Forgot your password?
typodupeerror
Australia Space Science

Fine-Structure Constant Maybe Not So Constant 105

Posted by samzenpus
from the variety-is-the-spice-of-life dept.
Kilrah_il writes "The fine-structure constant, a coupling constant characterizing the strength of the electromagnetic interaction, has been measured lately by scientists from the University of New South Wales in Sydney, Australia and has been found to change slightly in light sent from quasars in galaxies as far back as 12 billion years ago. Although the results look promising, caution is advised: 'This would be sensational if it were real, but I'm still not completely convinced that it's not simply systematic errors' in the data, comments cosmologist Max Tegmark of MIT. Craig Hogan of the University of Chicago and the Fermi National Accelerator Laboratory in Batavia, Ill., acknowledges that 'it's a competent team and a thorough analysis.' But because the work has such profound implications for physics and requires such a high level of precision measurements, 'it needs more proof before we'll believe it.'"
This discussion has been archived. No new comments can be posted.

Fine-Structure Constant Maybe Not So Constant

Comments Filter:
  • by Anonymous Coward on Sunday September 05, 2010 @06:52PM (#33484140)
    we need more research to tell if this is first or not.
    • Re: (Score:3, Insightful)

      this should get at least a +1 funny (:

    • by AnonymousClown (1788472) on Sunday September 05, 2010 @06:57PM (#33484178)

      we need more research to tell if this is first or not.

      I am unable to reproduce your results.

      • Re: (Score:1, Funny)

        by Anonymous Coward

        I am unable to reproduce your results.

        It worked for me; my post is also first.

      • I am unable to reproduce your results.

        This is slashdot: reproduction never happens.

      • by jonadab (583620)
        > I am unable to reproduce your results.

        Well, obviously. He didn't most any detailed methodology, or even any information about what controls he u sed. How is anybody supposed to set up an identical experiment in another lab?
    • by ultranova (717540)

      we need more research to tell if this is first or not.

      But rest assured, creationists/ID proponents don't need any more research to declare this as proof that carbon dating is incorrect. Which, shockingly enough, might actually be true, since the strength of fundamental forces define the rate of radioactive decay, and which nuclei are stable in the first place.

      Either way, I predict that Fundies Say the Darndest Things [fstdt.net] will have a flood of new Whiskey Tango Foxtrot -forthy quotes in short order.

      BTW. Your com

  • By comparing the light absorbed by the atoms in the gas clouds with the light absorbed by the same species of atoms on Earth, researchers can attempt to calculate the value of the fine-structure constant at different distances and times in the universe.

    The article doesn't say how they correct for radial movement (relative to us) in the gas clouds. A cloud experiencing time dilation should absorb different wavelengths of light.

  • by CheshireCatCO (185193) on Sunday September 05, 2010 @07:06PM (#33484216) Homepage

    This isn't the first time that some team has claimed this. Around 2000, someone made the same claim. I recall it not standing up when other teams checked it.

    Measurements like this have been done before and usually show a constant, er, constant to within experimental uncertainty.

    Note, for example, this paragraph buried at the end of the article:

    Nonetheless, the study “is as speculative as the previous claims,” asserts Patrick Petitjean of the Institute of Astrophysics in Paris, whose team has looked for variations in the fine-structure constant with the Very Large Telescope as far back as about 11.5 billion years ago and found none (SN: 4/8/04, p. 301).

    In other words, I wouldn't get excited at all yet.

    • "whose team has looked for variations in the fine-structure constant with the Very Large Telescope as far back as about 11.5 billion years ago and found none"

      WTF? These dudes were already alive 11.5 billion years ago??

      • Re: (Score:1, Insightful)

        by Anonymous Coward

        The farther out you look with a telescope, the farther back in time you're seeing.

        • Re: (Score:2, Funny)

          by Anonymous Coward

          the farther you look up, the farther you can see things going over your head

    • In other words, I wouldn't get excited at all yet.

      I'm not sure what to get excited about anyway. It's interesting, but the article didn't go into detail about how much this varies and how much this really changes the current understanding of the history of the universe. So far, it's just given imaginary legs to at least one denier of some kind in a response on the Science News web site.

      • Re: (Score:3, Informative)

        by CheshireCatCO (185193)

        If it varied at all, the result would be significant just on principle. It would mean that the laws of physics have varied in time, which is not something that most current models allow for. (At least, not that I've heard. I work on something a little closer to home, so I might have missed something.)

