Not Being Announced Tomorrow: Discovery of the Higgs Boson

By Sean Carroll | December 12, 2011 10:35 am

Tomorrow, Tuesday 13 December, there will be a couple of seminars at CERN presented by Fabiola Gianotti and Guido Tonelli, speaking respectively for the ATLAS and CMS collaborations at the LHC. They will be updating us on the current status of the search for the Higgs boson. The seminars will be webcast from CERN, and there should be a liveblog on Twitter that you can follow by searching for the #higgsliveblog hashtag (no Twitter account required). The seminars start at 14:00 Geneva time, so that’s 5:00 a.m. Pacific time if I do my calculations correctly. Of course there will be plenty of news coverage immediately thereafter, so don’t feel too bad if you sleep through it. Many places with LHC physicists (including Caltech) are also having their own local seminars. Should be exciting!

If you want to know why it’s exciting, after you’ve read John’s description of life in the trenches and Matt Strassler’s post about the multiple stages of hunting the Higgs and mine about why we need something like it, see even more recent posts by Matt, Jester, and Pauline Gagnon. Reader’s Digest version: not only are we being updated on the status of the search, there are believable rumors that the searches are actually seeing something — hints of a Higgs near 125 GeV, with better than 3-sigma significance from ATLAS and better than 2-sigma significance from CMS. But obviously rumors are no match for what actually happens.

All I’m here to tell you is: you should not expect to hear anyone announcing that we have discovered the Higgs boson. This will, at best, be a hint — “evidence for” something, not “discovery of” that thing. The collaborations realistically can’t claim to have actually discovered the Higgs, even if it’s there — they don’t have enough data. (CERN even issued a press release to drive home the point.) And in the real world, hints are sometimes misleading. That is: the experimenters will give us their absolute best judgment about what they are seeing, but at this stage of the game that judgment is necessarily extremely preliminary. If they say “we have 3.5-sigma evidence, which is quite suggestive,” do not think that they are just being coy and what they really mean is “oh, we know it’s there, we just have to follow the protocols.” The protocols are there for a reason! Mostly, that many 3-sigma findings eventually go away. This is one step on a journey, not the culmination of anything. (For Americans out there: it’s like a bill has been passed by the House, but not yet passed by the Senate, and certainly not signed by the President. Much can go wrong along the way.)

The journey of a thousand miles begins with a single step. It’s possible that tomorrow’s announcement means that we’re nearing the end of the journey, say at the mile-990 marker. But we can’t be sure, and there are no royal roads to particle physics. Patience! The excitement of not knowing for sure is what makes science one of the most compelling human stories.

CATEGORIZED UNDER: Higgs, Science, Top Posts
  • MPS17

    It’s true that many 3-sigma results go away, but I think one has to think carefully about this.

    Frequently, someone searches a large data set for some sort of anomaly, and finds one at the 3-sigma level, and this anomaly later disappears. This is because the person was searching “blindly” for anything anomalous, there were thousands or more of “potential” signals, and so one expected some 3-sigma results by chance. Like if I look for correlations between consumption of any substance and cancer, for thousands of substances, I’ll find some that correlate to 3-sigma.

    Yet, this is very different from when I have a theoretical reason for expecting a result, perform a study, and find it to 3-sigma. For example if I have a microbiological reason for thinking substance X causes cancer — or maybe I’ve seen it causes cancer in rats — then 3-sigma correlation with causing cancer in humans is good evidence.

    The question is, what category does the Higgs fall into? Granted, there is potential for Higgs to appear with a large range of masses, and that large range serves as large number of tests among which one expects some anomalous results. But now we’ve ruled out Higgs in huge swaths of the range. I personally think 3-sigma evidence for the Higgs in the remaining allowed range should probably be taking as roughly 3-sigma evidence. That is, given a 3-sigma result, I’m inclined to take 100 to 1 odds that it’s there.

  • Maki

    I’m getting Arsenic Life flashbacks all of a sudden. It’s exciting, but I’m not expecting to pop any champagne corks. They’ve certainly gotten the WSF office in a buzz of meetings though. Which I’m sure is the goal.

