Einstein Wept.

September 22, 2011 11:01 AM

Neutrinos discovered to be faster than light at CERN. If confirmed, these results will overturn a century of one of the most basic assumptions in modern physics. 'Thousands of experiments have been undertaken to measure' the speed of light 'ever more precisely, and no result has ever spotted a particle breaking the limit. But Antonio Ereditato of the Orion collaboration and his colleagues have been carrying out an experiment for the last three years that seems to suggest neutrinos have done just that.'

'The team measured the travel times of neutrino bunches some 15,000 times, and have reached a level of statistical significance that in scientific circles would count as a formal discovery.

But because the result is so unexpected and would wreak such havoc with our understanding of the Universe, the group is being particularly cautious. They have opted to put a report their measurements online to subject them to wider scrutiny, and will hold a seminar at Cern on Friday to discuss the result.'

posted by VikingSword (254 comments total)

88 MPH MOTHERFUCKERS
posted by nathancaswell at 11:02 AM on September 22, 2011 [17 favorites]

Uh, oh, I may be traveling back in ti
posted by Mental Wimp at 11:02 AM on September 22, 2011 [3 favorites]

My surprise is reaching levels of statistical significance.
posted by 2bucksplus at 11:05 AM on September 22, 2011 [18 favorites]

"Neutrino Speed" just doesn't have the same romantic quality, I'm afraid.
posted by Thorzdad at 11:05 AM on September 22, 2011 [1 favorite]

No....You can't tell me neutrinos exceed the speed of light and then send me to BBC. Where's a geek with a day job supposed to get real information?
posted by Shutter at 11:05 AM on September 22, 2011 [10 favorites]

I like how their response is kind of like "ugh, I wonder what we screwed up in our experiment".
posted by smackfu at 11:06 AM on September 22, 2011 [30 favorites]

I can't believe they fell for this! It's a really old trick. There's just another neutrino that happens to strongly resemble the first, and they dress the same, and the assistant in the short skirt distracts you when they hand off the baton.
posted by Tomorrowful at 11:07 AM on September 22, 2011 [71 favorites]

Thus ushering in a new wave of poorly thought out space opera utilizing terms like "neutrino drive" as an otherwise unexplained justification for interstellar travel.
posted by infinitywaltz at 11:07 AM on September 22, 2011 [19 favorites]

Warp speed, bitches!
posted by TheRedArmy at 11:07 AM on September 22, 2011 [2 favorites]

I, for one, welcome our new transluminal neutrinos.
posted by warbaby at 11:09 AM on September 22, 2011 [19 favorites]

I like how their response is kind of like "ugh, I wonder what we screwed up in our experiment".

Kind of refreshing, actually. There will be plenty of time to name new soldiers in Scalzi books after them once it's all sorted.
posted by OmieWise at 11:10 AM on September 22, 2011 [5 favorites]

No....You can't tell me neutrinos exceed the speed of light and then send me to BBC. Where's a geek with a day job supposed to get real information?

This is the website for the research.
posted by VikingSword at 11:10 AM on September 22, 2011 [5 favorites]

Whoa.

c might be getting a credit downgrade soon, too?
posted by saulgoodman at 11:11 AM on September 22, 2011 [10 favorites]

oh ... i thought you meant flying neutrinos
posted by msconduct at 11:12 AM on September 22, 2011

Writeup and related by theoretical physicist Matt Strassler.
posted by seanmpuckett at 11:12 AM on September 22, 2011 [8 favorites]

It's a really old trick. There's just another neutrino that happens to strongly resemble the first, and they dress the same

They should tag the neutrinos like they do for wildlife before they send them off.
posted by Hoopo at 11:12 AM on September 22, 2011 [4 favorites]

Imagine this is the discovery that finally makes Time Travel a reality.
posted by Skygazer at 11:12 AM on September 22, 2011 [1 favorite]

I can't have waited to will have wanted to have already read this article tomorrow when I will have time to do so yesterday.
posted by PapaLobo at 11:12 AM on September 22, 2011 [4 favorites]

Rounding error?
posted by axismundi at 11:12 AM on September 22, 2011 [1 favorite]

I know any number of border collies that can get from here to the nearest tennis ball faster than any neutrino you care to name.
posted by Wolfdog at 11:15 AM on September 22, 2011 [3 favorites]

Hoopo: They should tag the neutrinos like they do for wildlife before they send them off.

Yes, and attach criitter cams so we can see what they do in the wild. I've always wondered how neutrino's mate.
posted by Skygazer at 11:15 AM on September 22, 2011 [1 favorite]

Rounding error?

Says it's multiple billionths of a second, with 15,000 tests. That seems pretty verified.
posted by kafziel at 11:15 AM on September 22, 2011

The suckiest part of special relativity being proved wrong is going to be the "I told you so" from my 12 year old son, who insists you can just throw a ball from the surface of a photon to exceed the speed of light.
posted by DU at 11:17 AM on September 22, 2011 [29 favorites]

If confirmed, these results will overturn a century of one of the most basic assumptions in modern physics.

It's about damn time. Change is good.
posted by 3FLryan at 11:18 AM on September 22, 2011 [2 favorites]

Neutrinos discovered to be faster than light at CERN

Maybe the light at CERN is just really slow. Try turning the dimmer switch up.
posted by DU at 11:18 AM on September 22, 2011 [18 favorites]

Oh, great. Now I'm going to have to listen to relatives at thanksgiving saying "See? I told you so! We can go faster than light!"

On preview: I see DU has a similar problem. Except my father isn't 12.
posted by mrgoat at 11:18 AM on September 22, 2011 [1 favorite]

Maybe the track get shorter?
posted by swift at 11:18 AM on September 22, 2011

I already broke the news that c is not a constant, but rather the transformational limits of the system right here on MetaFilter.
posted by It's Raining Florence Henderson at 11:19 AM on September 22, 2011 [2 favorites]

Can someone who actually knows something give a guesstimate of how much theory this creates problems for? Are there theories that already account for this possibility? How much change is required for the main candidate big theories to accommodate this? Etc.

Just saying this changes something we thought we knew is actually not that illuminating.
posted by grobstein at 11:20 AM on September 22, 2011

Whew, the timeline has finally caught up to where I can tell y'all about time travel without fucking things up.

I just need to say this: Bob Smith in Decatur, Illinois: In 743 days, on your morning drive to work, DO NOT, I repeat, DO NOT take that shortcut you usually take. The lives of countless billions of people (human and otherwise) depends on it.

Thank you.
posted by kmz at 11:21 AM on September 22, 2011 [15 favorites]

Where is physicsmatt when you need him?
posted by Hairy Lobster at 11:22 AM on September 22, 2011

They have opted to put a report their measurements online to subject them to wider scrutiny, and will hold a seminar at Cern on Friday to discuss the result

Translation: they want to have an extra 24h to fuck around with unsupervised time travel.
posted by elizardbits at 11:22 AM on September 22, 2011 [3 favorites]

Tachyons.
posted by VikingSword at 11:23 AM on September 22, 2011 [1 favorite]

Have they done any drug testing on the neutrinos? I suspect steroids.
posted by never used baby shoes at 11:23 AM on September 22, 2011 [2 favorites]

Doesn't the speed of light vary with the density of the medium? Is it possible they've discovered that space is slightly denser, rather than the neutrinos being slightly faster?
posted by Grimgrin at 11:25 AM on September 22, 2011

Can someone who actually knows something give a guesstimate of how much theory this creates problems for?

c not being a constant, or not being the upper speed limit, screws up pretty much all physics larger than the Earth. A lot of physics smaller than the Earth.
posted by DU at 11:25 AM on September 22, 2011 [1 favorite]

Does anyone have a more real source than a BBC article? Science journalism being what it is, and the OPERA page being as inscrutable as it is, I'm curious about seeing the CERN guys actually say that this is actually what they've measured.
posted by kafziel at 11:27 AM on September 22, 2011 [1 favorite]

Can someone who actually knows something give a guesstimate of how much theory this creates problems for?

I don't work professionally, but I do have a relevant college degree, and if this is true, this doesn't create *problems* per se, but it shows us that our theories are incomplete. For the vast majority of what you do, Newtonian mechanics is good enough. Unless you work on GPS satellites (and need Relativity) or design cutting edge computer electronics (and need Solid-State physics/Quantum Mechanics).

Our GPS satellites are still going to work based on Relativity, even though this result means that Relativity doesn't cover all the bases. From a practical standpoint of your daily life, don't expect much disruption.

For the chalkboard physicists.... this would be HUGE. If someone can successfully explain this and modify our theories to account for it, it's probably Nobel Prize, household-name level huge.
posted by chimaera at 11:27 AM on September 22, 2011 [6 favorites]

Ha, I knew it!

We're just a bunch of, mostly, well dressed apes who have been studying the universe for the barest second of existence. There's a lot to learn, as the Vulcans keep telling me.
posted by Brandon Blatcher at 11:29 AM on September 22, 2011 [4 favorites]

I don't care if they made 15,000 measurements or 15,000,000 measurements: if they made a mistake in their experimental setup the data is flawed. The link above by seanmpuckett is good reading.
posted by exogenous at 11:29 AM on September 22, 2011 [2 favorites]

The writeup that seanmpuckett posted is worth a read, here's my tl;dr take on it: "Let's wait and see; neutrinos are hard to measure over short distances, and the 1987 supernova provided compelling evidence that neutrinos travel pretty much the speed of light, although it's possible these neutrinos are different from supernova neutrinos - also please stop saying six-sigma"
posted by RobotVoodooPower at 11:29 AM on September 22, 2011 [8 favorites]

I've always wondered how neutrino's mate.

With a particle and then a wave.
posted by griphus at 11:29 AM on September 22, 2011 [11 favorites]

Also, if this is true, expect the future (probably decades hence at least) to be very, very different as the science is understood and practical engineering applications of Faster-than-C physics becomes possible.
posted by chimaera at 11:30 AM on September 22, 2011

Proof positive that Gene Roddenberry was a time traveller from the distant future.
posted by El Sabor Asiatico at 11:30 AM on September 22, 2011

I like how their response is kind of like "ugh, I wonder what we screwed up in our experiment".

That is the correct response to new data that overturns settled theoretical results. It's why they repeated the experiment 15 000 times before even making tentative conclusions.
posted by justsomebodythatyouusedtoknow at 11:30 AM on September 22, 2011 [9 favorites]

Hey, with all these neutrinos speeding through the neighborhood and breaking who knows what other laws of physics, it's no surprise the Higgs boson packed it up and moved.
posted by localroger at 11:37 AM on September 22, 2011 [4 favorites]

I can't believe they fell for this! It's a really old trick. There's just another neutrino that happens to strongly resemble the first, and they dress the same...

Ahh yes. The old Patty Duke Uncertainty Principle.
posted by PlusDistance at 11:39 AM on September 22, 2011 [7 favorites]

Neutrinos move in simultaneous 4 day creation in 1 Earth roatation?
posted by d1rge at 11:42 AM on September 22, 2011 [5 favorites]

Is that a typo in the post? It's the OPERA collaboration, not Orion, no?
posted by auto-correct at 11:43 AM on September 22, 2011

I can't believe they fell for this! It's a really old trick. There's just another neutrino that happens to strongly resemble the first, and they dress the same...

Ahh yes. The old Patty Duke Uncertainty Principle.

They wuark alike, they tuark alike,
Sometimes they even quark alike --
You could lose you mind,
Nutrinooooos
Are two of a kiiiiind!

posted by Herodios at 11:44 AM on September 22, 2011 [14 favorites]

Now is the time to tell your programmer friends that researchers at CERN have discovered something faster than C.
posted by tykky at 11:44 AM on September 22, 2011 [23 favorites]

But because the result is so unexpected and would wreak such havoc with our understanding of the Universe, the group is being particularly cautious.

If I may be so bold as to say that all of math is based on theories. All of Physics too. We just assumed the speed of light was wicked fast, then they found something faster.

So this blows out of the water the theory that things get massively massive as they approach the speed of light, right?
posted by Gungho at 11:50 AM on September 22, 2011

Neutrinos are the Michelle Bachmann of the elementary particle crowd, only interacting weakly with reality.
posted by seanmpuckett at 11:53 AM on September 22, 2011 [36 favorites]

Neutrinos are the weirdest.
posted by Zerowensboring at 11:53 AM on September 22, 2011

C Equals the Speed of Light Squared Dyspepsia Blues... (in C Minor)

C's got the blues;
I'm sad to say,
Everyone knew that right away.

It used to be,
the lynchpin in the
Theory of Relativity

But now things have changed
That C done been found out
It's been messin' round
with Mr. Neutrino
Someone get me the Bean-o

I got the C equals the speed of light dyspepsia blues...



{Blues Trumpet Solo}
posted by Skygazer at 11:57 AM on September 22, 2011 [1 favorite]

$100 on experimental error, if anyone wants a piece.
posted by fraac at 11:58 AM on September 22, 2011 [1 favorite]

Pretty poor form to leak to the press before even putting a paper on the arXiv. Nanosecond timing is tricky.
posted by fantabulous timewaster at 11:59 AM on September 22, 2011

"DO NOT take that shortcut you usually take."

