The Psychiatrist in Search of a Science

British Journal of Psychiatry, 1972, Vol. 121, pp. 591‑8 (I); 1973, Vol. 122, pp. 625-36 (II); 1975, Vol. 126, pp. 205-24 (III) 

I: Early Thinkers at the Maudsley

   Until quite recently the prestige of Science was enormous, and in its glow basked innu­merable earnest workers who firmly believed that their meticulous ant‑like activities were advancing the health, wealth, welfare and spiritual ascent of man. In the last few years that prestige has quite suddenly collapsed. We do not deny the loftiness of aim which inspired our forerunners, but we have begun to realize that the secondary consequences of the most nobly motivated activities may lead to disaster. The main consequence of the advance of scientific knowledge has been the proliferation of technologies which have armed and powered a materialistic culture in the exploitation and progressive erosion of a fragile living environ­ment. The scientist of an earlier generation had some notion of seeking after an aspect of the truth, and relied in the main on his own insights and ingenuity. Today, all too often, he knows himself for an expendable member of a professional team, depending on the routine deployment of technical resources for the manipulation of an established paradigm.

    But if we look back to see what Science meant for our forebears, we might recapture some of their idealism. We would see what Science could be, a self‑forgetful journey of discovery into the nature of reality, the cosmos in all its multifarious wonder, and that infinitesimal part of it that is ourselves. The enterprise calls for attitudes now almost obsolete in scientific working communities: dedication, reverence for truth, humility. With those attitudes great advances were made in our understanding of nature and of ourselves; and where they still obtain, as in the study of the heavens, advance continues. In the main, however, scientific activities are directed towards short‑term ends, and are mission‑oriented, the more so the more advanced the field. If Science is going to help us out of the mess that the daughters of Science have made, such severely practical projects will still be the principal part of its business. Theo­retical and philosophical questions are only likely to get serious attention in the early days of scientific development in a new field, or when the search for knowledge and understand­ing runs up against a serious check.

   The practice of medicine and the healing arts relies on the medical and more generally the life sciences for the knowledge and under­standing which are needed for success. It is the unique feature of Science that in the long run errors are corrected, information grows, and new and more general theories take the place of ones which are more limited or ad hoc. This process goes on in psychiatry as elsewhere. We have, for instance, seen that one‑time scourge, general paresis, dealt with by more and more efficacious methods, from symptomatic begin­nings proceeding through infection with malaria, hyperthermia, antibiotics, to a point now where the aim is less frequently to treat than to totally prevent. These advances have depended on an increasing understanding of the prime cause of the disease: the tissue reaction to the specific parasite. Accessory causes, such as variations between individuals in constitutional suscepti­bility, have now only minimal importance. Aetiology is more significant than explanation. Similar increase in understanding and in power to treat and prevent have emerged in the mental pathologies accompanying the chemical deprivations and intoxications. Psychiatrists hope to increase the scientific element in the theories on which they work; and they include in their number not only those who heal but also those who claim to be single‑minded seekers after truth. Such claims have as regularly been combated, not only by others vis‑à‑vis the psychiatrists, but also among psychiatrists inter se. The arguments are ones which we should clearly understand, since it is only in that part of our work which is scientific that we can hope for that additive self‑corrective process of growth to go on which helps us to help others.


    This was the background to Mapother's Presidential Address to the Section of Psychiatry of the Royal Society of Medicine, delivered on 14 November 1933. His title was 'Tough or tender: a plea for nominalism in psychiatry'. He started from the thesis that the whole business of science is to account for events as a sequence in the phenomenal world; and he set out to controvert the position taken up by Bernard Hart in his Presidential Address to the same Section two years before. He quoted from it the following striking appreciation of the work of Sigmund Freud, particularly as a contribution to science:

Freud introduced dynamic concepts while still remaining within the psychological field. He did not confine himself to those first two stages of science, the observation and classification of phenomena, which had hitherto constituted the only ground in which the psychologist was allowed to play. He advanced to a third stage, construction of causal concepts designed to explain the observed phenomena. Moreover, these concepts were built out of psychological stuff and not handed over in despair to the ministrations of the physiologist and the chemist. This was an epoch‑making step, and one which if it can be established at once raises psychology to a level with the other scientific disciplines which serve psychiatry.

