Just a phrase they're going through

Jon Turney meets Evelyn Fox Keller, the physicist turned philosopher who is putting the words scientists use under the critical microscope.

At intervals over the past few decades, a cultural vacancy has been advertised for a science critic. Like a literary critic, a science critic would offer readings of scientific theories, interpret them in new ways, reveal meanings perhaps hidden from their authors. Since the mid-1980s, the writer most likely to fill the vacancy has been physicist turned biologist, turned feminist commentator Evelyn Fox Keller.

It is, of course, a role doomed to misunderstanding. Literature is fair game for critics. It can only be enriched by their work. But science, despite its vastly greater cultural authority - or maybe because of it - is mostly hostile to such efforts. Some scientists seem to regard cultural analysis as criticism only in the pejorative sense. Tell them that they use metaphors to ground their theorising, for example, and they take it as a slight rather than as an observation about how theorising is possible.

Evelyn Fox Keller, in a career which has shifted from trying to advance science to pondering what "counts" as an advance, has had more than her share of misunderstandings. She is a philosophical realist who is seen as a relativist, a woman identified with questions of gender and science who believes that a feminist science is a contradiction in terms, above all a lapsed scientist who is still passionately engaged with contemporary research and wants to influence how its theory and methods unfold.

In her four books to date, the latest a trio of lectures collected as Refiguring Life, she has covered a wealth of figures and topics, from Plato and Francis Bacon to evolutionary biology and the human genome programme. But her overriding concern has been with what shapes styles of thought in science; with the personal, psychological, or cultural influences that make certain styles accessible to some scientists but not to others.

Fox Keller, professor of history and philosophy of science at the Massachusetts Institute of Technology, has her sights set on changing science. Since her first book, about the Nobel prizewinning corn geneticist Barbara McClintock, most of her work has taken the linguistic turn, putting the words biologists use under the critical microscope. Her latest paper, for instance, asks how the meanings of "gene" have shifted. It is a story others have told, but Fox Keller moves it forward into the era of the human genome project. Today, she argues, the predominant metaphor of the gene as computer program is straining to accommodate the many ways a DNA sequence can be processed, and appears incapable of accounting for the way cells and their DNA work together in development. There is still a troubling gap between gene and organism, and as gene mapping extends further this will become clear.

Fox Keller began in the early 1960s as a physicist, but found as a graduate student that "it was not possible for a woman to do theoretical physics at that time at Harvard". She revisited the experience in a well-known essay on "The anomaly of a woman in physics", but by then she had also successfully pursued and then dropped a career in mathematical biology. This culminated in 1970 in an important collaborative paper on why slime mould cells decide to get together. But she was becoming conscious of a growing disaffection.

Her collaborator, Lee Segal, "was running round the country, getting a lot of very positive feedback, and I was in the kitchen. I had two small children. And I was getting more and more bored with it and he was getting more excited".

It occurred to her to begin a new inquiry into the problems of women in science. At first it was a psychoanalytic inquiry, then, with her study of McClintock, a biographical one. Her interviews with the late geneticist started out as the basis for a conventional biography, but then broadened to examine how her remarkable creativity was embedded in a particular research style which was at odds with that of most of her fellow biologists. McClintock, with her knowledge of each individual plant growing in her maize plot, her seemingly uncanny ability to read precise genetic changes from patterns of colour in corncobs, developed "a feeling for the organism" to a pitch rarely seen in the life sciences.

Some feminists saw the appeal of McClintock's approach to her science, but were annoyed to find nothing in the book to warrant the notion that this was feminist science. Fox Keller was quite clear: "If you knew her, that was just a joke. She was so hostile to any stereotypes of femininity."

But the issue went beyond gender. It was important to provide a working picture of a research style at odds with the reductionist habits then predominating. "At the time," she says, "I was confident that a feeling for the organism would lead you to better science." Fox Keller's more recent work on language and metaphor now appears middle of the road, at least in relation to some of those adopted on the wilder shores of science studies as it has grown up over the past 20 years. Nature exists, but we only have access to it through our representations; some kind of linguistic framework is a prerequisite for pursuing science at all.

The difficulty is in envisaging the alternative metaphors in ways which might make a difference. Sometimes it is easy to see how metaphors have already shifted - as in interpretations of fertilisation over the past 20 years - moving broadly from penetration to fusion. More challenging, though, is the perception that one set of metaphors is nearing exhaustion without a clear notion of what the next reframing might be. "I do not think the prevailing metaphor, the discourse of gene action, is capable of addressing the question, where do the 'specific value' or goals of biological organisms come from," she says.

She is looking to the new ideas about self-organisation coming out of computer modelling for inspiration. "My plan is to learn more about cellular automata. I'm sceptical, but I need to understand better." It sounds a plausible step. The patterns generated when simple rules govern how a set of elements in a computer display respond to the state of their nearest neighbours often look reminiscent of living forms.

She describes her career as revealing a problem with borders, between scientific disciplines, between doing science and commenting on it, between science and the humanities. The metaphors she tracks of course cross borders easily, slipping out of one discourse and into another. That is how they do their work. It is much harder for a writer to cross these borders, still less to try and manage the linguistic traffic. They are policed inside disciplines and institutions, and history and philosophy of science are cast as enemies by some scientists.

But she will go on speaking to scientists who are willing to listen, not just about history but about the future. "I want science to answer certain kinds of questions. I have very definite preferences, but people are entitled to legitimate differences in what they want. It is a social question: what do you want science for?" And that is just the question a science critic ought to be asking. Come to think of it, it would be nice if more scientists asked it as well.

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