I just noticed that Gordon Kane has recently published a second edition of his 2017 String Theory and the Real World. Columbia doesn’t seem to yet have full online access to the second edition, but one can already compare the two editions in a few places. For instance, on page 1-5 of the 2017 edition one reads
The LHC is now working in a region of energy and intensity where well-motivated theories imply superpartners could be seen by late 2018.
There is good reason, based on theory, to think discovery of the superpartners of Standard Model particles should occur at the CERN LHC in the next few years.
The corresponding first chapter of the latest edition has:
The LHC has so far just entered the region of superpartner masses predicted by compactiﬁed theories, which ranges from about 1.5 to ∼5 GeV (we’ll discuss that range later). Those values are the only physics predictions, rather than just speculations. The LHC will run with higher luminosity after an upgrade, beginning in late 2021 if pandemic work stoppages do not delay it. That increases the possibility of discovery, though not very much. A higher energy collider is needed. From what we know now, a collider with twice the LHC energy range would probably sufﬁce, and cover the region of gluino masses to about 5 GeV.
The concluding chapter of the 2017 edition tell us:
The compactiﬁed M-theory implies that three superpartners (and only three) will be observed at the LHC in the current three-year run (assuming the full integrated luminosity is achieved). These are the gluino and the charged and neutral winos.
Presumably he’s talking about the LHC Run II (2015-18) which did meet its luminosity goals, without any hint of the three superpartners. I don’t yet have access to the later parts of the 2021 edition to see what they say about this.
This isn’t the first time Kane has published multiple editions of ever changing “predictions” about supersymmetry. At one point I compared the 2000 and 2013 editions of “Supersymmetry and Beyond”, you can see the results here.
I am writing this comment for the interested reader, not for Peter who I know understands the situation perfectly well.
Those particles are not scientific predictions. They are calculations based on theories that have been fabricated for the very purpose of having some “new” physics in the energy range of question. I have worked in this field myself and this is literally how it’s done: You make up a theory and fiddle with the details until it shows a deviation from the established theories in a range that may become experimentally accessible soon. Then you claim this is a “prediction”.
As I have said now for more than a decade, this methodology is obviously unscientific and must be discontinued. It has become accepted practice because it’s easy and convenient and allows people to crank out a lot of papers quickly. But it’s a waste of time and money, and certainly not a reason to build a bigger collider.
Many people are afraid of criticizing high energy particle physics because they don’t understand the math. But you don’t need to understand the math to see that these so-called predictions have as a matter of fact not worked since the 1980s. You can look at Kane’s own publications to see that his predictions have been falsified over and over again and he keeps on changing them. (Which Peter has documented on this blog.)
This is not a problem of “it’s just hard to measure” (as it was, eg, with the neutrinos or gravitational waves). This is a problem of people who eternally amend their so-called theories. They live under the mistaken impression that just because a hypothesis makes predictions it is scientific. This is a trivial misunderstanding of the most rudimentary philosophy of science. Guesses aren’t scientific predictions. And piling lots of math on top of guesses doesn’t change the fact that they were guesses.
I am not saying this because I am against particle physics (or other research on the foundations of physics). To the very contrary, I am saying this because I care about progress in the field. This methodology has not lead to progress. Researchers in the field urgently need to clean up their act.
I think it’s important to note that what Kane is doing now is different than what most “string theorists” are doing. A lot of his book and recent writings are actually aimed at criticizing the string theory community for admitting that there is no way they can predict anything about LHC-scale physics. He’s arguing for models and calculations that most of the rest of the string community has essentially disowned. They’re content to take the position that if SUSY is found that proves string theory, if it’s not, that’s no problem for string theory.
What I find remarkable about the Kane story, in his Physics Today articles, his earlier popular books, and now these two editions is the repeated blatant, on its face violation of conventional norms of scientific ethics. Most other people making models of the kind you describe are careful to stick to “my model shows there’s a possibility the LHC will see something”, and not say “if my model is right, the LHC will see something”. This is because they realize that if they say the latter, when the LHC sees nothing they will be ethically bound to admit “my model is wrong”, with negative implications for their career, funding, etc.
Kane instead deals with the ethical problem by just ignoring it, never admitting anything and doing the same thing time and time again. This is not usual behavior. What’s remarkable is that those responsible for maintaining ethical standards do nothing about it: the editors at Physics Today let him publish more articles doing the same thing, the publishers of his popular books and now the IOP let him publish revised editions. This really is not supposed to happen.
I notice that while the old popular books carried endorsements from others in the field, the IOP ones don’t seem to. My impression is that privately most string theorists find Kane an embarrassment, but will never say so publicly because they have for a long time circled the wagons and will not criticize behavior of members of their own tribe, no matter how egregious.
“What’s remarkable is that those responsible for maintaining ethical standards do nothing about it: the editors at Physics Today let him publish more articles doing the same thing, the publishers of his popular books and now the IOP let him publish revised editions. This really is not supposed to happen. … most string theorists find Kane an embarrassment, but will never say so publicly because they have for a long time circled the wagons and will not criticize behavior of members of their own tribe, no matter how egregious.”
You appear to be calling for censorship to solve the string theory hype dilemma, which is a bit like the tactics used in totalitarian regimes (by Stalin against Trotsky, for instance). This puzzles me. Surely censorship is always wrong in a libertarian world? Surely, truth will conquer all and the publishers of BS will go bankrupt when the correct model emerges? Forgive me for asking, but isn’t the way to defeat evil, to fight weak arguments with reality instead of trying to ban weak arguments from being published?
“My impression is that privately most string theorists find Kane an embarrassment, but will never say so publicly because they have for a long time circled the wagons and will not criticize behavior of members of their own tribe, no matter how egregious.”