Um, if he usually takes it, is it a short cut? Isn't it his usual route?
posted by Splunge at 12:01 PM on September 22, 2011 [2 favorites]

My five year old son has been asking a lot of questions about physics (yeah, he is unusual that way) and just recently we have solidly established that the speed of light is the fastest thing possible -- Even faster than a motorcycle. Now I have to go back with my hat in hand. Ah, parenthood.
posted by dgran at 12:02 PM on September 22, 2011 [5 favorites]

It's actually a nail-biting move for them. They're saying "We've found something that we think invalidates a core principle of modern physics. It's much more likely that we've screwed up somewhere, but we can't see how. Here's our data, and now we wait for one of you to point out where we were bozos."
posted by bitmage at 12:04 PM on September 22, 2011 [4 favorites]

Science! IT WORKS, BITCHES!

(by accident, after a century of ostracizing cranks who challenge the prevailing consensus)
posted by Tell Me No Lies at 12:06 PM on September 22, 2011 [2 favorites]

So this blows out of the water the theory that things get massively massive as they approach the speed of light, right?

Not really. We can observe mass increases in the particle chambers, and that has always been entirely in line with current knowledge of Relativity. Electrons and protons get heavier exactly as the theory predicts.

Neutrinos are weird creatures anyway, so if true, this effect might be limited to them, or other peculiar or exotic circumstances (which could, almost literally, be anything: neutrinos with exceptionally high energies, neutrinos moving normal to the force of a gravitational field rather than parallel to it, neutrinos created from atypical (non-stellar) sources... could be almost any specific peculiarity).
posted by chimaera at 12:07 PM on September 22, 2011

1.21 Gigawatts = Energy consumed by 12 million Easy-Bake Ovens.

Energy Fun Fact #121:
Since their introduction in 1963, over 16 million Easy-Bake Ovens have been sold; each using a 100 watt lightbulb (sometimes two 60 watt bulbs). The total energy consumed by all those ovens? 1.6 BILLION WATTS (i.e. 1.6 GIGAWATTS)! Given a usage of about 1000 kilowatts per month, it would take the average American household over 133 years to consume that much energy!

Now THAT's a lot of wattage!
posted by not_on_display at 12:08 PM on September 22, 2011 [1 favorite]

Doesn't the speed of light vary with the density of the medium?

When physicists talk about the speed of light, they usually mean the speed of light in a vacuum, c, which is (thought to be) constant. Neutrinos are hardly affected by matter, especially over these relatively short distances, so you would expect the speed to be really close to c.
posted by dirigibleman at 12:10 PM on September 22, 2011

Most awesome, and I can’t wait to see how this will be harnessed by the new-agey handwavers.

From the people who brought you free will courtesy of quantum indeterminacy, the power of positive thinking courtesy of quantum entanglement, now: ... I dunno. Astral projection? Clairvoyance? Maybe just a greater chance of reptilian infiltrators in the Senate.
posted by Durn Bronzefist at 12:14 PM on September 22, 2011 [4 favorites]

The Bartender says "We don't serve tachyons here". A tachyon walks into a bar.
posted by alby at 12:15 PM on September 22, 2011 [88 favorites]

Imagine this is the discovery that finally makes Time Travel a reality.

"I was standing on the edge of my toilet hanging a clock, the porcelain was wet, I slipped, hit my head on the sink, and when I came to I had a revelation! A vision! A picture in my head! A picture of this! This is what makes time travel possible: the flux neutrinitor!"
posted by Clinging to the Wreckage at 12:16 PM on September 22, 2011

From the people who brought you free will courtesy of quantum indeterminacy,

Funny you say that, because there's an interesting piece on Scientific American right now that examines some of the actual scientific issues concerning free will in light of Quantum Mechanics. It's not necessarily just the loudmouth, wannabe-polymath down at the local coffee shop who think there's room for some legitimate speculation around those topics.

So does this mean we can travel backwards in time on a space ship made of neutrinos and finally kill Hitler?
posted by saulgoodman at 12:27 PM on September 22, 2011 [2 favorites]

Actually, I think it's fantastic and shows how smart they really are by saying "here, please find where we screwed up" instead of doing something a-la cold fusion, etc.
This way, if they DID screw up, it's a big win for the scientific method and crowd-sourcing data analysis; if the DIDN'T screw up, it's also a big win for crowd-sourcing AND they get a Noble prize.
posted by Old'n'Busted at 12:32 PM on September 22, 2011 [1 favorite]

Neutrinos, they are very small.
They have no charge and have no mass
And do not interact at all.
The earth is just a silly ball
To them, through which they simply pass,
Like dustmaids down a drafty hall
Or photons through a sheet of glass.
They snub the most exquisite gas,
Ignore the most substantial wall,
Cold shoulder steel and sounding brass,
Insult the stallion in his stall,
And, scorning barriers of class,
Infiltrate you and me. Like tall
And painless guillotines they fall
Down through our heads into the grass.
At night, they enter at Nepal
And pierce the lover and his lass
From underneath the bed—you call
It wonderful; I call it crass.

/mandatory
posted by saturday_morning at 12:34 PM on September 22, 2011 [8 favorites]

So does this mean we can travel backwards in time on a space ship made of neutrinos and finally kill Hitler?

Goddamnit I thought we'd been over this!?
posted by thsmchnekllsfascists at 12:34 PM on September 22, 2011 [7 favorites]

was the LHC used for this?

Sort of. It is the the Super Proton Synchrotron, used as a "pre-accelerator" for the LHC that produces the beam of neutrinos which are sent through the Earth to the Gran Sasso laboratory.
posted by alby at 12:36 PM on September 22, 2011 [1 favorite]

which are sent through the Earth

Physics is so cool.
posted by thsmchnekllsfascists at 12:37 PM on September 22, 2011 [4 favorites]

which are sent through the Earth

Physics is so cool.

Did I point out it's a 730km journey between the two labs?
posted by alby at 12:41 PM on September 22, 2011 [2 favorites]

I like how their response is kind of like "ugh, I wonder what we screwed up in our experiment".
posted by smackfu at 7:06 PM on September 22

Which is precisely as it should be.
posted by Decani at 12:42 PM on September 22, 2011

Hairy Lobster: physicsmatt is at a conference. physicsmatt is also sick as a dog. This makes physicsmatt extremely irritated. I'm going to ignore all the cheerleading from the side-lines on how everyone knew all those mean scientists were wrong all along, because... yeah.

So there's no paper out on this, and so there's not much for me to judge it on. If this rumor turns out to be true, my gut tells me that it's some systematic that they're not accounting for. Matt Strassler's writeup is very nice, and gets at the heart of the problem: we have VERY good information that neutrinos travel at c in a vacuum from SN1987, and so any new physics would need to explain both observations. It's much easier for me to believe that their clocks are miscalibrated or something like that. Otherwise there's some wacky energy-dependent or matter effect; both seem unlikely (though I'm sure some MOND-proponent will jump on the former). Also, there have been a lot of other neutrino beam-line experiments done in the past, and though they tend not to agree with each other very much on neutrino mass splittings or neutrino-anti-neutrino differences, I would have thought that they would place pretty strict bounds on superluminal neutrinos. That said, I don't know the experimental set-ups on the other detectors, so maybe it's just that no one checked. (The fact that neutrino experiments continually come out with unusual results that aren't seen by other experiments should also tell you that this particular field of experiment is damn hard)

Someone asked what would be wrong if this result was right. Well, to some degree, everything. Modern physics uses special and general relativity everywhere (even quantum mechanics, via quantum field theory), if that goes out the window, then we'd have to start from scratch. The problem with this is that modern physics is incredibly successful. Yes, dark matter, dark energy, the higgs and so on aren't figured out, but you only hear about that because that's the edges of our knowledge. If relativity was wrong, then I would generically expect the effects to have shown up before now, in ways we can't sweep under the rug. It hasn't (for example, if GR and SR are wrong, why does my GPS work?), so my initial reaction is to be distrustful of this result. Any new theory not only needs to explain this, it needs to explain why GR looked correct for so long.

Which means, to answer the question that is being asked a lot in this thread, maybe if this result were right you could build a FTL-spaceship, but the odds are that it would be prohibitively expensive anyway, since we know regular matter in at reasonable energies doesn't do FTL.

All of which is to say, dgran, that I wouldn't backtrack too quickly there in explaining things to your son. This result is almost certainly wrong, and then you'd have to backtrack on the backtrack. If it isn't, the end result will turn out to be something much more interesting than a blanket statement like "oh, some things can travel faster than light, other things can't."
posted by physicsmatt at 12:43 PM on September 22, 2011 [30 favorites]

¿ɹǝןʇıɥ ןןıʞ ʎןןɐuıɟ puɐ souıɹʇnǝu ɟo ǝpɐɯ dıɥs ǝɔɐds ɐ uo ǝɯıʇ uı spɹɐʍʞɔɐq ןǝʌɐɹʇ uɐɔ ǝʍ uɐǝɯ sıɥʇ sǝop os


Yes. But it's complicated.
posted by Skygazer at 12:45 PM on September 22, 2011 [1 favorite]

I keep posting in other threads, but this one moves so fast it's always at the top of my Recent Activity.
posted by OmieWise at 12:46 PM on September 22, 2011

I keep posting in other threads, but this one moves so fast it's always at the top of my Recent Activity.

This thread is moving so fast I feel like I'm moving backwards through time.
posted by Skygazer at 12:49 PM on September 22, 2011

I want to know how they synchronized their clocks.
posted by Chocolate Pickle at 12:50 PM on September 22, 2011

¿ɹǝןʇıɥ ןןıʞ ʎןןɐuıɟ puɐ souıɹʇnǝu ɟo ǝpɐɯ dıɥs ǝɔɐds ɐ uo ǝɯıʇ uı spɹɐʍʞɔɐq ןǝʌɐɹʇ uɐɔ ǝʍ uɐǝɯ sıɥʇ sǝop os

Yes. But it's complicated.

READ BULLETIN 1147, PEOPLE!
posted by Durn Bronzefist at 12:51 PM on September 22, 2011 [1 favorite]

d1rge: Neutrinos move in simultaneous 4 day creation in 1 Earth rotation

DILUTE! DILUTE!
posted by Greg_Ace at 12:51 PM on September 22, 2011 [1 favorite]

I...just can't drive....299792.458 km/s!!!!!!! -- Sammy and the Neutrinos
posted by OHenryPacey at 12:52 PM on September 22, 2011 [1 favorite]

Doesn't the speed of light vary with the density of the medium?

Sort of, but not really, and not in the sense that would allow you to draw a conclusion like:

Is it possible they've discovered that space is slightly denser, rather than the neutrinos being slightly faster?

Really (or, as I understand it (or, as I understand we understood it until today)), light only travels in a vacuum. At the lowest level, there's no such thing as the speed of light in anything other than a vacuum. Rather, in some medium (let's say air), most of the actual space between particles is vacuum. Light travels in that vacuum, between particles, at the (only) speed of light. But it might hit a particle, get absorbed by that particle, and then get emitted by that particle. That absorption and re-emission takes time, so on average, the light took longer to travel the same distance that it would have had the particle not been there. Hence the rough idea of light moving slower in a medium than in a vacuum.

But (assuming that idea is basically accurate) that wouldn't seem to allow for a medium wherein light travels faster than in a vacuum; it's still really only ever traveling at the (only) speed of light (in a vacuum). To average a faster speed, it would have to be re-emitted from a particle before it was absorbed by that particle.

Disclaimer: IANAP.
posted by Flunkie at 12:54 PM on September 22, 2011

Huh, so I was was messing around in the mad-science lab and there was this one switch that saidPhysics Overide:
DO NOT TOUCH!
And I wondered how long it would take before I started noticing random, real world weirdness after I flicked it.

Answer: less time that I expected. Or maybe it was more. If the letters from the future I've been getting are to be believed, things are just going to get more and more strange from this point forward.
posted by quin at 12:58 PM on September 22, 2011 [2 favorites]

The neutrinos better not try this shit in Texas.
posted by Camofrog at 12:59 PM on September 22, 2011 [2 favorites]

If the letters from the future I've been getting are to be believed, things are just going to get more and more strange from this point forward.

Hmm...
posted by Durn Bronzefist at 1:02 PM on September 22, 2011

tykky: Now is the time to tell your programmer friends that researchers at CERN have discovered something faster than C.

They'll just tell you to port all your equations to C++
posted by wenestvedt at 1:03 PM on September 22, 2011

Tachyons.

Searching for a distant star
Heading off to Iscandar
Leaving all we love behind
Who knows what dangers we'll find?
posted by homunculus at 1:13 PM on September 22, 2011 [1 favorite]

Some neutrino speed measurements were done in the 1980's, according to Wikipedia. They found "the speed of 3 GeV neutrinos to be 1.000051(29) c. While the central value is higher than the speed of light, the uncertainty is great enough that it is very likely that the true velocity is not greater than the speed of light." So there is a previous result in the range that CERN is seeing, just dismissed at the time.
posted by bitmage at 1:16 PM on September 22, 2011 [3 favorites]

Why would it be just a smidge different, if it is?
posted by fraac at 1:21 PM on September 22, 2011

Newtonian mechanics was incredibly successful and useful in its time (and still is), but eventually we determined that it doesn't accurately describe the universe for very large scales of mass or energy. In fact, this realization came about from an observation not entirely unlike this one: a small discrepancy in the orbit of Mercury. Newtonian mechanics couldn't explain it. It took a new system, general relativity, to accurately explain phenomena on cosmological scales. Similarly, we realized that very tiny scales of mass and energy obey an entirely different set of rules. It took a third system, quantum mechanics, to accurately describe those.

