Philip Anderson cautions scientists against omitting the human scale.
John Brockman is the impresario of a series of internet essays by, and interviews with, scientist-intellectuals. They are all members of what Brockman terms the "third culture" - after C. P. Snow's two cultures - and many have also written books on science for the lay reader. In T he Next Fifty Years , Brockman has selected two dozen of these scientists and asked them to surmise what the next half-century may bring, especially but not necessarily exclusively in their scientific fields.
I am aware that most exercises in futurology turn out to be thoroughly mistaken. This is all the more true when the predictions involve society as a whole. Brockman's authors, who are limited to an average of a dozen pages, mostly stick to their particular lasts, and make little effort to envisage the whole. Where they do predict societal trends, they seem to risk naivety.
The coverage is biased towards the cognitive sciences, in the generalised sense: there are no fewer than six psychologists, three neuroscientists and three artificial-intelligence researchers out of the 25 authors. The remaining 13 writers are a hefty number (five) of biologists, only four workers in the physical sciences - two physicists, an astronomer and a chemist - two mathematicians and two computer scientists. Since one of the physicists, Paul Davies, devotes his entire essay to the search for life on Mars, and the astronomer Martin Rees spends half of his article on the search for extraterrestrial intelligence, the coverage of hard physical, experiment-based science, especially of the terrestrial kind, is at best sketchy. Surprising, too, is the omission of so much of science: anthropology, geology and geophysics, agriculture, technology and engineering in general, and ecology and climatology. The last two particularly, it is all but universally accepted, will have a lot to say during the next 50 years. Does this bias reflect who writes well for the lay reader, or possibly that Brockman's authors are self-selected, or that he himself leans toward the current notion that the age of the physical sciences is past? Perhaps some of each; or perhaps as editor he wanted the tenor of the book to be upbeat, and on that account avoided the predictions of disaster that some of these sciences would present, and which in any case are often described and discussed.
Probably the only way to review a book as diverse as this is to pick out some notable chapters. The biologist Brian Goodwin begins well. He suggests that the way to predict the future is to take a really hard look at what is happening in the present. By way of illustrating this idea, he imagines looking ahead from 1600 at what would happen in the next 50 years. The year 1600 already contained the roots of the scientific revolution to come. One could also have seen the roots of the disastrous religious wars that were soon to break out and that would postpone this revolution. The parallel with today could hardly be clearer. After this excellent start, Goodwin, interestingly, goes on to emphasise a sort of Gaian holism.
John Holland, computer scientist and generalist, is one of the few to abandon his specialism to look at the whole. He stands out for being not entirely enamoured with computers, seeing prediction and planning as the computer's most socially important functions. In a rather broad discussion of our needs, he points to space and transport as crucial areas for improvement. Unlike others in the book, he notes - but then optimistically finesses - the demands of our ever-increasing population.
Another computer scientist, Jaron Lanier, emphasises the long record of over-optimism of computer advocates, in my view wisely. If processing power continues to develop as it has done, he warns, we will be living on "the planet of the help desks" because the development of software will lag behind. Lanier - and one or two others, notably David Gelerntner - do not accept that there is much, if any, progress, in that they ask: "Are you really better off than 15 years ago?", and believe that we have hit a "complexity ceiling". Lanier's proposal for avoiding this fate seems sensible: to abandon the monolithic style of bit strings and protocols, and to communicate between and with machines in a more "lifelike" manner.
Steven Strogatz, mathematician, points out that one of the inventions of the past half-century bids fair to become of ever-greater importance in the next: the computer experiment. He, like Lanier and his fellow mathematician Ian Stewart, laments the lack of a proper theory of complex systems, and sees this as an essential problem for the future. To some extent, Strogatz bypasses the complexity ceiling by proposing that we use computer experiments in the form of "agent-based modelling" to understand complex systems.
One of the more perceptive insights in the book comes from the psychologist Paul Bloom who proposes that our understanding of psychological phenomena must become clearly supported by explicit explanations in terms of neurological and evolutionary reality. For instance, he emphasises the biological basis of moral development as an almost untouched problem. Geoffrey Miller seconds this.
