A sensitive outsider, watching and reading the press and the physics blogs carefully, might detect an unusual level of irrationality surrounding the topic of supersymmetry, both in terms of what the press is reporting and in terms of what scientists are saying. Not that this is intrinsically abnormal. Although science is in the long run a largely rational process, scientists themselves are no more rational than anyone else. Just because you are well-trained to employ rational thinking doesn’t mean you are trained (or could be trained!) to use it exclusively. Indeed the best scientists, in my experience, tend to have a very individual (and often strikingly unusual) brew of abilities and quirks, with the abilities to think rationally and do mathematics joined to far less stereotypical talents and characteristics. Some scientists are always passionately certain that their ideas are correct. Occasionally they are right, though it is rare. Others are by nature highly skeptical, dismissive of fancy ideas, or fatalistic. All this is to say that there is plenty of rationally-unjustifiable exuberance and scorn in science. But the amount of it surrounding supersymmetry is a bit higher than usual. What’s behind it?
The reason for the current irrational discussion over supersymmetry is that a substantial minority of particle physicists — mostly theorists but including a significant number of experimentalists too — were long ago fully persuaded by circumstantial evidence that supersymmetry would show up at the Large Hadron Collider (LHC). These scientists view each of the circumstantial threads as strong, and the case made from weaving them together as nearly airtight.
Meanwhile, there have been others in the field who have been staunch opponents of supersymmetry, for various reasons. Their view is, typically, that the case for supersymmetry is woven out of little threads that are individually weak, and that the case is threadbare. And they’ve been waiting for years for their opportunity to pounce on the supersymmetry advocates, who’ve been dominant politically in some circles.
[In case you’re wondering, I, along with many of my colleagues, neither strongly advocate for nor denigrate supersymmetry. As someone who had nothing to do with the invention of the theory, but who has often worked on it over the years, I view the case in its favor as moderately plausible but hardly convincing. I don’t myself think this is a place where theoretical arguments will get you anywhere; to resolve the issue requires data…which is why that’s where I keep my focus.]
While there are many ideas for what might show up at the LHC, supersymmetry is the only one that has a significant number of supporters who basically viewed the case in favor as already closed, and its appearance at the LHC as virtually certain. That’s why the failure to find supersymmetry as of yet has received so much attention, whereas the failure to find signs of other speculative ideas (which I will cover on this site in coming weeks) has not.
In some previous articles I have argued that the standard search strategy for supersymmetry (to look for high-energy quarks, antiquarks or gluons [which make “jets” of hadrons] along with signs of new undetectable particles [which are inferred when observed particles appear to recoil against nothing, a phenomenon (mis-)named “missing energy”]) is based on three assumptions:
- in any process, the number of superpartners can only change by an even number;
- the lightest superpartner [which is stable, by assumption 1] is a superpartner of a particle we know (and therefore, to avoid conflict with other data, an undetectable neutralino or sneutrino);
- the superpartners that are affected by the strong nuclear force are significantly heavier than the other superpartners of known particles.
These assumptions are not random. They are linked to the circumstantial evidence that is cited in the case for supersymmetry. If you’re a strong supersymmetry advocate, you will probably argue that the elegant simplicity and predictability of the minimal version of supersymmetry, the fact that it provides a possible dark matter candidate particle [a stable neutralino], and its consistency with the idea of grand unification (the speculation that the strong nuclear, weak nuclear and electromagnetic forces are all manifestations of a single force, something that would be obvious at a collider with collision energies 10,000,000,000,000 times larger than those of the LHC) form a significant part of your case. So you won’t easily give up any of these three assumptions.
- For instance, if you relax assumption 1, you lose the possibility that supersymmetry provides the particle that makes up dark matter, because there are no longer any stable supersymmetric particles.
- If you relax assumption 3, you are potentially giving up a prediction of the simplest form of grand unification: superpartners that feel the strong nuclear force are heavier than those that only feel the weak nuclear and electromagnetic forces by a ratio that is related to the ratio of the strengths of the various forces.
