[Note Added: this afternoon the author of the Scientific American article made a few corrections. I leave it to you to judge for yourself whether he addressed the issues raised here.]
There’s been a little silliness floating around (sadly, in Scientific American, whose article contains at least two factual errors) unscientifically speculating that ATLAS’s new results on the Higgs-like particle, from data collected at the Large Hadron Collider [LHC], suggest there are two such particles rather than one. The mass measurement of this particle using the data when it decays to two photons, 126.6±0.3±0.7 GeV/c², is different, by 2.7 standard deviations, from the mass measurement obtained from its decays to two lepton/anti-lepton pairs, 123.5±0.9+0.4-0.2 GeV/c². So… huh… gee… maybe there are two Higgs-like particles, a lighter one which rarely decays to two photons and a heavier one which rarely decays to two lepton/anti-lepton pairs?
[Note Added: I should emphasize, lest anyone blame ATLAS for this implausible line of speculation, that in the ATLAS presentation last week, which was one of several presentations that morning, these two mass measurements were presented simply and responsibly, as results from data. Not a single speculative word was said about there being a hint of two Higgs particles. I don’t know who got the ball rolling on that idea, but it wasn’t ATLAS. And it’s not a plausible idea: see below.]
Take a deep breath. For not only would the two types of particles somehow have to be magically and implausibly arranged to mimic, at first glance and to a rough extent, a single Standard Model Higgs particle (the simplest possible type of Higgs particle), there’s another experiment, which unfortunately the writer of the Scientific American article neglected to consult.
ATLAS’s mass measurement from the events with two lepton/anti-lepton pairs also disagrees with CMS’s mass measurement obtained from the same type of events: 126.2±0.6±0.2 GeV/c². Two similar experimental detectors, same measurement, moderate disagreement. Nature is nature; there’s no way that ATLAS can be making one type of particle all the time, while CMS is making a different one all the time. So there is no evidence here, taking ATLAS and CMS together, favoring the existence of a separate particle with a mass of about 123.5 GeV/c² that decays to two lepton/anti-lepton pairs.
What is behind these discrepancies, then? ATLAS and CMS each have scarcely a dozen of these two lepton/anti-lepton events, and their extraction of the Higgs particle’s mass from each event is somewhat uncertain, which is why many events are required for a good mass measurement. When you still have small amounts of data, funny statistical fluctuations will often occur. We’ve seen this before; back in 1989, when the Stanford Linear Collider (SLC) produced its first few Z particles at the Stanford Linear Accelerator Center, the plot of the Z particle’s mass gave a double resonance peak, instead of the single peak that was expected. A brief moment of speculation occurred, but with more data the anticipated single peak structure emerged. I’ve heard at least one other similar story from an earlier decade. In fact ATLAS and CMS had a 2 GeV mass discrepancy when the first Higgs hints came in; that was just an effect of statistics. Combine a fluctuation of this form with a minor detector calibration problem, and you’ll get discrepancies like this.
Multiple types of Higgs particles are certainly possible; people have considered this scenario for decades, and I’ve written about it here, for instance. Efforts to search for a second type of Higgs particle have been going on since the discovery of the first one. But let’s not manufacture one out of thin air by looking selectively at the data; that’s not how reliable science gets done.
It took me over six months, following my article on molecules, to write the sequel, on atoms. These are just two in a series, intended to introduce the structure of matter to novice readers who want to learn what particle physics is about. Atoms aren’t the main focus; future articles will focus on electrons, on protons and neutrons, on quarks, and on the forces that hold these objects together. But the essay on atoms might be the hardest of the set to write (at least I hope so). The long delay reflects the challenges involved, and as my readers’ wise and helpful criticisms of Friday’s first version confirmed, I didn’t meet them on my first try.
So after some thought, I’ve made another attempt. Critique still welcome from anyone who wants to make suggestions.