        • I thought the anomaly described varied in space, rather than time? That is, things further away in one direction had a greater alpha, but ones in the other direction had a lesser alpha.

          • Farther away = farther back in time due to the finite speed of light.

            The size of the change seems to vary in space, but they're talking about a time variation in alpha as a function of time.

            • Yeah, I know how (cosmologically speaking) distance = time, I just meant that they're saying alpha varies in space as well. From the article:
              "Along one direction the fine-structure constant, which governs the strength of the electromagnetic force, grows slightly weaker with time, while in the other direction it grows slightly stronger. " ... which seems to suggest that the entire universe has some sort of axis along which alpha changes.

              • Yes, that's what I said:

                The size of the change seems to vary in space, but they're talking about a time variation in alpha as a function of time.

                We're in agreement.

                (What's even more concerning about this result is that if I read it right, stuff ~12 billion years ago had notably different alphas in the two directions. But at that time, the universe would have been much smaller, so those parts of the universe would have been closer together. That doesn't make it seem any more palatable.)

          • Yes, the title of the paper reads:
            "Evidence for spatial variation of the fine structure constant"
            and the abstract contains:
            "We previously reported observations of quasar spectra from the Keck telescope suggesting a smaller value of the fine structure constant, alpha, at high redshift. A new sample of 153 measurements from the ESO Very Large Telescope (VLT), probing a different direction in the universe, also depends on redshift, but in the opposite sense, that is, alpha appears on average to be larger in th

        • by Artifakt (700173)

          The standard model doesn't, but then the standard model doesn't predict what the fine structure constant has to be by any of the fundamental assumptions. The standard model says in effect "measure the damned fine structure constant and take it as measured, but don't expect to know why the universe is that way." Some string theories (and if I recall correctly, some brane theories) predict a specific value for alpha and some other constants that are just arbitrary by the standard model, but the trouble is, th

      • by Zocalo (252965) on Sunday September 05, 2010 @07:30PM (#33484342) Homepage
        The fine structure constant is given as being equal to "e^2/hc", so if the FSC is not a constant then one (or more) of the other values must also be a variable. Take your pick between:

        If any of those constants turned out to in fact be variable, or even a "constant" which has varied over the lifetime of the universe, then the implications would be profound to say the least.

        • by Black Parrot (19622) on Sunday September 05, 2010 @08:10PM (#33484554)

          The fine structure constant is given as being equal to "e^2/hc", so if the FSC is not a constant then one (or more) of the other values must also be a variable.

          Or maybe the "2" is changing?

          Not mathematically, but 1.999--> 2.001 or such.

          • Well, according to some guys on here last month, pi is dependent on physics, because 2*pi*r is the circumference of an Euclidean circle. But I see that there's a 2 in that expression also, so why not?

          • Units wouldn't work then, though. I don't think you can change that exponent without doing something to at least one other term.

          • Or maybe the "2" is changing?

            Not mathematically, but (...)

            Not mathematically, but linguistically. In the time before time, 2 was actually known by its true name, potato. It's only lately we humans who have screwed that up.

          • Re: (Score:2, Funny)

            by symbolset (646467)
            Wait, there's also the operators "^" and "/" to consider. Maybe those are the variables.
            • Funny? Actually, perhaps real. Operators are functions that map the operands to a result. If the mapping changes over time, the mapping may be non-constant with constant operands. Something to ponder when you have time ;-).
            • Re: (Score:1, Funny)

              by Anonymous Coward

              If C++ can overload operators, why couldn't Nature?

        • by BitterOak (537666)

          The fine structure constant is given as being equal to "e^2/hc", so if the FSC is not a constant then one (or more) of the other values must also be a variable. Take your pick between:

          If any of those constants turned out to in fact be variable, or even a "constant" which has varied over the lifetime of the universe, then the implications would be profound to say the least.

          Actually, it is possible to work in fundamental units in which hbar = c = 1, so the latter two can be considered constant by definition.

          • I thought that the fine-structure constant was dimensionless, so it always has a value of approxmiately 1/137. Does that mean in natural units (h=c=G=1), e must be the square root of 1/137?
          • by Idarubicin (579475) <allsquiet.hotmail@com> on Sunday September 05, 2010 @11:50PM (#33485728) Journal

            Actually, it is possible to work in fundamental units in which hbar = c = 1, so the latter two can be considered constant by definition.

