Science! It Marches On

By Sean Carroll | December 13, 2011 8:39 am

The news from Geneva this morning is in. Essentials: what we’re seeing is pretty consistent with the existence of a Higgs boson around 123-126 GeV. The data aren’t nearly conclusive enough to say that it’s definitely there. But the LHC is purring along, and a year from now we’ll know a lot more.

It’s like rushing to the tree on Christmas morning, ripping open a giant box, and finding a small note that says “Santa is on his way! Hang in there!” The LHC is real and Santa is not, but you know what I mean.

Here are the technical write-ups from ATLAS and CMS. For stories and some live-blogs, check out Philip Gibbs, Matt Strassler, Aidan Randle-Conde, Ken Bloom, or Jester. Or if you just want the bottom line sigmas, Jim Rohlf provides them. ATLAS gives 3.6 sigma local significance, 2.3 sigma global significance; CMS gives 2.6 sigma local significance, 1.9 sigma global significance (although CMS points to about 124 GeV, while ATLAS points to about 126, which might be important). The difference between “local” and “global” is that the first asks “if I were only looking at this one point in parameter space, how surprising would the result be?”, while the latter asks “what is the chance I would find this kind of deviation somewhere in parameter space?” Nominally the global significance is obviously more relevant, although one could argue that we have good reasons to expect that the Higgs is actually lurking right there, so the local significance isn’t completely cheating.

Let’s put it this way: if we were testing a theory that everyone thought was wrong, rather than one that everyone thinks is right, nobody would take these results as strong indications that the idea was correct. We have a strong theoretical bias that the Higgs exists and is somewhere close to this mass range, so it’s completely reasonable to think that we are seeing hints (tantalizing ones!) that it’s there, but wait-and-see is still the right attitude.

Here are the simplest plots I could find. First the full analysis from ATLAS (zoomed in on the interesting region), via Philip Gibbs’s blog.:

Then from CMS, via Ken Bloom:

These plots are complicated because they’re trying to tell you two things at once. The black curve is the data, the green/yellow bands are the expected ranges of the data at 1 sigma and 2 sigma. If all you want to do is ask whether we can exclude the Higgs in a certain range, just check whether the black band is below the value 1. But if you want to say you have evidence for the Higgs, you need the black line to wander above the yellow band (or higher, if you want more than 2 sigma [and you do]). So ATLAS sees something at 126 GeV, CMS is at least consistent with 123-124 GeV (although it doesn’t see much at 126).

As Sarah Kavassalis puts it, the real message today is that the LHC is working great. 2012 will bring another year of data, hopefully at even higher luminosity (so many more total events). The Higgs has been around for 13.7 billion years, it will still be there tomorrow.

CATEGORIZED UNDER: Higgs, Science, Top Posts
  • Pingback: Physics/Cosmology by mskauffman - Pearltrees()

  • tom

    “The Higgs has been around for 13.7 billion years, it will still be there tomorrow.”

    Has it? 😉

  • Ernest

    if the Higgs bosom is so fundamental to physics, why is it so hard to find such silly particle? is it ’cause is too small? or because is it too fast? for me is like they tell you there is a force call gravity all over the universe but they can’t find it.

  • John Peacock

    Sean: how can you say “the data aren’t nearly conclusive enough”? We’ve seen independent experiments yield results with global significances of 2.3 and 1.9 sigma, which sounds not hugely impressive (although I’ve seen the 2.3 reported as 2.5 – which is right?). But for random noise, these signals would not peak in the same mass bin – which they do. Looks to me like the formal global significance of the combined signal is 3.5-4 sigma. Barring someone goofing (always possible, whatever the statistics), that sounds like a clear detection. I know PP types like 5 sigma, but isn’t that because in the past they haven’t always been careful enough about local vs global probabilities?

  • Kappy

    What is ATLAS seeing at 114 GeV. That looks outside the expected range for background to me. Is that a known particle? I don’t see the same indication in the CMS chart.