A few of the articles beat the drum for the author's controversial position within their specialism, merely forecasting the triumph of that position. My scepticism about Judith Rich Harris' attempt to exclude the influence of parenting on development is strengthened by personal experience as a guinea pig for the psychological test-builders, when I learnt to mark false any multiple-choice answer that contained "always" or "never". Nonetheless, one appreciates Harris' argument that behaviour is deeply influenced by context. Paul Ewald's contention that all disease is infectious also fails to convince. Other articles that seem blinkered are those of Alison Gopnik on learning and Stuart Kauffman on the nature of life: they may be interesting and profound, and relevant as contributions to society's toolkit of scientific truth, but they are not very predictive.
Roger Schank, in discussing the future of education, tells amusing anecdotes about the excesses of the "great books" idea of education and teaching a fixed canon of literature. But in discussing computers in education, he then goes far along the information highway, seeming to assume that because in 50 years the facts will be available instantly "from the walls", one needs only to teach how to search for them and so the present structure of education as we know it will fade away into virtual schools. One of the few things that artificial intelligence work has taught us is that meaningful communication cannot take place without an agreed substrate, a factual context external to those who are doing the communicating. Someone who has all of knowledge available may know nothing.
Schank is only one of those bedazzled by the possibilities of computers and of modern biology. Several contributors say that "we" will have this or that gadgetry or medical technique: a chip containing our personal genome for Richard Dawkins, something similar for Samuel Barondes, silicon and steel spare parts for Rodney Brooks, brain scans for Nancy Etcoff, "swappable minds" for Marc Hauser. Who is "we"? Everyone in the populations of developed countries? Are we really going to split the human race irrevocably into the fortunate and the rest? Mihaly Cziksentmihaly goes the farthest, proposing to bring "comfort and joy" to mankind by means of bioengineering.
After such techno-optimism, Robert Sapolsky, ethologist and biologist, sounds a welcome note of commonsense. He surveys all of the diseases of the 20th century and remarks that probably the most widespread and debilitating was, is and will continue to be depression. But to be fair, the above pieces do contain thoughtful and humane remarks, such as Joseph Ledoux's prediction that there will be more study of individuality, and Etcoff's presumption that some form of psychotherapy will still be with us in 2050.
I have saved the physical sciences for the last. The chemist Peter Atkins seems not to make much of chemistry: "No new understanding" will appear, he predicts, only incremental advances. I suppose that can be made true by defining any new understanding in chemistry as belonging to physics or biology. Both Rees and Lee Smolin, on the other hand, by confining themselves almost exclusively to the relativistic extremes of the big bang and quantum gravity, have plenty to say, though most of it is speculative and unfalsifiable. Smolin proposes seven questions that he hopes can be answered, at least one of which strikes me as crucial: "What is dark matter?" We should deal with that before hoping for any real progress in cosmology. I also resonate with the problem of the wide variety of scales in physics - from subatomic to galactic. Rees, as an astronomer, must focus on cosmology as a major part of his bailiwick, but why doesn't Smolin, the book's only representative of physics (as it turns out), at least mention the other three-quarters or more of physics? If, with Dirac, he feels that "the rest is chemistry", he should say so. Instead, Smolin leaves it to the mathematicians to mention that there are exciting theories - often physical ones - of complex systems; and there will be more of these.
The sum of these pieces is a book that may not be anyone's useful or handy guide to the future, but is full of all kinds of intellectual effervescence, thought-provoking points of view and good writing aplenty. There is excellent science here, mind-stretching and accessible to general readers. Perhaps I am wrong to feel a bit disappointed by the almost complete absence of prediction of what we expect or hope will happen in the physical world at the human scale, and of how or whether the wonderful science discussed by the authors will affect a globally warmed and overpopulated world swarming with conflict. I would have liked to learn more about possible technical fixes for the problems that we - I mean all of us - undoubtedly face.
Philip W. Anderson, Nobel laureate, is professor of physics, Princeton University, US.
The Next Fifty Years: Science in the First Half of the Twenty-First Century
Editor - John Brockman
ISBN - 0 297 82925 4
Publisher - Weidenfeld and Nicolson
Price - £12.99
Pages - 301