Aside from the fact that I fell into a couple of pedagogical traps that anyone who’d taught chemistry would have known about, I also struggled to describe atoms briefly, clearly and accurately because their features are determined by quantum mechanics — that weird but fundamental behavior of our world that we don’t encounter in daily life but is essential to the structure of matter. What’s profoundly confusing to the non-expert (and somewhat confusing even for experts) is that electrons are, on the one hand, best described in many circumstances as point-like particles (much smaller than atoms, and smaller even than atomic nuclei) yet around atoms they are in some way spread out in a very non-particle-like fashion. Well, indeed, thinking of elementary objects like electrons as “particles” will get you into trouble; for one thing, they are really “quanta” of quantum fields, and in most circumstances they behave much more like waves. And yet it is essential to explain that one can try to measure their size — essentially by forcing them, through an appropriate experiment, to reveal whether they, like baseballs, rocks and dumplings, have internal structure.
Ok, I can’t even figure out how to write this paragraph clearly. There needs to be a way to explain this issue, one that is both moderately intuitive and based on accurate and clear physical reasoning…
A few commenters have complained that I’m too hard on science journalists, who have a tough job; it’s hard to explain difficult concepts in a few words. To paraphrase them: “if it’s so easy, you do it! Rather than merely complain about the erroneous TIME magazine paragraph on the Higgs boson, write your own explanation of the Higgs particle for readers of TIME magazine.”
Well, first of all, I have never once suggested science journalism is easy; far from it! A big part of the challenge is to find ways to explain complex ideas that are simple, compelling and accurate (and not two out of three.)
Second, I have written an article suitable for non-expert readers; it’s just over a page long, and is called Why the Higgs Particle Matters. It’s gotten about 30,000 hits; some people seem to really like it, so try it out on your friends.
And third, for those who point out that the above-mentioned article is much longer than a paragraph, and that I shouldn’t be so critical of the TIME journalist who had to fit so much into a such a small space, here is my version of the TIME paragraph: six sentences rather than five, but scarcely longer. I have borrowed the style and the feel of the TIME journalist’s writing, and I have removed some inaccurate content and replaced it with different accurate content.
Take a moment to thank the Higgs field for all the work it does, because without it, you’d explode. This field pervades the universe and supplies electrons (and many other particles) with their mass, thus preventing ordinary matter from disintegrating into a ghastly vapor. It was in the 1960s that British physicist Peter Higgs (and a few others) first posited the existence of this field. But it was not until last summer that two huge teams of researchers at Europe’s Large Hadron Collider at last sealed the deal by discovering a new particle — the Higgs boson — which confirms the Higgs field exists. You see, the particle is a consequence of the field wiggling a bit; and just as sound, a ripple in the air, can’t be heard unless there’s air in the room, there wouldn’t be Higgs particles to discover unless Higgs and friends were right all along about their famous field. Now the Higgs — as most particles do — decays in an instant to other particles, so it wouldn’t be able to attend the award ceremony; however, the scientists would surely be happy to appear in its stead.
Although not everything I’ve written here is 100% accurate — that would indeed be impossible in a paragraph for a wide readership — I believe none of it is fundamentally wrong (but my colleagues should feel free to complain!) Yes, science journalism is difficult; but is it really inevitable that profound errors concerning the science must appear in articles for the public?
Yes, it was funny, as I hope you enjoyed in my post from Saturday; but really, when we step back and look at it, something is dreadfully wrong and quite sad. Somehow TIME magazine, fairly reputable on the whole, in the process of reporting the nomination of a particle (the Higgs Boson; here’s my FAQ about it and here’s my layperson’s explanation of why it is important) as a Person (?) of the Year, explained the nature of this particle with a disastrous paragraph of five astoundingly erroneous sentences. Treating this as a “teaching moment” (yes, always the professor — can’t help myself) I want to go through those sentences carefully and fix them, not to string up or further embarrass the journalist but to be useful to my readers. So that’s coming in a moment.
But first, a lament.
