Until about twenty years ago, theorizing the origins of the universe was closer to philosophy or even to theology than to science. Now, however, it is possible to reach back scientifically into the very early moments of the universe and this has extraordinary implications. In this book Martin Rees presents an authoritative account of what is known and what is thought to be the case or might just conceivably be the case. Rees was persuaded by his editor to include a certain amount of speculation and this was sensible, otherwise the book might have been somewhat dry.
Rees is a good populariser of science. Unlike many such, he is himself a practising scientist of distinction, but he writes well and understands how to make a difficult idea clear. Nevertheless, the book requires close concentration from the reader, who needs to have a serious interest in the subject in order to persevere. At times the ideas seem to reach the limit of what can be made comprehensible with words rather than mathematics and the non-mathematical reader is left, perhaps inevitably, with the feeling that the full significance hasn't come across. One place where I felt this was the case was in the discussion of cosmic strings and the relation of these to magnetic monopoles.
A recurring theme throughout the book, with which it begins and also ends, is the occurrence of strange number coincidences which had to be there if life was to exist. Some are familiar to everyone (for example, the fact that ice floats, so that ponds don't easily freeze solid), while others are more abstract, such as the strength of the nuclear forces or the fact that the electron weighs so little compared with the nuclei of atoms. If these values were different stars would not exist or would have quite different properties, and the same is true of atoms. Some have seen in these things evidence of planning, of the hand of God, but Rees inclines to the view that our entire universe is just one item in the 'multiverse'. There may be innumerable other universes which are unsuitable for life but we, obviously, couldn't have evolved in them.
Rees is prepared to toy with the notion that life might exist in quite different forms, as envisaged by Fred Hoyle in his early novel 'The Black Cloud'. But perhaps the most far-out idea he touches on is the possibility that the next generation of nuclear colliders might cause a phase transition in the whole universe. Two such transitions are thought to have occurred in the evolution of the universe: one when the forces of electromagnetism and the 'weak' nuclear force emerged from a single 'electroweak' force, and an earlier one when all the fundamental forces except for gravity emerged from a single primeval force.
Is a third transition possible? The Harvard theorist Sidney Coleman suggests that it is, if the universe is 'supercooled', ready to change state when triggered by some local concentration of energy. This would, indeed be the ultimate catastrophe. The possibility isn't absurd; the only consolation is that it would happen at the speed of light so we'd know nothing about it. As Rees remarks, we can only hope that if there are any extraterrestrials with greater technical resources, they are suitably cautious.