Think of Charles Darwin, and the ideas that spring most obviously to mind are natural selection, and - if you haven't yet learned to hate the term - survival of the fittest. Natural selection rightly permeates modern biological thinking as an explanatory tool of enormous power across the whole field, from molecular biology to evolutionary psychology; but it is not the last word in evolution.

Darwin's "other" theory was sexual selection, the idea that mate choice (principally by females, but potentially by either sex) can cause evolutionary change by creating a differential in reproductive success between potential mates with more or less sexually attractive traits. In essence the theory is no harder to understand, nor any less superficially plausible, than natural selection, and indeed it is a part of "orthodox" modern evolutionary theory; but it definitely plays second fiddle to natural selection in the explanation of biological phenomena. (As recently as 2001, Richard Morris could write a book, The Evolutionists, about controversies in modern evolutionary theory and not once mention sexual selection.)

In The Mating Mind (2000), psychologist Geoffrey Miller argues that sexual selection has been responsible for a great deal of the evolution of the human mind from great ape to Homo sapiens. This is immediately an audacious and counter-intuitive suggestion. Most people are barely willing to accept that any kind of evolutionary approach can contribute anything to an understanding of the mind, and here is a little-known academic suggesting that the mind evolved by a process which even most biologists don't pay that much attention to. Such a daring hypothesis needs a powerful argument to back it up, and Miller's book is not only powerful but persuasive and immensely enjoyable.

Miller starts by arguing that traditional evolutionary explanations of the human mind, which have focused almost entirely on natural selection as their mode of explanation, are often very weak. For sure, some aspects of human cognition - vision, for example - are best explained by natural selection - better seers are more likely to survive. But then, vision is hardly a unique human trait. The features and products of the human mind unique to humans - humor, art, morality, science - do not yield to explanation by pure "survival of the fittest", "nature red in tooth and claw" thinking.

As Miller elegantly puts it, "Our minds are entertaining, intelligent, creative and articulate far beyond the demands of surviving on the plains of Pleistocene Africa. To me, this points to the work of some intelligent force and some active designer. However, I think the active designers were our ancestors, using their powers of sexual choice to influence - unconsciously - what kind of offspring they produced."

Miller then spends some time outlining the theory of sexual selection, including its history, and the various ways in which it is thought to operate. The theory is as old as Darwinism itself - it made an appearance, albeit a brief one, in The Origin of Species, and was the prime occupant of Darwin's thoughts for most of his life after that book. Darwin (the first evolutionary psychologist) thought it as important as natural selection, and indeed was using it to explain human evolution a hundred years before Miller, in The Descent of Man. But it met a cold reception by 19th century evolutionists and continued to be neglected, for various scientific and ideological reasons, even as Darwinian natural selection was being cast into stone by the Modern Synthesis in the 20th century.

Nevertheless, biologists have identified three main ways in which sexual selection can operate on species. "Runaway sexual selection", particularly associated with the mathematical biologist Ronald Fisher, happens when an originally arbitrary trait (generally in males) is attractive to the opposite sex and becomes, over the generations, more and more exaggerated as the most successful males in any generation will be those who display the trait slightly more than the population average. The quintessential example of this is the peacock's tail, which has become a glorious, useless adornment found on every male peacock simply because females in previous generations, for whatever reason, found the biggest and brightest tails most attractive. "Runaway" tends to produce characteristics which are exaggerated in just one sex (usually males) - the tails of female peacocks are fairly drab.

Sexual selection can also act on genuine indicators of genetic fitness. Some traits advertise a creature's fitness to potential mates by being dependent on good genes, which is often synonymous with "genes free from mutations". Strength, intelligence, even body symmetry are all things which can be damaged by genetic mutation and which thus, when genuinely present, demonstrate an individual's genetic fitness. Other traits advertise fitness by handicapping their own possessor in some way. For example, the males in some species of birds grow tails so long they interfere with flying and make it harder to escape predators. Males who can survive despite such a limitation must be strong and hence genetically fit. This is called the handicap principle and is closely associated with the Israeli biologist Amotz Zahavi.

