Lord May, you are the keynote speaker of the Latsis Symposium this year on the subject of EVOLUTION, IMMUNITY AND INFECTIOUS DISEASE. What is the focus of your lecture?

About 40 years ago, the US Surgeon General (1) expressed a view that was common in the medical community in the developed world at that time when he said: "the time has come to close the book on infectious diseases". Not only was this an extraordinarily parochial view, totally ignoring the morality and morbidity from infectious diseases in the developing world, but it was also a deeply foolish statement about the likely future of the developed world, in the light of the history of infectious diseases. And, over the subsequent decades, first HIV/AIDS, and hopefully to a much lesser extent the recent outbreak of SARS, make clear how foolish this view was.

But one legacy of the view exemplified by the Surgeon General has been that development of a sophisticated understanding of the nonlinear dynamical processes which characterise the engagement between populations of infectious agents (viruses, bacteria, protozoa, helminths) and their human or non-human animal hosts has greatly lagged behind the marvels of understanding the interaction, for example, between individual viruses and individual immune system cells, at the molecular level.

My talk will focus on the increasingly sophisticated understanding that is emerging of both the dynamics of disease transmission and control, and the population level, and – rather differently – of the way populations of viruses interact with populations of immune system cells within individual infected hosts. Particular attention will be given to the effects of heterogeneity in transmission processes, for example the role of "superspreaders" in sexually transmitted diseases. More generally, I will describe some of the counter-intuitive things that can happen (along with some current misunderstandings), when one deals with complex and heterogeneous networks of contacts among which infections are spread.

So far, immunology has only marginally concerned itself with evolution. In how far can adequate mathematical models support immunology?

There are two different questions here. The first is the extent to which mathematical models of interactions among populations of viruses and populations of immune system cells can illuminate aspects of the way the immune system works, as distinct from investigation at the level of detailed individual molecular events. It is rarely appreciated, even among workers on HIV/AIDS, that we still do not have a good understanding of why there is so long and variable an interval between infection with HIV and the onset of AIDS. This is in remarkable contrast to our marvellous understanding of how individual HIV viruses interact with individual immune system cells at the detailed molecular level. I believe mathematical models for the nonlinear dynamics, at the level of interacting populations, are increasingly playing a small but truly important part in helping our understanding of the overall phenomenon of HIV/AIDS, and that, as the discipline of immunology advances to a more mature phase, theoretical models will increasingly play such a part in the larger picture–always grounded upon and tested against experimental facts–in the way that is so familiar in more established disciplines like physics and chemistry.

The second question under this heading is the extent to which such mathematical models can illuminate evolutionary aspects both of the transmission of disease and of the workings of the immune system. Here again, the subject is in its beginning. Until recently, the conventional wisdom to be found in medical texts was that "successful infections evolve to become harmless". This was based on naïve and factually incorrect ideas about group selection. A more sophisticated analysis suggests that, over time, a newly emerging infection agent may become less harmful, or more harmful, or remain much the same, depending upon complex details of the way transmissibility interacts with damage done to the host.

You yourself have studied the spreading of HIV. Have any practical consequences followed from your work?

Together with a colleague, Roy Anderson, I was the first to publish an estimate of the likely demographic effect of HIV/AIDS in Africa. Sadly, this estimate has proved to be correct, even though it was criticized at the time as being much too pessimistic by the WHO and by the World Population Council (both of which organisations used mathematical models that looked much more realistic, because they had vastly more – although irrelevant – detail about aspects of the demography, but which on the other hand had a naïve and inadequate treatment of the infection process). Had this early estimate been immediately accepted, effective action might have begun a bit sooner.

But, more generally, it can fairly be said that work by Anderson and myself, along with many others, has been helpful in illuminating the basic biological and behavioural factors that can affect the overall epidemiological patterns of HIV/AIDS, and which can also point to the relative efficacy of different interventions. The most important factors, however, that distinguish those countries that have done a good job of controlling or ameliorating the epidemic and those which have not (South Africa being a conspicuous example of the latter) rely more on robust commonsense, particularly in the targeting of concern upon those "nodes of the network" which have the highest contact rates.

You are one of the world's leading mathematical biologists. In your opinion, where are the limits of a mathematical description of nature?

Einstein once spoke of the "unreasonable effectiveness of mathematics" in describing how the natural world works. Whether one is talking about basic physics, about the increasingly important environmental sciences, or the transmission of disease, mathematics is never any more, or any less, than a way of thinking clearly. As such, it always has been and always will be a valuable tool, but only valuable when it is part of a larger arsenal embracing analytic experiments and, above all, wide-ranging imagination.

Which important findings have been made on basis of mathematical biology?

