Faster than lightspeed achieved?

[michel123456]

How could one referee make a judgment on the work of over 150 scientists?

Edited October 20, 2011 by michel123456

[imatfaal]

How could one referee make a judgment on the work of over 150 scientists?

 

That was my feeling exactly - one/two/three referees would be unlikely to find an error in a few weeks and the journal would then publish and give a veneer of proof to the claim. By avoiding that route, i think, the OPERA collaboration came across as very honest and not publicity hunting

[pantheory]

How could one referee make a judgment on the work of over 150 scientists?

In this case they probably would have had several referees with one making final decisions. Referees represent the publication. The publication provides the referee with their written guidelines along with additional unpublished guidelines for conformance to publication preferences. Publication guidelines are based upon ownership and readers' preferences. Sometimes these guidelines are unspecified because they may favor particular theories, hypothesis, author-ships or institutions, etc. concerning their publication title and field of expertise. The referee can also appeal to the publisher for an exceptional case or inclusion.

//

Edited October 20, 2011 by pantheory

[md65536]

To clarify a vague idea that I'd posted earlier in this thread:

 

Doesn't the OPERA result (if verified) only show that the "group velocity" of the neutrino density exceeds the speed of light? Only the amplitude of the probability of detecting a neutrino has been measured and/or interpreted to be exceeding c. In this case, the actual neutrinos would be traveling at slightly less than c, but the probability of detecting them would travel faster. Essentially this would mean that a signal of sparse neutrinos would become easier to detect just before a dense group of neutrinos arrives.

 

If this is so, then it doesn't necessarily violate relativity. It is already known that a group velocity can exceed c without violating relativity. http://en.wikipedia..../Group_velocity

 

Also, it would still be impossible to use this to send information faster than c (because you can't determine the changing probability of detecting a neutrino based on a single detection of a neutrino. For 2 or more positive detections, I don't know how the probability would be calculated, but I'll note again that the determination of v > c is based on a best fit of a graph consisting of a lot of neutrino detections over relatively long times, and certainly as the timescale and number of detections decreases the certainty of a change in probability amplitude also decreases. My contention would require that it is theoretically impossible with the OPERA setup to detect a change in the probability amplitude within 60 ns after a measurement (detection of a neutrino)... so that a "dense group of neutrinos" would still arrive before you could detect any change in the probability that it is coming. I have no idea how this would be determined).

 

Still, this would be considered revolutionary maybe?, because it would demonstrate something like the wave function of some matter being affected by other remote matter. But I don't think it would invalidate any accepted theories that I know of.

 

 

 

This is all speculation and I only know slightly more about what I'm talking about than I did before, which still isn't a lot, but I'm still betting that this is the cause of the OPERA results (admittedly mostly because "I want to believe").

Edited October 22, 2011 by md65536

[md65536]

Doesn't the OPERA result (if verified) only show that the "group velocity" of the neutrino density exceeds the speed of light?

The oft-quoted-by-me Starts With A Bang blog has a write-up explaining the possibility of something similar to what I was describing, but with a lot simpler circumstances:

http://scienceblogs.com/startswithabang/2011/10/a_test_for_neutrinos_put_up_or.php

 

Of the 4 possibilities (listed in the link) of explaining the results, the one I was describing is this:

3. There's a bias in the detection of their neutrinos, and the pulse shape of the arriving neutrinos doesn't match the pulse shape of the things that created them.

 

I imagined the reason for this happening must be something weird like quantum entanglement, but Ethan explains how this can happen simply by having the pulse shape of the neutrino source a lot bigger than the pulse shape of the receiver (which it is because most neutrinos are not detected). In this case, the smaller detected pulse shape can be offset in time (either toward the past or future), yet it may still fit completely within the larger transmitted pulse shape. (Or something... I didn't actually read it yet! )

[imatfaal]

MD - I like Ethan's blog and his explanations of his four scenarios are easy to grasp; but the received neutrinos do 'look' to the naked eye to be fitting to a pulse shape (this could be a bit of academic paper legerdemain but I really hope not).