None of these systems are fundamentally right or wrong. They're just models that hold up under certain conditions. If the CERN results turn out to be accurate, it's not going to invalidate relativity, any more than relativity invalidated Newtonian mechanics. It just means that our repertoire of models will have to grow and/or adjust. And if history is any indication, whatever model comes next is just going to be weird as all hell.
posted by dephlogisticated at 1:27 PM on September 22, 2011 [4 favorites]

Thus ushering in a new wave of poorly thought out space opera utilizing terms like "neutrino drive" as an otherwise unexplained justification for interstellar travel.

Actually, I'm hoping it's true just to piss off all the Mundane SF adherents. ;D
posted by Celsius1414 at 1:27 PM on September 22, 2011 [2 favorites]

Can you use this to convey information? If not then it doesn't matter.
posted by humanfont at 1:30 PM on September 22, 2011

Surely some sort of error in measurement. If Einstein was right, the neutrinos would become infinitely more massive than the universe itself and require infinite energy to accelerate to that point. If you are an imaginary particle (what you get from the square root of a negative number, if that helps), sure, you can go faster - infinitely faster, in fact. The trick for imaginary particles is slowing them down to near the speed of light.
posted by Dodecadermaldenticles at 1:31 PM on September 22, 2011

And if history is any indication, whatever model comes next is just going to be weird as all hell.

(Hell is WEIRD AS FUCKING SHIT)




(and fucking shit is just scathingly, poke-your-eyes-out weird)
posted by 3FLryan at 1:46 PM on September 22, 2011

Flunkie: "

Light travels in that vacuum, between particles, at the (only) speed of light. But it might hit a particle, get absorbed by that particle, and then get emitted by that particle. That absorption and re-emission takes time, so on average, the light took longer to travel the same distance that it would have had the particle not been there. Hence the rough idea of light moving slower in a medium than in a vacuum. Disclaimer: IANAP."

I think I read something recently (might have been that "10 mindblowing things about the universe" video, actually, so "watched" would be correct in that case).

Anyways, I think he said something like in order for photons to get from the center of the sun to the surface takes a long time (thousands of years?) because of all the bouncing and absorbing and re-emitting on the matter of the sun, but once it's out and been propagated, it only takes 8 minutes to travel to Earth. Which shows the power of the vacuum vs bouncing and hitting everything...
posted by symbioid at 1:47 PM on September 22, 2011

Imagine this is the discovery that finally makes Time Travel a reality.

Dude, if time travel is ever going to happen, it already has.
posted by rokusan at 1:59 PM on September 22, 2011 [7 favorites]

Can you use this to convey information? If not then it doesn't matter.

It seems they couldn't be getting the results if information was not being imparted. The article is thin, but describes a particle gun in one country, and a particle detector in another country, and when they push the button on the gun, the detector lights up sooner than it should be able. Sounds like an FTL telegraph.

Well ok, it sounds like an error in the experimental setup. But further down the list of things it seems like, is FTL telegraph :)
posted by -harlequin- at 2:00 PM on September 22, 2011 [1 favorite]

I don't think we understand the gravity of the situation...

...or light speed. we don't understand gravity or light speed.
posted by acyeager at 2:02 PM on September 22, 2011 [1 favorite]

"and when they push the button on the gun, the detector lights up sooner than it should be able."
I'm making this bit up, but it sounds like a reasonable simplification.
posted by -harlequin- at 2:03 PM on September 22, 2011

Can you use this to convey information? If not then it doesn't matter.

Of course you could. Trivial one-bit example: Replace the traditional "white smoke indicates a new Pope has been elected" with "a beam of neutrinos indicates a new Pope has been elected".
posted by Flunkie at 2:04 PM on September 22, 2011

Sounds like an FTL telegraph.

Ansible.
posted by rokusan at 2:07 PM on September 22, 2011 [1 favorite]

Can you use this to convey information?

Sounds like an FTL telegraph.

Same thing, but without the dog.
 
posted by Herodios at 2:27 PM on September 22, 2011 [1 favorite]

Roland Emmerich's [i]2012[/i] was all caused by neutrinos suddenly acting all contrary, if I recall the parts before Los Angeles crumbling into bite-sized pieces correctly. It's going to be really embarrassing if that turned out to be prophetic.

Ah well. Better than Day After Tomorrow, at least. I'm just not up to trying to outrun a wave of cold in order to frantically burn books.
posted by Drastic at 2:31 PM on September 22, 2011 [1 favorite]

Imagine this is the discovery that finally makes Time Travel a reality.

It wasn't.
posted by The Bellman at 2:38 PM on September 22, 2011 [4 favorites]

"and have reached a level of statistical significance that in scientific circles would count as a formal discovery"
Statements like these are why us Bayesians should get on with the revolution and finish overthrowing those weirdo Frequentists.
posted by edd at 2:38 PM on September 22, 2011 [3 favorites]

There are some more juicy details here:
Can Neutrinos be Superluminal? Ask OPERA!
posted by Nyrath at 2:41 PM on September 22, 2011 [1 favorite]

"The Bartender says "We don't serve tachyons here". A tachyon walks into a bar."

My first tweet that anyone retweeted. Thanks guy!
posted by fraac at 2:44 PM on September 22, 2011 [7 favorites]

viXra, Nyrath? It's the home of cranks. Don't touch it with a 10,000 yard Lorentz-contracted bargepole.
posted by edd at 2:44 PM on September 22, 2011 [1 favorite]

Imagine this is the discovery that finally makes Time Travel a reality.

Dude, if time travel is ever going to happen, it already has.

It seems likely that time travel won't happen until someone builds a receiver.

If the CERN results turn out to be accurate, it's not going to invalidate relativity, any more than relativity invalidated Newtonian mechanics.

That sounds like either a semantic issue or a risky assumption. It's like saying that Copernicus didn't invalidate the Ptolemaic model --- they both gave very useful descriptions of the data they had, but the incompleteness of the Ptolemaic model led Copernicus to abandon it entirely.

Not saying that we are necessarily in the same situation here, but entire models have come crashing down before.
posted by Tell Me No Lies at 3:02 PM on September 22, 2011

It's the shoes. Don't know how models can walk in those things.
posted by It's Raining Florence Henderson at 3:08 PM on September 22, 2011 [4 favorites]

Somebody didn't measure the source's distance to the detector properly!
posted by Threeway Handshake at 3:11 PM on September 22, 2011

Lots of witty one offs here. I am surprised to say there is a more detailed and scientifically interesting thread about this on reddit. Link withheld, but yeah.
posted by datter at 3:13 PM on September 22, 2011 [2 favorites]

Why withhold?
posted by edd at 3:29 PM on September 22, 2011

It's tax free.
posted by It's Raining Florence Henderson at 3:31 PM on September 22, 2011

Hey! Who put all these rips in my space-time continuum?
posted by Sys Rq at 3:33 PM on September 22, 2011

I like how their response is kind of like "ugh, I wonder what we screwed up in our experiment".

That's science.
posted by zardoz at 3:39 PM on September 22, 2011 [1 favorite]

I really don't see what the big deal about time travel is, I've been doing it for years.

Though apparently I can only go in one direction, and at a fixed rate at that. Not really that useful of a skill actually.
posted by Hello, I'm David McGahan at 3:50 PM on September 22, 2011 [2 favorites]

There's no reason to believe that neutrino's travel faster than the speed of light, and a lot of reasons to believe that they shouldn't. I think it's exceedingly unlikely that this isn't a measurement error.
posted by empath at 4:06 PM on September 22, 2011

f the CERN results turn out to be accurate, it's not going to invalidate relativity, any more than relativity invalidated Newtonian mechanics.

That sounds like either a semantic issue or a risky assumption.

Relativity doesn't depend on slower than light travel. It works with tachyons. It's just weird.
posted by empath at 4:07 PM on September 22, 2011

Yes. Yes, he killed Blake and half New York. Excuse me, Rorschach. I'm informing Laurie ninety seconds ago.
posted by Scoo at 4:46 PM on September 22, 2011 [2 favorites]

Chang Kee Jung, a neutrino physicist at Stony Brook University in New York, says he’d wager that the result is the product of a systematic error. “I wouldn’t bet my wife and kids because they’d get mad,” he says. “But I’d bet my house.”

I'd bet my car that this result will be falsified by Xmas. Even if I had one.
posted by Twang at 5:49 PM on September 22, 2011

Here's the arxiv link to the OPERA result: http://arxiv.org/abs/1109.4897
Haven't read it yet, will drop back in later.
posted by physicsmatt at 6:22 PM on September 22, 2011 [1 favorite]

Faster-than-light travel discovered? Slow down, folks.
posted by carmina at 7:28 PM on September 22, 2011 [1 favorite]

As long as they're not decelerating to sub-light speeds, special relativity still holds true an Einstein is still right. They probably just have negative mass or something.
posted by Civil_Disobedient at 10:25 PM on September 22, 2011

I can't wait to check my e-mail over the next few days.

'See? This is how homeopathy works!'
'See? This is how The Secret works!'
'See? This is how God can exist but we can't see Him!'
posted by obiwanwasabi at 11:10 PM on September 22, 2011 [1 favorite]

Christ, this thread is worse than slashdot.
posted by ryanrs at 11:22 PM on September 22, 2011 [1 favorite]

Dude, if time travel is ever going to happen, it already will.

FTFY
posted by secret about box at 11:38 PM on September 22, 2011

I honestly think that snarky I'd-bet-my-whatever-they're-wrong comments in the media are disrespectful of the process. There are lots of uncertainties around the experiment, and around neutrinoes generally. The CERN experimenters have done exactly the right thing given the results they have. Snarky comments won't change the results one bit though they may alter the public's perception of the value of experiments like these.
posted by newdaddy at 3:54 AM on September 23, 2011 [2 favorites]

Little Neutrino.
posted by Splunge at 4:08 AM on September 23, 2011

newdaddy, all the betting I've seen are quotes from actual scientists, not the press. Making bets about results is a longstanding tradition in physics, and right now, there's a good chance to pick up easy money (though only if you make a bet with layman, no scientist will take this action).

Having read the OPERA paper, I have no idea what they did wrong (unsurprising, I'm a theorist), but they did do something wrong; I'll bet my whatever on it. This claim is beyond extraordinary, and so requires similar levels of evidence before people like me are going to believe it. The OPERA people clearly did due diligence, but there are a lot of steps in their timing chain, and frankly, I'll believe they fucked up something basic over neutrinos going FTL. Some other experiment will probably have to weigh in, but I don't know many that have that capability at the moment (though similar configurations are planned for NOvA at Fermilab, it's not up and running yet)

The problem in my mind is the breathless reporting saying "Einstein was wrong" (see post title here, though really I'm just annoyed with the Times and similar newspapers who employ people who should know better). This gives the wrong impression - namely that this is a confirmed phenomenon - and plays down the uncertainty and doubt in the community. And when the inevitable retraction comes there won't be nearly the press coverage.
posted by physicsmatt at 5:19 AM on September 23, 2011

Previously.

The link from Previously.

The shorter version:

Particle Chameleon Caught in the act of Changing
Geneva 31 May 2010.

Researchers . . . announced the first direct observation of a tau particle in a muon neutrino beam. This is a significant result, providing the final missing piece of a puzzle that has been challenging science since the 1960s, and giving tantalizing hints of new physics to come.

Several experiments since have observed the disappearance of muon-neutrinos, confirming the oscillation hypothesis, but until now no observations of the appearance of a tau-neutrino in a pure muon-neutrino beam have been observed: this is the first time that the neutrino chameleon has been caught in the act of changing from muon-type to tau-type.

We are confident that this first event will be followed by others that will fully demonstrate the appearance of neutrino oscillation”.
[. . . ]
“This is an important step for neutrino physics,” said CERN Director General Rolf Heuer. . . We’re all looking forward to unveiling the new physics this result presages.”

[According to the Standard Model], neutrinos have no mass. [But f]or neutrinos to be able to oscillate, they must have mass. Something must be missing from the Standard Model.

One possibility is the existence of other, so-far unobserved types of neutrinos that could shed light on Dark Matter, which is believed to make up about a quarter of the Universe’s mass.

posted by Herodios at 5:30 AM on September 23, 2011

CERN neutrino tutorial. Actual experiment starts at 20 minutes, although I guess they probably changed it once they noticed the timing error.
posted by fraac at 5:31 AM on September 23, 2011

Actual experiment starts at 20 minutes, although I guess they probably changed it once they noticed the timing error.

Right: now starts at 19.999999999999999999999999999999999999999999999999
9999999999999999999999999999999999999999999999999999999999999999999
9999999999999999999999999999999999999999999999999999999999999999999
9999999999999999999999999999999999999999999999999999999999999999999
9999999999999999999999999999999999999999999999999999999999999999999
999999 minutes.
posted by Herodios at 5:35 AM on September 23, 2011

Oh you!