  • Chris

    Oh good, 7 AM Central right when I’ll be having breakfast.

    I’m sure many people working at CERN have little side bets as to the mass as well. And if they announce a 3 sigma result I’m sure there will be a host of theoretical papers discussing the ramifications. I think particle physics demands a higher level of proof than biology. As for cancer genes, it is not as simple as just saying “you have this gene, you will get cancer.” There are a host of complicating factions which determine if that gene is expressed or not.

    As for particle physics, sure they are governed by probabilities and uncertainty principle of quantum mechanics, but they are well understood. And for that reason we need to know with 5 sigma confidence that the event is real and not some fluctuation in the detector.

  • Charlie

    MPS17 Says is absolutely correct above. This problem plagues medical research, where a common practice is to search for any darn correlation (of which there are an unlimited number by chance) and report it at p < 0.05, which is almost always spurious (the medical community really needs to adopt the physics threshold in this kind of study). On the other hand, a well-designed study testing a particular hypothesis can and should be publishable at p < 0.05 in most cases. These results are always provisional, of course, and a higher initial threshold may be warranted when the claim is particularly important or incredible.

  • Kevin

    Charlie: I agree, the medical community often needs to adopt stricter standards. A recent example of this: Size matters, and other lessons from medical genetics.

  • Jeff

    If CERN’s “hint” of a Higgs boson is like a bill passed by the House but not the Senate, does that mean that CERN has an uncompromising ideological agenda to prove the existence of a Higgs boson? 😉

  • Chris

    It’s those darn time traveling Higgs from the future trying to keep us from discovering them!

  • Adam

    @MPS17 — Actually, it’s even MORE important that you press for 5-sigma significance when you have a theoretical backing for the measured phenomena in question. Prejudging and all that whatnot.

    re: the medical debate — It’s very difficult to find millions of people to test on (which is what 5-sigma significance requires), at least ethically, where their backgrounds are identical. People just need to be informed about statistical significance, and how to interpret findings based on that.

  • Brendon J. Brewer

    “That is, given a 3-sigma result, I’m inclined to take 100 to 1 odds that it’s there.”

    You’re interpreting a p-value as a posterior probability. The odds on a 3-sigma result being correct depend on the assumptions but are typically more like 5-to-1.

  • MPS17

    Thanks for pointing out my naivete Brendon J Brewer; I should read more about probabilities and evidence and such.

    Yet I think your comment points to another aspect in which the analysis is tricky, because if you want to talk Bayesian probabilities for theories in light of evidence, you have the problem that we don’t have an alternative theory other than there is a Higgs. (I assume you are not talking about a spectrum of theories with Higgs but just uncertain mass.)

  • Woof

    990 mile marker?

    PLEASE tell me they don’t use MILES at CERN!

  • Albert Z

    Sigh, once again a toe appears from beneath the stage curtain and we are encouraged to think that it belongs to a beautiful tiny dancer. Alas, there are myriad other possibilities. In fact it looks suspiciously like the other “toes” that have appeared from beneath the curtain and then mysteriously disappeared.

    Don’t expect big news tomorrow.

    Just enough to keep the show going.

    Albert Z

  • BigBangBoom

    The lack of a Higgs Boson would in many ways be more exciting than finding one. Our current models have failed to find us the path to Fusion energy, move any closer to a Grand Unified Theory etc., etc. . A good solid Dead End might just about the greatest discovery (or non-discovery) possible. The motto “if it ain’t broke don’t fix it” is a two edged sword that means that real progress is only possible in moments of “failure”. Here’s to not finding the Higgs.

  • Pingback: Temps Higgs : 5 indicateurs à observer | Matières Vivantes()

  • Anadish Kumar Pal

    Faster than light Neutrinoes and Higgs both cannot coexist — either one has to be wrong. It’s DCE research and superluminal speed which has the potential of breaking current scientific barriers, rather than finding a nebulous statistical dual peak for a Higgs, which well could be due to many other anomalies, one that LHC could not decipher is that of the UFOs.