Having demonstrated that sexual selection is real, and important, Miller now goes on to apply it to the human mind. How can mate choice have resulted in the evolution of human nature as we know it? We can be certain than runaway sexual selection was not the only factor because the minds of males and females are fairly similar - if there had been a pure runaway process, we would expect males to be far more intelligent and creative than females, which we do not find to be the case. The principles of handicaps and fitness indicators, on the other hand, tend to produce much less divergence between the sexes in the traits that are being acted upon by mate choice.

Miller emphasises that the features of the human mind that we find so interesting, and that psychology is so bad at explaining - art, humor etc. - cry out for an explanation through sexual selection. They show many of the hallmarks of sexually selected traits - they have been carried to an extreme, they serve no apparent survival purpose, and they are unique to a single species. If intelligence is so important to survival, why have no other species evolved it to anything like the extent we have?

Miller calls up the image of pre-modern man, roaming the plains (not the caves or the jungles, by the way) of pre-historic Africa. According to contemporary reconstructions of Pleistocene life, pre-modern humans would have had a lot of spare time in between fending for their survival. (Although it is not vital to the issue, it is interesting to note that the common view today is that the "hunters" in hunter-gatherer societies were largely unsuccessful in their efforts; that hunting was undertaken mainly to impress women, and most of the food eaten by both sexes consisted of whatever had been gathered by the women while the men were out chasing animals they rarely managed to catch.)

In any case, so Miller proceeds, with all their free time, pre-modern men and women would have spent a lot of time simply talking to each other and observing each other's social behaviour. Courtship would have consisted to a large extent in verbal and social displays. The ability to be creative, witty, or socially adroit would have served as an indicator of a good brain and thus more generally of genetic fitness.

Miller selects four specific aspects of human behaviour that cognitive and evolutionary psychology have thus far (he feels) failed to explain - art, morality, linguistic creativity, and humor.

In the first case, for example, the ability to create good art and craftwork, or more generally to decorate and furnish objects attractively, requires a certain creativity and sensibility that indicates a good brain. Since art and decoration are, on the whole, functionally useless, they also act as an effective handicap.

In explaining these traits through sexual selection, Miller puts himself at odds not only with many schools of thought not associated with evolution - including most sociology and anthropology - but also, to a certain extent, with other evolutionary psychologists. Miller disagrees with Steven Pinker's conclusion, expressed in How the Mind Works, that these traits - which encompass perhaps the greater part of the human mind - are mere "by-products" of functionally adaptive traits such as vision. In explaining morality, meanwhile, he does not actively contradict the traditional explanations (proposed by, for example, Wilson and Trivers) of kin selection and reciprocity, but doubts that they can tell the whole story, as is sometimes asserted.

His thesis is thus an altogether daring one, but it is also, as I have said, powerfully argued. On first hearing the idea, that sexual selection is responsible for the evolution of human nature, one is skeptical; but on reading the book, Miller's careful, complementary arguments gradually build in power until the idea seems like second nature. The author is at pains to stress that the book is only an outline of how sexual selection may help evolutionary psychologists explain the human mind - it could hardly be otherwise, given that Miller is among the few who have pushed the idea this far. But it is a very convincing outline.

According to a naive interpretation of the idea, Miller appears to be rehashing discredited Freudian ideas about the ubiquity of sex drives in human behaviour - "it's all down to sex!" But there is an important difference here. To explain the evolution of a mental trait - be it an acute sense of humor or the desire to see social justice - in terms of sexual selection is not to say that the trait is "really" a form of sexual desire. The trait is genuine - it evolved because those who possessed it were more sexually attractive, but this does not make it any less real. Musical ability may have evolved through sexual selection, but this does not mean that Bach wrote music out of a desire to get laid - he wrote music out of a desire to write music. (Of course the split between evolutionary function and conscious desire is not perfect - every wannabe rock star knows that writing popular hits is a surefire way of attracting members of the opposite sex.)

The plausibility of Miller's hypothesis is high enough that this book, original as it is, deserves to be read by anyone with an interest in evolution or psychology. The writing is fluent, somewhat reminiscent of Steven Pinker (look out for the Led Zeppelin and Modern Lovers references in the book), and the arguments are clearly understandable to anyone with the patience to think them through. Miller has Richard Dawkins and Matt Ridley singing his praises, and if, as may well happen, his hypothesis turns out to be true, this could become one of the most important books yet written on evolutionary psychology.