Mooted speed difference is big over light years. As someone elsewhere said, first thing is to check whether anyone detected neutrinos 4 years before the 1987 supernova.
posted by fraac at 5:40 AM on September 23, 2011

So I'm going with mistaken identity over both measurement error and FTL, FTW.

 
posted by Herodios at 5:40 AM on September 23, 2011

Herodios, the fact that neutrinos have mass was known from 2000 or so. That quote is somewhat disingenuous, in that the Standard Model doesn't have neutrino mass, but can be trivially modified to allow for neutrino mass (the Standard Model is usually abbreviated in papers as SM, pendants will sometimes distinguish the massive-neutrino version as nuSM. We are a witty bunch). Though mu-tau oscillations had not yet been observed, it was pretty much known that it must occur, so that result didn't shake anyone's core belief system. Though when neutrinos where discovered to have mass, one of my undergrad profs did say that he would have preferred that the Sun was going out (one of the alternative explanations for the observed electron-neutrino deficit) over massive neutrinos. So clearly some people felt strongly about it.

kalessin, yes, it's a 2 10^-5 *c increase, so not very much. However, anything faster than c is a problem, and if it was factors of 2, we would have seen it already (and it would be pretty unambiguous). The timing errors are summarized on page 20 of the arxiv paper I linked to above. There are a lot, but they add up to only 7.4 ns out of the 60 observed. So either most of the errors are correlated, one error is MUCH larger than reported (and they checked all of them, they claim), or there is an additional error that they were unaware of. The paper is very good, by the way, and more readable than most. On a side note, Fig. 7 is pretty awesome. In checking the distance between CERN and Grand Sasso, they actually measured the plate shift due to the L'Aquila earthquake. Crazy.

fraac, Super-K was finished in April 1983, so a little after 4 years before 1987A. However, even if they were up and running at the time that any FTL-burst of neutrinos came through, it's not clear they would have tagged them. First, every experiment has a shake-down time, and if they were still in that, a unusually large burst of Cherenkov light in a short time might be attributed to experimental growing pains (some background that hasn't been nailed down) rather than real signal. Second, without a supernova in the sky to tell you that something is going on, a signal in Super-K might not mean much. The actual c-traveling neutrinos from 1987A only left 18 events in Super-K (which is pretty much exactly what the predictions about supernovas tell us should have happened, so not a lot of room for many FTL neutrinos to have been produced). If an order of magnitude fewer neutrinos came in the FTL burst, that would look just like an upward fluctuation in background at a random time. Pretty hard to search for, and if you looked, you'd have a large statistical uncertainty due to a trials factor.
posted by physicsmatt at 5:51 AM on September 23, 2011

Physicsmatt: I'm actually kind of sad that the adustments for neutrino mass are trivial, as I always took a certain amount of satisfaction from the old definition of neutinos as "nothing, spinning". [For those keeping score at home, we used to think they 'had' inertia, but no mass].

Still, massive thanks for weighing in and spinning the news with strangeness and charm.
 
posted by Herodios at 6:09 AM on September 23, 2011

kalessin, oops, must have missed your previous comment. Sorry about that.
posted by physicsmatt at 6:10 AM on September 23, 2011

kalessin, yes, it's a 2 10^-5 *c increase, so not very much.

Right, but at the distance of stars, the difference still works out to years.
posted by empath at 6:26 AM on September 23, 2011

as usual, xkcd reads my mind.
posted by physicsmatt at 7:00 AM on September 23, 2011 [2 favorites]

This seems to be the original article.
posted by Obscure Reference at 7:55 AM on September 23, 2011

physicsmatt: "Here's the arxiv link to the OPERA result: http://arxiv.org/abs/1109.4897
Haven't read it yet, will drop back in later."

Wait, S. Aoki is a particle physicist? I thought he was just a shitty hipster DJ.
posted by symbioid at 8:21 AM on September 23, 2011

It's D. Aoki who has done all the really good research.
posted by mikelieman at 8:43 AM on September 23, 2011

Ongoing live webcast at CERN discussing the results.

Physicists looking very tired right now.
posted by schmod at 8:47 AM on September 23, 2011

Herodios, well, let me elaborate on the differences between the nuSM and SM. (after all, I still haven't figured out OPERA, and we wouldn't want to have less real physics on MetaFilter than reddit and slashdot. That would just be sad).

Fermions (spin-1/2 particles, which are what "matter" is made of: quarks, electrons, muons, taus, and neutrinos are fermions. Force carrying vector bosons are spin-1, the higgs needs to be a spin-0 boson), can be either left-handed or right-handed. This refers to whether their spin axis is anti-aligned or aligned with their direction of motion.

What if they're stationary, you cleverly ask? Well, to be stationary means to have mass (ignoring this OPERA result, massless particle MUST travel at c, massive particles must travel < c). This means that a massive particle doesn't have a universally defined "handedness," after all you can always boost past a particle moving slower than c and see that the handedness switched (to see this, turn on a fan, look at it from one side, then the other. In one view it's spinning clockwise, the other is counterclockwise). So to be massive, a fermion needs two sets of components: a right-handed set and a left-handed set. If you don't have both, you can't be massive.

Now, it turns out in the Standard Model, for quarks and charged leptons (the electron, muon and tau), the left- and right-handed set of fields have different quantum numbers (in particular, they have different hypercharge and SU(2)_L representation). So none of these fields can be massive... unless there were something like the Higgs to "eat" the excess charge and bind the left- and right-handed fields together into a single massive object. This is why we say the Higgs "gives mass." The two different quantum numbers get "broken," and the conserved quantum is a combination of the two: the electric charge we know and love.

Now, neutrinos are different. In the SM, there doesn't appear to be any "right-handed" neutrino for the Higgs to bind to the left-handed neutrino, so for a long time, the default assumption is that neutrinos were massless. However, we discovered in the early 2000's that the neutrino flavors (electron, mu and tau, just like the charged leptons) oscillated: a electron neutrino traveling some distance has a chance of appearing as a mu or tau (and vice versa). This is what OPERA was actually looking for. It turns out that oscillating neutrinos requires each neutrino to have different mass (neutrinos are produced as a specific flavor, a "flavor eigenstate" but propagate as "mass eigenstates." Which I know doesn't explain much, but the quantum mechanics is actually pretty simple). Really, this requires the mass SQUARED to each to be different, and by measuring the oscillation, we know the mass-squared differences of the 3 flavors very well, without knowing too much about the mass. Which is sort of strange, if you think about it.

So, we know that neutrinos have mass. So there are one of two trivial changes you can make in the SM to fix the "no right-handed neutrinos" problem. First, you just add a right-handed neutrino. Well, this is fine, but right-handed neutrinos are hard to find, since it turns out they are "sterile:" they have no SM charges at all - making them even harder to see compared to their left-handed counterparts. This type of neutrino mass mechanism is called a Dirac mass. Of course, then you have a problem of explaining why the Dirac masses of neutrinos are so tiny, compared to everything else (< 1 eV, compared to 511,000 eV for the next lightest fermion, the electron).

The other mechanism is called a "Majorana mass," and what it uses is that the neutrino is the sole fermion in the SM that CAN be it's own anti-particle (it has no electric charge). The anti-partner of a left-handed particle is right-handed (and vice-versa), and for neutrinos, the particle and antiparticle can be tied together.

The current best guess in the community is something called the Seesaw Mechanism. Here, you add right-handed neutrinos. Those will typically have VERY large majorana masses between themselves (they don't gain mass through the Higgs, so there is no reason for them to be near the other fermions). At low energies, the combination of big majorana masses for the r-h neutrinos and "average" dirac masses between left and right-handed neutrinos induces a very small majorana mass for l-h neutrinos (a seesaw: as the r-h gets heavy, the l-h gets light). Again, this sounds much more complicated than it is; if you know how to find eigenvalues for 2x2 matrices, this is actually pretty simple.

So to get a massive SM neutrino, you just add some extra heavy particles to the SM (the nuSM), and you're set. The problem is that these things are so heavy (naive estimate is 10^14 GeV) and so weakly interacting no one has a good idea of how to look for them and (dis)prove the idea.
posted by physicsmatt at 8:54 AM on September 23, 2011 [7 favorites]

schmod, physicists always look tired. Coffee machines are our only required piece of lab equipment.
posted by physicsmatt at 8:55 AM on September 23, 2011 [2 favorites]

Physicsmatt finally made a post that I don't understand a word of. I have a lot of reading to do again.
posted by empath at 9:01 AM on September 23, 2011 [3 favorites]

empath, I was going to recommend the PDG review for neutrino mass http://pdg.lbl.gov/2011/reviews/rpp2011-rev-neutrino-mixing.pdf but the key bit here is Eq. 13.9, and they don't tell you why that's the right things to do. If I was home, I could recommend one of the intro quantum mechanics textbooks instead, but I'm not. It's a common example though. Basically, you need to know that the Hamiltonian operator is the thing that tells you how a system propagates in time, then split the flavor eigenstates into the mass eigenstates and evolve it forward in time.
posted by physicsmatt at 9:09 AM on September 23, 2011

Sounds like something from VX Junkies. I don't think I'm a stupid man but the weirdness of this universe far outpaces my ability to understand it. Glad there are folks like physicsmatt who are still able to keep up! The bit about "Majorana mass" made me think of video games though, so it's not a total loss on my part.
posted by The Winsome Parker Lewis at 9:10 AM on September 23, 2011

Yeah, I think it's the helicity part that I don't understand.
posted by empath at 9:34 AM on September 23, 2011

Is anyone else having problems w/accessing the PDF on arXiv? I keep getting some document error. It's frustrating. It starts to load fine, then bam, crash...
posted by symbioid at 10:55 AM on September 23, 2011

symbloid, maybe try the abstract link first, then get the pdf from there (both are working for me at the moment):
http://arxiv.org/abs/1109.4897
posted by physicsmatt at 10:57 AM on September 23, 2011

:( Nope - that's exactly what I was doing. Must be something with the version of Reader I'm using. Oh well...
posted by symbioid at 11:02 AM on September 23, 2011

I have a question: if neutrinos have mass, how is it that they travel at C?

Or is it that they travel near C?
posted by JKevinKing at 11:08 AM on September 23, 2011

Continued: Or is it that some types of neutrinos have mass and others don't?
posted by JKevinKing at 11:10 AM on September 23, 2011

JKevinKing, they travel near c. neutrino masses are known to be less than around 1 eV (see my previous post, we know the mass^2 differences well, not the masses). The lightest neutrino could be massless, we're not sure.

For a neutrino with 1 MeV of energy (like from the Sun), this means that their actual speed is given by (I'll assume their mass is 1 eV for simplicity, smaller mass just increases the speed, and beta is v/c)
E = gamma m
gamma = sqrt(1/1-beta^2)
-->
beta = sqrt(1-(m/E)^2) which is approximately (for m<<E)
beta ~ 1-(m/E)^2/2 or, for the numbers I picked, beta ~ 1-0.5*10^-12

So the deviation from c should be one part in about 10^12. The observed deviation at OPERA is around a part in 10^5.
posted by physicsmatt at 11:18 AM on September 23, 2011

empath, can you be more specific about where you lost the thread? I can try again from there.
posted by physicsmatt at 11:19 AM on September 23, 2011

And I had a thought (pure speculation) regarding squaring the results of the experiment with the observations of the 1987 supernova, assuming arguendo that the results of the experiment were not due to experimental error and reflective of a real phenomenon. (I know this is highly unlikely.)

Perhaps the neutrinos can temporarily travel at superluminal velocities, but over time the average velocity can't be greater than the speed of light; i.e., they can oscillate above and below C at very short time time intervals such that we can measure it over very short distances, but over the time it takes to reach us over cosmoological time scales, they would not reach us faster than light. The oscillation time might be very short, but maybe since they're traveling so fast, time dilation due to general relativity allows us to notice the effect.

This is pure speculation of course, and it's based on the very unlikely notion that the experiment was correct.
posted by JKevinKing at 11:28 AM on September 23, 2011

Right handed and left handed spin.

I understand 1/2 spins for fermions, etc, but I don't understand why the physics of a left-handed particle and a right handed particle should be any different. Aren't they rotationally symmetric? It seems like violates relativity if your particle behaves differently depending on your reference frame.
posted by empath at 11:29 AM on September 23, 2011

Like, if a particle is moving towards you and it's rotating clockwise to it's direction of motion, and you boost yourself so it's now moving away from you, and it appears to be now moving counterclockwise to it's direction of motion. It's still the same particle, yeah?
posted by empath at 11:31 AM on September 23, 2011

Thanks, Physicsmatt! :)
posted by JKevinKing at 11:33 AM on September 23, 2011

Oh - empath, also, a great book I thought did a fairly solid job of explaining stuff is A Theory of Almost Everything... He talks about left and right-handedness and symmetry breaking (would this involve the mexican hat potential and higgs?)
posted by symbioid at 11:34 AM on September 23, 2011 [1 favorite]

I honestly think that snarky I'd-bet-my-whatever-they're-wrong comments in the media are disrespectful of the process.

I think that scientists as a whole often forget how they are portrayed by the small percentage of them who have become media personalities.

