This idea had its roots in Darwin's Origin of Species. In its modern form it depends on Watson and Crick's discovery of the structure of DNA, because this reveals how information is embodied and copied in theorganisms we know about - those found on earth. Its precise formulation we owe to GC Williams' Adaptation and Natural Selection, published in 1966, and to Dawkins' The Selfish Gene.
Both books, especially the latter, can be read, and were meant to be read, by those with no training in biology. Unhappily, this has led many biologists to miss the fact that these books also made important contributions to evolutionary biology.
River out of Eden shows how the idea has developed in the past 20 years. The notion of information has become still more explicit: this is natural in a society dominated by information technology. It has also become clear that there have been major transitions in the way information is stored and transmitted - "replication thresholds". Examples are the origin of replicating molecules (equivalent to the origin of life), of sex (whereby information from different ancestors is combined), of multicellular organisms (each carrying many billions of copies of the information), and of human language (the invention of a new way of transmitting information).
Dawkins has an enviable gift. He can write books that are fun to read, yet which present fundamental ideas clearly. River out of Eden is no exception. The river of the title is the ever-branching stream of genetic information.
A peculiar difficulty faces anyone who writes about evolution for a wider public. Many readers do not want to accept what biologists are telling them. Most previous cultures have had myths about origins. Today, both cosmologists and evolutionary biologists are telling stories about origins. Understandably, people expect these stories to perform the functions previously performed by myths.
Oddly, they seem willing to accept a Big Bang. Perhaps they can imagine the creator lighting the blue touch paper: in any case, Big Bang has no obvious moral implications. But evolution by natural selection is another matter. As Dawkins remarks, so long as DNA gets passed on, it does not matter who or what gets hurt in the process. As he emphasises, Darwinism is a beautiful idea, in that it explains a multitude of facts with a minimum of assumptions, but it is a lousy myth. It is a scientific theory, not a guide to good behaviour. But inevitably, many readers will reject it because they dislike what they imagine to be its moral implications.
Dawkins is good at answering objections often raised to Darwinism - in particular what he calls the argument from incredulity. This asserts that there has not been time for natural selection to produce what we see around us, or that it is impossible anyway, because there are no useful intermediates between perfect structure and no structure. He points out that these are assertions, not arguments, and provides convincing answers. But my experiences suggest he is fighting a losing battle. If people don't want to believe something, they won't.
John Maynard Smith
The writer is professor of biology at the University of Sussex.