Who’s at fault here, and how did this happen? There’s plenty of blame to go around; some lies with the journalist, who would have been wise to run his prose past a science journalist buddy; some lies with the editors, who didn’t do basic fact checking, even of the non-science issues; some lies with a public that (broadly) doesn’t generally care enough about science for editors to make it a priority to have accurate reporting on the subject. But there’s a history here. How did it happen that we ended up a technological society, relying heavily on the discoveries of modern physics and other sciences over the last century, and yet we have a public that is at once confused by, suspicious of, bored by, and unfamiliar with science? I think a lot of the blame also lies with scientists, who collectively over generations have failed to communicate both what we do and why it’s important — and why it’s important for journalists not to misrepresent it. (more…)
The meaning of the title of Clara Moskowitz’s new article for the public, “Dark Matter Mystery May Soon Be Solved“, all lies in the word “may”. It may. It may not.
According to the article, “the answer to this cosmic mystery could come within the next three or four years, scientists say.”
I have to admit that this kind of phraseology, which one often sees in the press in reports about science, drives me a bit nuts. Which scientists? How many of them? You can’t tell from this line whether this is something that a group of three or four mavericks are claiming, or whether it is conventional wisdom shared by most of the community. And “the answer… could come…”? Interpreted literally it is content-free: yes, the answer could come in the next few years, or not — but you don’t need any scientists to tell you that. If one interprets it more optimistically — that it is intended to imply that the answer will very likelycome within the next three or four years — then I think it is far from clear what fraction of the experts will agree with that statement.
Rather than debate the claim, let’s start with the physics. What will determine how long it takes to discover what dark matter is made from? (more…)
Last week, I promised you I’d fill in the details of my statement that the recent measurement (of the rare process in which a Bs meson decays to a muon and an anti-muon — read here for the physics behind this process) by the LHCb experiment at the Large Hadron Collider [LHC] had virtually no effect on the constraints on any speculative theories, including supersymmetry, contrary to the statements in the press and by a certain LHCb member. Today I’m providing you with some sources for this statement.
A number of my colleagues have tasked themselves with keeping track of how measurements at the Large Hadron Collider and elsewhere are affecting certain subclasses of variants of the supersymmetry. They call themselves the “Mastercode Project”; here’s their website. They’re not the only ones looking at this, but among them is Professor Gino Isidori, whom I was talking to last week, so I’ve gotten this information from him. I quote from the MasterCode website regarding last week’s result from LHCb: “The new measurement provides a valuable new constraint on the supersymmetric parameter space, but the observation of a Standard Model-like branching fraction for the Bs→μ+μ– decay is quite consistent with supersymmetry. In fact, a Standard Model-like branching fraction of this decay was expected in constrained supersymmetric models like the CMSSM or NUHM1 (see, e.g., the recent MasterCode results for further details). As a result, the favoured regions in the parameter space of these models do not change significantly after the inclusion of the new constraint. ”
Now before I explain what this means, it’s important to have some terminology, running from most general to most specific.
Supersymmetry: the general theory that space and time are more subtle than they appear to be, and as a result, for every known particle in nature there is a corresponding superpartner particle with the property that either the particle is a boson and its superpartner a fermion, or vice versa. (Read more about supersymmetry here, and more about fermions and bosons here and here.)
CMSSM (Constrained MSSM): a much smaller subclass of variants of the MSSM in which the masses of the superpartners are assumed to be related to each other in particular ways (the details are technical and not essential, so I’ll skip them.)
NUHM1 (Non-Uniform Higgs Mass variant of the MSSM): another small subclass of the MSSM variants, slightly more general than the CMSSM.
Keep in mind that
ruling out the CMSSM or NUHM1 does not mean that the MSSM is ruled out;
ruling out the MSSM does not mean that supersymmetry at the TeV scale is ruled out;
ruling out supersymmetry at the TeV scale does not mean that supersymmetry is ruled out.
Among the many goals of the LHC is to find or rule out supersymmetry at the TeV scale. (It cannot hope to rule out supersymmetry altogether; that would presumably require a vastly more powerful collider that won’t likely be built for centuries, if ever.) It’s not enough to rule out the CMSSM, or the NUMH1, or even the MSSM. Similar statements apply for other speculative ideas that propose as yet unknown particles and forces; it’s not enough for the LHC to rule out just the simplest variants of these ideas.