            Oh, so close -- you just needed to look up one more Wikipedia article to get a hint about why your reasoning is faulty. There are indeed systems of so-called natural units [wikipedia.org] which assign a constant value of 1 to certain physical units. Yes, there are systems which define c and h-bar as 1, but there are also systems which define e to be exactly 1.

            Inconveniently, merely asserting a definition doesn't actually compel obedience on the part of the Universe. If I work in Stoney units [wikipedia.org], then I define e and c to be constant, so h-bar must be changing if the fine structure constant changes. In Schrodinger units, e and h-bar are constants, and c must be changing. The natural-unit systems only work properly if the assumption of constancy of their chosen fundamental constants is correct.

            • by BitterOak (537666)

              Inconveniently, merely asserting a definition doesn't actually compel obedience on the part of the Universe.

              As long as your definitions are consistent, though, you are not requiring the universe to obey anything. Take the speed of light for example. What does it mean to say that the speed of light is constant? It means that the distance covered by light in a specified unit of time will be constant. But if we define our unit of length to be the distance traveled by light in one second, then it will always be constant. My point is that two of the three quantities in the fine structure constant are arbitrary, a

        • ...the implications would be profound to say the least.

          Like free energy. If E=MC^2, then by converting matter and energy back and forth while C changes can lead to more than you started with. Of course, the theory that leads to E=MC^2 are based on the idea that C is constant, and break down if C starts changing. If the elementary charge changes, you could charge batteries while e was small, then discharge them while e was high. This would lead to more energy out then in. If the plank constant changed, well I'm too tiered to think about it now, but I'm sure the

          • by dumuzi (1497471)
            INAEB (I'm not an expert, but...) Your battery discharged while e is high may discharge more energy (shiver at the thought), but can you actually do more with that energy? If that battery is in a flashlight it now has to work against a higher e value to emit photons, in the end you may well end up emitting the same number of photons with the same wavelength, resulting in the same net change in entropy as if you had used your batteries while e was still small.
          • by Doc Ruby (173196)

            No, as the value of c changes (if it changes), the amount of energy changes with it to match. The FSC does not govern whether or not a closed system conserves energy.

            However, if FSC changes, then it might be possible that somewhere in the universe where the FSC is different than here that the matter is equivalent to much more energy, and even that the structure of matter is more unstable and easier to convert to energy. So in (completely speculative and possibly full of holes) theory, we could send a reacto

            • That's all probably true. What I'm saying is interesting consequences of fundamental constant change. What happens when you cross an FSC gradient? Well, a lot of paradoxes arise.
              • by Doc Ruby (173196)

                The specific interesting consequence you described is not a possible consequence of the FSC. There is no paradox. You're just not understanding properly what properties are dependent on the FSC, and which change with it.

        • by Dread_ed (260158)

          I seem to remember various and sundry stories at many times over the last 20 years in respectable science journals and online publications speculating that the speed of light has changed at times during the unfolding of our universe.

          If I remember correctly some postulated a "phase change" analogy, others spoke about the curvature of space, and still others about the density of matter, and I seem to remember some mentioning other fundamental constants fluctuating as well. Please feel free to comment about h

        • The fine structure constant is given as being equal to "e^2/hc", so if the FSC is not a constant then one (or more) of the other values must also be a variable.

          That does not follow. I don't know enough about the fine-structure constant to talk about the specifics here, but another possibility is that the theory (and equations) where the constant is used is incorrect and there should be some other expression there, not a constant.

          I'm not disputing that it could still be a very significant result, but it doesn't necessarily follow that e, h or c are variable.

        • c, kinda. The assumption that the majority of intersteller space is a vacuum is probably flawed. The effect of pressure in the formulae needs to be considered. If you assume that pressure in a black hole approaches infinity, the Planck constant (h) will also approach infinity. Unless of course, it (h) has properties like crystals which breakdown when certain levels of pressure are applied.
    • The further back they looked with the VLT, the larger alpha seemed to be—in seeming contradiction to the result they had obtained with the Keck. They realised, however, that there was a crucial difference between the two telescopes: because they are in different hemispheres, they are pointing in opposite directions. Alpha, therefore, is not changing with time; it is varying through space. When they analysed the data from both telescopes in this way, they found a great arc across the sky. Along this ar

      • Sure it needs verification, but I am also not surprised other institutes downplay what wasn't their finding.

        That's the problem, there are other tests out there and they don't show the effect. This isn't some new measurement, it's an old measurement (perhaps done better, perhaps not) that is showing a previously unseen behavior.