The people who present the public face of science are big on pretending that certain topics are 100% settled, no questions, not even the smallest trace of doubt. People who contradict them are cranks. This includes not only the no-possibility-of-exceeding-C issue but things like Evolution and Global Warming.

The fact that mass scientific consensus can be dead wrong is big news to the layperson, and the fact that scientists are clinging tightly to their beliefs despite contrary evidence presents a very different angle to the cultivated image of scientists as impartial interpreters of raw fact.

For geeks the big news is that a fundamental tenet of modern physics may be wrong. For lay people the big news is that a 100% proven scientific fact can be wrong, big time, and that scientists will circle the wagons to cling on to 'facts' that may have been disproven.

And here's the thing: it doesn't really matter if this turns out to be experimental error or not. For the public the new observations about the scientific community remain valid.

What does it mean? It means new life has been breathed into the causes of denying Evolution and Global Warming among other things. People who previously bought into the ". . . but it's science!" argument are now back in play.

Now you can argue (and I would) that science losing some of its mystique is a good thing; However, there's no question that this whole thing is a god send (and I'm sure there are people who believe that quite literally) to those who require fear, uncertainty and doubt to push their respective causes.
posted by Tell Me No Lies at 11:45 AM on September 23, 2011 [2 favorites]

Is anyone else having problems w/accessing the PDF on arXiv? I keep getting some document error. It's frustrating. It starts to load fine, then bam, crash...

The Adobe browser plug-ins blow chunks. I've found that disableing the embedded opening, saving the file, and opening it normally works well.
posted by mikelieman at 11:49 AM on September 23, 2011 [2 favorites]

thanks mikelieman, that did it!
posted by symbioid at 12:00 PM on September 23, 2011

The problem in my mind is the breathless reporting saying "Einstein was wrong" (see post title here, though really I'm just annoyed with the Times and similar newspapers who employ people who should know better). This gives the wrong impression - namely that this is a confirmed phenomenon - and plays down the uncertainty and doubt in the community. And when the inevitable retraction comes there won't be nearly the press coverage.

The Science News Cycle strikes again.
posted by dirigibleman at 12:08 PM on September 23, 2011

What does it mean? It means new life has been breathed into the causes of denying Evolution and Global Warming among other things. People who previously bought into the ". . . but it's science!" argument are now back in play.

I disagree. This doesn't have to be true. Especially if this provides an opportunity for people to see how science really works: to see that, unlike many religious orthodoxies, science is not so dogmatic that it refuses to accept changes even to its core assumptions in light of new evidence.

Your take on it just sounds like a rationalization for being reflexively dogmatic--to protect the reputation of science as a set of existing dogma, rather than to protect its integrity as a process for discovering and understanding reality.
posted by saulgoodman at 12:10 PM on September 23, 2011 [2 favorites]

At this point, has anyone yet been able to suggest a plausible alternative explanation for the results that seems to be credible, physicsmatt and others in the know?
posted by saulgoodman at 12:12 PM on September 23, 2011

lots going on at the moment. Sitting through an impromtu Q&A with an OPERA experimentalist at my conference. Back in a bit with any fun tidbits.
posted by physicsmatt at 12:23 PM on September 23, 2011 [2 favorites]

The problem in my mind is the breathless reporting saying "Einstein was wrong" (see post title here,

Err, that's an poor way of reading the title. The point of the title was that it can be read two ways - as seemed appropriate given that this finding might be (a) paradigm shattering or (b) an error; hence Einstein might weep because he's was wrong, but also - and this is the more immediate way of reading it, as a riff off of the "Jesus Wept" expression of exasperation, as Einstein might be *exasperated* after all the work he's done and the c constant being confirmed by countless experiments for decades - he'd weep at the ease all this is being claimed as "overturned" after one result not fully yet examined. Irony - like a fast subparticle zooming over some heads here.

And worrying about the quality of the science reporting of this in the pop press pales in significance - I'm deliriously happy, that it's being reported widely at all, in this age of anti-rationality and anti-science promulgated so relentlessly by certain sociopolitical forces.
posted by VikingSword at 12:26 PM on September 23, 2011 [1 favorite]

At this point, has anyone yet been able to suggest a plausible alternative explanation for the results that seems to be credible, physicsmatt and others in the know?

Systemic error.
posted by empath at 12:49 PM on September 23, 2011

Without a specific systemic failure mechanism to explain it, that non-explanation is about as practically useful and scientific as "God did it." I'll wait to hear from actual physicists on this one. (Boy, just keeping an open mind and holding multiple possibilities in tension without settling on one--is just not something we do very well anymore, it seems. That'd be the brain rot caused by all those sneaky-little sentient neutrinos invading our universe from the next one over, where c is attuned to be ever so slightly higher. These so-called neutrinos are Time and Space Invaders, people! Isn't it obvious? Lock up your clocks and calendars!)
posted by saulgoodman at 1:15 PM on September 23, 2011

It's not that hard - this seems to violate, so we wait for confirmation or none. If it's confirmed, then we seek a theoretical explanation. if not, then we can seek to explain it as a fluke or whatever... Or are you saying if we don't see a confirmation that we have to explain the wrongness of the results? Isn't that like trying to explain creationism?
posted by symbioid at 1:23 PM on September 23, 2011

Photons can spontaneously decay into a "virtual" electron/positron pair which then annihilate, reproducing the “original” photon. I don’t see why the pair could not interact with other photons which happen to be passing through (absorb then re-emit) in the interim. The absorb/re-emit process is one way light is slowed in a transparent medium.

Electrons could be slowed in a symmetric way by absorbing and re-emiting virtual photons. Could the measured value of c in a vacuum be subject to the foam of virtual particles out there? Then neutrinos, which interract far more weakly, could be travelling closer to the "true" value of c. Surely someone has accounted for this in the literature, or there is a simple argument that negates it.
posted by TreeRooster at 1:24 PM on September 23, 2011

saulgoodman: "Without a specific systemic failure mechanism to explain it, that non-explanation is about as practically useful and scientific as "God did it." I'll wait to hear from actual physicists on this one. "

No, systemic error is the most likely explanation. As edd alluded to above, it is incredibly more likely that this result is a fluke. Decades of experiments have established General Relativity and confirmed the c constant countless times. That means, using proper Bayesian reasoning, that this extraordinary claim needs extraordinary evidence to overturn the very-well-established current theory.
posted by -->NMN.80.418 at 1:35 PM on September 23, 2011 [1 favorite]

It's not actually likely that it's a fluke in the measurements, that's the 6 sigma certainty part of it. Either he's got something wrong with the distances or syncing the clocks or there is new physics here.

Could the measured value of c in a vacuum be subject to the foam of virtual particles out there? Then neutrinos, which interract far more weakly, could be travelling closer to the "true" value of c

c is actually a derived value from maxwell's equations.
posted by empath at 1:42 PM on September 23, 2011

No--I was just addressing specifically the "systemic failure" argument. It's plausible, and of course obvious.

But there has to be some evidence it actually happened in order to just discount the result. Science shouldn't just reject solid results it doesn't like because they might create more work for theorists, should it?

Most of the actual physicists I've read seem to be saying they can't really find any obvious errors in the way they measurements were carried out so far (though there may be subtle ones yet to come). But it's not particularly scientific just to say, in the absence of any evidence, well, there must have been some systemic error because our existing assumptions must be correct; we can just disregard these 16,000 measurements. Since it's going to be so hard to swallow the idea that c's been violated in the remote possibility that it is true, "confirmation" in this case could go on forever, unless an actual specific failure mechanism is found. And the effects of confirmation bias are going to have a very strong influence on our ability to think dispassionately and clearly.

The team measured the travel times of neutrino bunches some 16,000 times, and have reached a level of statistical significance that in scientific circles would count as a formal discovery.

I'm not betting against Einstein on any odds; just point out we don't have to (and really can't) conclude anything definite just yet.

How much more extraordinary do you get than actually measuring the c being violated? I don't think the standard of evidence should be supernaturally high, do you?
posted by saulgoodman at 1:43 PM on September 23, 2011

How much more extraordinary do you get than actually measuring the c being violated? I don't think the standard of evidence should be supernaturally high, do you?

I'd need to see a second experiment in a different location reproduce the results.
posted by empath at 1:48 PM on September 23, 2011

Most of the actual physicists I've read seem to be saying they can't really find any obvious errors in the way they measurements were carried out so far (though there may be subtle ones yet to come). But it's not particularly scientific just to say, in the absence of any evidence, well, there must have been some systemic error because our existing assumptions must be correct; we can just disregard these 16,000 measurements.

Most of the physicists I'm reading say its extremely likely that there is some systemic error, but they don't know what it is yet.
posted by empath at 1:49 PM on September 23, 2011

OK, talk over. No new surprises; the OPERA people are good at what they do, so there won't be smoking gun yet. Mood in the room is that the largest error might be the GPS transfer from CERN to Gran Sasso. Turns out the largest single known error is getting the distance down the tunnel; it's an operating highway and they can't easily close an entire lane of traffic to do the measurement right (that error is accounted for). My idle thought that the GR dilation could account for it is way too small, so I'm back to systematic errors.

Nice short comment by a senior theorist on what this would mean if correct. Short version: we'd need to break Lorenz invariance, and there are various ways to do that. No discussion of the causality problems of FTL, which are what bother me more, but it was good regardless.

VikingSword, my apologies for taking my grar out on your post title. I guess I'm preemptively annoyed at the life-time of crackpot emails I'll be receiving on this. Even if it's disproved tomorrow, people who don't understand modern science will be buttressing their arguments with it forever.

Tell Me No Lies, I had a long reply typed out, but I'm not going to go into this. I'm here to explain scientific results in my fields of expertise, and I've already strayed too much into commentary. Saulgoodman had a good reply, so I'll just clam up now.

saulgoodman, the "systematics" response is the right one at the moment. Yes, it's general handwaviness, but that's the automatic response to EVERY experimental results, and could be uncharitably interpreted as "I think they fucked up." The question is, of course, which systematic. No one knows yet, OPERA will have to let outside experimentalists crawl over their setup and figure out what went wrong where. If I had a better answer, I would be writing a note on it now for publication, but I'm not an experimentalist, and if I was, I don't have access to their equipment at the moment. Time will tell. But systematics isn't the end answer, it's the umbrella term for where we'll start looking (i.e, this probably isn't a theoretical issue, and they didn't fuck something up so obviously that any idiot could spot it).

And yes, I want a better answer than their 16K measurements. Systematic error means that they made the SAME mistake 16K times, so doing it over again doesn't help (for example, measuring my height 10,000 times doesn't help if your ruler is 0.1% long, and there have been million-dollar experiments that have made dumber mistakes). To be accepted, this experiment will have to be repeated by another, separate collaboration, preferably somewhere else (not using CERN, not in Gran Sasso), using a non-identical set up, analyzed by different people. If they both see the same effect, then physicists will take this seriously. That might take years (though Minos should be able to say something very soon), and until then, if theorists have a fun wacky idea using this, we'll write the paper, but it's not going to land anyone a job (or a prize) until the experiments weigh in. Before then, expert experimentalists will tear into every piece of the OPERA and nitpick them to death over it (this is cheaper than building a new experiment for every crazy result). This is the standard way that extraordinary claims need to be verified in particle physics.

TreeRooster. Good question; took me a minute to come up with why I don't think this will work (though I could be wrong). As you've correctly identified, c isn't the speed of light in a vacuum, it's the speed of a massless particle in the vacuum. So, you could be asking if the photon could have a mass induced by quantum loops. Experimentally, we know the photon is very close to massless (and certainly it's mass is less that of at least two out of 3 neutrinos), and formally there are some pretty strong theorems stating that the zero mass of the photon is protected from quantum corrections. Though I haven't looked to see which assumptions go into those theorems. Have to think a bit more about a loop induced index of refraction; I think those are related by a phase to the loops I'm thinking of, but I'm not 100% sure. If they are, they are also forced to be zero by various theorems about photons.

Still working on a reply for empath about helicity.
posted by physicsmatt at 1:57 PM on September 23, 2011 [1 favorite]

empath, Maxwell's equations give c in terms of the permeability of free space \mu_0. I guess I was wondering if the virtual quantum foam might determine this permeability!
posted by TreeRooster at 2:06 PM on September 23, 2011

I think some of us are hoping, beyond hope and with audacity, that these results are reflective of some new physics -- perhaps even leading us to guantum gravity, etc. This hope for me is similar to the hope that I win the lottery, or that others believe that there's a benevolent God; except that this hope for a new physics is even less likely, perhaps!

Please, physicists, don't take away our hope so harshly, you rascals, slapping our face with Reality!
posted by JKevinKing at 2:07 PM on September 23, 2011

But wouldn't we have measured that when we measured permeability?
posted by empath at 2:08 PM on September 23, 2011

(oops, sorry for all bold.)
posted by TreeRooster at 2:08 PM on September 23, 2011

There was at least one possible previous result in the same range mentioned up thread.