  • JJB

    Are 124 and 126 really distinct in these plots? What is the resolution of these analyses? CMS plots data every 1GeV, while ATLAS plots every 0.5 GeV in the range of 120-130. Yet the “bumps” on the curve are wider than that — the individual bins seem to be correlated.

  • Ryan

    The search for the Higgs is the only particle search I’ve paid attention too. Is this how all particle searches go? Was everyone on the edge of their seat waiting for the tau?

  • Sean

    John– 2.3 sigma is the number in the ATLAS paper:

    The signals *almost* peak in the same mass bin, but not quite. That could be important, or not. If you read some of the more detailed reports linked above (from people who are more expert than me), you see that there are various nagging problems to be resolved, although none of them seem like dealbreakers. Check out Matt Strassler’s informal poll of experimenters and theorists: almost nobody is convinced that the Higgs has been found.

  • drm

    I’m not a physicist but I do use statistics. It seems to me that both plots show systematic deviations from the expected over the whole range.

  • KWK

    drm makes a good point. Even if the Higgs is there, if it in fact has the SM cross-section then the black data points should relax to the center of the green band, and the current evidence will go away until the observed luminosity is great enough that the entire expected confidence band drops below 1. Taken at face value, though, it looks like the cross-section might be 3 or so times higher than the SM.

  • Marianne Brooks

    It’s exciting news but maybe also a justification for this hugely expensive, billion dollar, CERN project. I’ve read the main popular science book about the Higgs particle ( and will follow the news closely in the coming months to see if they have finally found it (or not).

  • Pingback: Midday open thread | Hotspyer – Breaking News from around the web()

  • nb

    Forgive me if this is terribly naive, but the two data curves don’t seem to have the same general shape. For example, the Atlas observations cross the expected line @ 136, but the CMS curve at that point is in the 2-sigma range and doesn’t show an appreciable dip there. The CMS does have a dip in the 128-132 range however. Will these two curves tend to resemble each other more with increased data, or are they never going to look very similar because of differences in the experiments themselves?

  • Hephaestus

    Santa’s not real?!?

    • Sean

      Finally! I’d been waiting for that comment.

  • Hans Kristian Eriksen

    I definitely agree with drm — the CMS plot looks weird. Even after taking into account correlations, it’s quite surprising that *all* data points lie above the mean (and most above 1 sigma) if the noise has been estimated correctly. And if the noise is biased low by 1 sigma, there isn’t much left of the “detection”.

  • Chris

    The photons transmit electromagnetism, gluons transmit strong force, W and Z transmit the weak force. So will the Higgs be responsible for a new force? And if so, what is it called?

  • Pingback: Midday open thread - Online Political Blog()

  • max

    Chris, the Higgs is pretty fundamentally different from the force-carrying bosons in the standard model. All of the (purportedly) fundamental bosons that we have discovered are vector gauge bosons. They are spin-1, and they can be thought of as the necessary ingredients to satisfy the various gauge symmetries. The Higgs, on the other hand, is a scalar particle. It doesn’t have a spin at all. It’s mixed up with the W and Z bosons in a somewhat complicated way (it gives them their mass via the Higgs mechanism), but it doesn’t create a new force of nature.

  • marcelo

    …”So will the Higgs be responsible for a new force? And if so, what is it called?”

    Ans: the divine force, of course, what else ! :) (just kidding).
    By the way, you forgot the gravitons—> gravitational “force”. They have spin 2

  • Anonymous_Snowboarder

    I’m in the unconvinced camp. Stats are weak (yet) and Atlas and CMS conflicting in certain channels. Whatever. What bothers me is that I really don’t think CERN should have done this dog and pony show over these ‘results’. Leaked or not, they are not worth the hype. Kind of like saying ‘the Redcoats are coming!” before they even put their coats on because, well, you know they are coming some day.

  • Sean

    Actually the Higgs does mediate a force, just like any other boson. As marcelo says, the graviton is spin-2, so it’s not only spin-1 bosons that give rise to forces. I mentioned this in a talk years ago:

    Like the weak force, the Higgs is so massive and short-ranged that it doesn’t lead to bound states between any known particles; unlike the weak force, it also doesn’t mediate the main decay channel for any known particles. So its force-ness is not very manifest.

