Critical Mass: How one thing leads to another by Philip Ball

Could the laws of physics really help us to solve our social problems? James Harkin admires an elegant argument but finds that it doesn't go quite far enough

More than a book, this is an intellectual curiosity. Philip Ball, a science writer of some distinction, has persuaded a publisher to let him argue, over 650 densely packed pages, that the laws of physics can help solve social problems.

Ball starts by presenting us with aerial views of the way a throng of visitors fills up a crowded art gallery, or the way that cars move along a crowded motorway. Because the direction which each art lover takes is determined by a subtle negotiation with each of his fellows, and because each driver on the motorway modifies his speed in response to what the driver ahead does, Ball concludes that there are some things which cannot be reduced to the sum of individual actions and which can only be understood within groups. The modern discipline of economics treats people as discrete individuals who are only out for what they can get. Within the "softer" social sciences like sociology, on the other hand, individuals are often stuck together in a kind of formless glue, mutually suffocated by the weight of culture and tradition.

Between those two caricatures, Ball suggests, lies a whole world of human interaction in which the activities of human groups cannot be explained by adding up the sum of their parts. Since physicists are accustomed to thinking in terms of the interaction between things rather than about things themselves, their methods can add something distinctive to our understanding of society. Collective social processes, in short, can be seen to have their own physical laws of shape and form.

Ball wants to press his "social physics" in the service of improving the human condition. "The more we can understand and predict the ways in which we instinctively want to move around our environment," he says, "the better we are likely to be able to create places where people feel relaxed, comfortable and considerately housed." He is not shy of drawing attention to his distinguished antecedents. The idea that the methods of physics and the other natural sciences might help us understand human affairs, he points out, was a natural outgrowth of the heady ambitions of Enlightenment thought. The abstractions and ruthless logic through which the classical political economists sought to understand society might be simplifications, he argues, but it is rather cheap to suggest that they are inhuman. The search for the mechanism which can explain how society works has been most enthusiastically pursued by liberals and radicals - men like Adam Smith and Karl Marx - whose concern was to understand society as a means of improving it.

Despite all the graphs and bar charts in this exquisitely produced and painstakingly researched book, there is the sneaking suspicion that we have been here before. In his recent book The Tipping Point, Malcolm Gladwell sought to explain how fashions change and crime rates suddenly vary using techniques drawn from psychology and epidemiology. In Butterfly Economics, Paul Ormerod used biology to much the same effect. Now Ball comes along to tell us, for example, that sudden changes in the way that crowds behave are akin to "phase transitions" in the world of physics. A long line of cars on a motorway, for example, is presented by Ball as a relatively stable physical body. But just as steam abruptly condenses to water, or snowflakes suddenly appear when water vapour freezes into ice, a random movement by one car can convert the whole entity into a congested state.

Ball writes patiently and eloquently about both the history of political economy and developments in modern physics, but there is a degree of recoil when he brings both together. The problem is that his notion of a "social physics" is capable of being pushed a long way, but was always destined to run out of steam. Some of the most exciting chapters in this book are inspired not by physics at all but by game theory, an exotic outpost of mathematics which has proved useful in social theory. The prisoner's dilemma game, for example, is the most intractable logical puzzle of the last 100 years, and might be used to deduce everything from the role of the State to the dynamics of social movements. But here again Ball spends too long in the laboratory, and takes us away from developing a logical theory of society which can be tested by empirical confirmation. He forgets that society has a logic of its own, and the laws of physics can be no more than an inspiration in the battle to decipher it.

The problem is not that Ball's attempt at a physics of society is too ambitious, but that it is not ambitious enough. Whereas the frigid templates of modern economics tend towards reassuring equilibria, Ball's idea of "phase transitions" is biased in the direction of social anarchy and the idea that small events can have disproportionate consequences. But all this goes with the peculiar terrain which Ball and his peers have made their own, the analysis of crowding and the statistical tyranny of small decisions. Their preoccupations suggest that society is a terrifying, amorphous mass, subject to random and violent fluctuations, and that change can only be achieved at the margins, with a wave here and a nod there to ease things along. This kind of "social physics" is deeply timid, nothing more than the science of signposting.

Physics and the other natural sciences, Ball records, have been expanding into the territory of the social sciences for the last 20 years. It would be tempting to attribute that to the arrogance of physicists. But, as Ball's survey of academic literature shows, much of the plundering of the laws of physics has been done by social scientists who have tired of their own techniques. "Over the past two decades," he tells us, "something extraordinary has been happening in this field of science. Tools, methods and ideas developed to understand how the blind material fabric of the universe behaves are finding application in arenas for which they were never designed, and for which they might at first glance appear ridiculously inappropriate. Physics is finding its place in a science of society." Ball would be hard put to divine a natural law which can tell us how that came about.

The logical problems upon which the classical political economists went to work have largely been resolved or sidelined. The pressing questions at the beginning of the 21st century - among them the way that we approach risk, the state of the environment, the progress of globalisation and the effects of the politics of human rights - are very different, but just as prone to riddle and wrong-headedness. Ball's search for a "social physics" is a rousing call-to- arms, and an elegant answer to the shallow tradition of British empiricism, for whom everything beyond the immediately observable comes as an uninvited surprise.

If too much of his arsenal is dedicated to understanding the motion of crowds rather than the broader forces in which we are all embedded, it need only be said that scientific enquiry begins by trial and error. The right answers to many of the logical problems thrown up by Ball surely exist. It is up to us to ask the right questions.