But my point is what's the value in rooting for a particular conclusion? Or in forming and stating a likely-outcome in advance? I don't think there's any. It just muddles things. Why is it so hard to leave it at, "let's wait and see"? I think the very apparent impulse to pre-judge the conclusion makes people suspect scientists are really just a different kind of religious cult. I know better, and you know better, but it seems understandable why non-scientifically inclined folks would read it that way. The obvious eagerness to rule out the results beforehand doesn't make the science look more secure from the outside, it makes it look less so. I think you'll get more cranks pestering you with religious nonsense in the future from insisting there must be a mistake somewhere before the work is done. Just my three and a half cents, there.
posted by saulgoodman at 2:10 PM on September 23, 2011

empath, to answer the 2nd post first:
The fact that you can boost past a particle means it's moving slower than light (this current madness not withstanding). So being able to boost means it's massive to begin with. Sounds like you're worried that you haven't done anything TO the particle during that boost, so how can it "change?" It's a mind-bender, I know, and I'm trying to come up with a better answer than a quantum handwave. I haven't thought about in this language in a while, so maybe this isn't accurate, but when you perform relativistic boosts in quantum mechanics, you're not guaranteed that the states that were present in one frame are the same as the states present in another.

Hawking radiation, for example, can be understood by boosting the vacuum state from the curved spacetime near the event horizon to the flat spacetime at infinity. The radiation is just "the vacuum" in one frame appearing as a "particle bath" in another. I'm sure that the left-handed creation operator in one frame (which says "you found a left-handed particle") becomes a right-handed creation operator after the boost. That doesn't tell you WHAT the properties of that other particle is though, just that the boost has taken one particle and turned it into something else.

Let me try to convince you it's true by going to the first question: why should the two helicities be different at all? This is just one of those things that you can't know ahead of time, you need to measure what the Universe decided to do.

It is possible that electrons could have left- or right-handed pieces that were the same. Such particles would be called "vectorial" as opposed to "chiral," and they wouldn't need the Higgs to have mass. So in a universe with vector masses for fermions, you (and the rest of us) would be much happier, because left and right are just different examples of the same thing. However, when we discovered the weak interaction, we discovered that ONLY one handedness (the left) interacted with the W boson (the Z boson interacts with both, but couples different to left- and right so I'll stick with the W). So the vector solution isn't right.

When a W decays, it must decay into left-handed particles (it can't talk to right-handed, after all). Decays conserve linear momentum, energy, charge, and angular momentum. So, since it can only decay into left-handed fields, the particular directions the decay products can fly out in are constrained, compared if it could decay into left and right. Looking at the decay products, we saw this chiral nature (I'm simplifying, I think historically, it was the Z coupling that showed this first, but let's stay with the simple story).

So the universe decided to be a bit of a bastard and go with the chiral theory (for the electro-weak interaction; strong force doesn't care about handedness). In fact, it's maximally chiral: right-handed fields have zero coupling. Though, in a way, this is far more beautiful and intricate than a vector matter sector. There are these quantum effects called "anomalies" that need to vanish for a gauge theory (a theory for quantum fields that carry forces) to be stable. These anomalies would automatically vanish for vector theories, but for our real world, that's not guaranteed. Instead, you have to calculate them out, and you discover that... they all vanish. But it's really hard to add new things to the theory and not ruin this anomaly cancellation, and since the anomalies care about all the forces (including gravity) it's beautiful in a weird, confusing way.

Hopefully, that gets at both why it's not a problem that left and right handedness are different, and also how we know. Does that help at all? If it does, I can go on a bit about how the Higgs makes it look like both helicities of an electron (for example) look the same to us. And yes, it involves mexican hats.
posted by physicsmatt at 2:13 PM on September 23, 2011

"what's the value in rooting for a particular conclusion?"

FTL satisfies the Rule of Cool.
posted by fraac at 2:13 PM on September 23, 2011 [1 favorite]

I'm not even hoping for the result to be confirmed; I've been a slavish Einstein fan-boy since I was a kid and don't ever relish seeing his ideas proved wrong (let alone the standard model up-ended). But I'd like to see a clear difference between how scientists behave and how young earth creationists do when their beliefs are challenged; it makes it easier to contrast the two worldviews and to make the case for science.
posted by saulgoodman at 2:15 PM on September 23, 2011

Thanks physicsmatt! In another conversation I was told that both photons and electrons can be slowed down by interactions but they are not going to be slowed down by the same amount. The same poster went on to say that the speed of electrons is measured as precisely c--which I have heard before--but am not certain is true.
posted by TreeRooster at 2:16 PM on September 23, 2011

I've been a slavish Einstein fan-boy since I was a kid and don't ever relish seeing his ideas proved wrong (let alone the standard model up-ended). But I'd like to see a clear difference between how scientists behave and how young earth creationists do when their beliefs are challenged; it makes it easier to contrast the two worldviews and to make the case for science.

Science is based on skepticism, not believing every result in every experiment.
posted by empath at 2:20 PM on September 23, 2011 [1 favorite]

Your take on it just sounds like a rationalization for being reflexively dogmatic--to protect the reputation of science as a set of existing dogma, rather than to protect its integrity as a process for discovering and understanding reality.

Well the point I was trying to make is that this steady stream of people coming out and saying "I'll bet my electron microscope that this is a bogus result" sounds exactly like people being reflexively dogmatic.

I was just trying to place the discussions happening in the lay press in a larger context and consider some of the implications in that context. I don't know that there's anything to be done about it, but I think it's worth understanding the place that science holds in general public discourse.
posted by Tell Me No Lies at 2:20 PM on September 23, 2011 [1 favorite]

I've been a slavish Einstein fan-boy since I was a kid and don't ever relish seeing his ideas proved wrong (let alone the standard model up-ended)

The standard model didn't exist when Einstein was alive, and FTL travel doesn't break relativity. It just causes weird results.
posted by empath at 2:21 PM on September 23, 2011

saulgoodman, the previous result is only 1+a bit sigma high. That's not evidence for anything, even with this new result. 3 sigma is the minimum for "interesting" and 5 for discovery. There are thousands 1-2 sigma results in particle physics; and that's statistically what we'd expect.

The reason I'm so convinced this is wrong is just that it flies in the face of several very accurate previous measurements (mainly the supernova one), and the fact that, as a theorist, I have particular set of prejudices- for lack of a better term- about what is and isn't true in the world I live in. General relativity is one of the strongest I have, for extremely good reasons. For lack of a better way of explaining it, it's just too pretty to be wrong.

Now, what does this mean? It means, that when you tell me your experiment has proven GR wrong, my initial, very likely-to-be-right response is "you fucked up, go back and try again." Is this the sign of the harsh physics thought-police? If you want to take a dim view, ok, I'm a physics fascist. I also will be right at the end of the day, and I will change my mind when you come back and say, "no, this is my answer, and this guy checked my work. And that girl built a new experiment and it agrees." Then I have to leave GR behind and find something new.

Another reason that we scientists are so skeptical is that crazy happens a lot. Since most results don't get this much attention, I don't think that the public gets a good idea about how often highly statistically significant results come down the pike (less from the big collider experiments, but they have more internal checks). Neutrinos alone have a host of totally contradictory results that make zero sense. They usually turn out to be wrong, but it's expensive to check them, and it takes a long time (they are being checked, slowly).

So we have to have a filter. We need some guiding principle that tells us "this new result is worth spending 3 years of your life on, and that one isn't." Guessing wrong means no citations, no tenure, no fame and glory (and we clearly are in it for the money, power, and women/men), wheareas being too cautious means missing out on the hot new topic. I recently have been getting into relatively heated arguments at conferences because I think some 3-4 sigma results are worth spending time and effort on, and other people disagree - so it's not that I'm particular conservative in this. However, this case, it doesn't pass my sniff test. But fortunately, there are a diversity of opinions, and it passed the test for someone, and now there's going to be a rush for someone else to show they're wrong. And the great thing about experiments designed to prove someone wrong is that they (very rarely) prove someone right.

So take that, and then wander in as an outsider and ask me what I think this is. I'll answer "it's wrong," because that's what I think and I'm not going to give a bullshit answer about that (though I am apparently willing to spend a few hours on this topic here, so there's that). Maybe that's the wrong thing to say, but I'm not convinced it is. I think it's wrong, I have and continue to explain why I think it's wrong, and we'll find out why they're wrong eventually.

Or I'll be wrong, they'll win a Nobel prize, and I'll get something crazy and unexpected to work on for the next few decades - but that doesn't make the reasoning I used to say they were wrong in the first place incorrect.
posted by physicsmatt at 2:32 PM on September 23, 2011 [70 favorites]

So, having read the paper and watched the webcast, I have basically the same opinion that I did when I knew only which collaboration was involved. This is a group of very clever and competent physicists who know exactly what they are doing — they just don't yet know exactly what they are doing wrong.

From the above comment about the scale of the effect, it's clear that the mainstream news coverage has not made clear that the claimed effect is small. The two laboratories where the neutrinos are emitted and detected are about 700 kilometers apart, and so the expected travel time for the neutrinos is about two milliseconds — that is, 2 000 000 nanoseconds. The claim is that the neutrinos arrive about 60 nanoseconds too early, only a few parts per million. This is still completely inconsistent with special relativity and causality, since s.r. implies that another observer, moving in the right direction beyond a certain speed, would see these neutrinos detected first and emitted later. But the effect is small.

Another blogger somewhere (possibly Motl) pointed out that 60 ns is also 60 feet, a relatively small error to make measuring the large distance between the two laboratories. This was discussed extensively during the presentation this morning, though. The group claims to know the distance from source to detector to within 0.2 meters (about eight inches) and I don't have any serious reason to doubt them. (I may be disagreeing with physicsmatt about the mood in the room after the webcast, or maybe he is talking about a different room.)

This leaves the question of the timing. I don't question the establishment of a pair of clocks, at CERN and at OPERA, which are corrected to keep the same time to within one nanosecond. But I wonder about the time that it takes for the signals from the data acquisition systems at the two facilities to reach to the two synchronized clocks (about 10 000 nanoseconds at CERN, and about 40 000 nanoseconds at OPERA). These times are presented with small error bars, but not described in detail — there must be another paper somewhere.
posted by fantabulous timewaster at 2:35 PM on September 23, 2011 [1 favorite]

fantabulous timewaster: I'm not at CERN, we just had a OPERA guy in the room and he got dragooned. Haven't been free to watch the webcast yet.

TreeRoster, did you mean neutrinos rather than electrons? Electrons have a mass of 511 keV, they all go slower than light. Neutrinos, as I've just said, also should be moving slower than c, but apparently the little bastards didn't get the memo.

And yes, any effect on the index of refraction will be particle-dependent. There's actually an "index of refraction" for neutrinos that is well known, it's called the MSW effect, and it causes certain flavors of neutrinos to be regenerated as they travel through the Earth (basically, the Earth has electrons but not muons and taus, so electron-neutrinos have a different potential energy in the Earth than the other flavors). I don't think it noticeably slows the neutrinos though, so maybe index of refraction was a bad analogy in this contextw

OK. posting way too much, and need to go be social at the conference. Hope I've been helpful, sincere apologies to anyone I offended at the end there, and I'll try to check in in a few hours to answer any new physics questions. Bye!
posted by physicsmatt at 2:40 PM on September 23, 2011

So does this mean we can travel backwards in time on a space ship made of neutrinos and finally kill Hitler?

Hitler Who?
posted by double block and bleed at 3:34 PM on September 23, 2011 [5 favorites]

Even if it's disproved tomorrow, people who don't understand modern science will be buttressing their arguments with it forever.

I've already seen it used to show that the "theory" of anthropogenic climate change IS NEXT, YOU BASTARDS!
posted by sneebler at 4:36 PM on September 23, 2011 [2 favorites]

Einstein's theory of generally relative global warming is totally bullshit, too, because the lamestream media don't want you to know that neutrinos aren't even American, duh!
posted by It's Raining Florence Henderson at 4:45 PM on September 23, 2011

I've already seen it used to show that the "theory" of [...]

... annnnnnd we're back to people claiming theories are actual truths again.
posted by Tell Me No Lies at 5:06 PM on September 23, 2011

I can only speak from the perspective of the molecular bio field, but I think a lot of people don't realize just how frequently published results turn out to be completely wrong. Sometimes even results from big name labs, published in big name journals. When doing lit review, it's distressingly common to see dozens of papers supporting a given conclusion and another dozen disproving it.

Our tools, methods, and scientists are imperfect. You'd quickly lose your mind if you took every result as conclusive. So skepticism is usually in order, especially when the result in question contradicts a well-established theory. And as far as well-established goes, it's hard to overstate just how far up the food-chain relativity is. It eats lesser theories like pizza-flavored Combos. It's the 1000lbs gorilla you'd be stupid to bet against.

It's not dogmatism that causes scientists to doubt results like these, it's just epistemological common sense.
posted by dephlogisticated at 5:32 PM on September 23, 2011 [8 favorites]

The way I see it: "I bet my microscope" translates as: "If this is true, I won't mind giving up my microscope while celebrating the new physics."

Personally, I'd like to see the neutrino race done over, only shooting through something less dense than the earth. Like, air, or outside the air even. I just have this thing when I visualize some things, and I get a freaky notion that something like a neutrino might get some kind of a boost when going through dense matter. Like a repulsion force that can only affect the target at relativistic speeds, in the direction in which it is already moving.

Then do-over again, through the earth. Even further than the current OPERA experiment.
posted by Goofyy at 3:09 AM on September 24, 2011

Ha! Tesla was there first!