  • Chris

    @19, 20 and 22
    Thanks. I’ll have to crack open my (seriously out of date) particle physics book over the Christmas break.

    Also I did know about gravitons, but I was just mentioning particles which have been discovered and gravitons are still in the hypothetical category.

  • marcelo

    Ok, wait wait (it’s precisely the whole point of this blog) although very likely to be considered as discovered soon, I would still put the Higgs as hypothetical too: strong evidence doesn’t mean a discovery, as Dorigo said. The gravitons have not been discovered yet as quantum particles, however, the binary pulsar leaves no doubt about the existence of gravitational waves, which are also to be detected in a more direct fashion in the forthcoming years.

  • Michael Pierce

    ok, I’ll bite. Why does the 1 (or 2) sigma region extend the same amount above and below the expected values? Wouldn’t you expect the plus/minus sigma to get distorted by the log scale of the plot?

    When I try to read off approximate numbers, it clearly looks like an asymmetry. Take the top plot, left side for instance, an expected value of 2 with a 2sigma low of around 1 and a high of almost 4.

  • Jason Dick

    Just a small comment. Because of our prior theoretical expectation to see a Higgs, and given the exclusion of most of the allowed mass range, to me it seems rather likely that this is real. While normally a detection of this magnitude of an unexpected particle should be treated with extreme skepticism (as in nearly all 2-3 sigma events go away), the fact that we have a strong prior reason to believe that the Higgs should be here means, to me, that it is more likely than not that this is real. I wouldn’t want to say it’s 90% likely, but perhaps 60%-70% likely. But in any event we will know for sure in a few years.

    So yeah, I agree with Sean here.

  • Pingback: Science Tuesday: from the Higgs Boson to the ID Bozos…. « blueollie()

  • Bob

    With almost 2.5 sigma, doesn’t that still mean a confidence level of about 95%? Granted that doesn’t mean certainty, but I’d think it at least means probably.

  • Pingback: Better than Average Links – Week of Dec 11, 2011 | Scrub Physics()

  • Pingback: Mi resumen de la conferencia ATLAS+CMS sobre la búsqueda del bosón de Higgs en el LHC del CERN « Francis (th)E mule Science's News()

  • Thorny

    Those data are also quite consistent with a Higgs at 145 GeV (almost 2 sigma, that’s 90% 😛 ). I think there’s a tiny little bit too much hype around this, so let’s just wait a year and not try to see the black cat in the dark room.

  • John Kerr

    Santa is real, though not at 5 sigma (yet).
    “Anything that can happen, does happen” – Brian Cox.

  • Pingback: Science! It Marches On | Cosmic Variance | Full Power News()

  • Noel25

    Have we not learned anything from the past year or so with several other “hints” of important particle physics discoveries that never panned out? The same thing has been true of the dark matter searches in which several “hints” have turned out to be mirages e.g CDMS II, Pamela, etc.

    My speculation is that since many of the breakthrough confirmations that many physicists have been hoping for a long time have yet to occur, there is a sort of desperation kicking. Hopefully the desperation will end soon so a refocusing of efforts in other theoretical directions can more easily occur. I predict that by this time next year what typically has happened will happen again: nothing, and we will be laughing about all of the ado over “hints” of a Higgs c. December 2011.

  • Qubit

    It can’t be as easy as 123. It should be a lot lower than 123, I mean come on what kind of a universe wants to exist! Or even needs to?

  • Albert Z

    Sean: “The Higgs has been around for 13.7 billion years, it will still be there tomorrow.”

    So you assume it actually exists?

    Before we have adequate proof?

    Are you absolutely sure?

    I would say the “Higgs conjecture” has been around for decades, and whether or not it has already exceeded its shelf-life remains to be seen, perhaps by summer 2012.

    Let’s at least present a modicom of scientific objectivity to the public.

    Albert Z

  • Pingback: Sean Carroll says: Science! It Marches On « UnEntangleMe()

  • J Gregory Moxness

    I offer an explanation of a Higgs @ 124.443 GeV


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