I believe this is true to a large extent, but there are exceptions. During the “Why trust a theory” conference in Munich 2015, David Gross called out Kane for the “prediction” of the Higgs mass, and there was very vigorous discussion by Kane, Gross, Dieter Lust, and a couple of other people in the audience. I attended it, and later commented the event on your blog, here:
Afterwards, the recording of Kane’s lecture became available, here:
The “fun” part happens between slides 31 and 32 of Kane’s talk. 🙂
First of all, I’m not a libertarian and don’t want to live in a libertarian world.
A healthy science literature needs editors and referees, and here I don’t think they were doing their jobs.
I do remember that, but that case was a bit different, since Kane was claiming for himself personally a huge (bogus) scientific achievement.
Sure I agree that Kane is an extreme case and of course there are pretty much no string theorists supporting him, but his methods exemplify those used by an entire community. The idea is that you fabricate some fancy explanation for why your particular guess is a good one, and then you do a series of impressive looking calculations which you call “predictions”. That lives up to the scientific standard of today and you’ll easily get it published in good journals because they can’t think of a reason to reject it. The reason they should reject such guesses is, of course, that just because a hypothesis is testable doesn’t mean it’s scientific.
What’s happened with Kane, I believe, is that since he’s waded into science communication he was forced to express himself clearly which resulted in big claims that now come back to haunt him. Most people in the field will make sure their predictions have loopholes so that they can be amended should need come. This isn’t hard to do since those predictions were guesses to begin with. Since Kane is only a particularly egregious example of a much more general problem, I find it actually somewhat unfair to pick on him particular.
The real issue is the entire community which, as you note, doesn’t move a finger to improve the situation. That’s because (to use a German idiom, if you excuse) they’d saw off the branch on which they sit. Producing these useless predictions is the only thing they know what to do. They can’t discontinue it because what else would they make money with? This is of course also why they constantly complain that I’m supposedly just complaining and not making constructive suggestions. Totally wrong of course, if you read my book (or listen to my lectures) I have explained very clearly that they should focus on resolving mathematically well-defined inconsistencies rather than pulling guesses out of thin air. That, however, is difficult and would make their lives harder, so they won’t do it. It’s rather simple economics.
As I have mentioned before, this is the same problem they had in psychology with their sloppy ways of assessing statistical significance. (a) The problem was known for a long time. (b) It stalled scientific progress but they kept on doing it anyway ( c) They justified it to themselves and others by claiming that it’s a generally accepted standard.
The difference is that psychologists have taken steps to solve their problem. Particle physicists haven’t.
I don’t think it is possible to actually prove that someone is “pulling guesses out of thin air”. That is just as much speculation as what you are alleging. The only evidence we can go on is what they say and write. And the only standards we can apply are how well motivated are their arguments. Are their ideas based on well established observations? Or are they based on well accepted assumptions of science? After that, it’s only a matter of whether the math is correct. Perhaps these are the criteria that should be used by publishers. It seems obvious, for example, that string theory would not pass, since it is not based on observational data and the assumption that particles are strings or the like is not well motivated, IMO.
I predict that String Theory will continue to, er, thrive until some new discovery or idea appears in the current great experimental and theoretical desert. Theorists got to theorize somehow, even if it is only about how many string angels can dance on the head of a pin.
May be there are too many models in the market, with few new ideas and with some regions of parameters being excluded by experiments, but the problem started with the master, who after writing the SU(2)xU(1) model proposed next the SU(3)xU(1) model …
On the other hand, Sabine and Peter, is it fair to judge the whole particle physics community by the “sins” of one of its subgroups? Take for instance the work done on QCD corrections to Higgs production, with amplitudes, twistors and more into the game, it seems to me a solid pice of research, don’t you think so?
I’ve never criticized “particle physics” in general, and Sabine’s critique is clearly aimed at a certain mode of research, quite different than the sort of thing you mention.
The Kane case is an extreme one. Other people working on this kind of speculative model are careful to not make the kind of repeatedly falsified claims he does. The institutional problem his behavior raises is not so much about the field of research but about those institutions publishing his claims (Physics Today, his book publishers), They should be well aware that these claims have been repeatedly falsified in the past and that he does not acknowledge or address this.
I know this blog is a big fan of Lubos Motl. I feel like handing out some damning evidence of the futility of string theory from his own thesis advisor, Tom Banks.
Here is his home page at Rutgers. As you can see in the second-to-last paragraph, he says:
” I’ve consequently decided to abandon this area of research and shift my focus to condensed matter physics. The study of quantum gravity, inflation, black hole information, and “string theory”, suffers from the absence of well defined mathematical models, which define the theory in asymptotically flat or de Sitter space in a non-perturbative manner. It also suffers from a lack of contact with experiment, which is not expected to be ameliorated in the near future. Until further data becomes available on supersymmetric particles, dark matter, or finer details of the primordial fluctuation spectrum, these fields will be in the realm of uncontrolled conjecture. 15 years is enough time to spend on them.”
What’s astounding about this (I’m not exactly sure when he switched focus) is that in the late 90s he was co-author of the BFSS matrix model which was supposed to be the much-sought definition of M-Theory. Lubos created his own matrix model for non-perturbative string theory as an undergraduate, Banks was thoroughly impressed, Lubos became his doctoral student, and the rest is history.
So some time ago, Banks realized that all that impressive work on matrix models and other such speculation was just that, speculation, and he moved on to more concrete areas of physics that are connected to experiment. (I’m not sure if anyone else in the past has brought up Banks’ change of heart on this blog, so apologies if this is redundant.)