Increasingly often, modern science is being attacked by people accusing it of failing to honor its own rules. The opening salvo was Smolin's The Trouble with Physics
, which caused some damage; it was taken seriously within the scientific establishment and widely read outside it. Other authors attempting to do the same thing have been less successful. Non-experts found Woit's Not Even Wrong
too mathematical and demanding; Berlinski's The Devil's Delusion
was written off as creationist propaganda; Kragh's elegant and meticulous Higher Speculations
somehow failed to have an impact; and Unzicker's Bankrupting Physics
came across as an ignorant rant. But I have a feeling that Baggott is about to score another hit. He has worked as a scientist himself, appears well acquainted with the facts, writes nicely, and presents his material in a balanced and responsible way.
The book is divided into two halves. "Farewell to reality", says Baggott in the title, and very properly starts by considering at length the questions of what "reality" is, how science might presume to be able to find out anything about it, and what methods it can legitimately use. He considers numerous cases from the history of science and philosophy, and concludes that it is hard to frame clear general rules. The borderline between physics and metaphysics is not as sharply defined as one would wish (quantum mechanics, or for that matter Newton's idea of absolute space), and it is not straightforward to define what it means for a theory to be empirically testable (it was surprising that Dirac's ideas about anti-matter were soon validated, and experts believed for a long time that no one would ever be able to detect neutrinos). Nonetheless, he argues that there is broad agreement on what constitutes normal scientific reasoning, even if there is a substantial gray area around it.
He then spends four chapters presenting an overview of modern physics, concentrating on the fields where controversy has arisen: quantum mechanics, particle physics, relativity and cosmology. In the final chapter of the first half, he explains why the currently accepted mainstream picture must be incorrect or incomplete: he is particularly worried by the quantum measurement problem (Schrödinger's cat), the unknown nature of dark matter, the lack of explanation for the masses of the elementary particles, inability to make quantum mechanics coherent with relativity, and the apparent fine-tuning of the universe's physical constants.
In the second half, Baggot goes on to talk about the speculative attempts physicists have made over the last 40 years to address known gaps: supersymmetry, superstring theory, the multiverse (both the Many Worlds Interpretation and eternal inflation), the Cosmic Landscape and the Anthropic Principle. He argues that these are, simply, not science as it is generally known: they involve piling one assumption on top of another, with no empirical evidence at all to buttress it. When the whole shaky tower is complete, we have the theoretical edifice which Susskind presents in The Cosmic Landscape
. Superstring theory (no empirical support) builds on supersymmetry (no empirical support), and gives rise to the vast Cosmic Landscape of different potential variant theories (no empirical support). Eternal inflation (no empirical support) means that all these different variant universes actually exist. The Anthropic Principle (according to Baggott, not even a scientific idea), then allows us both to argue that we have a solution to the fine-tuning problem, and
, simultaneously, to claim that the fact that the fine-tuning problem has been solved somehow validates the other links in the chain; circular reasoning at its finest.
I had the same reaction as Baggott when I read Susskind's book: this doesn't make sense. I was startled by Susskind's assertion that the choice was between the Landscape and some version of Intelligent Design, and even more startled to see Richard Dawkins endorsing it. Baggott doesn't like ID any better than Susskind or Dawkins do, but he is is equally unimpressed by the Landscape. He honestly says that he doesn't know what the answer is. Over the last 500 years, science has had an incredible track record of explaining the physical world without recourse to supernatural explanations, and one would expect it to succeed here too. So far, though, it's completely unclear how it's going to manage that feat. The universe looks like it has been designed, but no one knows why, and the people who say they do know are the ones you should trust least.
If you haven't yet got involved in this fascinating debate, Baggott's book is an excellent place to start. Check it out.