Now if it turns out that supersymmetry is part of nature, rather few of my colleagues expect the variant we find to be contained within the CMSSM or NUHM1; and personally (though I’m probably in the minority) I have long doubted that it would be contained within the MSSM. Nevertheless, it is instructive to look at how LHC data is impacting the CMSSM and the NUHM1 subclasses of supersymmetry variants. One just must be careful not to over-interpret; the exclusion of most variants in the CMSSM is not an indication that most variants of TeV-scale supersymmetry as a whole are excluded.
Now in this context, let’s see how the new measurement that was announced last week affects the CMSSM and the NUHM1. In Figure 1 is a plot showing the allowed variants of the CMSSM and the NUMH1, as a function of two quantities: on the horizontal axis, MA, which if large is (approximately but essentially) the mass of four of the five Higgs particles in the MSSM, and on the vertical axis, tan β, the ratio of the values of the two non-zero Higgs fields that are required in the MSSM. In solid red and solid blue are the one-standard-deviation and two-standard-deviation allowed regions after the new LHCb measurement is accounted for; any variant of the theory not sitting inside the blue region is excluded by the data. The dashed bands show the same thing before the new LHCb measurement. Since the dashed and solid blue bands are right on top of each other, you see there’s almost no effect at all. That’s what was behind my claim last Friday.
Fig. 1: Constraints on the CMSSM (left) and NUHM1 (right) subclasses of supersymmetric theories, before and after the HCP conference of last week. The quantities on the horizontal and vertical axes are explained in the text. In both plots: solid red (blue) give the constraints at one (two) standard deviations; variants outside the blue curve are excluded. Dashed red (blue) are the same limits before the new LHCb measurement. Notice there is almost no change.
But please, don’t misinterpret what I’m saying (or my colleagues) as suggesting that the LHC’s data has had no impact on the list of possible variants of supersymmetry! Far from it! Many variants are excluded, and many popular (but not necessarily more likely) subclasses of variants of supersymmetry have been pushed into regions that many would consider corners. The only statement in Figure 1 is that the new LHCb measurement didn’t make these corners smaller. But to see how things have changed since before the LHC began, look at Figure 2, which shows how the LHC as a whole — all the measurements from LHCb, ATLAS and CMS taken together — have affected the CMSSM and NUMH1 since 2009. (The CMSSM and NUHM1 also make assumptions about where dark matter comes from, so even effects of the dark matter measurements from the XENON100 experiment are included here.)
Fig. 2: as in Figure 1, except that the dashed lines give the constraints on the CMSSM and NUHM1 before the LHC began taking data, and the solid line gives the constraints after the data taken through early summer 2011 was analyzed. Notice the scale on the horizontal axis is different from that of Figure 1.
Figure 2 is a similar plot to Figure 1 — but this time, solid blue and red indicate the impact of LHC data as of summer 2011, and the dashed blue and red indicate the situation before the LHC started. Now compare the dashed blue line in Figure 2 (before the LHC) with the solid blue line in Figure 1 (now); note the scale on the horizontal axis is different!. You’ll see that in the CMSSM it was possible before the LHC to have MA as low as 350 GeV/c², but now it has to be over 900 GeV/c², which many would consider a rather high value. In the NUHM1 there’s been a similar shift from 150 to about 300 GeV/c², not yet so high but still a significant increase. And meanwhile, while almost any value of tan β from 2 up to 60 was allowed before the LHC, this number is now limited to a smaller range. For example, if MA were below 900 GeV/c², then the CMSSM would be excluded and the NUHM1 would be allowed only for tan β < 30 or so. This upper limit on tan β is mainly caused by the similar LHCb measurement presented back in March (and mentioned by me on Friday), and by similar ones from the CMS, CDF and ATLAS experiments.
But clearly there are plenty of variants within the NUHM1 that remain viable. And the NUHM1 is not representative of the full range of possibilities within the MSSM, so even if the NUHM1 were excluded, we’d still have a long way to go to exclude the MSSM, much less all of TeV-scale supersymmetry. In short, it’s neither all nor nothing.Yes, a lot of progress has been made; LHC data (and data from other sources) have ruled out a lot of variants of TeV-scale supersymmetry. But no, we’re not yet close to ruling out the full range of variants.
