Literary intellectuals are so ignorant about science, some scientists are claiming, that it is best to avoid their 'uninstructed' mediation and write directly to the public. Jon Turney is unconvinced.

Scientists who write books are playing a special role in late 20th-century culture. Unlike mere science writers, they really know what they are talking about. Unlike mere literary intellectuals, they know about something interesting.

This is the idea which animates one of the most influential figures in popular science publishing, the American literary agent John Brockman. He became famous in the late 1980s for winning dollar advances in the hundreds of thousands for many well-known scientists. But Brockman aspires to be more than an agent, and has lately been promoting his authors as the bearers of the only authentic news of our times, news about how the world really is.

It is a notion which obviously appeals to the judges of the Rhone Poulenc science book prize in Britain. Three of this year's short-listed titles - which will yield a winner next week - are by Brockman authors: Richard Dawkins, Paul Davies and Ian Stewart. Both Dawkins's River Out of Eden, widely tipped as a winner, and Stewart's Nature's Numbers are titles in the ambitious Science Masters' series, which Brockman sold as a package deal in 50 countries and 18 languages, a series to which Davies has also contributed.

³ÉÈËVRÊÓƵ

A fourth book on the list, John Carey's anthology The Faber Book of Science, carries a preface which argues a similar case to Brockman. Carey compares Dawkins's Blind Watchmaker with Martin Amis's collection of stories Einstein's Monsters, and suggests that the essential difference between them is that "from the viewpoint of late 20th-century thought, Dawkins's book represents the instructed and Amis's the uninstructed imagination". One is knowledgeable, the other ignorant. This is strong stuff for the Merton professor of English at Oxford.

But Brockman goes further. In a collection of interviews with his favourite scientists, he argues that we are seeing the emergence of a third culture, a term once used by C. P. Snow. According to Brockman, in the United States traditional intellectuals are "quite often proudly (and perversely) ignorant of many of the truly significant intellectual accomplishments of our time". Fortunately for the rest of us, the complementary third culture "consists of those scientists and other thinkers in the empirical world who, through their work and expository writing, are taking the place of the traditional intellectual in rendering visible the deeper meanings of our lives".

³ÉÈËVRÊÓƵ

Snow originally hoped that the third culture would be a group who bridged the gap between the two cultures he famously defined in his Rede Lecture in 1959, literary intellectuals who were on speaking terms with the scientists. Brockman sees it differently. His third culture is a group of scientists who write directly for the public, cutting out the middleman and more or less ignoring the rest of literary culture. In this group he places biologists Dawkins and Stephen Gould, physicist Paul Davies, cognitive scientist Marvin Minsky, philosopher Daniel Dennett and computer designer Daniel Hillis. The same list has been signed up to produce Science Masters titles.

It is hard to know whether to take this idea of a new intelligentsia seriously. We do seem to be in a time when it is fashionable again for scientists in some fields to write books for the wider public. And some of them are undoubtedly producing fine books.

But despite this, the first objection to Brockman's thesis is that there are not all that many scientific authors who can produce books that sell. In the US, Brockman is well-known for high-priced failures. Astronomer George Smoot used Brockman to get him a $400,000 advance for his Wrinkles in Time - on the big bang and its aftermath - which sold under 20,000 hardback copies in the US. The brilliant cosmologist Alan Guth had to back out of a $250,000 contract when he found he could not produce the promised book . And physicist and complexity guru Murray Gell-Mann's The Quark and the Jaguar went through several ghost writers and two publishers before it finally appeared, to mixed notices.

Nor is there much merit in the notion that having a real scientist as an author gets the reader any closer to a view of "how things really are". There is no better example of this than Tulane University cosmologist Frank Tipler's The Physics of Immortality, sold by Brockman to Doubleday in the US for $500,000, and a monumentally silly book if ever there was one. If Brockman and Carey maintain that scientists are in a position to disclose the truth about the world and our lives in it, they neglect the rather obvious point that all an individual scientist can offer is still an interpretation of the facts, informed by their own values.

³ÉÈËVRÊÓƵ

A further doubt about the effects of the rise of "third culture" authors on the health of popular science publishing is that scientists, trading off their undoubtedly hard-won expertise, tend to write essentially the same book over and over again. Dawkins is the most obvious example with five books to date (including his latest, Climbing Mount Improbable) which are all basically commentaries on the ways of selfish genes. Paul Davies has written many more, but the repeats are becoming noticeable. Even Stephen Gould shows a growing tendency to write the same four or five essays in his ever-accumulating collections.

Finally, the most serious reservation is that such books add up to a one-dimensional view of the scientific enterprise. For all that they sometimes relate internal disagreements among scientists, scientific authors all believe strongly in the progressive cast of their collective work, and the virtues of it continuing. While it can be inspiring to hear about the way science is going from the researchers themselves, we also surely need more critical views about where science and technology are going, and how we might respond. The best books about the human genome project, for example, are not by scientists but by writers with a bit more distance from the research who can take a cooler look at its ultimate worth.

There is, though, one aspect of the current trend which is significant for what it tells us about the state of science. As Daniel Hillis says, many of the scientists who write popular books do so "because there are certain kinds of ideas that have absolutely no way of getting published in the scientific community." This is probably putting it too strongly, but there are a number of figures who are using popular books to give their more unorthodox ideas a wider airing.

These aside, the books which are Brockman's stock in trade seem less like the new wave than a recovery of a notion at least 100 years old. As historian Leslie Howsam has pointed out, the Science Masters' series is strongly reminiscent of the International Series, launched in 1871 at the behest of T. H. Huxley, John Tyndall and Herbert Spencer.

³ÉÈËVRÊÓƵ

So perhaps Brockman's master plan is not so much a novelty as a return to Victorian values. But the history does suggest that his newest venture may be on to a good thing. The International Scientific Series lasted 40 years and clocked up over 100 titles. Now that might be enough to establish the third culture as something more than one entrepreneur's fancy.

Jon Turney is Wellcome fellow in the department of science and technology studies, University College London.

³ÉÈËVRÊÓƵ