 

I think a (hidden to experimenter) selection of particles from the front section of a bell shaped distribution could also have a bell-shaped distribution - it's a curve that appears in nature all the time; but the pulse shape in CERN is irregular and jaggedy - and the built up set of readings at OPERA seems to match that irregular jaggedy shape. I cannot see how (without the pulse shape being a simple shape that occurs naturally) the neutrinos in the first section of the pulse are accurately mimicking the shape of the whole pulse. And both pulses seem to match the build up neutrino readings - so a lucky coincidence is less likely

 

Secondly I believe the length of the pulse generated at CERN was accurately repeated at OPERA - if the speed up is caused by a only a leading edge being detected then the pulse length should be shortened (changed) and I don't think it was.

 

my money is still on a systematic error in the details of the experiment - there is a miscalculation or delay on when they believe the neutrinos start their trip ie they start 60ns before expected (or finish it later). But the guysngals at OPERA know the set up better than anyone and they havent been able to spot it - and I doubt anyone else will be able to.

[md65536]

MD - I like Ethan's blog and his explanations of his four scenarios are easy to grasp; but the received neutrinos do 'look' to the naked eye to be fitting to a pulse shape (this could be a bit of academic paper legerdemain but I really hope not).

 

I think a (hidden to experimenter) selection of particles from the front section of a bell shaped distribution could also have a bell-shaped distribution - it's a curve that appears in nature all the time; but the pulse shape in CERN is irregular and jaggedy - and the built up set of readings at OPERA seems to match that irregular jaggedy shape. I cannot see how (without the pulse shape being a simple shape that occurs naturally) the neutrinos in the first section of the pulse are accurately mimicking the shape of the whole pulse. And both pulses seem to match the build up neutrino readings - so a lucky coincidence is less likely

 

Secondly I believe the length of the pulse generated at CERN was accurately repeated at OPERA - if the speed up is caused by a only a leading edge being detected then the pulse length should be shortened (changed) and I don't think it was.

 

my money is still on a systematic error in the details of the experiment - there is a miscalculation or delay on when they believe the neutrinos start their trip ie they start 60ns before expected (or finish it later). But the guysngals at OPERA know the set up better than anyone and they havent been able to spot it - and I doubt anyone else will be able to.

Disclaimer: This discussion is over my head.

 

Is it possible that the irregular shape can be described with (essentially constructive) interference of several simpler (bell-shaped) curves, each showing the same apparent temporal offset?

 

To violate causality, it would have to be possible to determine the shape of a pulse of incoming neutrinos before the neutrinos defining the shape of the pulse arrived. I think that if an irregularly shaped pulse (consisting of many regular "sub-pulses") arrived, where each of the sub-pulses was detected near the leading edge of the sub-pulse, but still completely within the larger (taller, but not wider) source sub-pulse, then it might be impossible to detect the shape of the irregular pulse (ie. to detect the length and intensity of any sub-pulse section of the whole pulse). I suppose the question is, if you made the source pulse more distinct, say a square wave with short durations, what would the detected pulse look like? I suspect that it would look smoothed for some reason, still like bell-curves. But I also suspect that somehow those curves would still fit within the curve of the source pulse, at least to the degree that any information could be extracted before the actual information-carrying neutrinos arrived.

 

 

 

 

Anyway, the blog post mentions that OPERA is planning an experiment (with results "in the next few weeks") that could rule out this and another explanation. That's #3 and #2 respectively, of this list:

 

1. There was a systematic error in their measurements, and their measurements are simply systematically off by 60 nanoseconds (or thereabouts).
2. The errors are much larger than they claim, and they're not actually measuring the arrival time of these neutrinos to their claimed accuracy.
3. There's a bias in the detection of their neutrinos, and the pulse shape of the arriving neutrinos doesn't match the pulse shape of the things that created them. Or...
4. They really did break the speed of light, and the laws of physics don't work the way we think they do, and in your face, Einstein!

The list seems ordered by likelyhood, and it would be rational of me to also suspect #1 or #2 over #3, but I guess I just want it to be #3.

Edited October 27, 2011 by md65536

[morgsboi]

Does this experiment of moving particles faster than the speed of light disprove Einstein's theory of Energy=Mass*Speed of light^2? Because the laws of physics say that nothing can exeed the speed of light and if it did then it would travel to the future because the laws of physics won't allow it and therefor will make it go slower but traveling to the future.