- If you relax assumption 2, you’re not necessarily losing anything, but dark matter might not work out so well anymore, and you’re certainly making life a lot more complicated. In particular it often becomes much harder to make definite predictions.
So over the 20 years preceding the LHC, those who were strongest defenders of supersymmetry made these assumptions and pushed hard for the particular search strategy that was pretty sure to find supersymmetry consistent with these assumptions. And that’s the search that — although it is not yet complete — has so far turned up empty.
A number of those who detest supersymmetry, and some of those who have not spent a lot of time understanding supersymmetry’s many variants, have responded with the scornful claim scornfully responded with the claim that supersymmetry is now largely ruled out by the LHC, generating a lot of blog commentary and press articles. This is understandable as a reaction against those who said it would certainly be found easily at the LHC. But though understandable, I feel it is a baseless claim. It’s one thing to say that the particular version of supersymmetry most loved by the faithful is in trouble; it’s quite another to say that the general idea of supersymmetry at the LHC is out the door. There’s a lot more work to do, and it’s far too early to declare the case against supersymmetry closed, just as it was far too early before the LHC started to declare the case in favor of supersymmetry closed. The data still needs to speak, and we still need to listen.
Fortunately, many at the ATLAS and CMS experiments pay little or no attention to this controversy, and have been listening closely to the data. A number of very interesting searches have been done in the past few months looking for supersymmetry without one or more of the above assumptions. And so the task of excluding many more variants of supersymmetry is well underway.
I’ll be detailing some of these searches this week. The first one should be available later today. Keep an eye out for updates.
20 Responses
Truly when someone doesn’t know then its up to other people that they will assist, so here it happens.
Great article! We will be linking to this particularly great post
on our website. Keep up the great writing.
Matt and Mark,
To give me a clue on my second question, are you basically saying that if you impose SU(5)-invariant boundary conditions on the soft SUSY-breaking masses at the GUT scale, e.g. m_q = m_u = m_e and m_d = m_l, where the subscripts denote superpartners, and evolve these down to the electroweak scale by the renormalization group, then the gluino and squark masses come out larger than the slepton, chargino, and neutralino masses, by factors related to the ratios of the gauge coupling constants at the electroweak scale?
And Mark is saying that that falls apart if you abandon the SU(5) mass relations at the GUT scale?
“a significant number of supporters… who viewed [susy’s] appearance at the LHC as virtually certain”
Really? I am surprised because I can only think of two people who I would put in that category (they are both mentioned in the comments). Talking to susy phenomenologists who have spent their careers working mostly on mSUGRA I get 50-50 quoted as the chance of susy showing up at the LHC.
Dear Matt,
Regarding point 3) what exactly do you mean by “the simplest form of grand unification”? Could you give an example?
🙂 I knew I was opening a can of worms here. Grand unification, like supersymmetry or extra dimensions or any other great idea, has many, many variants. If you don’t think hard about those variants, and keep your focus on the simplest versions of the idea, you get very simple predictions. If you do think a bit harder, you realize that those lovely predictions are often lost in models that are just one step more complicated… which is disappointing, but is true of every similar case I can think of.
For those who say that elegance and simplicity are the criteria all theorists should use in choosing which models to consider, I can only point out that the Standard Model itself fails that criterion on several counts.
To go beyond this requires a full introduction to Grand Unification and its variants. Someday this site will host a page or two on this subject. But I am afraid it’s not my highest priority right now, since we won’t be testing it experimentally at the LHC… not directly, anyway. I will try to get back to it.
Thanks, Matt! Well, I guess I just wanted to point out that even for the simplest vanilla susy GUT models, assuming just the MSSM with light gauginos and higgsinos at least, the question of GUT-scale boundary conditions for the susy breaking parameters (and correspondingly the EW scale spectrum) and the question of gauge coupling unification are not related. For example, in string-inspired gravity mediated susy breaking models, one generically obtains the universal tree-level contrubution being comparable to the anomaly mediated contributions. As a result, one has a rather peculiar set of GUT scale of boundary conditions for the gaugino mass parameters which may lead to a lighter gluino, perfectly consistent with the gauge coupling unification.
