Not Even Wrong: The Failure of String Theory And the Search for Unity in Physical Law 
At what point does theory depart the realm of testable hypothesis and come to resemble something like aesthetic speculation, or even theology? The legendary physicist Wolfgang Pauli had a phrase for such ideas: He would describe them as "not even wrong," meaning that they were so incomplete that they could not even be used to make predictions to compare with observations to see whether they were wrong or not.
In Peter Woit's view, superstring theory is just such an idea. In Not Even Wrong, he shows that what many physicists call superstring "theory" is not a theory at all. It makes no predictions, even wrong ones, and this very lack of falsifiability is what has allowed the subject to survive and flourish.
Not Even Wrong explains why the mathematical conditions for progress in physics are entirely absent from superstring theory today and shows that judgments about scientific statements, which should be based on the logical consistency of argument and experimental evidence, are instead based on the eminence of those claiming to know the truth.
In the face of many books from enthusiasts for string theory, this book presents the other side of the story.
Reviews
The level of the book should be noted. Since the mathematics involved is well over the head of the lay reader, one has to trust Woit's descriptions of what various mathematical techniques do, and how they relate to physical theories. As Woit details in his chapter "The Bogdanov Affair", the ideas are sometimes so arcane that even experts can't tell what might be valid research, and what might simply be a hoax. Woit does his best to help the reader, however, by including juicy biographical details of physicists' lives, great quotes in the chapter headings, and good lists of both technical and popular references for the concepts in each chapter.
Now, what about string theory? Is it in trouble? Is it, um, unraveling? Or getting, um, tied up in knots? Perhaps it is. The author reports on his own doubts about it. An obvious problem is the lack of experimental verification. We see references to this problem from a number of people who have shown terrific insight about the physical world, such as Richard Feynman.
Still, there are some good reasons why many bright people have done so much work with strings. If theories that employ zero-dimensional particles aren't giving good results, it makes sense to try one-dimensional strings. And I think the work that has been done in assessing what such theories would look like has not been in vain. But is it leading in the right direction? I'm not so sure about that. Scientists have to try new things to solve problems. Sometimes, they clearly increase human knowledge, and the progression of advances from the first use of the Lie Group SU(3) by particle physicists in the early 1960s through the development of the Standard Model is a good example of that. Woit explains much of this extremely well.
And, of course, sometimes, new ideas do not lead anywhere, at least at first. That's what science is all about.
There is some material about how the scientific community operates. Do scientific fads get established, and if so does all this retard progress? Well, sure, we're all only human, and sometimes that can happen. But I think that there are always plenty of scientists who are willing to try new ideas, especially if they feel that too many of their colleagues are off on a wild goose chase. So I do not really share some of the author's concerns here. And there is no way I would insult the world's string theorists by comparing their work (or commitment to scholarship) to that of the Modern Language Association!
There is one other point I would like to comment on. Woit does discuss the anthropic principle. This is, at its simplest level, the statement that the universe we live in has to be one in which life could exist. That's obviously true, of course. But it can be extended to imply that there have been (or are, or could be) a vast multitude of other universes or realities with other properties, many of which could not support creatures such as ourselves. And I think this is quite fair. It is reasonable philosophically as well. If we regard all of Reality as the Universe since the Big Bang, we are making a rather provincial claim that Reality is confined to the products of the Big Bang, an event which occurred less than 20 billion years ago. That's less than five times the age of our planet! I think it is at least as reasonable to hypothesize that there may be much more to Reality than that!
Woit complains that some of the references to the anthropic principle simply make excuses for the lack of predictive power of string theory. And that may get some people to think that the anthropic principle is some sort of cop-out, like invoking Gods or Goddesses to explain why the parameters we see have the values they do. But that would be unfair. It really does not advance Science to simply give up, ascribing the nature of the Universe to the whims of an inscrutable Deity. But it is a scientific question to ask if some physical parameter is constrained to have the value it does or can have (or could have had) some other value. It might well be that scientists will conclude that some of these parameters appear to have arbitrary values. That could lead to a search for mechanisms that might vary these parameters. And that in turn might even lead to the discovery of some mechanisms that might explain some of this. So I think Woit's reaction to the anthropic principle may be unnecessarily harsh.
In spite of my occasional disagreements with the author's reactions to the state of particle physics today, I highly recommend this book.
