jump to navigation

Interpretation of multi-muons! November 3, 2008

Posted by dorigo in news, physics, science.
Tags: , , ,

The CDF authors of the study which is causing ripples in the blogosphere have published tonight a second paper, where they try to interpret the excess of events with large impact-parameter muon tracks within a phenomenological model of new physics. You can find their paper here.

In short, they try to fit the observed muon multiplicity within narrow cones, as well as their quite peculiar kinematic characteristics, with the decay of a heavy object which produces a cascade of long-lived particles, ending with a multi-muon signature.

The paper was born as part of the other document (see the story in the post below), but was extracted from it and published separately since this was the best way to proceed promptly to a publication of both. As you see by checking the arxiv entry, this second preprint has only the names of the very authors of the study on the multi-muon anomaly.

I will have more detail on the physics later…


1. Anonymous - November 3, 2008

I don’t see any plots of the vertex probability of the multi-muons. Are there any? (& if so, how do they compare with, e.g., the vertex probabilities of b jets?)

2. Matti Pitkänen - November 3, 2008

The lightest state with 3.6 GeV mass would correspond to the mass of neutral leptopion in TGD based model. Depending on whether colored neutrino has same mass as tau-neutrino or tau, the mass of charged state tau-pion is 1.8 or 3.6 GeV.


3. Tony Smith - November 3, 2008


1 – In a thread based on your Halloween 2008 blog entry about CDF multimuons, you said:
“… I [ Tommaso ] would estimate odds that this is new physics at … One percent …
a paper by Arkani-Hamed and collaborators … appears quite extraordinarily to have foreseen the above signature of new physics, in a very timely fashion! …”,
that paper by Arkani-Hamed (and Weiner) is not only the subject of widespread serious discussion in the physics community, but is on the Cornell arXiv at hep-ph/0810.0714 along with a related paper at hep-ph/0810.0713
in this thread based on your 3 November 2008 blog entry about CDF interpretation of multi-muons, you said:
“… CDF authors [ Giromini et al ] of the study which is causing ripples in the blogosphere have published tonight a second paper [ it is on the Cornell arXiv at hep-ph/0810.5730 ], where they try to interpret the excess of events with large impact-parameter muon tracks within a phenomenological model of new physics …”
the Giromini et al paper is also the subject of widespread serious discussion in the physics community and published on the Cornell arXiv.

2 – In comments about Fermilab data analysis on a thread based on a 5 September 2007 entry in your blog, you said:
“… the two plots [ CDF 1994 and D0 1997 tquark semileptonic histograms ] … show single bin fluctuations at 140 GeV. I [ Tommaso ] read 8 events in the CDF one, in the face of a background plus signal totalling probably 2.5, and 5 in the D0 one, with about 2 from bgr+”standard” ttbar … the probability of such a fluctuation .. is of the order of 4-sigma. …”,
my physical interpretation, as evidence that the t-quark is not a simple-minded single-state thing, but rather a multi-state t-quark – Higgs – Vacuum system,
has led to vicious personal attacks on me even though I personally have done nothing but offer physical interpretation of publicly available data.
Further, for the past several years I have been blacklisted by the Cornell arXiv.

It is especially ironic that the physical interpretation of Giromini et al in hep-ph/0810.5730 , based on 1-percent probability, which is widely substantively discussed and published on the Cornell arXiv, calls for
“… three new states … heavier states cascade-decay into the lighter ones …”,
my model, based on 4-sigma probability, calls for viewing the T-quark as a 3-state system involving known particles: the T-quark and the Higgs as a quark condensate
but has led to vicious personal attacks and blacklisting.

since the physics community is happy to have serious substantive discussion, and publication on the Cornell arXiv, of physical interpretations by Arkani-Hamed et al and Giromini et al based on a 1-percent probability of new physics
the physics community reacts to my physical interpretation based on a 4-sigma probability of new physics with hostility and blacklisting,

it is not likely that I will be able to follow your advice [ given on the thread of your Halloween blog entry ] about advocating full openness and transparency with respect to physics collaborations (such as Fermilab, CERN, etc),
which advice was stated as
“… forget it … we are not yet prepared to it …”.

Tony Smith

PS – The present global financial Debacle of the Dollar is a very clear example of what can result when a collaboration (in the Dollar case, the financial hedge-fund-banker community) operates without the constraint of full openness and transparency – see a pdf slide show at
It is a shame (but maybe not surprising, given the number of hedge-fund quant personnel that were hired from the physics community) that the physics community is also “not yet prepared” for full openness and transparency.

