Twelve invaluable attempts at elucidation leave plenty for us all to argue about

A symposium in 1999 brought distinguished participants to Jesus College, Cambridge, to discuss scientific explanation. One outcome of the debate was this collection of 12 essays. There are few readers to whom they can all be recommended. Severe thinkers may enjoy getting their teeth into what the philosophers say, but for many, even of scientific bent, the logic may prove tough. The going is not so heavy at the biological and sociological end, but there is little that does not demand serious study, debate and disagreement; that is the whole point - the problems are ultimately insoluble or at best unsolved. The book explains almost nothing but tries to elucidate what is meant by explanation.

Every parent knows the pitfalls; answers to the question "why?"

are met by a further "why?" and you feel the ground falling away as you slide down the slope of infinite regress. An old story tells how a child asked at school about the causes of the Great War replies that if Eve hadn't eaten the apple there wouldn't have been a war. Peter Lipton, in the first essay, discusses in more up-to-date terms both the value and the futility of this kind of reductionism. Yet reductionism is a leading principle of science, carried to an extreme in physical science; matter can be understood in terms of atoms, atoms in terms of protons, electrons and other particles, and these in their turn in terms of the probably unobservable quarks.

Before sinking to the depths, let us stay awhile in shallower water. David Hanke disposes of one type of explanation, the teleology that, often through negligence, attributes the evolution of complex and efficient life forms to innate purpose: "This is one reason why nature developed macromolecules," he quotes, not on this occasion from Richard Dawkins, who elicits admiration as an expositor but also criticism for falling back too readily on the attribution of purpose. As a biologist remarked in the old strait-laced days: "Teleology is like a mistress whom one cannot live without but dare not acknowledge in public." In Hanke's view, Darwin's survival of the fittest means no more than survival of that which survives; the insidious intrusion of purpose into our thinking may help, but ultimately stultifies, imagination.

Accounting for evolution and other apparent manifestations of design simply in terms of the purposeless operations of chance is but one example of successful explanation by the voluntary acceptance of a basic principle without deeper questioning. Few scientists think twice about it. Peter Atkins makes clear the wealth of understanding available to chemists who accepted primitive concepts of valency long before quantum mechanics showed how they might come about and at the same time enhanced their power. And not only chemists; most physicists are confident that quantum mechanics, in the form it had reached by 1940, is beyond reproach when used to account for the properties of ordinary matter, down to atomic level. They are aware they cannot picture what is going on among the particles, but in their professional lives it is a rock-solid foundation.

This is the science that underlies modern technology and can still enthral the majority of scientists whose business is with things we can touch and see, but the media rarely find it exciting enough. It is the restless probing of mysteries that makes good copy; so here we read Sir Martin Rees to learn what is known of cosmic truths, and Steven Weinberg for the opposite extreme to the inconceivably small. They do a very good job of it, eschewing long words and clever nonsense. All the same, their story is so strange that we are tempted to throw it all overboard and sink back into the warm stupor of teleology (aka divine providence).

In sub-atomic physics the trouble began 80 years ago with the necessity to abandon determinism. The rules for calculating what may be expected in an observation have never been bettered, but they forbid asking for details of what goes on between one observation and the next; events do not follow a path, they simply occur. Theoretical models of great subtlety aim to bring these rules into harmony with relativity and the force of gravity, but the more they purport to explain, the less they correspond to ordinary ideas of reality; a string in 11 dimensions may be the key, but to most of us it lacks appeal.

On the cosmic scale, different problems arise. A mere four non-Euclidean dimensions may be tolerable, with a Big Bang to start the universe from nothing, but all is not then plain sailing. How did the universe happen to develop on a long enough timescale, and with all the variety of chemical elements needed for us to evolve and observe it? It looks as if the numbers that control the properties of fundamental particles were very carefully adjusted to this end, and yet we know no reason why they should not have other values. Is this divine providence again, or do we (as has been seriously suggested) occupy the only suitable candidate out of a multitude of slightly different universes, coexistent and without intercommunication?

Deeper probing seems to offer no resolution, and beyond these mundane mysteries lies the greatest and most important of all - how to fit conscious mind into the scientific picture? The essays by philosophers take considerable trouble to reach what I regard as an almost obvious answer - it cannot be done. The vocabulary of science is designed to cope with observations that are common to all and to exclude the individual's private knowledge of his own existence, his personal observations and thoughts. These are the very essence of understanding: without them there would be no science, but paradoxically it is by refusing to include them within its structure that science has made such progress.

It can examine what the nerves and brain do when the subject says he is thinking, but the thought itself is not available. For the scientist, the conclusion is summed up in the remark quoted here by Colin McGinn: "The mind-body problem is wide open and extremely confusing." When the best we can manage with non-thinking matter is a formulation so abstract that we cannot visualise its meaning, to assert, as some scientists do, that we are poised to understand mind seems little short of impertinence.

It is something of a relief, after plumbing the abyss of scientific ignorance, to turn to Jack Goody's essay on anthropology, particularly his comments on the shift from pseudo-historical theory of cultures to the observation and interpretation of current behaviours. Here is, as yet, no reductionism, no attempted explanation in terms of fundamentals. The concern with individual attitudes and social customs is as far as one can get from modern physical science and adds further nuances to the problem of providing explanations. But it was never the intention of the original meeting to explain anything, and the broad coverage of the book only enhances its value to a serious reader; to repeat that at every point there is much to argue about is a very positive recommendation.

Sir Brian Pippard is emeritus professor of physics, Cambridge University.