The Life of the Cosmos Lee Smolin Weidenfield and Nicholson.
New Scientist 19 July 97

IMAGINE a discussion of Darwin's theory of natural selection, Penrose's twistor theory, Lovelock and Margulis's Gala hypothesis, Conway's Game of Life, Mandelbrot's fractals and Prigogine's nonequilibrium thermodynamics, intimately intertwined with gauge theories in elementary particle physics, string theory, cosmology, general relativity, quantum mechanics and topological quantum field theory. Add philosophy, art, theology, politics and sociology, with homage to the relational view of space and time proposed by Leibnitz in contrast to the discredited absolute space and time of Newton.

Smolin has the ambitious goal of explaining the problems that confront the new generation of theoretical physicists and cosmologists, and proposes his own ideas towards their solution. He has undoubtedly made an impressive sweep across most of the unanswered questions in fundamental physics, but it is difficult to form a judgment about the parts of his story with which I have only a passing familiarity. Let me look first at a part that I am very familiar with-namely, elementary particle physics and quantum mechanics. Curiously, in his discussion of the standard model of the weak and electromagnetic interactions, Smolin makes no mention of Sheldon Glashow, the first to write down such a model and who shared the Nobel prize with Steven Weinberg and Abdus Salam. The omission of a reference to Glashow seems significant because Glashow is a well-known opponent of the string theory approach that Smolin embraces enthusiastically Smolin is honest, however, in his account of the present arbitrariness in string theory and its inability to predict the particle masses we see around us, along with its lack of "any realistic expectation of soon receiving guidance from experimental physics".

The centrepiece is a popular exposition of Smolin's Darwinian approach to cosmology, which he calls "cosmological natural selection". The basis for the theory is the apparent need for the twenty or so parameters of the standard model to be fine-tuned to an incredible degree of accuracy over a huge range of magnitudes, so that life, the Galaxy and the Universe can exist at all. It is widely believed that our Universe contains a huge number of black holes whose interiors correspond to unobservable regions. According to general relativity, all matter inside the black hole collapses to a point singularity. Smolin argues that quantum effects may come into play to remove this singularity and the collapsing black hole "bounces" back into an expansion. Smolin suggests that inside each black hole is a new expanding Universe with black holes. He then assumes that the twenty-odd parameters of the standard model change a little bit at each bounce and, as in Darwinian evolution, the "fitness landscape" of all possible universes evolves until universes in which the parameters correspond to a state of maximal black hole production predominate. Supposedly, the theory is testable by playing God and twiddling the set of parameters we observe in our Universe: we can then attempt to see whether or not such changes lead to more or less black hole production. Perhaps I am at odds with a new, postmodernist world, but such "random assumption" physics reminds me of the once-great University of Salamanca. In the 18th century, only one faculty set an examination with a single question repeated every year: "What language do the angels speak?" Fl

Tony Hey switched from particle physicists to parallel computing at the University of Southampton. He is author of Einstein's Mirror (Cambridge)