4. dorigo - November 3, 2008

Tony, be careful. My 1% estimate is a gut feeling about the chance of this being new physics as opposed to some conventional background. It has NOTHING to do with the significance of the numerical excess of multi-muon events, which is HUGE (of the order of 10 sigma).


5. Not Even Wrong » Blog Archive » The Circus Begins - November 3, 2008

[…] Over at Tommaso Dorigo’s blog there’s a short posting about Giromini et. al., and an exchange with Nima Arkani-Hamed, who claims to have had no inside knowledge of the CDF […]

6. Matti Pitkänen - November 3, 2008

A further observation about the masses suggested in the article, which were 3.6 GeV, 7.3 GeV, and 15 GeV . As I noticed 3.6 GeV is the TGD prediction for the tau-pion mass (2 times tau mass in good approximation). In good approximation the masses come as powers of 2: exact prediction would be 3.6 GeV, 7.2 GeV, 14.4 GeV) .

p-Adic length scale hypothesis predicts that allowed mass scales come as powers of sqrt(2). Several p-adic scales appear in low energy hadron physics for quarks and this replaces Gell-Mann formula for low-lying hadron masses.

Therefore one can ask whether these masses correspond to tau-pion with p= M_k=2^k-1, k=107) and its scaled up variants with p=about 2^k, k= 105, and k=103 (also prime).


7. Alejandro Rivero - November 3, 2008

The question, to me, is which should be the decay width of Upsilon (1S) into these particles. Y(1S) has a mass of 9.4 GeV, very in the target range.

8. dorigo - November 3, 2008

#1. Vertex probability: are you talking about the primary vertex ? The probability is 0.000000000000000000000000000. These are displaced tracks. If you look at the impact parameter plots of the former post, you see hundred bins, and the resolution on the impact parameter is a third of a bin’s width.

#2., #6. Hi Matti, I am happy that you seem to be able to fit this observation into your model. I remain sceptical both of your model and of the CDF signal in equal measure, but I stay vigilant.

#7. Ciao Alejandro, this is in fact a very important objection. I will try to submit it to Paolo.

Cheers all,

9. Anonymous - November 3, 2008

Hi Tommaso,

Sorry if I was unclear — I meant that when you have several ghost muons in a single event, if they were due to the decay of a single beyond-SM particle, the muons should all vertex up (unless there were some other long lived particle that somehow separated the vertices — perhaps even more unlikely, no?). So what is the vertex probability distribution for ghost muons in events with multi-ghost-muon jets? This is something that is worth checking, no?

10. Tony Smith - November 4, 2008

Tommaso, I have been “careful”.

In part 1 of my comment,
I quoted you as saying that you estimate odds that the CDF multimuon events represent new physics at One percent.
Now you say “… the numerical excess of multi-muon events, which is HUGE (of the order of 10 sigma) …”,
you ignore the very substantial probability that the CDF multimuon events may not be real,
which is probably the real reason you quoted One percent instead of 10 sigma as your estimate that new physics is involved.

In part 2 of my comment,
everyone (including you, me, and the consensuses at CDF, D0) agrees that there is ZERO probability that ANY of the tquark semileptonic events in the CDF 1994 and D0 1997 histograms are not real.

Therefore, taking into account the relative probabilities that the relevant events are real or not,
I stand by my comment and I believe that it is in fact very careful.

Tony Smith

PS – In fact, in a 1 November 2008 comment on your blog Haelfix said
“… I don’t see why they would see such a large excess with the integrated luminosity CDF has utilized … I mean the excess is really unusually large …”
you replied on that date
“… yes the effect is large, it seems to imply we are dealing with a very common phenomenon. That is one of the reasons why most of us still think at QCD fake muons of some kind as the most likely source of the anomalous muons …”,
as of 1 November 2008 you were explicitly saying that the “HUGE” “numerical excess of multi-muon events” led you (and others at CDF) to “think … QCD fake muons of some kind …[are]… the most likely source of the anomalous muons”.

What has happened between 1 November and 3 November to change your mind and to regard the “HUGE … excess” as a positive indication of 10 sigma significance?

11. dorigo - November 4, 2008

Hi Anon,

ok, I get it now. Yes, it was of course checked. The impact parameter of multiple muons traveling together is uncorrelated. They not only do not vertex, but they seem unrelated. So, if they are due to some physics process, this is a cascade of different bodies, not a single vertex.

Hi Tony,

I understand better what you meant now. Well, one should not confuse the numerical significance with a proof that something is exotic. The multi-muon signal is very large, but its interpretation in terms of non-standard physics is not straightforward. My 1% estimate is a very, very personal interpretation of the situation, and no meaning should be attached to it other than discussing how I get my own gut feelings. A statistical significance of 10 sigma can be had by 20 events with zero background, or by 70000 with a large background. The two cases are really different. In the first case, one can be sure that each and every event is a good representation of a signal of something, in the second one does not have any handle, since the “signal” is buried in large backgrounds and no single event can be taken as an example of the signal.

