In the Dark

Front Cover

I couldn’t resist including the following which is taken from the Times Literary Supplement (March 28, 2008, No. 5478). I get the TLS mainly for the crossword, and was chuffed when they actually published a review of my book From Cosmos to Chaos, published by Oxford University Press in 2006, and which was also reviewed in Nature and Physics World.

Between you and me the book developed out of a number of bits and pieces about probability theory I had written over quite a long time but never published. I cobbled them together in a rush and the book is a bit of a mess, really. Had I had more time it might have been more coherent. Perhaps. And it didn’t help that OUP didn’t allow me to correct the proofs, so there are lots of typographical errors. Anyway, reviewers have been very generous, particular Jim Bennett (Director of the Museum of the History of Science at Oxford) who wrote the review from which I have taken the following excerpt. And in case you think I edited out the bad bits, that’s not true. He actually reviewed four books in one article and I’ve just taken the bit about mine.

We can turn from inclusiveness and caution to the refreshingly opinionated writing of Peter Coles in From Cosmos to Chaos. This is really a book about probability theory and its application to different branches of science, but Coles is a Professor of Astrophysics, and cosmology is one of the most evident strengths of his book. Here again we learn much besides about our author: he supports Newcastle United, follows cricket and is keen on gambling. His is the only book of these four that has any formal mathematics to speak of, and we are encouraged not to give up at the first hurdle. He also uses illustrations from card games and seems oblivious to the fact that his fascination with contract bridge is just as likely as his affection for mathematical formulae to put readers off.

Coles’s preferred methodology in probability theory is Bayesian, based on an assignment of probabilities, understood as degrees of reasonable belief, to possible outcomes, rather than deriving them from frequency-based statistics.

This preference is carried throughout From Cosmos to Chaos and its epistemological implications are readily embraced. The “standard model” in particle physics, for example, is not absolutely right, but is currently the best bet among the alternatives. The Big Bang is not certain but the best available model given the present array of observational data. A “Theory of Everything” will not, pace Stephen Hawking, reveal the Mind of God – that is “silly” – it will be the most economical description of the universe and a good way of saving paper. The concept of entropy has a “subjective” aspect, not in the sense that anyone can use it as they choose, but because it arises from “the way we manage our knowledge about nature rather than (being) about nature itself’. Here there is a genuine engagement between the scientist, the historian and the philosopher of science.

Is it his approach based on the assignment of reasonable belief that has liberated Coles to express such clear preferences and opinions on all manner of theories? He is good on the difficulties and inadequacies in quantum mechanics, and charming in telling us that, having been warned beforehand that the subject would be confusing, he studied it for three years before realizing “what was the correct way to be confused about it”. He is amazed that the Copenhagen interpretation (where an act of measurement compels a realization of one or other physical states that existed previously only within a distribution of probabilities) could have been embraced seriously by so many highly intelligent people; but he has even less time for the multiverse, and contends that “in the gap left by the failure to find a sensible way to understand quantum reality there has grown a pathological industry of pseudo scientific gobbledegook”.

Coles suggests that the probabilistic descriptions given by quantum mechanics may simply arise from its incompleteness and he sees potential in a Bayesian approach, where quantum states are understood as states of knowledge rather than states of reality. He is pessimistic about the value of string theory: its apparent unconcern for predictable outcomes sets it outside scientific practice, while its plethora of possible accounts of our universe, known as the “string landscape”, would be better called the “string scrap-yard”.

Coles’s mathematics is not always easy to follow, but it seems to occupy its proper place, with the voice of the physicist helping us to position and appreciate it even without full understanding. In the chapter on the Big Bang, for example, the general reader may not understand all of the technical accounts, but she will get a real sense of what cosmology is and the kinds of claims it makes. These are not dogmatic but offered with a kind of realistic integrity and concluded by a series of “open questions” – fundamental but not yet answered. In the last chapter, probabilistic reasoning is applied to questions closer to everyday life, such as medical statistics and expert witnessing, and in a final – seemingly incongruous but enjoyable – addendum Coles addresses the breakdown of trust between scientists and the public. This does not arise from his subject and seems to be there just because the author – characteristically, one feels – had things he wanted to say. While bemoaning decline in the distribution of science understanding, he also berates the baleful effect of the scientific zealot, insisting that the scientific approach is pragmatic rather than idealistic. Coles urges scientists to engage honestly with the public and educationalists not to dumb down the school curriculum.