    Read today, this passage oppresses one with the melancholy undertones of dramatic irony; psychoanalytic psychology is no nearer to being on a par with the other scientific disciplines which serve psychiatry than it was when it was so acclaimed by Bernard Hart. While he hoped that psychological phenomena could (and should) be reserved from causative explanation by physiologists and chemists, we now realize that we shall get nowhere without their help.

    Mapother also quoted another passage in which Hart spoke of the unprecedented successes then being obtained by the psychoanalyst in the treatment of the neuroses, and perhaps no less in the psychoses. Mapother denied that there was any demonstrable progress in treatment. The desirability of explaining psychological phenomena solely by psychological causes, so strongly felt by Hart, meant nothing to him. He objected to the purely psychological Freudian concepts on the grounds that they were animistic, and so were primitive and a pre‑scientific way of thinking. Nevertheless, he mentioned animistic concepts in contemporary physics as a com­parable manifestation of the Zeitgeist. He too comes under the shadow of dramatic irony, since the sub‑atomic physicist of today seems as prone as ever to help himself out with just such animistic counters and symbols.

    Mapother maintained that the living form of conflict and controversy in psychology and psychiatry, indeed in science, lay between nominalism and idealism. He began by dis­missing materialism out of hand. He described this as 'the doctrine that the physical world has independent reality; that nothing exists except matter and its movements and modifi­cations and that the phenomena of conscious­ness are wholly due to the operation of material agencies'. He asserted (quite unphilosophically) that this philosophical theory was 'defunct‑if indeed it ever existed, except as a mythical butt'. He would also have nothing of idealist doctrines. He would not accept the so‑called psychological determinism of Freud, and said that to affirm the entire independence of a mental set of phenomena from all others is not determinism in the scientific sense. He regarded the psychologies of McDougall, Freud and Hart as all being animist psychologies, all based metaphysically on an idealist conception, con­cerning themselves almost exclusively with subjective subject matter with the aim of giving a self‑contained account of the sequence of sub­jective processes. While he admitted that there was plenty of observational data in psycho­analysis, he insisted on the relative paucity in all three psychologies of observational matter in proportion to the prominence of interpretation.

    So Mapother found himself driven (one feels, rather reluctantly) into adopting nominalism as the only tolerable way out. Nominalism he described as the view that universals or abstract concepts are mere names without any corre­sponding realities. Phenomena are the only objects of knowledge, and by phenomena are meant the immediate processes of perception. For the nominalist, matter and its supposed constituents‑molecules, atoms, electrons‑are conceptual fictions. Similarly, mind is a con­ceptual fiction: the thoughts themselves are the thinker. Words are just symbols. A 'thing' is merely a term for a recurrent pattern of con­current sense impressions; abstract nouns are the terms for the very fact of recurrence of such concurrent patterns. He said:

the goal of science is the production of formulae summarizing the maximum number of past phenomena in the simplest, most concise and most frugal manner possible, and enabling us to foretell the sequence of future phenomena with the maximum economy of thought.

    Methods of observation (1) must be public, (2) must permit of discrimination how far any observation is truly impersonal, and (3) must ultimately lend themselves to quantitative results.

     Mapother's standpoint is closely similar to that of a modern scientific positivism. Ernst Mach held, according to Passmore (1968), that:

if the scientist is led by its [atomic theory's] 'successes to suppose that atoms have a reality of their own then he is crossing the boundary which marks off the fruitful fields of science from the marshy wastes of metaphysical speculation. Absolute space, absolute time, causality even, so Mach thought, must go the way of atoms. In Nature there is neither cause nor effect; Nature merely "goes on". A developed science will express its conclusions as functional relationships; aseptic formulae replace the "causal links" of metaphysics.

     This is a chilling formulation. If atomic and subatomic particles are no more than mental constructs, they do not have the provocativeness of a supposedly physical existence, with elements of paradoxicality, mystery and refractoriness, which suggest and demand further enquiry. Karl Popper (Conjectures and Refutations, 5969) (CR) has pointed out that the advance of research, which is based on the idea of concrete physical realities, stands in conflict with Machian scientific positivism:

a theory may be speculative, and based upon the fundamental (Parmenidean) principle that the world as it must be understood by argumentative thought turns out to be different from the world of prima fade experience, from the world as seen, heard, smelled, tasted, touched; and... such a speculative theory may nevertheless accept the empiricist ''criterion'' that it is the visible that decides the acceptance or rejection of a theory of the invisible (such as the atoms). This philosophy has remained fundamental to the whole develop­ment of physics, and has continued to conflict with all "relativistic" and "positivistic" philoso­phical tendencies.

