Books: Birth pangs of a baby universe

Will Big Science disappear up its own black holes? The Life of the Cosmos by Lee Smolin, Weidenfeld, pounds

We are often reminded that nothing in biology makes sense except in the light of evolutionary theory. In this bold yet modest book, Lee Smolin, a theoretical physicist at Pennsylvania State University, seeks to persuade us that Darwin's theory of natural selection is also the best way of understanding the existence of the universe itself. In his view, the cosmos and in particular the physical laws that control it have evolved by natural selection. It is for this reason that the universe is so hospitable to life. Tweak the properties of the fundamental particles, or the forces between them, by just a fraction, and the world as we know it would fall to bits - all the variety, order and complexity we see would simply vanish. The universe, so it appears, has been finely tuned for the existence of galaxies, stars, subatomic particles and, of course, living things capable of observing it.

At the heart of Smolin's astonishing idea are black holes. All you have to do is add the essential ingredients of natural selection - reproduction, variation and competition. Smolin suggests that every time a star collapses to form a black hole, spacetime itself is crushed out of existence and reshaped. A new universe is born; and with each birth the basic laws of physics emerge slightly different. So each baby universe is not a perfect replica of its parent, but a mutated form.

From among a huge population of competing universes, ours has evolved to maximise its production of black holes, and so of baby universes. And it just so happens that the kind of things - stars, carbon and complex organic molecules - that are ideal for making black holes are also ideal for making life. In other words, the laws of physics have evolved to maximise the reproductive success of the universe.

Not surprisingly, physicists have other ways of explaining why the universe is the way it is. Most popular is the anthropic principle, the idea that we inhabit one of an infinite number of universes, each with different constants. Smolin sees this idea as a cop-out that offers no testable predictions. Other physicists take refuge in non-scientific explanation: that a divine creator adjusted the constants of nature so we could evolve. Others still hanker after a "theory of everything".

Certainly, Smolin believes that a new view is required to unite the principles of quantum mechanics and general relativity. But he doubts that a unified theory can be encapsulated in a single formula, especially not one which assumes the laws of nature are absolute. Rather, he argues that the conditions we require for our existence are compatible only with a relational idea of space and time that takes into account the whole universe.

This is an immensely thought-provoking and thoughtful book, which tackles some of the deepest problems in physics. Along the way, we gain a clear overview of current thinking across a broad range of subjects - relativity, quantum mechanics, black holes, particle physics, ecology, the origin of life. Sadly, Smolin has been let down by his publishers. The book abounds in verbiage and typographical errors.

Popular science books, unlike the cosmos, really do benefit from the fine-tuning of an editor. Which is not to say the book won't be widely read. Think of Stephen Hawking's A Brief History of Time, hardly a paragon of stylish writing. Smolin's ideas, unlike Hawking's, have the added virtue of being intuitively attractive, for they provide a self-contained historical explanation for why we are here without appealing to any external agent or mechanism. In cosmology, as in biology, the beauty of Darwinism as a unifying theory is that it can explain a multitude of facts with a minimum of assumptions.

But there are snags. We don't know for sure what goes on inside black holes, and, even if they can give birth to baby universes, we could never observe their offspring. Also, the laws of self-organised complexity that scientists discern in computer simulations, and which Smolin draws on heavily to explain how galaxies evolve, may in fact bear only a sketchy correspondence to what happens in the real world. He does, though, stress that his theory can be tested and disproved. But the theory is next to useless at giving meaningful predictions of future events. When it comes to experimenting with universes, we are stuck with a sample of one.

Smolin is nevertheless at pains to distinguish between fact and speculation. In any case, his idea is by no means the most extravagant put forward by cosmologists. Respectable scientists have fantasised about quantum jumps from one universe to another (usually through wormholes), about parallel worlds, and even about whether one can create a universe inside a test tube. Reading this hugely inventive book, one is inclined still to concur with the late, great atheist biologist J B S Haldane, when he suggested that "the universe is not only queerer than we suppose, but queerer than we can suppose".