Please note that I’m not telling you this because I’m some devotee of supersymmetry who believes deeply in his heart that we’ll someday find it, and is trying to persuade you not to give up. I’m just laying out for you the facts on the ground. Do you imagine that I’m happy that a long, painful slog lies ahead, during which particle physicists — theorists and experimentalists — will painstakingly cover all the possible variants of supersymmetry, and slowly but surely determine whether or not supersymmetry is absent at the TeV scale? Don’t you think my life and that of my colleagues would be a lot easier if we could snap our fingers and with one or two quick measurements settle the question of whether supersymmetry is a fact of nature or not? Unfortunately, things don’t work that way. You should simply ignore the irresponsible grand statements you will see in the press and on various blogs; indeed, sweeping remarks are a sign of careless thinking, and you should beware. The truth is that only through very hard work — by the experts who make the measurements, by those who advise them on which measurements to make, and by those who do the calculations that are the ingredients for studies like the MasterCode Project — can we hope to settle profound questions about nature.
According to the BBC, it was a heavy blow. According to a member of the LHCb experiment quoted in the article, it put the theory “in the hospital.” The reality? Nobody even suffered a scratch.
When I chatted with him later, Isidori reminded me exactly what LHCb reported in March, and how it compares to what they report now.
March: LHCb reported that at most 4.5 per billion B_s mesons decay this way (at 95% confidence)
November: LHCb reports about 3.5 per billion B_s mesons decay this way, and at 95% confidence the rate is at least 1.1 per billion and at most 6.4 per billion.
Notice that the new measurement raises the upper limit on how often this process occurs. This upward shift is not an indication of a problem; it’s probably just an ordinary statistical effect that arises from having small amounts of data. But it makes the constraints from this measurement — on the many variants of supersymmetry, and on other theories — a little weaker, if anything. Most of the supersymmetric (and other) models that would be constrained by this measurement lead to a higher rate than in the Standard Model (where it is predicted to be about 3.2 ± 0.3 per billion). So a higher upper limit means fewer of these variants are excluded by the data. All in all, the constraints on supersymmetric (and other) models are little changed, and perhaps somewhat weaker than they were in March. Isidori and his colleagues have worked this out, and I’ll try to get details from them next week.
No need to take my word for it, or even Isidori’s. Professor Michelangelo Mangano of CERN, apparently having spoken independently to other experts, made exactly the same point on page 41 of his summary talk concluding the conference. (By the way, it was a great talk, and I recommend that experts read it.)
Well! So much for the big BBC headline! [Most likely the public will never learn about this; a news report describing more accurately what this really means for supersymmetry etc. probably will not appear at all on the BBC, and even if it does, it certainly won’t get a big attention-grabbing headline. It’s sad that this inherent bias in media reporting ensures the public gets an unhealthy dose of incorrect scientific information and rarely gets the antidote.]
None of this at all diminishes LHCb’s accomplishment! They’ve made a great measurement, for which they deserve big congratulations. A round of applause, please!! But let’s not overstate its immediate impact. As the measurement becomes more precise, its impact will gradually become greater, and even more so when it is combined with similar measurements from ATLAS and CMS.
But meanwhile, there were powerful and truly new constraints on supersymmetry (and other theories) reported at this conference, and they came from ATLAS and CMS, in their searches for effects from superpartner (and other) particles. I told you about a small number of these searches a couple of days ago (by the way, I learned meanwhile that CMS has a search that is similar to the one I mentioned from ATLAS involving bottom quarks) and maybe I’ll point out a few others next week, if I have the energy.
I’m always amused at how very reasonable remarks so often generate attacks from unreasonable people. I wrote a perfectly ordinary post about what one does and doesn’t learn from LHCb’s important new measurement at the Large Hadron Collider [LHC] (and in fact I overstated the significance of the result — more on that later), and somehow I touched off a mini-firestorm. Well, that just indicates how essential it is to have calm people expressing sensible points of view. When people become so politicized that they can’t distinguish propaganda from science, that’s not good.