This could disprove a lot of theories.

http://www.guardian.co.uk/science/2011/sep/22/faster-than-light-particles-neutrinos

So did these particles go through time?

[swansont]

Does this experiment of moving particles faster than the speed of light disprove Einstein's theory of Energy=Mass*Speed of light^2? Because the laws of physics say that nothing can exeed the speed of light and if it did then it would travel to the future because the laws of physics won't allow it and therefor will make it go slower but traveling to the future.

This could disprove a lot of theories.

http://www.guardian.co.uk/science/2011/sep/22/faster-than-light-particles-neutrinos

So did these particles go through time?

 

No. There is so much confirmation of relativity that, at best, there would be a tweak to the theory rather than discarding it. The things that worked up until this point still work the same way.

 

However, the experiment is not confirmed, so this is all very premature.

[alpha2cen]

The experiment would be done very accurately, but the measured time range is too short.

How accurately can we measure the starting time? Are there any time lag to detection from the neutrino starting point to the time recorder? To detect the signal, enough neutrinos or something must hit the detector.

I hope the neutrinos really did break the speed of light. But...

[imatfaal]

No. There is so much confirmation of relativity that, at best, there would be a tweak to the theory rather than discarding it. The things that worked up until this point still work the same way.

 

However, the experiment is not confirmed, so this is all very premature.

 

If and when the result is confirmed - when other experimenters have replicated the results; will this not be seen as a "perihelion of mercury" sort of moment?

 

Newtonian gravity was fantastically accurate but we learnt of flaws - Einstein's relativity is even more accurate and perhaps we are now learning of the first flaws.... To continue your phrasing - Newtonian gravity still worked up to that point, still works within limits, and is still used for the vast majority of problems; Einstein's theories were not just a "tweak" so why would this new experimental data only require a tweak. I worry that any tweak will turn out to be more like Vulcan or Epicycles; and that what will be needed is closer to the understanding that Newtonian gravity is a local limited case of a more complex theory.

 

And I do realise that this is far from confirmed and the clever money is on systematic error - but I do have a niggle with the "only a tweak" response that you share with a great many physicists (although at the back of my mind I worry that its because you and they know more than I do...)

[alpha2cen]

How long is the time-lag at the CERN to detect the signal?

 

time line ----------------|||||-----------------------------|||||--------------------------------------------------------------|||||-------------------||||--------------------->>>

 

event ------------neutrino leaving---------neutrino leaving recorded-----------------------------------------neutrino arriving-------neutrino arriving recorded

 

Is the time-lag very short?

Edited October 28, 2011 by alpha2cen

[swansont]

If and when the result is confirmed - when other experimenters have replicated the results; will this not be seen as a "perihelion of mercury" sort of moment?

 

Newtonian gravity was fantastically accurate but we learnt of flaws - Einstein's relativity is even more accurate and perhaps we are now learning of the first flaws.... To continue your phrasing - Newtonian gravity still worked up to that point, still works within limits, and is still used for the vast majority of problems; Einstein's theories were not just a "tweak" so why would this new experimental data only require a tweak. I worry that any tweak will turn out to be more like Vulcan or Epicycles; and that what will be needed is closer to the understanding that Newtonian gravity is a local limited case of a more complex theory.

 

And I do realise that this is far from confirmed and the clever money is on systematic error - but I do have a niggle with the "only a tweak" response that you share with a great many physicists (although at the back of my mind I worry that its because you and they know more than I do...)

 

GR reduces to Newtonian gravity, and KE = 1/2mv^2 is still accurate out to many decimal places for object moving at v << c. We did not throw those (and related) equations away with the introduction of relativity. Similarly, we will not throw out E=mc^2, which is what morgsboi was asking. Anything that would modify the theory will still reduce to these equations under a wade spectrum of conditions. That's what I mean by tweak — relativity does not simply get tossed, the way that the luminiferous aether, caloric theory and phlogiston did.

 

There's a possibility that there would be a modification to the theory that is an exception for the weak interaction or specifically for neutrinos. I don't know enough about particle physics to know how that would work.

