The Boundaries of Science

Working Copy: Second Impression August 1947, First published 1939 by Faber, 268 pp, Preface + Six chapters + Index. (Out of Print.)

Click for full text and download options from The Internet Archive


1. The Philosophy of Science

2. The Sociology of Science

3. The Limits of Science

4. The Psychology of Psychology

5. Theory and Practice in Psycho-Therapy

6. Fact, Motive and Intention


This is the most genuinely clarifying and illuminating book on the philosophy of science I have ever read. And compared weight for weight with later volumes by others on the subject its achievement is all the more remarkable.

Macmurray opens by saying that the creation of science is the act of Western civilisation in its modern period. It is the peculiar contribution of Western civilisation to human life. The creation was necessary for there to be the possibility of a fully human life for the whole community of mankind. With the beginnings of the creation of a scientific psychology the creation of science is nearing its completion. This is not to say that the work of science will have been completed — just the creation of science itself as an instrument.

Macmurray gives an illuminating account of the nature of scientific and mathematical procedures. Scientific hypotheses are the product of human fantasy or imagination. They are retained as working hypotheses if they are shown to be valid. They are shown to be valid if they enable the scientist correctly to predict how things in the world will behave, and if they enable him to construct laws of the behaviour of things. Even if hypotheses are in this way shown to be valid this does not mean that they are necessarily correct pictures of the real constitution of things. They might be or they might not. But the scientist is not primarily interested in giving a correct picture of the constitution of things. He is primarily interested in gaining a reliable knowledge of how things behave.

These observations apply as much to a scientific psychology as to any other part of science. Macmurray chooses thoroughly to investigate philosophically the psychological sciences because in them science is reflected back upon itself and the apparent paradoxes which lurk beneath the surface of all science are encountered more forcefully and explicitly in psychology. Science is reflected back upon itself in psychology in the sense that in this discipline the observer finds himself part of the subject matter which is being observed. Science here is studying human behaviour which includes the behaviour of the scientist producing science. The paradox of psychology is that at first glance it seems condemned to produce theories of a type which if they were true could not be true. Macmurray draws attention to the distinction between an event and an action. An event just happens. An action only occurs when an agent tries to implement an intention. We make use of the distinction in daily life when we try to draw a contrast between an accident we have caused and a deed which we fully own as ours. Courts of Law are concerned with this distinction. If we killed the man accidentally then it is manslaughter. If we’d intended seriously to injure or kill him then it’s murder. However, a person external to the event relying merely on sensory observation cannot observe the difference between an event and an action. Intentions cannot be sensorily observed. (Hence we can deceive one another.) To reach a conclusion he has also to rely on circumstantial evidence and intuition based on his general experience of life (which will include his own experience of implementing intentions).

Now science is always in the position of relying on a third person observer who is relying on sense experience alone. When confronted by human behaviour it can only see a series of events for which it will attempt to find causes. Like the rest of science it will produce imaginatively constructed hypotheses in order to be able to predict (and ultimately to control) these events. Now events are something which have happened. What’s happened is in the past. So science is always studying the past. The past cannot be changed. So scientific accounts cannot be anything other than deterministic. And this will apply to the account the psychologist gives of the scientist producing science. Which will make the beliefs possessed by the scientist as the result of pursuing his science appear to be beliefs he could not help but hold. And similarly those of a second scientist holding an opposing set of beliefs. Now this is not what the scientist wanted. He wanted knowledge, to discover truth, to find a method of checking his own and his colleagues’ results for error. In a fully determined realm the possibility of error cannot arise. Hence what was meant by saying above that the paradox of psychology was that it seemed to produce theories of the type which if they were true couldn’t be true. So we have the strange situation where we can see that science is apparently successful but where it cannot produce a non-paradoxical account of itself.

Now the solution to the apparent dilemma of science is to focus on the types of questions which agents can ask and to realise that the questions they ask depend upon their intentions. Philosophy, for example, and science both ask questions of the same world — but different questions. The answers to scientific questions give us instrumental knowledge — that knowledge of the world we need to have if we are to implement our intentions fully and successfully. It is easy to see how, for example, physics, chemistry and biology can provide this but can psychology, especially in view of the comments made above about its theories? Yes, it can. And this is because human beings have to build up whole layers of habitual behaviour, both physical and mental, as a basis for their higher order intentional behaviour. This habitual behaviour will have its regularities, both within the individual and across individuals, and provide the psychologist with something to observe and make predictions about. This will be useful instrumental knowledge to human beings wishing to control the behaviour of themselves and others. Whether it gets used by the master wondering how to get the best out of his slaves, or by the community of free men wondering how they can best be useful to one another is not primarily a scientific question. And remember, the determinism of science is built into the method — not discovered by it. Science does not give access to intentions or value. One requires other forms of human reflection for this.

The book derives from the 1936 Deems Lectures but is not a publication of them. Macmurray states that he has omitted reference to religious psychology partly because his treatment of it in the lectures seemed inadequate. One would like to have known more! What he has to say in the last chapter about intention, knowledge of value and past human wisdom (and how it was embodied in the religious traditions of mankind) catches the interest and leaves one wishing he had more to say. A glance at the List of Contents will give an idea of the range of subject matter covered. The work’s concluding message could be seen as urging us to take seriously in its own right the phenomenon of intention and what it might reveal to us about an aspect of our experienced world. We should not be too ready to assume that intention can be reduced without remainder into something less. We should not, in the words of J.B. Priestly, ‘be talked out of our own experience’. Such a postulate is no more outrageous than those which have to be granted to make science possible.

Macmurray concludes with the sentence: “I find it difficult to believe that the accompaniment of the decay of religion in Europe by the breakdown of social cohesion and the relapse into barbarism which we are now witnessing, is a matter of purest coincidence.”