Kane's attitude is the more scientific one
Yesterday, I mentioned Gordon Kane's paper based on his talk in Munich. Today, I noticed that
Among those, I was most interested in
The main claim that ignited the battle was Gordy's assertion that M-theory on \(G_2\) holonomy manifolds with certain assumptions had predicted the mass \(m\approx 126\GeV\) of the Higgs boson before it was discovered; see e.g. Gordon Kane's blog post in December 2011. David Gross responded angrily. I've tried to understand the calculation "completely" with all the details and so far, I have failed. I feel that Gordon would have been able to compress the calculation or its logic if it were a clean one.
On the other hand, I partly do understand how the calculation works, what the assumptions are, and I just find it plausible that it's entirely accurate to say that with those assumptions including some notion of genericity, M-theory on 7D manifolds does produce the prediction of a \(126\GeV\) Higgs without fine-tuning. This statement surely isn't ludicrously wrong like many of the claims that I often criticize on this blog and some very careful researchers (importantly for me, Bobby Acharya) have pretty much joined Gordy in his research and in the summary of their conclusions, too.
Gross' and Kane's attitudes to the exchange were dramatically different. Gordon was focusing on the assumptions, calculations, and predictions; David was all about polls and the mob. "Gordon, you will agree that most of us wouldn't agree that M-theory has predicted the Higgs mass." And so on. Yes, no, what of it? If there's some important enough prediction and you have missed it or you don't understand it, it's your deficiency, David. If a majority of the community doesn't get it, it's the mistake of the members of the community. None of these votes can settle the question whether it's right for Gordon to say that M-theory has made the Higgs mass prediction, especially if most of these people know very well that they haven't even read any of these papers.
(By the way, Gordon phrases his predictions for the superpartner masses as predictions that have gone beyond the stage of "naive naturalness" which is how the people were estimating the masses decades ago. These days, they can work without this philosophy or strategy – as David Gross often categorizes naturalness.)
I think that David was acting like an inquisitor of a sort. The mob doesn't know or doesn't like that you have made that prediction, so you couldn't have done so. Well, that's a very lame criticism, David. With this approach of a bully, I sort of understand why you have sometimes endorsed the climate hysteria, too.
Also, I disagree with one particular claim by Gross, namely his assertion that the situation was in no way analogous to the prediction of Mercury's perihelion precession by general relativity. That was a prediction that would have killed general relativity if it had been falsified. Nothing like that is true in the case of Kane's M-theory predictions, Gross says.
Now, this claim is just rubbish, David. First of all, just like in the case of many of the string/M-theoretical predictions, the precession of Mercury's perihelion wasn't a full-fledged prediction but a postdiction. The precession anomaly had been known for a very long time before general relativity was completed. Einstein has only used this postdiction as a confirmation that increased his psychological certainty that he's on the right track (his heart has stopped for a second, we have heard) – and Gordon and his collaborators have arguably gone through totally analogous confirmations that have strengthened their belief that their class of compactifications is right (and string theorists – like reportedly Witten – have surely gone through the very same feeling when they learned that string theory postdicted gravity, and perhaps other things). At least, I don't see a glimpse of a real difference between the two situations.
Second, on top of this problem with David's argumentations, it's simply not true that any of these predictions or postdictions would have killed general relativity to the extent that they would convince Einstein to abandon it. One could be afraid that we need speculations about Einstein's thinking to know what would have happened if the confirmation hadn't taken place. Fortunately, we know what Einstein would have thought in that case – because someone has asked him:
When asked by his assistant what his reaction would have been if general relativity had not been confirmed by Eddington and Dyson in 1919, Einstein famously made the quip: "Then I would feel sorry for the dear Lord. The theory is correct anyway." Famously enough, in the first edition of the Czech Elegant Universe by Brian Greene, your humble correspondent translated "dear Lord" with the Czech word "lord" indicating Eddington. The quote makes sense in this way as well, doesn't it? ;-) I wasn't too aware of God, the other guy whom Einstein may have had in mind.
But back to the main topic.
Einstein would have definitely not abandoned general relativity. If the bending of light weren't observed, he would look for other explanations why it wasn't – abandoning GR wouldn't be among his top choices simply because the theory is beautiful and theoretically robust but was still able to pass some tests of agreement with the well-known physics (the Newtonian limit etc.). Today, many string theorists are actually more eager to abandon string theory for possibly inconclusive reasons than Einstein has ever been willing to abandon relativity.
The only possible kind of a difference between the two theories' predictions (GR and M-theory on \(G_2\) manifolds) is the fact that we think that GR is sort of a unique theory while M-theory on \(G_2\) manifolds, even as the "class of generic compactifications on 7D manifolds that Gordon has in mind", is not quite as unique. Even within string theory, there exist other classes of vacua, and even the \(G_2\) compactifications could be studied with somewhat different assumptions about the effective field theory we should get (not MSSM but a different model, and so on).
However, this difference isn't a function of purely intrinsic characteristics of the two theories. GR seems unique today because no one who is sensible and important enough is pushing any real "alternatives" to GR anymore. But these alternatives used to be considered and even Einstein himself has written papers proposing alternatives or "not quite corrected" versions of GR, especially before 1915.
My point is that in a couple of years, perhaps already in 2020, the accumulated knowledge may be such that it will be absolutely right to say that the situation of GR in 1919 and the situation of M-theory on \(G_2\) manifolds in 2016 were absolutely analogous. By 2020, it may become clear for most of the string theorists that the M-theory compactifications are the only way to go, some predictions – e.g. Gordon's predictions about the SUSY spectrum and cross sections – will have been validated, and all the reasonable people will simply return to the view that M-theory is at least as natural and important as GR and it has made analogous – and in fact, much more striking – predictions as GR.
In other words, the extra hindsight that we have in the case of GR – the fact that GR is an older theory (and has therefore passed a longer sequence of tests) – is the only indisputable qualitative difference between the two situations. I think that every other statement about differences (except for possible statements pointing out some particular bugs in the derivations in Gordon et al. papers, but Gross has been doing nothing of the sort) are just delusional or demagogic.
Sadly, the amount of energy that average members of the scientific, physics, or string community dedicate to the honest reading of other people's papers has decreased in recent years or a decade or so. But whenever it's true, people should be aware of this limitation of theirs and they should never try to market their laziness as no-go theorems. The fact that you or most of the people in your room don't understand something doesn't mean that it's wrong. And the greater amount of technical developments you have ignored, the greater is the probability that the problem is on your side.