[Cap'n Refsmmat]

In related news, the experiment is being re-run with a different setup to gather better data:

 

http://www.bbc.co.uk/news/science-environment-15471118

[imatfaal]

GR reduces to Newtonian gravity, and KE = 1/2mv^2 is still accurate out to many decimal places for object moving at v << c. We did not throw those (and related) equations away with the introduction of relativity. Similarly, we will not throw out E=mc^2, which is what morgsboi was asking. Anything that would modify the theory will still reduce to these equations under a wade spectrum of conditions. That's what I mean by tweak — relativity does not simply get tossed, the way that the luminiferous aether, caloric theory and phlogiston did.

 

There's a possibility that there would be a modification to the theory that is an exception for the weak interaction or specifically for neutrinos. I don't know enough about particle physics to know how that would work.

I think the difference is more on the meaning of tweak than anything else. I view the aether, phlogiston, Vulcan etc as tweaks (addons to make failing theories follow reality) but it is clear that that is not what you are saying - and I think I am in agreement with your view once the difference in expression is cleared up.

[swansont]

In related news, the experiment is being re-run with a different setup to gather better data:

 

http://www.bbc.co.uk/news/science-environment-15471118

The MINOS people are going to do it as well. I just attended a colloquium on the MINOS findings and neutrino oscillations where the "FTL" experiment was also discussed; USNO is going to make sure the timing is done properly.

[imatfaal]

The MINOS people are going to do it as well. I just attended a colloquium on the MINOS findings and neutrino oscillations where the "FTL" experiment was also discussed; USNO is going to make sure the timing is done properly.

 

so cool!

 

I do hope you aren't embargoed from sharing a few bits of information throughout the process.

[md65536]

The experiment would be done very accurately, but the measured time range is too short.

How accurately can we measure the starting time? Are there any time lag to detection from the neutrino starting point to the time recorder?

Unless the scientists have missed something or made an error, the measurements are precise to within 10 ns. A list of the various uncertainties is shown here: http://scienceblogs.com/startswithabang/2011/09/are_we_fooling_ourselves_with.php

 

If and when the result is confirmed - when other experimenters have replicated the results; will this not be seen as a "perihelion of mercury" sort of moment?

I don't think so, because the "perihelion" observations confirmed the predictions of an existing theory, whereas in this case there is no contending theory (to my limited knowledge) that predicted the OPERA results. The OPERA observation is not confirming any expected previously made prediction.

 

Say for example we had a viable theory that unified quantum mechanics and relativity, and that that theory could predict the OPERA results. Then it could be a similar kind of moment. Purely speculatively, that could still happen. If the OPERA results are not due to error, there should be some theory that will explain the results, and that theory might be revolutionary the way that relativity was. Even then there's still the difference in the order that the theory and the confirming observations happened in, compared with the "perihelion moment".

 

 

[Iggy]

Unless the scientists have missed something or made an error, the measurements are precise to within 10 ns. A list of the various uncertainties is shown here: http://scienceblogs.com/startswithabang/2011/09/are_we_fooling_ourselves_with.php

 

 

I don't think so, because the "perihelion" observations confirmed the predictions of an existing theory, whereas in this case there is no contending theory (to my limited knowledge) that predicted the OPERA results. The OPERA observation is not confirming any expected previously made prediction.

 

Say for example we had a viable theory that unified quantum mechanics and relativity, and that that theory could predict the OPERA results. Then it could be a similar kind of moment. Purely speculatively, that could still happen. If the OPERA results are not due to error, there should be some theory that will explain the results, and that theory might be revolutionary the way that relativity was. Even then there's still the difference in the order that the theory and the confirming observations happened in, compared with the "perihelion moment".

I think the perihelion precession was first noticed in the mid 1800s. The light-bending observation happened in the other order... i think.

 

EDIT...

 

imatfaal, the conclusions section (page 75) repeats what you were saying, I know you guys came to an understanding, but I thought you might find interesting,

http://www-itp.particle.uni-karlsruhe.de/~schreck/general_relativity_seminar/The_confrontation_between_general_relativity_and_experiment.pdf

Edited October 28, 2011 by Iggy

[md65536]

I think the perihelion precession was first noticed in the mid 1800s. The light-bending observation happened in the other order... i think.

Oh that's right, I must have been thinking of the observations of deflected starlight during an eclipse http://en.wikipedia.org/wiki/Tests_of_general_relativity#Deflection_of_light_by_the_Sun

 

 

 

So yes, I think that if the OPERA observations are correct and a "paradigm shifting" theory explains it, this could be part of something like a "perihelion of mercury" moment. But do we consider that "moment" to be when the observations are made, or when there is a theory that predicts it?