        Also, yes, it's true that other groups are liable to downplay the finding (although that happens less than you might think, speaking as an astronomer myself), but the team themselves do the opposite. Why believe one is biased and the other isn't?

    • This isn't the first time that some team has claimed this. Around 2000, someone made the same claim. I recall it not standing up when other teams checked it.

      Check out the Economist article, it's better. The difference there was that the telescopes were in different hemispheres. They thought previously that they were measuring a variation back in time (and the follow-up study in the southern hemisphere showed no such measurement back in time when they looked), but it turns out that they're measuring a vari

      • Right, but other groups haven't found any variation in time or in space. It's possible that they looked in exactly the wrong directions, I suppose, but that's kind of unlikely.

        I'm not saying that this result is wrong. I'm just saying it squares badly with other measurements and it's more likely at this point that something is else confusing the result, either experimental error or some other effect.

        • Right, but other groups haven't found any variation in time or in space. It's possible that they looked in exactly the wrong directions, I suppose, but that's kind of unlikely.

          Indeed, but that's the suggestion of the article.

          I'm not saying that this result is wrong. I'm just saying it squares badly with other measurements and it's more likely at this point that something is else confusing the result, either experimental error or some other effect.

          They're saying now that they have multiple observations of a

          • Right, but other groups haven't found any variation in time or in space. It's possible that they looked in exactly the wrong directions, I suppose, but that's kind of unlikely.

            Indeed, but that's the suggestion of the article.

            Fair enough (and it may be true), but it's not reasonable to expect people to readily believe the result if that's their explanation. Low probability things happen all the time, but we shouldn't be expected to run out and believe them right away. Which is pretty much all I'm saying when I suggest skepticism.

            They're saying now that they have multiple observations of a half-dozen quasars on different telescopes and have calibrated the error between them.

            Having now looked at (but not carefully read) the paper, it also looks like their signal is pretty erratic. Measurements in the same part of the sky given very different results. It looks like they'v

            • Measurements in the same part of the sky given very different results. It looks like they've teased out trends statistically

              Ah, interesting. I hate to disbelieve well-done statistics, but perhaps they can work on getting the noise out of their sensing. Thanks.

              • Stats are an interesting thing in science, especially in physics and astronomy. We're all trained in a few stats, but after those ones, the more you do the less many of us trust your result. Part of it is that those fields generally don't need a lot of statistical teasing to produce clear results and part of it is unfamiliarity, but another big chunk is the soft nature of statistics: a given stat is rigorous, but which ones give you the best information is kind of hard to say so you can go shopping around

                • I appreciate the problem. Even some of physics itself can be statistical in nature, and certainly various sensing gear used in experiments uses statistical filters to screen out noise. Yet, it seems like we only accept statistical models if we're pretty sure they're true. It would be lovely if we had rigorous methods to accept or discard statistical conclusions based on the math itself. As you quite correctly point out, all too commonly we're left wondering if the statistics are correct, misleading, or

    • by Artifakt (700173)

      That earlier team would be at the University of New South Wales under John Webb in 1999.

      It is interesting however, that Cosmologists/Astrophysicists seem to be far from universal agreement on one point:

      What else would have to vary if Alpha (the Fine structure constant) varies.

      Alpha can be described in terms of other physical constants, via any of three equations that each use some of these values:

      e = the elementary charge;
      = h/2,

    • by ZeRu (1486391)
      Indeed, this is not news. My particle physics professor mentioned it in 1998, though I still don't understand how a constant can change, and more importantly, why it does..
  • by nyri (132206) on Sunday September 05, 2010 @07:20PM (#33484282)

    Evidence for spatial variation of the fine structure constant [arxiv.org]

    An evaluation from a practicing physicist would be appreciated.

    • Re: (Score:1, Redundant)

      by Americium (1343605)
      The summary's point was the even if the paper is correct, there may be systematic errors in the data. I.E., these guys downloaded the data from the telescope repositories, this data might have errors.
    • Re: (Score:2, Insightful)

      by Anonymous Coward

      Let's just say that if called upon to review this paper I would have a lot of comments on techniques and assumptions. They are also sending it to a journal where it is likely they will find reviewers who don't understand the details of the instruments used.