  • Nirdosh

    Sean, there is quite a chance that the media will misinform the public about the bill passed by the house.
    Anyway, its a good feeling that once in a while “physics” does not only get a huge attention but also gets pure hype. We really are a part of a scientifically aware (or informationally manipulated) civilization.

  • Josh

    3 Sigmas is exciting stuff. Between this and the faster than light neutrinos, I had to learn all about this sigma business. It makes me want to make jokes about synergy, and retreating to move forward.

  • Pingback: Physics and astronomy: worth billions?()

  • Pingback: Higgs-mania Day « In the Dark()

  • KWK

    So, two-ish sigma significance each for ATLAS and CMS–but with slightly different best fit masses. How much of a problem is that? It seems to me that at least one of the two low-significance excesses will have to go away, unless their energy resolution is worse than they claim–but that seems like an even bigger problem.
    Also, with regard to posterior significance, experimentalists who are careful will specify the number of trials they performed and what effect that has on their result. In such cases a 3-sigma post-trials result should be interpreted straightforwardly as the usual 99.7% certainty.

  • Pingback: The God particle and Tim Tebow |

  • Avattoir

    “Faster than light Neutrinoes and Higgs both cannot coexist”.

    Oh, noes!

    Hey, Pal*, I think you’ve got some ‘splainin’ to do (and not just on why you apparently think that ‘neutrinos’ decay into “neutrinoes”).

    We know quantum tunnelling is real, so does that fact mean the Higgs boson cannot exist? And if not, then how is that different from what you are asserting?

    *I’m serious here, not just seriously handicapped in the limits to my understanding of HEP-th, but having a serious problem with understanding the basis for your assertion.

    But besides all that, this is the first ever decent opportunity I’ve ever had to start a blog comment with “Hey, Pal”!

  • steven johnson

    I’m rather confused. One detector has an excellent indication of something massing 123 GeV and another has an excellent indication of something massing 126 GeV. Are we supposed to expect that another year’s run of data will provide a five-sigma (practically certain) indication of something massing 123-126GeV? Which I rather expect is far too imprecise to count as a discovery, given LHC’s technical capacities.

    Aren’t these figure self-canceling, instead of reinforcing? Or, to put it another way, wouldn’t data from the two experiments be a lot more convincing if they really found the same figure? The articles seem to be assuming 123 and 126 are effectively the same figure. Suppose next year they say the figures indicate (with the requisite five sigmas,) one experiment finds 122 GeV and the other 127 GeV?

  • Claire C Smith

    “This problem plagues medical research, where a common practice is to search for any darn correlation (of which there are an unlimited number by—- >chance”)<– – 4. Charlie.

  • Frank M DiMeglio

    Here is the clear, definitive, complete, and best proof that the Higgs search is a total and failed waste of money and time:

    Here is the true, general, and fundamental unification of physics, AND this is what Einstein’s gravity lacks and what dreams FUNDAMENTALLY demonstrate, include, and unify:

    1) Instantaneity (and fundamentally).

    2) Truly and fundamentally equivalent and balanced inertia and gravity (both at half energy/force strength.) Gravity is reduced to the extent that inertia is increased.

    3) Fundamentally balanced and equivalent attraction and repulsion.

    4) Gravity and inertia are both FUNDAMENTAL to distance in/of space.

    None of you can refute this. THIS IS GIGANTIC NEWS IN PHYSICS!

  • Amrit Sorli, Space Life Institute

    no good no higgs boson

  • Pingback: The last days … of 2011 | The Zingularity()


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

About Sean Carroll

Sean Carroll is a Senior Research Associate in the Department of Physics at the California Institute of Technology. His research interests include theoretical aspects of cosmology, field theory, and gravitation. His most recent book is The Particle at the End of the Universe, about the Large Hadron Collider and the search for the Higgs boson. Here are some of his favorite blog posts, home page, and email: carroll [at] .


See More

Collapse bottom bar