John D. Barrow: The Universe That Discovered Itself (1987, 2000)

Edition: Oxford, 2000 (revision of The World Within the World, 1987)
Review number: 1001

Barrow's book, an updated version of The World Within the World, is a philosophical look at the history of science and contemporary scientific ideas with a rather unusual slant. It takes a list of nine statements about the laws of science and how they relate to the underlying reality of the universe, and then sees a general trend up to the work of Newton to establish these statements, followed in the twentieth century by the opposite trend with the development of new theoretical frameworks very different from the Newtonian one. These statements are things like "Space and time exist" or "The world can be described by mathematics", and are a set of basic philosophical assumptions about the universe, informing scientists' attitudes to physical theories.

The Universe That Discovered Itself is aimed at the experienced reader of popular science. Even though brief explanations are given, it would be difficult to follow without a previous acquaintance with relativity, quantum mechanics, string theory, and the anthropic principle among other ideas. With a familiarity with these concepts and an interest in the philosophical, there is much pleasure to be obtained from the book. The title refers to the thought that we, as part of the universe, have discovered a great deal about it, and is particularly appropriate given the quite lengthy discussion of the role of the observer in quantum mechanics and particularly quantum cosmology.

The presentation is typical of Barrow, with each section enlivened by interesting and frequently amusing quotations, including the following anecdote. In an Oxford physics viva in the 1890s, a student was asked to define electricity. His response was that he did know but had forgotten, to which the examiner drily replied, "How very unfortunate. Only two persons have ever known what electricity is, the Author of Nature and yourself. Now one of them has forgotten".

I'm not sure how radical a revision was made to the earlier book, and there are some sections which seem to be less up to date than others, which is a pity. Still, I found The Universe That Discovered Itself a fascinating exploration of the philosophy behind modern physics.

John Barrow: The Origin of the Universe (1994)

Edition: Phoenix, 1995
Review number: 649

Part of a series of explanations of important areas of current science by leading science writers, The Origin of the Universe would be an ideal place for to start for a reader with virtually no scientific background who wants to try to understand something of current thought on the subject. It is concise and simple, admirably written, and has the odd point of interest even to a voracious reader of popular science books. It reminded me of the science books of Isaac Asimov, and is a worthy successor to his books on cosmology, written skilfully in the same spirit yet up to date. I found Asimov's books fascinating when I was a teenager - they were all about the interesting science they didn't teach at school - and I think that The Origin of the Universe could have the same effect on an interested teenager today.

John D. Barrow: Impossibility: The Limits of Science and the Science of Limits (1998)

Edition: Oxford University Press, 1998
Review number: 483

In what is almost a response to John Horgan's The End of Science, Barrow examines the limitations of scientific thought from several different points of view with the aim of working out what science can say about what it cannot say. He skims quickly over some of the problems Horgan talks about, such as the increasing economic cost of scientific experimentation; these limitations are not scientific in nature (non-scientific events such as a change of government may change their nature) and there is little that can be said about them beyond acknowledging their existence.

Barrow is far more interested in the limitations inherent in modern scientific theories, such as the impossibility of knowing what happens outside the edge of the visible universe. He concentrates on the less well known ideas, rather than ploughing once again the well worn furrow of the popular account of relativity and quantum mechanics. His final section is a brief but sensible account of Gödel's Incompleteness Theorem and its relationship to physics. The problem with this relationship is that it is only possible to determine its nature when the more basic question of how mathematics is embodied in the universe is answered. If mathematical physics is only a description of patterns in the universe, for example, then there is not necessarily any connection. Even if sufficiently complex mathematics is in some way embodied in the universe - you need to have arithmetic with both addition and multiplication - then it is not at all clear what the physical version of a Gödel Undecideable Sentence would be (it would depend on the precise nature of the embodiment, for a start).

Barrow is less polemic than Horgan, more interested in the nature of the various types of scientific impossibility than in ramming home the point that there are limitations to science. Barrow is much more pro-science than Horgan - he is after all a research physicist - which means that his book is less excitingly iconoclastic but perhaps more informative. (The structure of the book also helps here; Horgan's is organised around interviews with prominent scientists which means that his main philosophical points are hidden behind personalities.)

John D. Barrow & Frank J. Tipler: The Anthropic Cosmological Principle (1986)

Edition: Oxford University Press, 1996
Review number: 355

It may seem that there is not very much to be said about the anthropic principle, that it is an interesting sideline in the philosophy of science which may have a minor role in explaining why the universe is the way it is. To Barrow and Tipler, it has formed the peg around which a seven hundred page book can be written, one which takes the reader on a survey of cosmology, theology, the future of the human race, and the existence of extra-terrestrial intelligence. It is a fascinating book, occasionally rather on the mathematical side for a popular science book.