[DrRocket]

GR reduces to Newtonian gravity, and KE = 1/2mv^2 is still accurate out to many decimal places for object moving at v << c. We did not throw those (and related) equations away with the introduction of relativity. Similarly, we will not throw out E=mc^2, which is what morgsboi was asking. Anything that would modify the theory will still reduce to these equations under a wade spectrum of conditions. That's what I mean by tweak — relativity does not simply get tossed, the way that the luminiferous aether, caloric theory and phlogiston did.

 

There's a possibility that there would be a modification to the theory that is an exception for the weak interaction or specifically for neutrinos. I don't know enough about particle physics to know how that would work.

 

In the unlikely event that the result of the experiment is validated and a neutrino has actually exceeded c, then relativity does get tossed out.

 

This would not entail just a minor tweak in response to an isolated troubling bit of data, but rather result in the direct refutation of one of the central axioms of special relativity which carries over to general relarivity, invalidating the logical structure on which the theory is built -- either the speed of light is not constant (and maybe more profoundly no speed is constant), or the laws of physics vary with reference frames. When you further consider that the point of quantum field theories is consistency with special relativity, you see the deep implications.

 

On the other hand, the implications of the usual set of axioms for relativity have otherwise been spectacularly successful, so I think the likelihood of the experiment being upheld is very small. But the implications cannot be minimized in the event that it is upheld.

[swansont]

so cool!

 

I do hope you aren't embargoed from sharing a few bits of information throughout the process.

 

I won't be directly involved and unfortunately I doubt we will get advance notice of result that we'd be permitted to share. But I can describe what they plan to do, since AFAIK the talk I saw is not any kind of secret.

 

They will use a series of much shorter pulses and they have a neutrino detector at both ends of the experiment, so they are not just trying to fit the leading and trailing edges of the pulses and compare them to the pulse of particles that created the neutrinos (which is what I think OPERA did). MINOS will use a large number of narrow pulses spaced at 19 ns and the target distance is almost the same, so they should be able to confirm or exclude the 60 ns value. If they can drop it down to the same 19 ns window, that's within the error bars of zero.

 

The OPERA data the speaker showed pointed out the scale of the problem — the pulses are at least 10000 ns wide, and the leading and trailing edge are around 1000 ns

fig 12 in http://arxiv.org/pdf/1109.4897

 

You can see the noise and the claim is that the fit is good to the level of one of the minor divisions on the graph, which implies they (OPERA) have a lot of statistics to beat down this noise, and that the multiple runs can be combined to make the statistics useful.

 

So it seems that the narrower pulses from MINOS will answer some of the major questions, like is this a pulse-reshaping problem that only gives the illusion of advancing 60 ns. I had not fully appreciated the width of the pulse when that possibility was first raised. It seems much more plausible now that I see that the pulses are 10 times wider than I had thought.

 

(The speaker also explained mixing angles and neutrino oscillations in optics terms, so I finally have a grasp of what that means.)

 

In the unlikely event that the result of the experiment is validated and a neutrino has actually exceeded c, then relativity does get tossed out.

 

But we have 100+ years of experimentation that agrees with relativity. That data cannot have changed to disagree with it. (unless this acausal relation has a wide reach). Nuclear reactors still work, as do nuclear reactions of the type that produced the neutrinos. I don't see how any new physics can give us any other answer than what we already have observed to be true. It will have to reduce to the same formulas that exist in relativity.

 

This would not entail just a minor tweak in response to an isolated troubling bit of data, but rather result in the direct refutation of one of the central axioms of special relativity which carries over to general relarivity, invalidating the logical structure on which the theory is built -- either the speed of light is not constant (and maybe more profoundly no speed is constant), or the laws of physics vary with reference frames. When you further consider that the point of quantum field theories is consistency with special relativity, you see the deep implications.

 

On the other hand, the implications of the usual set of axioms for relativity have otherwise been spectacularly successful, so I think the likelihood of the experiment being upheld is very small. But the implications cannot be minimized in the event that it is upheld.

 

What if it's found that one of the neutrino flavors has m^2 < 0? I don't see how that tosses relativity.

