« June 7, 2006 | Main | June 9, 2006 »

Thursday, June 8, 2006

Reading List: Not Even Wrong

Woit, Peter. Not Even Wrong. London: Jonathan Cape, 2006. ISBN 0-224-07605-1.
Richard Feynman, a man about as difficult to bamboozle on scientific topics as any who ever lived, remarked in an interview (p. 180) in 1987, a year before his death:
…I think all this superstring stuff is crazy and it is in the wrong direction. … I don't like that they're not calculating anything. I don't like that they don't check their ideas. I don't like that for anything that disagrees with an experiment, they cook up an explanation—a fix-up to say “Well, it still might be true.”
Feynman was careful to hedge his remark as being that of an elder statesman of science, who collectively have a history of foolishly considering the speculations of younger researchers to be nonsense, and he would have almost certainly have opposed any effort to cut off funding for superstring research, as it might be right, after all, and should be pursued in parallel with other promising avenues until they make predictions which can be tested by experiment, falsifying and leading to the exclusion of those candidate theories whose predictions are incorrect.

One wonders, however, what Feynman's reaction would have been had he lived to contemplate the contemporary scene in high energy theoretical physics almost twenty years later. String theory and its progeny still have yet to make a single, falsifiable prediction which can be tested by a physically plausible experiment. This isn't surprising, because after decades of work and tens of thousands of scientific publications, nobody really knows, precisely, what superstring (or M, or whatever) theory really is; there is no equation, or set of equations from which one can draw physical predictions. Leonard Susskind, a co-founder of string theory, observes ironically in his book The Cosmic Landscape, “On this score, one might facetiously say that String Theory is the ultimate epitome of elegance. With all the years that String Theory has been studied, no one has ever found a single defining equation! The number at present count is zero. We know neither what the fundamental equations of the theory are or even if it has any.” (p. 204). String theory might best be described as the belief that a physically correct theory exists and may eventually be discovered by the research programme conducted under that name.

From the time Feynman spoke through the 1990s, the goal toward which string theorists were working was well-defined: to find a fundamental theory which reproduces at the low energy limit the successful results of the standard model of particle physics, and explains, from first principles, the values of the many (there are various ways to count them, slightly different—the author gives the number as 18 in this work) free parameters of that theory, whose values are not predicted by any theory and must be filled in by experiment. Disturbingly, theoretical work in the early years of this century has convinced an increasing number of string theorists (but not all) that the theory (whatever it may turn out to be), will not predict a unique low energy limit (or “vacuum state”), but rather an immense “landscape” of possible universes, with estimates like 10100 and 10500 and even more bandied around (by comparison, there are only about 1080 elementary particles in the entire observable universe—a minuscule number compared to such as these). Most of these possible universes would be hideously inhospitable to intelligent life as we know and can imagine it (but our imagination may be limited), and hence it is said that the reason we find ourselves in one of the rare universes which contain galaxies, chemistry, biology, and the National Science Foundation is due to the anthropic principle: a statement, bordering on tautology, that we can only observe conditions in the universe which permit our own existence, and that perhaps either in a “multiverse” of causally disjoint or parallel realities, all the other possibilities exist as well, most devoid of observers, at least those like ourselves (triune glorgs, feeding on bare colour in universes dominated by quark-gluon plasma would doubtless deem our universe unthinkably cold, rarefied, and dead).

But adopting the “landscape” view means abandoning the quest for a theory of everything and settling for what amounts to a “theory of anything”. For even if string theorists do manage to find one of those 10100 or whatever solutions in the landscape which perfectly reproduces all the experimental results of the standard model (and note that this is something nobody has ever done and appears far out of reach, with legitimate reasons to doubt it is possible at all), then there will almost certainly be a bewildering number of virtually identical solutions with slightly different results, so that any plausible experiment which measures a quantity to more precision or discovers a previously unknown phenomenon can be accommodated within the theory simply by tuning one of its multitudinous dials and choosing a different solution which agrees with the experimental results. This is not what many of the generation who built the great intellectual edifice of the standard model of particle physics would have considered doing science.

Now if string theory were simply a chimæra being pursued by a small band of double-domed eccentrics, one wouldn't pay it much attention. Science advances by exploring lots of ideas which may seem crazy at the outset and discarding the vast majority which remain crazy after they are worked out in more detail. Whatever remains, however apparently crazy, stays in the box as long as its predictions are not falsified by experiment. It would be folly of the greatest magnitude, comparable to attempting to centrally plan the economy of a complex modern society, to try to guess in advance, by some kind of metaphysical reasoning, which ideas were worthy of exploration. The history of the S-matrix or “bootstrap” theory of the strong interactions recounted in chapter 11 is an excellent example of how science is supposed to work. A beautiful theory, accepted by a large majority of researchers in the field, which was well in accord with experiment and philosophically attractive, was almost universally abandoned in a few years after the success of the quark model in predicting new particles and the stunning deep inelastic scattering results at SLAC in the 1970s.

String theory, however, despite not having made a single testable prediction after more than thirty years of investigation, now seems to risk becoming a self-perpetuating intellectual monoculture in theoretical particle physics. Among the 22 tenured professors of theoretical physics in the leading six faculties in the United States who received their PhDs after 1981, fully twenty specialise in string theory (although a couple now work on the related brane-world models). These professors employ graduate students and postdocs who work in their area of expertise, and when a faculty position opens up, may be expected to support candidates working in fields which complement their own research. This environment creates a great incentive for talented and ambitious students aiming for one the rare permanent academic appointments in theoretical physics to themselves choose string theory, as that's where the jobs are. After a generation, this process runs the risk of operating on its own momentum, with nobody in a position to step back and admit that the entire string theory enterprise, judged by the standards of genuine science, has failed, and does not merit the huge human investment by the extraordinarily talented and dedicated people who are pursuing it, nor the public funding it presently receives. If Edward Witten believes there's something still worth pursuing, fine: his self-evident genius and massive contributions to mathematical physics more than justify supporting his work. But this enterprise which is cranking out hundreds of PhDs and postdocs who are spending their most intellectually productive years learning a fantastically complicated intellectual structure with no grounding whatsoever in experiment, most of whom will have no hope of finding permanent employment in the field they have invested so much to aspire toward, is much more difficult to justify or condone.