This, for the layperson who can’t read physics shorthand, is Mark’s way of saying — that he essentially agrees. He’s giving a particular example of how you can easily get around the predictions of the simplest versions of grand unified theories (“GUTs”).
Hi, I’ve looked carefully through the above articles, and can’t see that they contain any reason why a neutralino other than the LSP should be undetectable. If a non-lightest neutralino was heavier than Z, could it not decay to a lighter neutralino by emission of a Z, which would be an observable signature? And if a non-lightest neutralino was heavier than the sum of the masses of W and a light chargino, could it not decay to a W plus that chargino, and the W would be detectable, and the chargino would either decay in the detector to say the LSP plus a charged SM particle that would be observed, or the chargino itself would be seen in the electromagnetic calorimeter, and maybe also in the muon system if it was penetrating enough?
Looking back at what you wrote above, I read “an undetectable neutralino”, in assumption 2, as indicating that neutralinos are always undetectable, which is perhaps not what you meant. Is that correct?
That’s right — the meaning of what is stated in assumption 2 is that if assumption 1 is correct, the lightest superpartner is stable — but from experiment, any unknown stable particles must not affected by either strong nuclear or electromagnetic forces — and the only superpartners of known particles that satisfy that criterion are neutralinos (partners of photons, Z particles and Higgs particles) and sneutrinos (partners of neutrinos.) A neutralino that is not stable may decay to visible particles, but then it isn’t relevant for assumption 2.
I’d be grateful if you could briefly explain a couple of points from above, (or suggest a ref):
(1) Why are neutralinos undetectable?
(2) Why does the simplest form of grand unification predict that gluinos are heavier than neutralinos by a ratio related to the strengths of the forces?
For question 1; the answer is contained in my supersymmetry pages, start here: http://profmattstrassler.com/articles-and-posts/some-speculative-theoretical-ideas-for-the-lhc/supersymmetry/ You will find other insights here: http://profmattstrassler.com/articles-and-posts/largehadroncolliderfaq/how-to-figure-out-what-happened-in-a-microcollision/
For question 2; this is complicated and not obvious. I can’t give you the answer without teaching you something about quantum phenomenon that won’t fit in a comment. Consider this a missing point that I’ll explain at some point, but I am afraid you you won’t see it on this site for quite some time.
Since I may be someone Matt has in mind here, let me just make clear my position on some of this. I should also make clear that I’m largely in agreement with his description of the situation.
I started grad school in 1979, am now middle-aged (just turned 54). For pretty much my entire career, SUSY extensions of the Standard Model have in some sense been the dominant paradigm for what is supposed to come after the SM. While I’ve personally a lot of interest in SUSY as an abstract mathematical idea whose significance we may someday understand, the kind of SUSY QFTs that are studied as phenomenologically viable have always seemed to me very unconvincing. I wrote quite a bit about this in my book: it’s a complicated story, but the bottom line to me is that SUSY extensions don’t explain much and have no significant evidence pointing to them, while adding a problematic level of complexity. Long before the LHC turned on, there was a long history of failed SUSY “predictions”. If this were a good idea, there’s lots of places evidence should have shown up earlier.
Instead of acknowledging this, a large group of theorists heavily promoted the idea that SUSY was the answer to the problems of the SM, and that it would show up at the LHC. This often came as a package with superstring theory: LHC-scale SUSY was held up as the answer to the (accurate) argument that string theory predicted nothing. Physics Today published two articles by Kane promoting this over the years, Arkani-Hamed lectured about how we’d know soon after LHC turn-on whether SUSY was there, Ellis produced detailed plots showing specific predictions about where SUSY would appear if it were to explain what it was supposed to. Examples go on and on…
Once initial results started shooting these claims down, I’ve been pointing this out, and, perhaps at times demonstrating a “scornful” attitude about the combination of forcefully made “predictions” (based on over-hyped and thread-bare arguments) and current unwillingness to admit failure. As far as I’m concerned, the sooner an end is brought to the three decades of dominance of the subject by an unpromising idea, the better off we’ll be.