[md65536]

In the unlikely event that the result of the experiment is validated and a neutrino has actually exceeded c, then relativity does get tossed out.

I agree that this is unlikely -- exceedingly unlikely -- because the result of the experiment is a discrepancy in the distribution of measurements from what was expected, which is what the OPERA team reported. The result of the experiment isn't that a neutrino exceeded c (though that is one of the simplest interpretations of the result). This means there's a possibility that the results can be validated and no neutrinos exceeded c.

 

It will have to reduce to the same formulas that exist in relativity.

Any new observations that disagree with relativity will probably only do so due to some peculiar aspect that doesn't apply to existing observations (and so doesn't invalidate them), such as m^2 < 0 as you suggested.

If relativity doesn't work, a new theory would be needed.

Relativity would still be an accurate model, it just wouldn't have a "range of validity" that applies to the new observations.

So it could be that in the future, relativity is "thrown out" the same way that Newtonian gravity was. That is: Not at all. It's just that its range of validity doesn't apply to everything.

 

 

That said, if relativity implies that something is impossible but it turns out that it is possible, then there are more problems with the theory than just limited range of validity. I agree with DrRocket that relativity would have to be thrown out (or completely overhauled).

 

 

However I don't think anyone has anything to worry about here. Wondering about this now is like asking "What would it mean if this strange thing I just saw was actually something impossible?"

 

 

 

What if it's found that one of the neutrino flavors has m^2 < 0? I don't see how that tosses relativity.

Would this allow v > c? Could it allow a violation of causality, without actually breaking relativity?

[DrRocket]

 

 

What if it's found that one of the neutrino flavors has m^2 < 0? I don't see how that tosses relativity.

 

 

 

The foundational axioms are: 1) The laws of physics are the same in all inertial reference frames, and 2) the speed of light is the same in all inertial reference frames.

 

It can also be shown, and we discussed this in another thread, that if information can be sent at superluminal speed in one frame, then one can find other frames in which causality is violated. A superluminal neutrino, which is detectable, would provide a violation.

 

Now, relativity might well survive failure of causality. But it can't survive falsification of the basic axioms. That tosses out the whole thing on a logical basis.

 

If [math]m^2<0[/math] then somebody will have to figure out what in the hell that means, without hand waving, and I think that will be quite a challenge. They will also have to explain how neutrinos are produced in the experiment, since with imaginary mass the minimum speed should be c, and energy decreases with increasing speed. Note that appearance of tachyons in a theory is generally viewed as evidence that the theorist has screwed up. But that is secondary.

 

The fact that special relativity is supported by a mountain of evidence actually makes the problem more difficult. When Newtonian physics gave way to relativity, it was because it failed to agree with observation. But in its place we had a theory that both matched experiment and reduced to the Newtonian theory for velocities <<c. One simply has the Lorentz group in place of the Galilean group. Here we have a case where a purported massive particle (with positive real mass) is claimed to have exceeded c. A logical consequence of the 2 axioms of relativity is that this is impossible. What is under attack is the very foundation of the theory. There is no competing theory that is "almost relativity except for neutrinos".

 

Note also that the various quantum field theories are relativistic. They are formulated to be consistent with special relativity, so you have to be very careful in using them to either explain or resolve this issue. Reasoning can become circular in a hurry.

 

The nature of the challenge makes it a really big deal. But it also makes it very very unlikely that the experimental results are valid, precisely because of the empirical evidence that has been amassed in support of relativity.

 

Note that the inferences of the experiment are purely statistical. They did not measure the transit time of any individual neutrino. Basically they have a time history for a bunch of initiation events and a time history for a smaller bunch of reception events. The difference between the means suggests superluminal transit, but there are a lot of issues to be considered.

 

BTW I had a conversation with a high energy physicist last night, He told me that a paper was prepared for formal publication but that a third or so of the "authors" refused to sign. So there is a lot more to be done before confidence should be placed in the experiment -- no surprise there.

 

This deserves serious review. Because of the potential implications. But even more importantly because the likely invalidation would involve subtle considerations that will be important in future experiments.

[Lukas!]

Can somebody please tell me what's going on with the faster than light neutrino? For such a big claim, you'd think it would be in the news more...