Note: This is being published simultaneously here and at Scientia Salon. Discussion will be at the Scientia Salon site.
Last month’s Strings 2014 conference in Princeton included two remarkable talks by prominent physicists, both of whom invoked philosophy in a manner unprecedented for this kind of scientific gathering. On the first day, Paul Steinhardt attacked the current practice of inflationary cosmology as able to accommodate any experimental result, so, on philosophical grounds, no longer science. He included a video clip of Richard Feynman characterizing this sort of thing as “cargo cult physics”. On the final day, David Gross interpreted Steinhardt’s talk as implicitly applying to string theory, then went on to invoke a philosopher’s new book to defend string theory, arguing that string theorists needed to read the book in order to learn how to defend what they do as science.
The book in question was Richard Dawid’s String Theory and the Scientific Method, which comes with blurbs from Gross and string theorist John Schwarz on the cover. Dawid is a physicist turned philosopher, and he makes the claim that string theory shows that conventional ideas about theory confirmation need to be revised to accommodate new scientific practice and the increasing significance of “non-empirical theory confirmation”. The issues of this kind raised by string theory are complex, so much so that I once decided to write a whole book on the topic (Not Even Wrong). A decade later I think the arguments of that book still hold up well, with its point of view about string theory now much more widespread among working physicists. One thing I wasn’t aware of back then was the literature in philosophy of science about “progressive” vs. “degenerating” research programs, which now seems to me quite relevant to the question of how to think about evaluating string theory.
I’ve written a bit about the Dawid book and earlier work of his (see here and here), although as for any serious book there’s of course much more to say, even if I lack the time or energy for it. Recently an interview with Dawid appeared, entitled string theory and post-empiricism, which summarizes his views and makes some claims about string theory critics which deserve a response, so that will be the topic here. In the interview he says:
I think that those critics make two mistakes. First, they implicitly presume that there is an unchanging conception of theory confirmation that can serve as an eternal criterion for sound scientific reasoning. If this were the case, showing that a certain group violates that criterion would per se refute that group’s line of reasoning. But we have no god-given principles of theory confirmation. The principles we have are themselves a product of the scientific process. They vary from context to context and they change with time based on scientific progress. This means that, in order to criticize a strategy of theory assessment, it’s not enough to point out that the strategy doesn’t agree with a particular more traditional notion.
Second, the fundamental critics of string theory misunderstand the nature of the arguments which support the theory. Those arguments are neither arbitrarily chosen nor uncritical. And they are not decoupled from observation. String theory is indirectly based on the empirical data that drove the development of those theories string theory aims to unify. But more importantly for our discussion, the arguments for the viability of string theory are based on meta-level observations about the research process. As described before, one argument uses the observation that no-one has found a good alternative to string theory. Another one uses the observation that theories without alternatives tended to be viable in the past.
Taking the second part of this first, Dawid seems to be claiming that Smolin and I don’t understand what he calls the “No Alternatives Argument” (discussed in detail in his book, as well as in this recent paper). In response I’ll point out that one of the concluding chapters of my book was entitled “The Only Game in Town” and devoted explicitly to this argument. To this day I think that a version of such an argument is the strongest one for string theory, and is what motivates most physicists who continue to work on the theory. The version of this argument that I hear often privately and that has been made publicly by theorists like Edward Witten goes something like:
ideas about physics that non-trivially extend our best theories (e.g. the Standard Model and general relativity) without hitting obvious inconsistency are rare and deserve a lot of attention. While string theory unification hasn’t worked out as hoped, we have learned a lot of interesting and unexpected things by thinking about string theory. If they see a new idea that looks more promising, string theorists will shift their attention to that.
This is a serious argument, one that I tried to seriously address in the book. Beyond that, more naive versions of it seem to me to have all sorts of obvious problems. Of course, if you really can show that alternatives to a given model are impossible, that’s a convincing argument for the model, but this is rarely if ever possible. Working scientists beating their heads against a hard problem are always in the position of having “no alternatives” to some flawed ideas, until the day when someone solves the problem and finds the alternative. The only example I can recall seeing from Dawid of a successful example of the “No Alternatives Argument” is the discovery of the Higgs, and I find that very hard to take seriously. Pre-2012, the Standard Model was a very precise and exhaustively tested theory, providing a huge amount of indirect evidence for the Higgs. There were plenty of alternatives (technicolor, SUSY, etc.), all much more complicated and with no evidence for them. Making a “No Alternatives Argument” for a theory with overwhelming experimental evidence behind it is something completely different than trying to do the same thing for a theory with zero experimental evidence.
As for the other mistake that Dawid thinks string theory critics make, that of believing in some unchanging notion of empirical theory confirmation, the first thing to point out is that of course every theorist is well aware that one can can’t just demand experimental predictions and confirmation for ideas, that one spends basically all one’s time working on better understanding ideas that are far from the point where empirical confirmation comes into play. The second thing to point out is that I agree completely with Dawid that as experiments become more difficult, one needs to think about other ways of evaluating ideas to see if they are going anywhere. The last chapter of my book was devoted to exactly this question, arguing that physicists should look carefully at how mathematicians make progress. Mathematics is certainly “post-empirical”, and while logical rigor is a constraint, it is not one that necessarily points mathematicians to fertile new ideas. There is a long history and a deeply-ingrained culture that helps mathematicians figure out the difference between promising and empty speculation, and I believe this is something theoretical physicists could use to make progress.
The epigram from that last chapter though was something that kept going through my head when thinking about this, a line from Bob Dylan’s “Absolutely Sweet Marie”:
But to live outside the law, you must be honest.
Yes, theoretical particle physics is in a stage where empirical results are not there to keep people honest, and new and better “post-empirical” ways of evaluating progress are needed. But these must come with rigorous protections against all-too-human failings such as wishful thinking and Lee Smolin’s “groupthink”, and I just don’t see those anywhere in Dawid’s proposal for new kinds of theory confirmation.
I’d like to thank Massimo Pigliucci for the opportunity to write something here at Scientia Salon, and hope it will generate an interesting discussion. Contributions from philosophers to this kind of debate in physics I believe are very much needed, on this issue and others. Don’t even get me started about the multiverse…
Update: Frank Wilczek, unlike Gross and Schwarz, is not a fan of Dawid. From Twitter:
Wheeler: “Mass Without Mass”, “Charge Without Charge”, “Field Without Field” http://bit.ly/1p2cWQv
Dawid: “Physics Without Physics”
Update: Sabine Hossenfelder blogs about this here, with a response from Dawid here. Dawid writes
My claim is that a strong record of empirical confirmation in a research field under certain conditions can increase the probability of the viability of other theories in the field that have not been empirically confirmed. The currently predominant philosophical conception of theory confirmation (Bayesianism) equates confirmation with the increase of the probability that a theory is true or viable. For that reason I speak of “non-empirical theory confirmation”.
This seems to just be an argument that HEP theorists have been successful in the past, so one should believe them now, an argument in conflict with the argument that things have changes due to the difficulty in getting relevant experimental data.