    We can take a modern example from gene theory to illustrate Popper's point. Genes started as purely theoretical constructs, specified only in a dispositional or relative way; the purpose of the construct was to help one to think purposefully about the observations. Then they became associated with the chromo­somes, of which they were thought to be a part. Then it was found possible to associate them with individual bei on the chromosomes, so that they could be arranged in a linear order. By then, there were few workers who had any doubts of their physical reality. And lately DNA theory has connected them more closely, via messenger RNA, with the function of blue­printing the building of proteins. But even now they are to some extent theoretical, since we cannot say that they have been directly observed. The instrumentalist view of genes, that they are no more than operational definitions, can no longer be sustained. O'Neil (1969) writes: 'Scientists act as though hypothesized entities (terms) were real.   They do so in the way they test such hypotheses.'

    What this comes to is that Mapother's nomi­nalism cannot be faulted logically, but it fails in giving an incorrect description of the way scientists think, and it fails heuristically. As Popper has said (CR), scientific theories are serious attempts to discover the truth. By scientific work we discover further worlds, beyond the world given us by the immediate responses of our senses. Scientific theories are genuine attempts to describe these worlds. All these worlds share reality with our ordinary world, and are, if you like, equally real aspects or layers. It would be wrong to say, as Edding­ton did, that this table here is only an appear­ance, and is 'really' empty space in which atoms are vibrating; but equally wrong, and wrong in the same kind of way, to say that the atoms are conceptual constructs only, instruments for description and prediction, and nothing more. Popper (CR) writes:

Instrumentalism is unable to account for the pure scientist's interest in truth and falsity. In contrast to the highly critical attitude requisite in the pure scientist, the attitude of instru­mentalism (like that of applied science) is one of complacency at the success of applications. [1]

     Scientific theories are not the same as techno­logical rules of computation. Theories are tested by attempts to refute them; technological rules are not refutable, but merely become inadequate for certain purposes. Newton's theory has been refuted, but it can be and is systematically used for instrumental purposes within the limits of its applicability. Instru­mentalism cannot account for scientific progress.

    Mapother's proposal fails, then, to give a satisfying picture of the human mind at work in trying to understand the world around, and it fails to give that kind of foundation which feels firm enough to step off from into the unknown. If the psychologies of McDougall, Freud and Hart are to be deemed unsatisfactory, it will not be for the reasons that Mapother gave.


    We now come to an entirely different attempt to relate psychiatry to science, this time along nihilistic and pessimistic lines. Three years after Mapother's address, on 25 November 1936, Golla answered him in the Eighteenth Maudsley Lecture, under the title of Science and Psychiatry. He said:

The proposition that I wish to put forward is this: Scientific method alone is incapable of dealing with the whole of the personality, and in so far as we attempt to describe psychological processes in terms of scientific epistemology we will arrive at definitely false conclusions.

    In cases of maladjusted personality, in which pathological methods had failed to demonstrate a cause, Golla disputed the possibility of a description in terms of science. Physical and chemical terms prove inapplicable. One cannot proceed by analogy from the behaviour of the single neurone. He poured scorn on psychiatrists who would attempt to express psychology in terms of neuronic function. He attacked mental mechanisms, psychological atomism, the reflex arc model of behaviour. Even at the neuro­physiological level we have to regard the nervous system as an organic whole and not as an integration of reflex arcs. The concept of the conditioned reflex, he said, had proved to be of very little utility to the psychologist.

    For Golla, the advent of Gestalt psychology came like a refreshing breeze. He formulated its significance like this:

There exist natural circumstances in which what happens in the total is not conditioned by the nature of the parts or their mode of combination, but, on the contrary, what occurs in any part of this whole is determined by the inner structural basis of the entirety.