So, here’s where I do disagree with Matt. Yes, there are plenty of still viable SUSY models, but there is very little in the way of reasonable argument for why you should take them seriously. SUSY searches have received a huge effort from the experiments, and surely that will continue. The danger isn’t that SUSY searches will be prematurely dropped. It’s that attention and effort will be continue to be wasted on something with no promise. Looking at some talks about SUSY searches that involve discussions of how to optimize the triggers for various of the remaining SUSY models makes me nervous. Better to have more people giving up on SUSY searches and turning their talents to finding some more promising analyses to try.
Peter writes: “It’s that attention and effort will be continue to be wasted on something with no promise. Looking at some talks about SUSY searches that involve discussions of how to optimize the triggers for various of the remaining SUSY models makes me nervous. ” This isn’t a zero-sum game. You can’t divide searches at the LHC up by models; you have to divide them up by strategies — by the phenomenon they are looking for. (See http://profmattstrassler.com/2011/08/29/some-comments-on-theory-and-experiment-at-the-lhc/ ) In many cases, optimizing the triggers and analysis techniques to look for other supersymmetric models also increases the sensitivity for various non-supersymmetric models — just as the original supersymmetry search strategy was sensitive to many classes of non-supersymmetric models. (Indeed my recent paper http://arxiv.org/abs/1107.5055 with Lisanti, Schuster and Toro is about a strategy, not about a specific class of models.) So I think you are not organizing the problem correctly — it is not either/or.
This complex issue deserves a longer discussion and will eventually require its own page.
Please read carefully, and I must politely request that you not put words into my mouth.
Clearly thinking supersymmetry *might* be wrong (as I do) is not scornful. Nor did I even say that only people who detest supersymmetry might say it (indeed, I would say it too.) But your wording is different from mine. These differences in wording are very important to the meaning of the statement.
I said: “A number of those who detest supersymmetry, and some of those who have not spent a lot of time understanding supersymmetry’s many variants, have responded with the scornful claim that supersymmetry is now largely ruled out by the LHC,”; notice the use of the word “number”, “some”, etc., and the phrase “largely ruled out by the LHC.” Those are the only statements I made. Again, they do not correspond to your paraphrase of my meaning.
I would agree with you that someone who studies the data (as I have) and finds no direct evidence for supersymmetry at any collider (and I see none) would conclude that supersymmetry might well be wrong. But it also still might be right. The evidence against it is still not strong enough for a conclusion. Clearly that is neither a scornful nor an unreasonable point of view. To draw an inference from existing data that supersymmetry is largely ruled out would be, yes, premature. It might, nevertheless, turn out to be correct. We’ll soon find out. Let us have some patience.
You said “…the scornful claim that supersymmetry is now largely ruled out by the LHC”. So you have explicitly stated that the claim “supersymmetry is now largely ruled out by the LHC” is a “scornful” claim. These are your words and they are quite ridiculous. No-one is putting words in your mouth except for you; please take some responsibility for them.
I do see the wording issue. I should have said “make the claim scornfully”. Some have made the claim scornfully. Others have made the claim less scornfully. I guess you would argue you are one of them.
But scornful or not: though we are all entitled to our opinions, we should not confuse opinion and knowledge.
By studying the data, and finding no evidence for supersymmetry (at LHC, Tevatron, or any collider, or in any other experiment), some have concluded that supersymmetry at low energies might be wrong. But you say that such a thought is “scornful” and must be said by someone who “detests” the theory. Wait, what?? How about saying that such people are “drawing inferences, perhaps premature, from the existing data”, rather than saying they are “scornful”?