You doubt? We have picture!
posted by warbaby at 7:23 AM on September 24, 2011

Ha! Tesla was there first!

Except that quote is two years younger then neutrino theory, which Pauli proposed in 1930, in 1932 the neutron was discovered, but it's not credited to Tesla.
posted by delmoi at 8:18 AM on September 24, 2011

Physicist Michael Brooks on CERN scientists' neutrino discovery: It's quite possible they took a shortcut through a wormhole.

How did CERN get their hands on a Stargate?
posted by homunculus at 4:22 PM on September 24, 2011 [2 favorites]

but that doesn't make the reasoning I used to say they were wrong in the first place incorrect.

I agree. Your reasoning is quite sound, and I actually do get why there needs to be extraordinary proof in a case like this before these results are fully accepted, but I also think there must be some way to figure out what's actually happening here, and if that ends up turning everything on its head, so be it.

As I said, I just wanted to emphasize that I think it really helps if people can see that scientists will take credible challenges to their orthodoxies seriously, and I really do think having the public see this process play out publicly will ultimately make people more--not less--confident that the scientific method works and that science really is more than just another kind of religious faith (I know, I know--it's ridiculous to even have to worry about that, but these are strange times).
posted by saulgoodman at 5:04 PM on September 24, 2011

(And that's also why I see it as a good thing there are other labs gearing up right now to take a crack at reproducing these unlikely results.)
posted by saulgoodman at 5:09 PM on September 24, 2011

I cater on the side, and a few months ago I worked a conference event for the Natural Philosophy Alliance. Their major defining feature is a hatred of "physics fascists" (like physicsmatt, apparently), and a general skepticism towards general relativity. They were, by and large, some of the most miserable people I've ever served food to. So even though I don't have an ideological dog in this fight, I'm pulling really hard for GR to still be totally valid.
posted by codacorolla at 8:25 AM on September 25, 2011 [5 favorites]

Except my father isn't 12.

He could be if we can travel faster than light.
posted by philip-random at 11:27 AM on September 25, 2011 [1 favorite]

"Hey, we need a photo for this CERN neutrinos story, what'll we do?"

"Relax, guys. I've got this."
posted by Rhaomi at 12:37 AM on September 26, 2011 [6 favorites]

OK, back from the conference, no longer sick, and thus significantly less annoyed. Wanted to drop a couple of links here, from Sean Carrolland Lisa Randall who are, in my opinion, two of the very best physics communicators out there today (Lisa is also one of the top 2 or 3 theoretical physicist of her generation as well). She has a very good discussion as to the implications of this result (and also a bit of a plug for her new book as well, which I haven't read. But her previous popular science book was good, so...)

Sean's post also leads me to a mea culpa; apparently FTL doesn't HAVE to lead to time travel, but only because neutrinos going FTL presumably means Lorentz invariance (the space-time symmetry that underpins general relativity) is broken. With that gone, there could be different laws for different types of objects.

It's not that time-travel and communicating with the past can't happen when Lorentz violation occurs, it just that we don't know what set of physical laws the Universe picked. In some, I'd guess that time travel was possible, but the point of the post is that in others, it doesn't have to be. I'd argued in previous discussions in metafilter that FTL always implied causal violations; it's nice to know that there are some ways around that, though to be honest, I'm still having trouble coming up with a non-GR worldview that is anywhere near as internally self-consistent. That said, no one else has either, and I still think this result is wrong, so it's not that I'm trying very hard.
posted by physicsmatt at 6:51 AM on September 26, 2011 [2 favorites]

One physicist blogger has noted the doubts as to the exact uncertainty about when exactly neutrinos are actually sent from the CERN source. This uncertainty could be the source of the error, if the result is incorrect.

Recall that OPERA was conceived to observe tau neutrino appearance, not to measure the neutrino speed, and indeed there are certain aspects of the experimental set-up that call for caution. The most worrying is the fact that OPERA has no way to know the precise production time of a neutrino it detects, as it could be produced anytime during a 10 microsecond long proton pulse that creates the neutrinos at CERN. To go around this problem they need a statistical approach. Namely, they measure the time delay of the neutrino arrival in Gran Sasso with respect to the start of the proton pulse at CERN. Then they fit the time distribution to the templates based on the measured shape of the proton pulse, assuming various hypotheses about the neutrino travel time. In this manner they find that the best fit is for the travel time is 60 nanoseconds smaller than what one would expect if the neutrinos traveled at the speed of light. However, one could easily imagine that the systematic errors of this procedure have been underestimated, for example, the shape of the rise and the fall-off of the proton pulse have been inaccurately measured.

http://resonaances.blogspot.com/
posted by Catfry at 8:21 AM on September 26, 2011

I sort of doubt that is the explanation, but I could be wrong. The pulse that creates the neutrinos is actually a beam that's traveling directly toward Gran Sasso at very close to the speed of light (though, since it's comparatively massive mesons that are in the beam, it's moving somewhat slower than the speed than neutrinos are supposed to travel). If anything this should ADD to the time that it takes to travel, not decrease. Though resonaances might also be saying that they don't know the time that the pulse is kicked out of the accelerator, which I would definitely agree is a possibility. I like the idea of redoing this with shorter bunch times.

The shape fitting is also a worry for me. When the OPERA guy was talking on Saturday, I did get a chance to ask if they fit times to both the rise and fall-times separately, and they don't. The neutrino pulse has a specific shape, and they match that template, but they don't allow it to be longer or shorter in time. I have no idea how adding that degree of freedom into the fit could account for this observation, but it would be helpful for me to see that they're fitting what they think they are.
posted by physicsmatt at 8:47 AM on September 26, 2011

Apparently, Fermilab's Minos experiment in 2007 also measured FTL neutrinos, and they're now redoing the experiment with more events and greater precision.
posted by empath at 8:58 AM on September 26, 2011

Could it also be phase/group velocity confusion?
posted by empath at 9:01 AM on September 26, 2011

empath; yes, but not statistically significant. It's same magnitude as the OPERA result, but since their timing had a 30 ns error bar and the distance was roughly the same, this is expected and shouldn't be read-into too much.

re: phase/group distinction, I don't think it matters here; usually the way you get "FTL" travel in the lab is to have a standing wave that you then sneakily arrange to change all at once in a way that looks like the signal propagated faster than light. The analogy is to imagine lining up a bunch of people, giving them all very accurate watches and then telling them to all raise their hands at 12:00 exactly. To an observer, the wave will "propagate" instantly, but all the communication beforehand means no information traveled FTL. Here, that doesn't work, since the neutrinos are being measured to physically move faster than c, so no trickery seems possible.
posted by physicsmatt at 9:40 AM on September 26, 2011

IANAP, but isn't phase/group velocity confusion mostly the venue of time-travel cranks? I'd hope a medium-sized group of well-trained physicists would understand the difference (admittedly I don't, but I know one can "travel" faster than light without actually imparting information/sending particles FTL).

Re: Breaking Lorentz invariance. This is over my head really, but I can understand different particles having different "speeds of light," so maybe neutrons have a slightly wider cone. I can see how they wouldn't see each other as travelling backwards in time, but what when they hit things with a normal speed of light? Why would we not see time travel then?

Also, it's neat to see someone smart posting on HuffPost. After reading a couple anti-vax/pro-homeopathy "articles" I more or less gave up on the site as a whole.
posted by BungaDunga at 9:42 AM on September 26, 2011

Motl has an interesting post about non-commutative field theories here.

Basically he suggests that gravitons might actually also be faster than light, and would set the limit. What's the record on gravity wave measurements? If it's true, and neutrinos are also faster than light, then it seems we should have a rather a precise prediction of how much sooner at minimum that a gravity wave would hit us than the light from the event would.
posted by empath at 9:58 AM on September 26, 2011

BungaBunga, mostly the confusion comes up in the popular press after some AMO guys specifically build an experiment to get "FTL light." It's sort of the physics equivalent of stage magic, and starts making you suspicious that they're just doing it to annoy theorists by indirectly encouraging cranks. My first thought on seeing the OPERA headlines (and this metafilter thread) was "oh, not this shit again." Surprises abound.

And yeah, I'm not necessarily buying what the Lorentz-violating paper is selling, so I can't mount a coherent defense of it here, but at least someone is thinking about it. I do have many of the same questions you just asked, so you're definitely not crazy.

empath, gotta run, so I can't do my homework on that, but I remember posting a gravity speed-measurement here a few physics-threads ago. It used quasar measurements, so maybe that's a place to start checking.

Catch you later, if the day allows.
posted by physicsmatt at 10:00 AM on September 26, 2011

Why would we not see time travel then?

You can always look at things moving backwards in time as things moving forward in time in the opposite direction.
posted by empath at 10:01 AM on September 26, 2011

I really hope time travel isn't possible--at least, in practical terms, because if it were, we'd destroy our own history a million times over to try to make a buck or to try to control events for various self-interested reasons, and the end result would be that time would just be another socioeconomic and political battlefield. As much as we humans fight over control of the present, I can't even imagine the horrendous uses we'd put time travel to.

Luckily, it seems extremely unlikely any new workable time travel technologies will somehow come out of this result, even if it isn't eventually found to be the result of systemic error.
posted by saulgoodman at 10:05 AM on September 26, 2011

As much as we humans fight over control of the present, I can't even imagine the horrendous uses we'd put time travel to.

The cabal have been time travelling for years, and it's never caused any problems so far caused nothing but endless grief so far it's never caused any problems so far caused nothing but endless grief so far it's never caused any problems so far caused nothing but endless grief so far it's never caused any problems so far caused nothing but endless grief so far it's never caused any problems so far caused nothing but endless grief so far...

*There is no time travelling cabal.
posted by It's Raining Florence Henderson at 10:31 AM on September 26, 2011 [4 favorites]

In modern physics, hand-waving is also hand-particle-ing.
posted by Galaxor Nebulon at 11:16 AM on September 26, 2011 [1 favorite]

that doesn't make the reasoning I used to say they were wrong in the first place incorrect

Yeah, far from being "biased," this reasoning is actually pretty principled. As other people have alluded to, it's really a type of Bayesian reasoning that I think is actually pretty common in science. Hear me out (some light algebra follows):

A p-value of 1e-6 (the "6 sigmas" part above) actually has a weird and kind of counterintuitive definition. It is saying that given that this result is random chance (the "null hypothesis", or H₀), the probability of seeing this result R is 1e-6. In mathier notation, this goes: P(R|H₀) = 1e-6)

But the quantity most people are really interested in isn't P(R|H₀), it's the reverse: given the result, how likely is it to be a fluke? That's something different: P(H₀|R).

And actually, there's a formula to get that quantity: if you apply Bayes' rule, you get P(H₀|R) = P(R|H₀)P(H₀)/P(R). So it's just the p-value scaled by P(H₀)/P(R). P(R) is sort of a cryptic thing to write (what's the probability of the result?), but if we expand it it makes a little more sense. P(R) = P(R|H₀)P(H₀) + P(R|H₁)P(H₁), where H₁ is the alternative hypothesis.

Unfortunately, that quantity is hard to measure in most real cases. But even without actually measuring anything, we might notice that if the alternative hypothesis P(H₁) is really unlikely to begin with (close to 0), and the null is really probable (P(H₀) is close to 1), then the quantity you get is:

• P(H₀|R) = P(R|H₀)P(H₀)/P(R)
• P(H₀|R) = P(R|H₀)P(H₀)/[P(R|H₀)P(H₀) + P(R|H₁)P(H₁)]
• P(H₀|R) = (1e-6)(~1)/[(1e-6)(~1) + (whatever)(0)]
• P(H₀|R) = (~1e-6)/[(~1e-6) + 0]
• P(H₀|R) ~= 1

So even though we had a six sigma event, if the alternative hypothesis is actually impossible then our measurement must have been wrong -- much like if your thermometer were to read 400 degrees Fahrenheit you probably wouldn't say "I must have been wrong about the weather this whole time." Of course, the alternative hypothesis isn't completely impossible, but in this case the null hypothesis follows from GR and so we have overwhelming evidence to support it, which means P(H₀) is going to be very, very close to 1.

tl;dr, prior beliefs matter in science, but this doesn't make it any less rigorous.

Further reading, focusing on this phenomenon as it applies to things like drug trials and genetic association studies.

(As a postscript I'll just say that one other problem with taking the 6 sigma thing at face value is that the measurements from a single study are often correlated -- same machines, same researchers, etc. If there is a consistent measurement error it's probably going to apply to every measurement, so it's going to look very, very significant.)
posted by en forme de poire at 11:53 AM on September 26, 2011 [2 favorites]

You can always look at things moving backwards in time as things moving forward in time in the opposite direction.

Right, but I thought that was just an interpretation of antimatter? ie if you flip the time axis around, the equations for matter will pop out a description of antimatter (hurray Feynman etc). But I don't think antimatter actually goes back in time.