    What this amounts to is that scientific attack, being analytic, is bound to fail when faced with the problems of human psychology. Even though we may substitute for the reflex arc model a more satisfying descriptive psychology, we can still only apply scientific methods to our concepts by an analytic procedure that is in­admissible ex Fp'pothesi in a psychology where the parts depend on the properties of an organized whole. Science, being a process of analysis and classification, cannot deal with anything as a concrete whole. No amount of scientific data could help us ultimately to know a personality, self‑evidently so if we think of someone we know intimately. Our power to predict future con­duct or to understand history depends on the qualification of all experience by the conceptions of value and purpose. These are not concepts that can be deduced from science. He felt, indeed, that there are some things which should be sacrosanct from scientific enquiry:

We have found a way from the dead world of science to the living world of purpose and values where ethical, aesthetic and religious considera­tions do not need to be deliberately excluded in the interests of a highly abstract account of conduct, whether normal or psychotic… When the grosser forms of the older religious superstitions and the authority of its priesthood decayed before science and historical exegesis, mankind celebrated its freedom by taking to itself a new superstition in the shape of scientific materialism, and a new priesthood of professors just as arrogant and intolerant as their pre­decessors.

    Golla ended his discourse on a somewhat mystical note. Explanation, he said, must always be a transcendant principle. Thus the success of a religious system will depend on its power to offer an explanation; perhaps this is the direction from which we have most to hope in our attempts to comprehend personality. Mal­adjusted personalities, or some of them, can be envisaged as 'due to discord between the purposive activity implied in all personal conduct and the values imposed by the universe considered not only in its social but its ethical and aesthetic aspects'. He hoped that psychi­atrists would be able to hand down the great humanistic tradition to successors who would be not only scientists but men and women of culture and spiritual insight.

    From this we can see that Golla's psychology would have been looked on by Mapother as vitalistic, just as Golla looked on Mapother's attempted founding of psychiatry on neurophysi­ology and mental mechanisms as not only doomed to frustration but also a kind of bar­barism. Golla's nihilism is unjustified because he has mistaken, or has at least misstated, the way in which science works. The processes of analysis do not have to be founded on firmly based elements; the concepts employed may be themselves global, even vaguely envisaged, and made workable by operational definition. Golla was asserting both the necessity and the impossibility of reductionism, when it is at no time indispensable and at some times is possible. Whether an ultimate reductionism is an impossi­bility or not is still an unsolved problem a generation later than Golla, but a problem which has largely lost its urgency.

    Largely, but not entirely. Golla's rejection of reductionism has recently been echoed by the neuropsychologist, Richard Gregory (1971). He asks whether we can understand organisms purely in terms of the interactions of their cells, and whether we can understand societies in terms of the interactions of their people. The answer must be 'no' to one or another or both of these questions, since if in both cases the answer were 'yes' then we should be able to describe our society in terms of its sub‑members, all our cells, which is absurd. Cells, organisms and societies must each be described with the aid of different concepts; and the concepts appropriate to one kind of gestalt cannot be 'reduced' to those used for the others without loss of understanding. The argument can readily be accepted, but it does not lead to the conclusion that Gregory wishes to draw. We may, from what he says, conclude that un­reduced concepts may be essential for scientific work at a given level, not that their reduction is impossible. The tasks of higher mathematics are carried out, for instance, with the aid of theorems which constitute massive building blocks; but all these theorems can be reduced to the primitive concepts of basic mathematics, such as emerged from the analysis of Russell and Whitehead. It would not be merely un­economic, it would be a sheer impossibility to continue work at the higher level without these lower‑level foundations.

     Reductionism is not an immediate tactic for the tasks in hand, but more in the way of an ultimate strategy. We can, in fact, postpone the aim of explaining wholes from an understanding of their parts, i.e. the aim of reductionism, and let it fulfil itself in its own time. Not so long ago it would have been unimaginable to explain the physical properties of water, a higher‑level whole, from the known properties of hydrogen and oxygen. With modern conceptions of atomic structure this is no longer so. But the reductionist bridge from higher to lower levels requires bridge‑heads, that is the product of a mass of work done mainly at the lower level, and available only after the more fundamental science has conquered a wide territory. Needless to say, it is not necessary to wait so long to do scientific work at the higher level. While waiting, perhaps without enthusiasm, for the reductionist bridge, the worker on the higher level makes do with higher level concepts, such as behavioural concepts and not neurophysiological ones. Analysis goes on perhaps without or with only tentative contacts with lower level concepts.

    Reductionism is now an unfashionable con­cept, and in some areas causes a smart inflam­matory reaction, a fate which has met the gene theory of schizophrenia, for instance. The respect in which reductionism is held is the lower the higher in the hierarchy of sciences that we go. At the highest level of all, in sociology, it is usually emphatically rejected; the reasons are, perhaps, not far to seek. Every young science establishes its independent worth by operating with concepts referent to its own field, and there is a tendency to reject the applicability of lower level concepts. There is an unreasoning fear that if they are allowed to intrude the whole new growing science might be taken over by strangers coming up from below stairs. This is a delusion: in the main the higher level concepts are the ones appropriate to their field, and will only be made more precise if they can be shown to be translatable into lower level terms (in much less concise and manageable form). While guarding his working tools, the higher level scientist should not reject the possibility that one day the concepts with which he operates may prove to have such translations; it is a future to which he should look forward with hope.