The problem, to state it in a manner more inflammatory than the measured tone of the author, and in a word of my choosing which I do not believe appears at all in his book, is that contemporary academic research in high energy particle theory is corrupt. As is usually the case with such corruption, the root cause is socialism, although the look-only-left blinders almost universally worn in academia today hides this from most observers there. Dwight D. Eisenhower, however, twigged to it quite early. In his farewell address of January 17th, 1961, which academic collectivists endlessly cite for its (prescient) warning about the “military-industrial complex”, he went on to say, although this is rarely quoted,

In this revolution, research has become central; it also becomes more formalized, complex, and costly. A steadily increasing share is conducted for, by, or at the direction of, the Federal government.

Today, the solitary inventor, tinkering in his shop, has been over shadowed by task forces of scientists in laboratories and testing fields. In the same fashion, the free university, historically the fountainhead of free ideas and scientific discovery, has experienced a revolution in the conduct of research. Partly because of the huge costs involved, a government contract becomes virtually a substitute for intellectual curiosity. For every old blackboard there are now hundreds of new electronic computers.

The prospect of domination of the nation's scholars by Federal employment, project allocations, and the power of money is ever present and is gravely to be regarded.

And there, of course, is precisely the source of the corruption. This enterprise of theoretical elaboration is funded by taxpayers, who have no say in how their money, taken under threat of coercion, is spent. Which researchers receive funds for what work is largely decided by the researchers themselves, acting as peer review panels. While peer review may work to vet scientific publications, as soon as money becomes involved, the disposition of which can make or break careers, all the venality and naked self- and group-interest which has undone every well-intentioned experiment in collectivism since Robert Owen comes into play, with the completely predictable and tediously repeated results. What began as an altruistic quest driven by intellectual curiosity to discover answers to the deepest questions posed by nature ends up, after a generation of grey collectivism, as a jobs program. In a sense, string theory can be thought of like that other taxpayer-funded and highly hyped program, the space shuttle, which is hideously expensive, dangerous to the careers of those involved with it (albeit in a more direct manner), supported by a standing army composed of some exceptional people and a mass of the mediocre, difficult to close down because it has carefully cultivated a constituency whose own self-interest is invested in continuation of the program, and almost completely unproductive of genuine science.

One of the author's concerns is that the increasingly apparent impending collapse of the string theory edifice may result in the de-funding of other promising areas of fundamental physics research. I suspect he may under-estimate how difficult it is to get rid of a government program, however absurd, unjustified, and wasteful it has become: consider the space shuttle, or mohair subsidies. But perhaps de-funding is precisely what is needed to eliminate the corruption. Why should U.S. taxpayers be spending on the order of thirty million dollars a year on theoretical physics not only devoid of any near- or even distant-term applications, but also mostly disconnected from experiment? Perhaps if theoretical physics returned to being funded by universities from their endowments and operating funds, and by money raised from patrons and voluntarily contributed by the public interested in the field, it would be, albeit a much smaller enterprise, a more creative and productive one. Certainly it would be more honest. Sure, there may be some theoretical breakthrough we might not find for fifty years instead of twenty with massive subsidies. But so what? The truth is out there, somewhere in spacetime, and why does it matter (since it's unlikely in the extreme to have any immediate practical consequences) how soon we find it, anyway? And who knows, it's just possible a research programme composed of the very, very best, whose work is of such obvious merit and creativity that it attracts freely-contributed funds, exploring areas chosen solely on their merit by those doing the work, and driven by curiosity instead of committee group-think, might just get there first. That's the way I'd bet.

For a book addressed to a popular audience which contains not a single equation, many readers will find it quite difficult. If you don't follow these matters in some detail, you may find some of the more technical chapters rather bewildering. (The author, to be fair, acknowledges this at the outset.) For example, if you don't know what the hierarchy problem is, or why it is important, you probably won't be able to figure it out from the discussion here. On the other hand, policy-oriented readers will have little difficulty grasping the problems with the string theory programme and its probable causes even if they skip the gnarly physics and mathematics. An entertaining discussion of some of the problems of string theory, in particular the question of “background independence”, in which the string theorists universally assume the existence of a background spacetime which general relativity seems to indicate doesn't exist, may be found in Carlo Rovelli's "A Dialog on Quantum Gravity". For more technical details, see Lee Smolin's Three Roads to Quantum Gravity. There are some remarkable factoids in this book, one of the most stunning being that the proposed TeV class muon colliders of the future will produce neutrino (yes, neutrino) radiation which is dangerous to humans off-site. I didn't believe it either, but look here—imagine the sign: “DANGER: Neutrino Beam”!

A U.S. edition is scheduled for publication at the end of September 2006. The author has operated the Not Even Wrong Web log since 2004; it is an excellent source for news and gossip on these issues. The unnamed “excitable … Harvard faculty member” mentioned on p. 227 and elsewhere is Luboš Motl (who is, however, named in the acknowledgements), and whose own Web log is always worth checking out.

This is the three hundredth posting on Fourmilog since its inception in October 2004.

Posted at 00:12 Permalink