Forget supersymmetry — because none of my remarks have anything to do with this theory in particular, and the theory doesn’t deserve the excessive attention it’s getting. Take any theory: call it Theory X. Extra dimensions; compositeness of quarks and leptons; non-commutative spacetime; grand unification; your-theory-here. The idea behind theory X may be very clever, but as always, there are many variants of theory X, because an idea is almost never precise enough to permit a unique realization. Each variant makes definite predictions, but keep in mind that detailed experimental predictions may very well differ greatly from variant to variant.
Now, here is a logical fact: one of two options is true.
Option A: One variant of theory X is “correct” (its predictions agree with nature) while all other variants are “wrong” (disagree with nature)
Option B: All variants of theory X are wrong.
Nature is what it is; there are no other options (and this is not the place for a discussion about this basic scientific assumption, so pace, please, philosophers.). [More precisely about option A: the space of variants is continuous, so the correct statement is that an arbitrary small region in this space is correct; you can put in the correct calculus vocabulary as you like. I’ll stick with the imprecise language for brevity.]
For either option, as more and more data is collected, more and more variants of theory X will become “dead” — excluded because of a disagreement with data. Therefore — obviously! — a reduction in the number of live (i.e. unexcluded) models always takes place over time. And this has absolutely no bearing on whether, at the end, all variants of X will be dead, or one (or perhaps several very similar ones) are still alive.
And thus it makes absolutely no sense to describe, as a “blow to theory X” — in particular, to the idea behind theory X — a measurement that excludes (“kills”) even a big fraction, but not virtually all, of the variants of theory X. It’s certainly a blow to those variants; in fact, it is a fatal blow for them. But it does nothing to distinguish between Option A and Option B. It only tells us that if Option A is true, the variant of X that will be alive at the end is not among the ones that have just been killed.
This isn’t rocket science, folks. It’s logic. [Well – As a commenter points out, it’s not “logic” in the strictest sense; but it is basic scientific reasoning.] And if we take theory X to be the Standard Model itself, I’ve just described its history. (more…)
As I warned you to expect, there wasn’t anything particularly dramatic in the new results, reported at the HCP conference in Kyoto, on the Higgs-like particle. There just wasn’t enough new data relative to what was available in July for that to happen — a bit less than double for most measurements. That said, there were some interesting developments. Here’s are the summary plots from ATLAS (left) and CMS (right); things are a bit more consistent with the Standard Model Higgs hypothesis (especially for decays to tau’s, b’s and W’s) than they were before, but the precision of the measurements is still pretty low.
The measurements looking for signals of a Higgs particle at ATLAS (left) and CMS (right). For each process, the dot is the measurement of the rate relative to the Standard Model expectation, and the horizontal bar gives the one-standard-deviation uncertainty. Note all measurements are within two standard deviations of expectations. Some measurements (photons for CMS and ATLAS, ZZ* –> 4 leptons for ATLAS) were not updated for this conference. For details, see the previous post.
Also notable: that the Tevatron experiments’ measurement of Higgs decaying to bottom quarks is now smaller than before and more in line with Standard Model expectations; that the ratio of W to Z coupling strengths is consistent with a simple generic Higgs [i.e custodial symmetry works with uncertainties of 30%]; and at 2.5 standard deviations this is not a parity-odd spin-zero particle, while it remains consistent with being the parity-even spin-zero particle that we’d expect a simple Higgs to be.
First things first — based on the questions I’m getting from reporters and other non-experts, I need to emphasize right away that it is very unlikely that anything dramatic is going to happen here. ATLAS and CMS only just co-discovered this particle in July; the amount of data that is available now is only about twice as much as was available in July, and that is nowhere near enough for the big questions about the Higgs-like particle to be resolved. We will not know after today whether it is a Higgs at all, whether it is a Standard Model Higgs or not, or whether any particular speculative idea beyond the Standard Model is now excluded. What we are likely to see is some incremental changes in what we know.
Knowledge about nature does not come easy. We discovered the top quark in 1995, and we are still learning about its properties today — there were over a dozen talks about the top quark earlier in this conference. And we will still be learning important things about the Higgs during the coming few decades. We’ve no choice but to be patient.