by its very nature, dynamic, i.e., involves change, this change must be orderly which means that it must be heading in a given direction.Back to top
The mechanism of control
Cybernetics has probably contributed more than any other discipline to the unification of Science by demonstrating that control, at all levels of organisation, is achieved in the same way that, in fact, the basic cybernetic model is of universal application.
Data are obtained, transduced and interpreted. A hypothesis or model is postulated and projected back onto the data, followed by a modified hypothesis and a further projection. Each time the hypothesis is made to fit better with the general model of the system – either by modifying the former or the latter. This can be repeated over and over again, and in this way there will be a continual monitoring of a series of even better hypotheses formulated after successive accretions of information.
This process gives rise to a damped system, i.e., one in which errors are progressively reduced. If interpretation is taken as tending towards a position of four-dimensional equilibrium, i.e., along an equilibrium course – which we can represent by a straight line – it will in fact take the form of a series of oscillations of ever-diminishing size – tending towards the reduction of errors – i.e., the development of an ever better representation of the system.
On the other hand, if this mechanism does not function properly, i.e., if the system gets out of control, then the oscillations will increase in size. This of course cannot continue indefinitely, the discontinuities would become increasingly insupportable and the system would eventually collapse – just as is happening to our society today.
At this point it might be worth noting that for two million years or so, human social systems displayed considerable stability. Unstable social systems appear to have been largely confined to recent times, i.e. to the period following the Neolithic revolution.  Even during this period, traditional societies which have succeeded in remaining outside the orbit of mainstream civilisations have continued to display considerable stability. Such stability can only be achieved in one way, and that is by the operation of a control mechanism of the type described above.
This mechanism is provided by its culture, of which an essential component is a specific world-view or model of the society’s relationship with its environment, a corresponding goal-structure and a means of achieving it. 
Science appears to be an attempt to replace the cultural information which makes up traditional worldviews which is very different in the case of each traditional society with a single organisation of information which should theoretically serve each of them equally well. It is an attempt, in fact, to substitute objective for subjective information as a basis for control.
Such a substitution has many implications, which I shall look into in this. article. First of all, however, let us consider what can conceivably justify it.
Epistemologically, the answer is fairly obvious. Traditional information does not qualify as knowledge. It is only true vis-à-vis a largely subconscious and very subjective model and not vis-à-vis a conscious objective one.
It involves reference to such things as Gods and spirits whose presence is empirically unverifiable, while it establishes a strange set of cause and effect relationships between men’s ritual activities, the behaviour of these Gods and spirits and day to day biological-social and ecological events, which can be shown to be ‘irrational’.
Objective truth must be the overriding criterion for judging the validity of information. We all assume that this must be so.. . . but why? On what is this assumption based? I shall show in this article that this is epistemologically unjustified, in fact that it is a pure act of faith.
If cultural information is organised subjectively rather than objectively, it provides a society with a very restricted view of its environment, that which it has so far required for its own specific adaptive purposes. If information be organised objectively, it is assumed that it provides a faithful reproduction of the outside world, which should provide the basis for a much wider range of adaptations, enabling a society in possession of this objective information to adapt to all possible eventualities.
For this to be so, a number of obvious conditions must clearly be satisfied:
- The first is that objective information can be organised to constitute an effective model of Man’s relationship with his environment.
- The second is that this model can effectively be made use of to control this relationship.
- The third is that this would enable Man and his society to adapt to important and improbable changes, i.e., that there are no other limits to their potential for adaptation. I shall show in this article that all these assumptions are false.
Science has not provided so far a means of controlling society
We are living in an age in which public policy has, for the first time, been largely influenced by Science. In spite of this, society is increasingly out of control, hence unstable, and the discontinuities we are subjected to are on an ever increasing scale, so much so that we are well on the way to social and ecological collapse. On strictly empirical grounds one cannot avoid the conclusion that Science has been a failure.Back to top
Objective truth does not appear to be the relevant criterion for judging the value of information used by natural systems. Genetic information is not objective. It is specific to a particular individual as a member of a particular species. Cultural information is not objective either. In fact one can go further and say that as the biosphere has evolved out of the primaeval dust, as complexity and variety have built up, so has the corresponding subjective information determining this evolutionary development. To substitute a single organisation of objective information for the multitudinous organisations of subjective information, which are at present used to control the behaviour of natural systems, would be to reverse evolutionary trends and hence to foster informational entropy.
It could be objected that objective information also displays order or negentropy. This is true only in the sense that the technosphere displays order. Order is but another word for organisation. Things are organised for a particular purpose to satisfy a particular goal. Random organisation is a contradiction in terms. Now the goal of the technosphere is very different from that of the biosphere. The former is designed to provide Man, one of the myriad forms of life which inhabit this planet, with the maximum comfort and convenience, whereas the latter being concerned with the maintenance of its overall stability is concerned with catering for the requirements of all, not just one of the forms of life which constitute it. They all have an essential function to fulfil within it.