I do not think I changed my mind with respect to the multi-muon signal between nov 1 and 3. I always said I think this is background, but the chance that it is BSM is non zero and this makes it exciting. Sorry if I was not clear.


12. Tony Smith - November 4, 2008

Tommaso, thanks for your clarifying comment.
In light of it,

my case 2 (a multi-state T-quark) has roughly a 4-sigma “signal” based on 20 really-real events that you can individually clearly study “with zero background”
my case 1 (the CDF multi-muons) have 70,000 candidate events “with a large background” in which the candidate events are “buried in large background and no single event can be taken as an example of the signal”.

In case 1 (CDF multi-muons) I think that your “gut feelings … very, very personal interpretation of the situation” is very important in trying to understand what is going on, because of the possibility
(not present in case 2 where each and every event can be clearly studied)
that the 70,000 events may not even be real events at all,
but may be some indication that detector details, QCD analyses, etc may not be well understood.
Your position, as someone who really knows how the equipment works and also the details of the techniques of analyses,
means that your “gut feelings” should be described in detail to the world (either in a paper or on your blog etc).
Your “gut feelings” are probably a subconscious sum-over-histories path integral of all those aspects of the situation,
and people (including me) are very interested in all the components of that path integral, and the weights you give to each of them in doing your subconscious path integral construction of your “gut feeling”.

Tony Smith

PS – To me, the situations have analogies in the financial system:

1 – case 1 (CDF multi-muons) seem to me to be like the huge amount of “money” (actually around $500 trillion) created by hedge funds etc in the unregulated “shadow banking system” over the past several years out of derivatives,
which seem to be turning out to be of little or no real value,
just as the hugely numerous CDF multi-muon events might turn out to be not real.

2 – case 2 (Fermilab semileptonic Tquark events) seem to me to be like the old Gold Standard Dollar that really had real value.

13. Matti Pitkänen - November 7, 2008

Dear Tommaso,

a comment about the opening angle of the cone with respect to the direction of the first muon.

There were three models to be considered.

a) Virtual pi^(107) decaying to leptohadrons.

b) The formation of coherent state of k=107 pions heated to QCD like plasma state and producing lepton jets as analogs of quark jets.

c) Real pi( 103) pion decaysing to 3 k=105 pions and pi^0(105) decaying to k=107 pions. The two charged pions decay to tau-nu and mu-nu pairs for k=105 and mu-nu pairs for k=107 cases. This is the option fixed by the consistency with CDF model and should produce lepton jets as a consequence of the peculiar reaction kinematics forcing the reaction products to be almost at rest. I found that this is indeed the case.

The strong decay for p-adically scaled up neural tau-pion pi(103) with mass equal to 8 m(tau) to pi^0(105)+pi^+(105)+ pi^-(105) followed by similar decay of pi^0(105) to k=103 tau-pions predicts that secondary muons resulting from the decay of second charged pi(105) have opening angle 45 degrees and those resulting from decays of charge pi(107) have opening angle 28.7 degrees. Measured opening angle was 36.8 degrees.

It is essential that the masses of colored neutrinos are small, most naturally same as that of ordinary ones, and that the masses of pi^0(103) and pi^(105) are slightly above the threshold allowing the decay to 3 pions occur with 3 resulting pions almost at rest (otherwise the second charged pion must be virtual and decay electroweakly and on mass shell condition is lost and decay is much more slower).

The model is consistent with all what I know at this moment at both qualitative and quantitative level (assuming that CDF model with masses coming as powers of 2 catches the quantitative aspects). CDF anomaly gives new support for two basic notions of TGD: p-adic length scale hypothesis (actually not hypothesis anymore) and the notion that color is not spinlike quantum number but corresponds to CP_2 partial waves so that both leptons and quarks have colored excitations and QCD like dynamics should appear in variety of p-adic length scales.

I have not yet estimated the total rate for the production of pi^0(103): this requires the use of the formulas developed for electropion production cross section. This obviously gives the final killer test for the model.

For more detailed explanation see my blog
my blog.


14. CMS and extensive air showers: ideas for an experiment « A Quantum Diaries Survivor - February 6, 2009

[…] order of likelihood. For sure, if an exotic particle like the 300 GeV one hypothesized in the interpretation paper produced by the authors of the CDF study of multi-muon events (see the tag cloud on the right […]

Sorry comments are closed for this entry

%d bloggers like this: