Popular commentators on evolution, such as Richard Dawkins, have become overly enamoured with the idea of the gene. Genetics is certainly the most powerful mechanism of evolution and was unknown in Charles Darwin's time but although we have learned much from sequencing DNA, the idea of the gene does not explain everything about the living world and certainly not about the human world. However, just as Herbert Spencer used the notion of the "survival of the fittest" to explain why some people are rich and others are poor, so Dawkins argues that culture has genes, too - self-replicating particles of information that he calls "memes" (think of the dumb jokes and "viral" videos that proliferate on the internet).

If all evolution happens for the sake of proliferating selfish genes, then everything we see in living creatures has to be useful and practical. But that's not at all how Darwin saw it. He envisioned as at least two distinct processes: natural selection and sexual selection. The former concerns the survival of the fittest. The latter, however, is an aspect of evolution that is too often overlooked today.

After writing On the Origin of Species, Darwin was perplexed by the marvellous phenomena that natural selection could not easily explain. "The peacock's tail," he wrote in a letter to his colleague Asa Gray, "whenever I gaze at it, makes me sick." Did he not appreciate the splendour of such feathered beauty? On the contrary - he just could not justify it as a useful or adaptive trait. Many of the features of living creatures caused a similar queasiness in him: long, complex bird songs, garish patterning such as zebra stripes and the elaborate artworks created by bowerbirds in New Guinea (sculptures made out of twigs and branches, decorated with carefully arranged piles of flower petals, snail shells and dried-up insect larvae) - what practical use could any of this behaviour have?

Such wondrous aspects of animals' lives need not be seen as merely useful. Darwin's second major book, The Descent of Man, dealt with sexual selection. Over hundreds of pages he catalogued those features of living creatures, usually but not always males, that evolved simply because females happened to prefer them. The peacock's tail evolved because peahens found it beautiful, and generation after generation the more beautiful display won out. In most species such ornament is tempered by practical constraints, but in this one, the females ended up wanting the male with the most extravagant and magnificent display.

You could call this process aesthetic selection. We see it all around us and it is an aspect of evolution that most people find genuinely fascinating, though they are often disappointed to discover that scientists have tried to downplay its significance. How can science best deal with things that seem impractical? Take the elegantly-designed feathers of the waxwing, the huge unicorn-like tooth of the narwhal or the outsized antlers of the now-extinct Irish elk - there's no practical reason for these features to have evolved. Inside the brains of the female animals who selected for these traits lies a definite sense of aesthetics that serves to define such features in the species over thousands of generations. Where animals appear outlandish or do peculiar things such as singing 24-hour-long songs in the case of humpback whales, or building complicated works of art in the case of bowerbirds, sexual selection can be the explanation. Evolved traits need not always be useful.

The standard view of sexual selection among Darwinians today has moved far away from what Darwin intended. For nearly a hundred years, science avoided taking sexual selection seriously, perhaps because it was embraced enthusiastically by nature writers such as Wilhelm Bölsche. Consider this passage from his book Love Life in Nature (1898):

An animal is as if bewitched during loving-time. In all its feelings it belongs to another dimension . . . For a brief period of intoxication it is a citizen of another world sky-high above the ordinary cares of life. Something in the animal reaches out beyond the individual, to the life of the species, which wanders over generations, over millennia . . . The time of love's feelings becomes a time of liberated aesthetic life, a time of beauty.

Today, many biology textbooks tell us that sexually selected beauty is ultimately practical, in that it is all for the good of propagating the species. Others assert that the magnificent peacock's tail is a kind of "handicap" that the male must carry around like a burden to demonstrate that he is still strong and fit enough for the important activity of mating. The gene is supposed to be more important than the beautiful trait that it makes possible. The bowerbird's painstakingly constructed bower is supposed to be a kind of "honest signal", an artistic sign demonstrating that its maker has the best genes in the entire population.

Darwinians have tried to turn sexual selection into a subset of natural selection and they have done it using a method based not on research but on faith. They believe, unquestioningly, that any trait that nature has evolved must be useful and practical, even when there is little evidence for it.

Darwin himself said that male bowerbirds build species-specific styles of artwork because the females evolved to become arbiters of taste. They are the art critics who have dictated the attributes of each species' defining style. The beautiful features and behaviours produced could have taken an entirely different form. Wind the clock back a few million years, start again and you would get radically different creatures with wholly distinct features. Nothing about the life forms we have ended up with was inevitable.

Any unified theory of evolution has to be able to appreciate beauty, without explaining it in such a way that its allure is lost. Darwin was able to accomodate wonder in his writing but it is this unalloyed beauty that fundamentalist Darwinians today seem to miss. They are like Marxists who oversimplify Marx for their own purposes, narrowing the range of forms of knowledge that can be trusted to explain the world.

Yet a handful of biologists are beginning to take sexual selection seriously on its own terms. Among them is Richard Prum of Yale University, who recently figured out for the first time the true colours of a feathered dinosaur, using a blend of genetic analysis, sexual selection and amathematical theory of feather aesthetics. Were these particular colours useful? Not necessarily, but they are indeed beautiful. The British scientist James Windmill has written about the loudest sound-to-body ratio of any animal, the thrumming generated by the lesser water boatman, a tiny underwater insect. Why does this creature make such a racket? Perhaps for no practical reason at all save runaway sexual selection (this millimeter-long bug makes the sound by whacking its penis against its abdomen). It may be that nature evolves such amazements simply because it can. Traits such as this can be beautiful, though not always practical. The most scrupulous scientists are always on the lookout for the next amazing phenomenon, not hidebound by a narrow ideology that ignores things that it cannot explain.

Polls show that disturbingly large numbers of people refuse to believe in evolution. Only 40 per cent of Americans trust the scientific consensus that today's organisms evolved from previous forms by natural selection. Britain fares slightly better, with 50 per cent signing on. Those figures might be bigger if biologists were better at explaining why nature is so beautiful, and at showing that science can enhance our sense of wonder rather than diminish it.

David Rothenberg is professor of philosophy and music at the New Jersey Institute of Technology. His new book "Survival of the Beautiful: Art, Science and Evolution" is published by Bloomsbury (£14.99)