Both organisations of matter are in fact in competition with each other since the former can only maintain itself by extracting resources from the latter and consigning to it its waste products. From the point of view of the biosphere the technosphere constitutes waste or randomness or in fact entropy. Similarly from the point of view of that organisation of subjective information associated with the biosphere, objective information constitutes entropy. It has played no role in building up the biosphere on the other hand without it there would be no technosphere.Back to top
Why should this be so?
In the biosphere, a system could adapt either by modifying its environment to satisfy its own requirements or by itself undergoing change to satisfy these requirements of a changed environment. In practice both strategies are resorted to. We have seen that the justification for Science must be to permit rapid and radical adaptation which a traditional culture would not do. When rapid and radical change is required however the former strategy is the only possible one. It is quicker and easier to adapt man to living in a cold climate, for instance, by building him houses with central heating, double glazing, etc. than by bringing about those physical changes in him which would enable him to support the cold weather. Thus in practice objective information simply permits environmental change. If this environmental change were not regarded as desirable there would be no reason for replacing traditional subjective information with scientific objective information. Objective information can thereby serve but as a basis for reducing the order of the biosphere of which we are part.
Practical people will not accept so strongly worded a condemnation of the technosphere. They will insist that by clever management it is possible to reconcile the two, to achieve some sort of a compromise between the rival goals of the biosphere and the technosphere. Besides it will be maintained, human ingenuity, combined with omnipotent Science must be able to compensate for the reduced stability of the biosphere, by devising means of correcting the resultant discontinuities.
The failure of Science could be rectified if our knowledge were improved, if Science were reorganised, if, for instance, it could be unified to provide us with a single model of the biosphere in its interrelationship with the technosphere.
This too was my feeling twenty years ago. In fact, I spent many years working out a general model of behaviour which led to my book (unpublished) The Theory of a Unified Science. Since then, however, my views have changed. This is why.
Back to top
The Empiricist thesis
Firstly it would mean abandoning the empiricist thesis which underlies science. According to Empiricist epistemology, knowledge can only be obtained by observation. This thesis is manifestly false. The child’s mind at birth is far from being the ‘tabula rasa’ which this would imply; for instance, a child is, and must be, in possession of a rudimentary model of its relationship with its environment. This has been shown to be true experimentally. Fantz,  for instance showed that chicks are born with a tendency to peck at certain objects rather than others – but we do not need such experience to prove this point.
If a species is to be stable which means if it is to display continuity, i.e., remain stable, then its behaviour must be based on information which itself displays continuity. Each generation must inherit information which reflects the experience of its ancestors going back into the mists of time. That is why genetic information is so stable. If it were more plastic and could permit adaptations to changes based on the experience of one or two generations only it might give rise to adaptations to what could turn out to be freak conditions unlikely ever to recur – such as the situation created by our industrial society. Its behaviour would display no continuity. It would be unstable – and hence could not survive.
Building up information means improving the model so that it may serve as a basis for ever more adaptive behaviour. This improvement, contrary to what Empiricists may think, does not simply consist in accumulating more data but also in organising it within the model.
The establishment of a new relationship between two of the variables of a model, for instance, permitting an improved explanation of certain observable data must provide information without involving the simultaneous detection of data – thinking – in other words.
The notion that all information is only built up by observation is thus simply not true. However one can go further than this. What is the status of that information which is built up by observation?
The first feature is its subjectivity. Detection is a directive process. It does not consist of randomly accumulating data, but rather of isolating from all possible data that minute fraction of it which appears relevant to a system’s behaviour pattern. Thus, as Judson Herrick points out:
“The skin is sensitive to mechanical vibrations up to 1,552 per second, but beyond that point feels only a steady push. The ear is aware of sound travelling by wave lengths of 13mm up to 12,280mm, but does not hear sounds below or above these limits. The skin is aware of heat-waves only from .0008 to .lmm long. The eyes take cognizance of light waves from 0008mm to .0004mm, but miss electric waves, ultra-violet waves, x-rays, gamma-rays and cosmic rays, running from wavelengths of .0004 to .000,000,000,008mm”. 
This first genetically determined selection is complemented by a further ontogenetically and largely culturally determined one. Thus at any given moment we can detect but a minute fraction of those data which we are genetically equipped to detect – those which may appear relevant to our behaviour in those particular circumstances in which we find ourselves at a given moment.
This leads one to the essential consideration that detection can only occur on the basis of a pre-existing model, more precisely two models – a genetic one and one which might be called a cerebral or cultural one, in terms of which is calculated the relevance of different data to a system’s behaviour pattern – yet another reason for rejecting the Empiricist thesis.
Data in fact are detected and interpreted in order to confirm or invalidate the model postulated. One might however ask on what basis is the model postulated? The answer is that which appears most probable in the light of the model one has built up during the course of one’s experience.
A brain, like any other organisation of information made use of by natural systems, acts among other. things, as a probability calculator. Back to top