If something's moving forward in time in the opposite direction, causality has to be reversed too (oh look, that bullet collided with gunpowder, decreasing entropy and compressing kinetic and thermal energy into chemical bonds...) for it to behave like it's going back in time, right? at which point it basically is just going back in time. </handwavey>
posted by BungaDunga at 1:14 PM on September 26, 2011

On reflection, I'm doubling down on a distance error.
posted by Twang at 4:28 PM on September 26, 2011

Twang, I think that's the least likely thing to be wrong, just because that accuracy of measurement is apparently standard in that field (but really, the whole thing is an open question at this point, so you certainly could be right, and I can find some physicists who would agree with you). I'd think the best odds are on the shape of the neutrino pulse being modeled wrong, and that's causing an error in the best-fit for the time. The next most likely explanation in my opinion is an error in the propagation time from the GPS receivers at CERN and Gran Sasso.

Alternatively, when they dug Gran Sasso, they accidentally punched a hole through the multiverse and now they're measuring the physics of the world of Arbe. This would also explain the DAMA/Libra experiment.

I just got back from dinner with some other physicists, and we got asked about OPERA by a nearby diner at the end of our meal. It's everywhere!
posted by physicsmatt at 6:43 PM on September 26, 2011

A night out asked about the OPERA, huh? Did it go a little something like this?

physicsmatt: Do you follow me?
Nearby Diner: Yes!
physicsmatt: Well, you better stop following me, or I'll have you arrested.
posted by It's Raining Florence Henderson at 6:49 PM on September 26, 2011 [4 favorites]

Well, he shook my hand until all the silverware fell out. Then there was a chase scene and a musical interlude.
posted by physicsmatt at 8:16 PM on September 26, 2011 [5 favorites]

I'd need to see a second experiment in a different location reproduce the results.

That's why I don't believe there's life on Earth. ;)
posted by obiwanwasabi at 3:01 AM on September 27, 2011

Ideas I have heard (with varying levels of seriousness) from physicists here:
1) build a giant lens from rock to see if we can defocus neutrino beams (as rock apparently could have a neutrino-index of refraction < 1)
2) Look for said index of refraction by looking for total external refraction of neutrinos off of stone slabs (that was from the same guy as 1, both semi-serious)
3) build a neutrino observatory underground on the moon and point a particle accelerator at it
4) build a neutrino observatory/particle accelerator on the floors of major stock exchanges and make a killing in arbitrage (this one is mine). The effects of 100-kilowatt particle beams on day traders is left as an exercise for the reader.

Anyway, in seriousness the MINOS people claim that they can address this in a few months. I'm not sure if they need to collect more data or just need to reanalyze their old stuff. My guess is the former, combined with some additional efforts to get timing errors down.
posted by physicsmatt at 11:30 AM on September 27, 2011

I will personally subsidize option 4.
posted by It's Raining Florence Henderson at 11:48 AM on September 27, 2011 [1 favorite]

I thought it'd be a winner. Frankly I don't know why the possibilities of FTL communication and time-travel in abusing the hell out of the market haven't been investigated more carefully.

Finally, I have a way to go to Wall Street AND remain in physics.
posted by physicsmatt at 12:05 PM on September 27, 2011 [1 favorite]

Well, he shook my hand until all the silverware fell out. Then there was a chase scene and a musical interlude.

Lemme guess. Yakety Sax, right?
posted by Mental Wimp at 2:48 PM on September 27, 2011

...we might notice that if the alternative hypothesis P(H1) is really unlikely to begin with (close to 0)...

What's the probability that P(H1) is less than δ, for δ positive and close to zero? 1 − ε? ;>

The calculation only works if you're willing to put a value on things like P(H_i), where i has a pretty big range. Because you really need P(~H₀), and there are a lot of H_i that ain't H₀. You have to be willing to integrate over all of them to get P(~H₀), after you describe their underlying pdf and, of course, figuring out P(R|H_i) for each one of them. I'm not sure it's safe to say that the value of ΣP(R|H_i)P(H_i) is close to zero.
posted by Mental Wimp at 3:24 PM on September 27, 2011

Mental Wimp: That's why usually a Bayesian will run off two theories and produce a Bayes factor. In this case though, while there's a large number of possible explanations if neutrinos are travelling faster than light, none have substantial competitiveness and the total is still close to 1 that the result is an error.

(Incidentally, you don't ever let it reach 1 unless it's a logical certainty - this is Cromwell's rule)
posted by edd at 4:00 PM on September 27, 2011 [1 favorite]

MW: even in the absolute worst case where you assume P(R|~H0)=1, though, this only ends up mattering if P(~H0) is greater than the p-value (you basically end up with 1e-6 / [1e-6 + P(~H0)]). Given that measurements with correlated errors tend to produce more extreme results/p-values (and it's likely any errors would be correlated here, since the observations are from a single study with the same equipment, design, and scientists), and given how much empirical and theoretical evidence for H0 there is (P(~H0) after all being 1 - P(H0)), I'm still comfortable saying my posterior belief in GR is strong.

Anyway, my point is more that, presented with a strong, well-tested theory and a small set of apparently contradictory data, it's not only understandable but also a scientifically principled and helpful stance to be skeptical of the data instead of the theory.
posted by en forme de poire at 11:13 PM on September 27, 2011

(P.S., don't mean to suggest that you disagree with that last part, of course!)
posted by en forme de poire at 11:19 PM on September 27, 2011

So long, and thanks for all the quarks: As an announcement is made in Europe that may shake physics to its foundations, America’s largest particle accelerator is to be switched off
posted by homunculus at 5:37 PM on September 29, 2011

Yes, tomorrow will be a sad day at work. On the other hand, there will be cake.

Ther are now about 20 papers on this on arXiv, mostly theorists doing ridiculous things to Lorentz symmetry (the symmetry of space-time that underlies Einstein's theory of relativity) just to prove they can, several on how badly all proposed theories do in fitting all the data, one on how difficult it is to synchronize clocks over a non-trivial distance in a varying gravitational potential (this one is interesting), and one Nobel laureate on how this is crap. Let it never be said we didn't take this seriously.
posted by physicsmatt at 9:35 PM on September 29, 2011

I'm going to criticize the intersubjective rhetoric being employed in discussions such as these. Either an explanation or result is right or wrong, and either the verification task is worth the trouble or not. But to say that "I think X is wrong" means too many things to too many people: taken literally, the statement becomes a conflict between belief and alief; taken metaphorically, others would have to "get" your metaphor.

Clarity in language is a powerful way to defuse irrelevant lines of discussion. It's more precise and less subject to misinterpretation (e.g. especially by those who do not share your frame of mind or worldview) to say, regarding any thing, that a) I don't know, but am extremely skeptical, for the following reasons, and b) nobody else can know at this time, for the following reasons.

And this I have known from experience. /endrant
posted by polymodus at 10:47 AM on October 5, 2011

Seems that the scientists didn't take into account the relativistic motion of the GPS satellites used for timing. "If it stands up, this episode will be laden with irony. Far from breaking Einstein's theory of relatively, the faster-than-light measurement will turn out to be another confirmation of it. "
posted by exogenous at 12:56 PM on October 14, 2011 [2 favorites]

I find it pretty hard to believe that they failed to consider something as basic as that.

Still, sillier things have been missed.
posted by Tell Me No Lies at 2:04 PM on October 14, 2011

Hmmm, interesting. The paper making that claim raises a couple of my red flags for arXiv crackpottery, notably with the hundred-year-old references.

The claim is that the velocity of a GPS satellite in a high orbit comes in linearly, like v/c, which is a few parts per million, the size of OPERA's speed-of-light deviation. Usually clock speeds in relativity come in quadratically, like (v/c)^2, which is a few parts per trillion.

Pretty sure the CERN/Opera group used local atomic clocks at each end of the experiment. The synchronization between the two clocks was measured three times over two years and found to drift by about 2 ns. Pretty hard to hide a clock in there with a frequency that's off by 10 parts per million --- that clock would lose 10 000 ns every second.
posted by fantabulous timewaster at 10:20 PM on October 15, 2011

Guys, thanks for bringing this up, as I wouldn't have bothered to have read this article carefully otherwise, and there's some interesting relativity in it.

fantabulous timewaster, you're correct that Lorentz contractions usually have v show up in terms of (v/c)^2 (though it's 1/sqrt(1-v^2/c^2), so the actual dependence is 1+v/c for small v). However, it isn't the Lorentz contraction that's at issue here (which is what I thought he was doing the first time around). That is to say, the author ISN'T arguing that the distance measured by the satellite between CERN and Gran Sasso is the equivalent of 32 ns shorter (so ~10 ft less) than the distance measured by a ground observer. Instead, he's saying that, every time a neutrino is sent towards Gran Sasso, it has slightly less distance to cover because, according to the satellite, Gran Sasso is moving towards the neutrino.

The reason this gives a different answer from classical mechanics is because velocities don't add in relativity the same way that velocities add at speeds much lower than light (or, they do add that way down here in the slow zone, we just can't notice). Both the ground observer and the satellite will see a photon moving at c (and a neutrino moving very very close to c), but the satellite will also see the target moving towards the neutrino (if relativity wasn't correct, the satellite would see the neutrino moving slower towards the target).

Now, I'm not sure this is the solution, because we were told by OPERA that they have taken all the relativistic effects into account, but they weren't too clear in which effects they meant. So it's possible that this was already accounted for (it's also not clear to me whether this is a double counting of effects I know they already took into account, but that's on me). OPERA has been very open though, so I'm sure that'll be addressed soon. As Tell Me No Lies says, this is a silly thing to have missed by some measures, but we do sometimes do that.

Also, one of the reasons I like this is that it gives a way for GPS to be correct for an individual but still screw up in OPERA. One of the arguments that it couldn't be GPS to explain the OPERA anomaly is that, if it was, your GPS would be out alignment. However, one of the things relativity teaches us is that it's really hard to coordinate two clocks that are in different frames. So GPS could tell you where you are just fine, but still have trouble when it tries to tell two people where they are relative to each other. That isn't an endorsement that this is the correct explanation, just that it passes an important test for me.
posted by physicsmatt at 8:26 AM on October 16, 2011

I'm not sure what GPS has to do with anything after figuring out where the two sites are the first time?
posted by empath at 8:32 AM on October 16, 2011

They're also keeping the clocks synchronized at both ends, so whether this effect matters at all seems to me to depend on how exactly that is being performed.
posted by physicsmatt at 8:48 AM on October 16, 2011

All right! I knew I could count on this thread to still be here, thinking this stuff through!

We should know, at some point soon, whether this particular effect was properly accounted for in the original results, ja? So, making too much of the news either way might be premature until at least the minimal confirmation steps have been taken (i.e., asking the team/reviewing their results to see if this effect was accounted for or not).

All the same, it would be great if this explanation held up--especially given the nice, punchy "relativity defends itself" news marketing angle. So it's going to be real tempting to think it is.

Luckily, it sounds like the idea can be confirmed/refuted fairly easily: either they did or didn't account for this effect, since the existence of the effect itself seems non-controversial (I'm assuming it is--is that a correct assumption, physicsmatt?).
posted by saulgoodman at 8:43 AM on October 17, 2011

saulgoodman, I'm hoping that by today or tomorrow someone who knows the timing technique a bit better will weigh in on arXiv (there was another proposed explanation that relied on a particular mechanism for transferring the timing data between CERN and Gran Sasso that was responded to pretty definitively a few days later with a "that's not how we did that"). The more I think about it, the less likely this explanation seems; it's not that the math is wrong, I just don't think that's how OPERA measures the time of flight, but again, I don't know for sure.

I think the media angle of "relativity defends itself" is behind the legs on this particular story; while this COULD be the right answer, it's no more likely than several others that have been seriously discussed (and possibly quite less likely than one or two others).
posted by physicsmatt at 8:59 AM on October 17, 2011 [1 favorite]

I dimly remember that the originally published modern measurement of c was off by a small amount, and that subsequent experimental results in publications clustered around this erroneous value, until someone published a corrected result, and then other experimental estimates clustered around that one.

It was in an American Scientist article, IIRC. The point of the article was that in corroborating research, it is important to not only question unexpected results, but to also question the expected result. Too often experimentalists quit looking hard at their experimental methods once they reach an expected result. The advice of the author was that one should always challenge the result by tearing down the experiment and building it up again from scratch, as one does instinctively when getting an unexpected result.
posted by Mental Wimp at 1:41 PM on October 17, 2011

(though it's 1/sqrt(1-v^2/c^2), so the actual dependence is 1+v/c for small v)

physicsmatt, I agree with everything else you wrote, but you'll be deprived of the impossibly useful binomial expansion as long as you make this error. The approximation is (1-x)^n = 1-nx; there's never any √x.
posted by fantabulous timewaster at 9:43 AM on October 19, 2011

Here's a recent reply to the article from the Netherlands proposing the relative frames solution. Via John Baez, who has made a pretty hefty bet on Einstein holding up.
posted by TreeRooster at 9:32 AM on October 20, 2011 [1 favorite]

This paper appeared in the comments to TreeRooster's link. It seems to claim that it has contradicted the conclusion that OPERA found FTL neutrinos. But then, it seems to depend upon other theory being correct, but IANAHEPP (I am not a high-energy particle phsysicist).

Thus we refute the superluminal interpretation of the OPERA result. Furthermore, we appeal to Super-Kamiokande and IceCube data to establish strong new limits on the superluminal propagation of high-energy neutrinos.

Well, sure, if you use Super Koolaid with Ice Cubes, you can get any result you want. Sheesh.
posted by Mental Wimp at 11:40 AM on October 20, 2011