    However, while reductionism reigns in the basic sciences, as we proceed from lower levels up the hierarchy we see an increasing tendency to separatism. Separatism becomes almost inevitable when workers at a lower level take little interest in what is going on on the floor above. Physicists for a long time bothered themselves little with chemistry, but in the end a whole new science of physical chemistry had to be built up on the mezzanine. We can probably look forward to Departments and Chairs in molecular physiology, physiological psychology, etc. However independent two sciences may seem to be, there is some informa­tional feed‑back from the higher level as the workers there respond to what is going on under their feet. In psychiatry we have suffered very much from the scepticism and suspicion with which our problems have been regarded by those medical scientists whose fields are basic to our own.

    However inevitable it is, however necessary a certain degree of separatism may prove to be, in the long run it is bound to prove unsatisfying. For a fundamental and enduring advance, the empiricist has to explain the phenomena of a higher level as implied by the simpler and more secure laws of a lower level. More and more the biologist is having to build his foundations on molecular biology; and the psychologist is seeing to it more and more that his findings are meaningfully related to neurophysiology, endo­crinology, chemical physiology and genetics. This dialectical process has been described and discussed by O'Neil (1969).

   The separatists, who have taken their stand on there being fundamental discontinuities between the sciences, who in the past thought of a gulf between inorganic processes and the organic processes of living matter, and later of a gulf between the bodily processes of living orga­nisms and their mental processes, and still later of a gulf between the behaviour of individuals and that of communities and societies‑stage by stage these separatists have had to see these gulfs being spanned by bridges, fragile and filamentary to begin with, but later by secure highroads.

     Workers at all levels naturally object to the view that the phenomena with which they are concerned are 'nothing but' more elaborate forms of what is being observed at a lower level. There has been a tendency to adopt a compro­mise between separatism and reductionism, which O'Neil proposes to call 'emergentism'. On this view, for instance, the properties of living matter emerge only when non‑living matter is arranged in certain ways. The new properties are those of the arrangement, the higher‑order structure, the gestalt, the whole which has been created from non‑living parts. 'It is argued that not only are these properties of the whole not present in the components but also that they are not deducible from propositions about the properties of the components' (O'Neil).

    As science advances, the position of the emergentist gets more and more eroded. It can no longer be maintained that the properties of H2O are not to be deduced from the properties of H and O atoms. Similarly some biological events are on the way to becoming deducible from lower level considerations. To permit the reduction, with explanations of higher level phenomena being translated into forms implied by lower level laws, certain supplementations are required. This will be first in the definitions of terms, 'gene' for instance being translated into a run of treads in the DNA staircase. By this the higher level term is not deprived of any of its existing meaning, but gains something new. Secondly there have to be supplementary laws or hypotheses, e.g. of a statistical kind. Thirdly, there has to be work done towards a union of discourse between workers on both levels, requiring a degree of goodwill, for the hypo­theses thrown up at the higher level to find firm bases below.

    The reductionist and the emergentist agree upon the unity of the sciences, and find them­selves at variance with separatists of all kinds, vitalists, dualists and mystagogues who, like Golla, posit an ultimate limit to the compre­hension of observable reality. In biology and psychology, emergentism is likely to hold the field for a long time to come, even if reductionism offers important hopes for the future. Emergen­tism is a model for present understanding; reductionism is too poorly armed to make a good model, but it is something different, a research strategy for the future. If these two views are likely to be with us as far as a prophetic eye can see, separatism, which is a kind of obscurantism, seems to be doomed.


    The question whether psychiatry is or can aspire to be a science is an important one. Two answers are examined, given respectively by Edward Mapother and Frederick Golla in the mid‑thirties of this century. They are both regarded as insufficient, in different ways, and ways which are relevant to present‑day under­standing.

[1] This penetrating remark gives us a glimpse into one root cause of the ills of our society, earlier referred to, insofar as they are traceable to the dominance of applied over pure science, of complacency over self‑criticism, of hubris over humility.