We commonly explain what people (and perhaps animals) do by assuming they have some purpose in doing it, some idea in mind which guides their activity towards their goal in an intelligently flexible fashion.
We say, for instance: Why is Mary buying arsenic? Maybe she intends to poison Martha, but more likely she needs it to get rid of the rats. Why did the chicken cross the road? To get to the corn on the other side. What made Jill insult Joan? I don't know, but she must have had some reason. As these examples suggest, we rarely use the word 'purpose' itself; perhaps this is why the Mock Turtle got the word wrong when assuring Alice that 'No wise fish would go anywhere without a porpoise. ... Why, if a fish came to
me, and told me he was going a journey, I should say "With what porpoise?" '. But we assume that we can explain or make sense of what people do by referring to their intentions, goals, aims, interests, ambitions, desires, wants, motives, needs, in a word, to their purposes.
Despite this universal acceptance of the concept of purpose in our everyday thinking about why people do what they do, it is not universally accepted by theoretical psychologists. On the contrary, throughout the history of psychology this concept has been, and remains, one of the most controversial of all. (It follows that the position argued in this article would not be endorsed by all current psychologists, although it is an increasingly accepted view. Examples of more 'reductive' approaches are given by the neurophilosopher Paul Churchland and the ethologist David Macfarland.) People do not merely disagree about purpose, but get more than usually heated in discussing it, often ascribing gross irresponsibility of one sort or another to their opponents in debate.
Some psychologists refuse point-blank to admit the concept of purpose into their theories, regarding it as not merely unhelpful but positively mystifying. Others are content to use it as a convenient shorthand, but believe that purposive explanations of behaviour could in principle be replaced by complicated stimulus–response or neurophysiological explanations in which the concept of purpose would not appear. Yet others insist that psychology must give a central role to purpose, that action and experience cannot possibly be explained without it; usually, they add that what it is to be a
human being cannot be understood without this notion, so that to reject it is to adopt an essentially dehumanizing image of mankind.
Theoretical resistance to purpose arises primarily from its close connection with the
mind–body problem. It is difficult to understand how purposes can function as guiding factors in behaviour, for how can an idea affect bodily action? (There is no help in identifying the purpose with the actual goal state, rather than with the subject's idea of the goal; for the goal state is always in the future until the action has been completed, and is often not achieved at all. How can a not-yet-existent state, which may in fact never exist, cause anything to happen here and now?) The
behaviourists, with the exception of
Tolman, rejected purposive explanation, because they avoided all reference to
consciousness, subjectivity, ideas, or mind. And neurophysiology, at least at first sight, seems to leave no place for purpose, since it deals with brain cells and brain functions whose physical description does not involve reference to ideas.
However, with the work of
Kenneth Craik, neurophysiology gained the concept of a cerebral model, or inner representation of the world. According to Craik, we possess physical mechanisms in the brain that function as inner models of the world by which we perceive, think about, and act upon the environment. Explanations of psychological phenomena must refer to these models, since it is only via these internal representations that action and experience can take place. But inner models may sometimes be misleading: the fact that there are no unicorns does not prevent people from developing internal representations of such mythical beasts. Similarly, an inner model may represent a state of affairs that does not yet exist, or a plan of action that cannot be fully carried out; the person can nevertheless use this model to guide present behaviour in various ways. So there is no radical difficulty in understanding how it can be that a person acts with the purpose, or goal, of finding a unicorn even though this purpose can never be achieved.
Cerebral models correspond to the
'schemas' described by the psychologist
F. C. Bartlett. They are essentially subjective, in that they constitute the psychological subject's view of the world — and of itself. This is why action and experience possess what
Brentano and
phenomenologists call
intentionality. Many psychologists and philosophers have claimed that intentionality could not possibly be explained in terms of a physical system, so that physiology is in principle incapable of helping us understand how psychological phenomena arise. In any case since very little is known about the detailed physiological basis of cerebral models, neurophysiologists cannot yet offer precise explanations of specific psychological characteristics.
But the concept of internal representation has entered
artificial intelligence, the science of writing computer programs which enable computers to do the sort of things that are done by human minds. Programs using different inner models of cubes, for instance, recognize (and misrecognize) cubes in different ways and on different occasions. (See
object perception;
pattern recognition.) Each program has its own view of the world, and its behaviour (saying that
this is a cube whereas
that is not) can be fully understood and explained only by reference to its inner schemata or models. Even though electronic engineers know precisely how the underlying mechanism works, it is the representational functions of the program which explain the 'psychological' characteristics of the programmed computer. Whether or not every aspect of human psychology could in principle be simulated on a computer is irrelevant here. The important point is that, even though the computer's 'physiology' is fully understood, its 'psychology' can be explained only in terms of its subjective (internal) models of the world.
By analogy, then, even if neurophysiologists knew everything there is to know about the brain,
psychological explanations would still be required to understand psychological phenomena. So psychology is not reducible to physiology, if by that is meant the claim that with sufficient physiological knowledge we could stop talking about purposes, ideas, beliefs, mistakes, and the like. Psychology is reducible to physiology only in the quite different sense that psychological features like purposes are generated by cerebral mechanisms, rather than being mysteriously inexplicable features outside the scope of science. A faith in the second kind of reducibility need not entail a faith in the first kind. It follows that the concepts in terms of which we express everything specifically
human about human beings would still be needed, even if we understood in detail how purposes are embodied in brain mechanisms.
(Published 2004)— Margaret A. Boden
Bibliography- Boden, M. A. (1972). Purposive Explanation in Psychology.
- Churchland, P. M. (1989). A Neurocomputational Perspective.
- McFarland, D. J. (1996). 'Animals as cost-based robots'. In Boden, M. A., (ed.), The Philosophy of Artificial Life.
- Woodfield, A. (1976). Teleology.