      I'm not saying its definitely wrong, but it definitely needs a peer review cycle or two before publication. If published in this form, the editor will get some "feedback" on the editorial process. Longer term, I and a lot of other people would like to

    • by CheshireCatCO (185193) on Sunday September 05, 2010 @11:16PM (#33485532) Homepage

      Wow. Check out Figure 5. It's hard to believe there's any systematic signal in their data, they have blue and red points mixed up all over the sky. They claim there's an excess of one color in one hemisphere and the other color in the other hemisphere, but it's not very strong signal at all. This isn't my field (I work on Saturn's rings), but my first reaction is amazement that this is serious enough to be talking to the press about.

      (I'm also more than a little put off by the fact that this hasn't been accepted, evidently, merely submitted to a journal. And not even an astrophysical journal like AJ or ApJ which seem like a far better fit for this subject matter and the audience.)

      • Re: (Score:3, Funny)

        by Inthewire (521207)
        They seem to be in good shape. Thank you for your efforts.
      • I work on Saturn's rings

        Wow - cool place to work!

      • This isn't my field (I work on Saturn's rings),

        Do you have a long commute? How is traffic out there?

    • by Chalnoth (1334923)
      A reasonable evaluation was already pointed out in the summary to the article, by Max Tegmark, who is a practicing physicist, and who pointed out that he would need more evidence to be convinced it's not just systematic errors. I would personally go even further and say it's almost certainly systematic errors. Just looking at the abstract makes this patently clear, as they find the opposite effect claimed in previous studies. They claim large statistical significance, of course, but when your result chan
      • by mrsquid0 (1335303)

        The paper does not do a great job of addressing potential systematic errors. What concerns me is that the dipole that they claim to find aligns with the direction of the Great Attractor. This makes me think that there may be a subtle systematic effect in the radial velocity measurements.

  • by PolygamousRanchKid (1290638) on Sunday September 05, 2010 @07:32PM (#33484354)

    This week's "The Economist" has a good article on this: http://www.the-economist.com/node/16930866 [the-economist.com]

  • by Gruturo (141223) on Sunday September 05, 2010 @09:05PM (#33484906)

    Ok, I guess 9 years is acceptable for a dupe [slashdot.org], and tbh I didn't even read the article, in /.'s finest tradition, so it might be an actual new development :-)

    Kinda sure there was some piece of news on the subject from around 2005-2006 too, but can't find it atm. Meh, google-fu weak at 3am, should sleep, work in under 5 hours.

    • I just skimmed both articles. It seems like they have submitted new data and analysis from the same long-running study. We'll need further analysis and a peer review cycle to be sure whether or not it qualifies as an actual dupe.

    • I think my story submission posted two hours before this one made the difference a lot clearer that the evidence points to a variation over space now, and not just time. (yes, I'm annoyed at being rejected, hah - and we had to waste several +5 informative to get the info I included in my summary like the arxiv link). It took an annoying amount of effort to track down the actual paper, too. I first read the story in 3quarksdaily.com [3quarksdaily.com] linking to the Economist story, then found the researcher's UNSW site whic
  • How come all these semi-science articles have to quote someone else saying 'needs more proof' etc.? The primary researchers almost invariably say the same themselves. Is the science not worth wasting the ink if it can't be made to appear as if it's an argument? Being skeptical yourself is good. Someone else being skeptical is trivial. It's one thing to interview someone else if they have something to add, but to do it just to hang a name on the preplanned 'controversial' portion is st00pid to the point of i

    • Researchers need to publish ....

      This means they have published, and it's got quoted, and it will be referenced in other articles, this means they get paid and can continue their research ..If it is found to be true or not is not important .... But if it is then they can publish that as well ....

      • Whether it's true or not is, of course, important - both to society at large and for the (longer term) viability of the careers of the scientists in question - but can scientists publish and have careers, even if they are wrong?

        Absolutely, and they *should* be able to do so. If anything, the current system over-emphasizes being right over doing the important experiments which are absolutely requisite to scientific progress; even if the results come out inconclusive, you have to try different thing
  • Duplicate joke for a duplicate claim:

    Planck's constant (h) increased in value this morning to roughly 50 joule-seconds, sending the DJIA to a 95% confidence interval between 0 and 15,000, and increasing the wavelength of a penny moving at a brisk walk to a value on the order of it's own diameter, so that macroscopic, every day objects behave as waves instead of billiard balls. Tennis players in central park (whose velocity could not be determined as of this printing) may have been alarmed to find tennis ba

Life would be so much easier if we could just look at the source code. -- Dave Olson

Working...