Time does not permit an encyclopedic list here. I will single out just one. The emerging scientific discipline of "chaos", which recognizes that even when we have rigidly deterministic rules or laws which we understand, we may nevertheless not be able to make useful predictions in practice because the dynamical workings of nonlinear systems can, under many circumstances, behave in ways which are effectively undistinguishable from random processes. This represents one important finding, which has emerged as much from mathematical biology as from anywhere else.

Latsis-Symposium on "Networks of contact, and their influence on the population biology of infections": Keynote lecture from Lord Robert May of Oxford. large

More specifically, the beginnings of the recognition of chaotic behaviour lie back in the 19th century. They began to come into sharper focus with Lorenz’ work on a simplified metaphor (a 3-dimensional system of differential equations) for meteorological patterns. But it was the recognition of chaotic behaviour in the simplest of 1-dimensional difference equations, explicitly in the context of theoretical biological studies in ecology, that really moved chaos centre stage in the mid-1970s. It is amazing that several centuries of traditional focus upon continuous systems had enabled us to evade the simple consequences found in basic nonlinear first-order difference equations, and it is even more remarkable that this recognition emerged from ecologists (Jim Yorke, George Oster, myself and others) looking at such equations in the practical context of problems in population biology.

You are strongly engaged in environmental protection. How alarming is the situation according to your models?

One does not need mathematical models to be alarmed about diminishing biological diversity. At the global level, estimates of the rate at which bird and mammal species have become extinct over the past century, compared with average background extinction rates across the 600 million-year sweep of the fossil record, reveal a recent acceleration in extinction rates (if indeed birds and mammals are typical) of something like 1,000 above background. This is the kind of acceleration in extinction rates seen in the Big Five Mass Extinctions in the fossil record (such as the one which ended the dinosaurs).

This does not mean that we are going to see rapid extinction of a fair fraction of all species over the next few decades, rather, at current and likely accelerating rates, as indicated by various different lines of argument, these extinctions are unfolding on century-long time scales. But, make no mistake, we are indeed standing on the breaking tip of a sixth great wave of mass extinction in the history of life on Earth. Such conclusions are based on factual observation, coupled with rather physics-like lines of argument comparing recent "species expected lifetimes" with those documented on average in the fossil record.

Do you believe that anticipated risk can change the attitude of politicians and populations in general?

One of the greatest, and most important, unsolved problems in evolutionary biology is how one may evolve and maintain cooperative or "altruistic" behaviour. It may seem obvious that if paying a small individual price purchases a much larger group benefit which one can enjoy as a member of the group, then there should be no problem. But the difficulty is that "cheating individuals" who reap the benefit without paying the cost, will – by virtue of avoiding the cos t– do better in evolutionary terms (leaving more descendents), and thus cheating behaviour will eventually prosper at the expense of the group benefit. If one is dealing with a small band of relatives, then one can get around this problem. But this no longer holds for the kind of larger aggregates of cooperative behaviour found in humans. Eventually, one can see that a framework of laws and regulations, all bound up with the associated problem of the evolution of language, may be able to keep such cooperative behaviour in place once it has appeared, but the reasons for the original appearance of such behaviour are still not well understood.

In other words, we do not understand why it is that we should, on occasion, take actions for the common good even though it means each of us giving up something. So it is even harder to understand how we may act together for the benefit of a seemingly distant future. Climate change unfolds on the time scale of decades; biodiversity is diminishing on the scale of centuries; feeding still-growing populations is not yet as crucial and may not be for the next fifty years.

In summary, we are really forced with the need to act now on behalf of the future, not least because in nonlinear systems small actions today are vastly more important than big actions tomorrow. Yet we do not have a fundamental evolutionary understanding of the social systems that bind us together. Small wonder that it is easy to be pessimistic about the future: we can anticipate the risks, but that does not necessarily mean that any past evolutionary experience has adapted us to doing anything about these risks, when many of the consequences lie beyond our personal horizon.

Last, but not least, we come to your award of an honorary doctorate from the ETH Department of Environmental Sciences. What does this honorary doctorate mean to you? Do you have a special relationship with ETH?

My PhD supervisor at Sydney University was Robbie Schafroth, who was Pauli’s Assistant in the mid-1950s at the ETH. Schafroth was, indeed, the first person to observe that a charged Bose gas would be a super conductor. This redefined the problem of a theory of superconductivity – one of the grails of physics at the time – to the problem of explaining how there could be effectively bound electron pairs in superconductors. Had he lived, then justice would surely have demanded that Schafroth share the prize with Bardeen for superconductivity (the prize that eventually went to Bardeen Cooper and Schriefer). Sadly, he and his wife were killed in an accident in a small plane in the Australian outback in 1959. He was just 42. But for this he would have taken up the Foundation Chair in Theoretical Physics at the University of Geneva a few months later and I would have accompanied him as his Assistant. So his death completely changed my life, although I would like to think that his mentorship, and wonderful example both as scientist and person, permanently shaped my own life. My Honorary Degree from the ETH thus has very special meaning for me.