The anthropic principle, as discussed by Barrow and Tipler, comes in three varieties, with a "Final" form as well as the more familiar "Weak" and "Strong" versions. The Weak Anthropic Principle is hardly contentious. It merely says that the existence of carbon based life is an observed fact, so that the universe must have properties which make such life a possibility. Barrow and Tipler make as strong a case as is possible for the explanatory power of this idea, but I still feel that it is limited. It may explain, for example, that the universe has to be large even if the Earth is the only planet containing life (to have expanded for long enough for galaxies to form and supernovae to occur to create some of the elements we require), but not why the universe happens to be this large. All the principle states is that if it weren't, we wouldn't be here to observe the fact. However, most of what can be inferred from it doesn't actually require the presence of life; the example I've referred to could be deduced just as logically from the existence of uranium. (Life is a sufficiently complex phenomenon that it requires a large collection of such pre-conditions, so the anthropic principle is a convenient summary of many similar explanations.) It also involves the deduction of causes from effects, and that is something which requires a great deal of care, to say the least.

The stronger versions of the anthropic principle are far more contentious, and more closely related to the design arguments used to "prove" the existence of God from the appearance of design in the universe. (These arguments are summarised in an excellent historical overview which forms the first chapter of the book.) The standard strong principle says that life must evolve at some point in the history of the universe, rather than that it just has evolved. As Barrow and Tipler point out, this means that life can be said to be part of the "purpose" of the universe in some way, and this doesn't make much sense without the deduction that life must at some point have a measurable effect on the whole cosmos. This point leads into a lengthy discussion of just what this effect could possibly be, which is fascinating but extremely speculative. The first point made is that it is very difficult to imagine any way in which a species confined to a single star system could affect the universe. So interstellar travel is a necessary development, and that requires intelligence. This is the motivation behind the authors' formulation of what they call the Final Anthropic Principle, which states that intelligence must at some point arise and never die out.

The discussion of how interstellar (and, indeed, intergalactic) travel could be developed is fascinating and seems convincingly feasible. Their ideas are based on the theoretical von Neumann machine, which is basically a machine which can create replicas of itself. A von Neumann machine could be made a space probe that seeks out a star likely to have the resources to enable replication (using a strategy based on analysis of the Polynesian colonisation of the Pacific islands), and then copies itself. Given sufficient processing power to be considered intelligent and a sufficient density of planetary systems - considered likely in current astronomy - this would amount to colonisation of the galaxy by intelligent systems over a period of several thousand years.

In fact, these arguments are sufficiently convincing that they are used to support the idea that there is no more advanced race of beings in the galaxy than humanity, because we should now have been contacted by probes of this sort. (Even if they did not want to directly contact other forms of life, the action of such a probe on reaching the solar system would probably be detectable.) The idea that we are alone in the galaxy, however, contradicts the equally convincing "Copernican Principle", that there should not be anything particularly special about the Earth - we are just a small planet orbiting a typical star in a typical part of the galaxy. The only way to reconcile this with the idea that a society only slightly more advanced than we are would have contacted us - and Barrow and Tipler estimate that von Neumann probes will be economically viable in a few centuries at most - is to argue that some catastrophe almost always destroys a civilisation between two and six hundred years after the Industrial Revolution (or its equivalent). This pessimism may seem justified in a society facing possible nuclear devastation, social disintegration and ecological disaster.

The authors do not dwell on this. It is really a long - and fascinating - digression. The main thread of the argument is rejoined with a discussion of the end of the universe in which some form of intelligent life has basically colonised the whole, and is trying to circumvent in some way the 'heat death' predicted by thermodynamics. This part is necessarily very speculative (cosmologists do not even agree on the broad details of how the universe will end), but certainly represents just about the only feasible way in which life could affect the whole universe.

The Anthropic Cosmological Principle is an extremely complex book, and is exactly the kind of science book I enjoy, finding a peg to discuss a large number of fascinating ideas that turn out to be connected despite appearances. The earlier chapters, about the Weak Anthropic Principle, are solid expositions of material which I've seen before (and which will probably be familiar to most people with an interest in the philosophy of science). The later writing, about the stronger principles, contains much less well-known science. I suspect that these versions of the anthropic principle are probably wishful thinking, the outcome of the desire to feel that we are significant. This doesn't invalidate much of the science contained in the book, which is an excellent one.