Whitehall's superconductor

Robert May tells Martyn Kelly how he helped bring order to scientific chaos

The Government's new Chief Scientific Adviser is, if the headline writers are to be believed, about to bring chaos to the Cabinet Office. Robert May is an old hand at chaos, having brought the mathematical expertise of a theoretical physicist to bear on knotty problems of population ecology in the early 1970s. Looking back, from his office in the zoology department at Oxford just before taking up his appointment, he traces this eclecticism to his "inappropriate" training at the University of Sydney, which he joined in the mid-1950s as an undergraduate and finally left in 1972 as professor of theoretical physics.

Ten years before May finished his degree, there would have been little opportunity to do a PhD in Australia. That May had the opportunity to stay was largely thanks to an entreprenurial academic at Sydney called Harry Messell who coaxed benefactions out of businessmen such as Kerry Packer to build up the physics department. With these funds he was able to attract a trio of remarkable theoretical physicists: Robbie Schafroth - formerly assistant to Nobel Laureate Wolfgang Pauli - who became May's supervisor, plus John Blatt and Stuart Butler, students of two other seminal physicists, Vicki Weisskopf and Rudi Peierls. "They were still close enough to their own supervisors," May recalls, "to have that ethos of the golden age of theoretical physics in the 1920s and 1930s when people thought that they could work on any problem under the sun. The training that said that I could work on one thing today and another tomorrow was something that I immensely enjoyed".

May's thesis was concerned with superconductivity, a subject on which Schafroth had already made a significant contribution. Put simply, protons and electrons ("fermions", or particles with spins of one half) do not superconduct. Schafroth's idea was that pairs of electrons bound together to act effectively as a single particle, a "boson" (or a particle of integer spin) would superconduct at low temperatures, by virtue of the phenomenon of "Bose-Einstein condensation".

"It redefined the problem of superconductivity to the problem of trying to understand how fermions are bound into pairs," explains May. "With that insight Schafroth, Blatt and Butler decided to try and produce an understanding of how you get the charged pairs and they produced a complicated theory which I then worked on for my thesis".

At coffee one morning during his PhD Schafroth made a comment to May about a theoretical two-dimensional Bose gas (a gas composed only of bosons) not being able to superconduct. "I went away," May remembers, "to satisfy myself that the two-dimensional Bose gas didn't condense at low temperatures". This ability to condense at low temperatures was what made a theoretical three- dimensional charged Bose gas a superconductor. "I saw that it didn't, but saw that it only just didn't," he went on. What happened, then, if it was charged? May tried a few more calculations. In practice, he discovered, he could not tell a two-dimensional charged Bose gas from a superconductor.

All of this had taken place late one evening at his mother's house in Sydney. Still on a "high", he set off on one further line of investigation and showed that, in two dimensions, the specific heat versus temperature of a Bose gas and a Fermi gas was identical, despite the two having totally different physical properties. At morning coffee the next day he told Schafroth and the others about his findings. "Everybody around the table said that couldn't be true." Robbie Schafroth went away to check and found that May was right. That same afternoon he wrote to tell Pauli about the discovery that his first-year graduate student had made.

May's first paper came out of that intensive evening's calculations. "My first paper (Physical Review vol. 115, pp.254-262, 1959) was this little theorem that says that two-dimensional Fermi and Bose gases have identical specific heats, and then I wrote a second paper showing that the charged Bose gas in two dimensions is a sort of superconductor. Neither were great discoveries; they really just seemed like mathematical jeu d'esprit because they are two-dimensional and the world is three-dimensional."

Apart from a postdoctoral period at Harvard and later study leaves there and at Caltech, May remained at Sydney, eventually achieving a personal chair, in theoretical physics. The switch to biology came shortly afterwards. He was searching for "real world" examples to use in a course on mathematical methods, aware that conventional mathematics was more concerned with elegance and rigour than with intuitive feel for how a system works. "I stumbled in 1970 on the then interesting model in ecology formulated by Charles Elton here in Oxford which said that complicated ecological systems are more stable," May recalls. "He had a set of arguments about that which were interesting but shaky." May started playing with the simple equations the ecologists used, building up from simple predator-prey systems to more complicated models. An ecologist at Sydney put him in touch with Dick Southwood at Imperial College and Robert MacArthur at Princeton. "I blundered into that subject at a time when not just stability and complexity were at issue but also wider questions about the dynamical properties of these models. A whole lot of problems were beginning to be formulated in the idiom of theoretical physics by people who didn't have the mathematical skills to pursue them." One thing led to another and he was eventually offered the chair in ecology at Princeton left empty by the early death of MacArthur.

So why another career shift? "People say to me 'how can you abandon an interesting career in research to do something administrative?' with all the academic contempt for administration that results in so much of academia being so badly administered. I say that I don't see it as being all that different." The training of a theoretical physicist comes in useful again: "The world is complicated but not all the complexities are important all the time," he explains. "The real trick is to intuit what are the subset of things that are important and on this basis to formulate a tentative understanding in unambiguous terms. You then pursue where that tentative understanding leads, test it against the facts, and usually circle back to refine the original assumptions. This basic process holds whether you are trying to understand superconductivity, or the causes and consequences of biological diversity, or how best to translate Britain's excellent science base into industrial strength ".

Chief scientist? He could be a management guru.