Monday, January 06, 2014

Trinary star won't disprove the equivalence principle

Today, science writers were busy to write about the newest death of a pillar of modern physics, in this case Einstein's theory of relativity: Google News, Physics World, BBC, dozens of others...

Video by Anne M. Archibald, a PhD student and a co-author of the discovery

All this buzz is based on a newly published article,
A millisecond pulsar in a stellar triple system by Scott Ransom and 20 co-authors (Nature)

arXiv copy of the paper (free PDF)
which reports some observations of a star system called PSR J0337+1715. In the very abstract, they decide to "hire" this multiple star as a tester of the "strong equivalence principle" and the media copy (and usually amplify) this point.

I don't think it's a sensible way to talk about this cool system. In fact, I think it's unfortunate that such a cool "family of celestial bodies" gets abused for promotion of this kind of broken physics.

First, what is the system? It is composed of three stars. One of them is a pulsar i.e. a rotating neutron star. It is a millisecond pulsar and its presence makes this new system so special. The pulsar's mass is 1.438 solar masses (the relative errors are 0.01% or so for the masses).

This "Sun" has two lighter companions which are nevertheless much heavier than "planets". Both of them are white dwarfs (a remnant of a star composed of electron-degenerate matter). As the animation at the top shows, one of them is very close to the pulsar ("the Sun") and the orbital period is 1.6 days (much shorter than Mercury's). Its mass is 0.198 solar masses.

The other white dwarf is very far from the two bodies I have mentioned. Its orbital period is 327 days – pretty much the same one as the Earth's orbital period known as a year. And its mass 0.410 solar masses – much heavier than the Earth but it doesn't really matter.

The three-body problem is notoriously unsolvable even in the "simple" theory written down by Isaac Newton. The behavior is chaotic and cannot be expressed in terms of elementary functions (and not even in terms of not so elementary functions!). This causes more trouble than just complexity; soon or later, the chaotic relative motion of the three bodies is likely to lead to some collisions.

Because of the hierarchical distances (the distance between the pulsar and the lighter white dwarf is much shorter than their distance from the heavier white dwarf), the collisions can't occur too early here. The distant white dwarf "feels" the pair of the pulsar and the lighter white dwarf as if it were a single body. And the relative motion of the pulsar and the nearby white dwarf is also near-elliptic, resembling the Earth-Moon motion as both of them are orbiting the Sun.

Křemílek and Vochomůrka, two ethnic Czech racially white dwarfs, were found at orbits around a pulsar. Check at least one episode of this cartoon!

Some more careful measurements of this very interesting system may confirm the general theory of relativity in a more nontrivial and newer way than other celestial systems that have been observed. This trinary star may also identify a deviation from general relativity. (Some media have already announced the falsification of GR; not so fast!)

But I find it highly unreasonable for them to guess that such a deviation would immediately imply the falsification of the "strong equivalence principle", a cornerstone underlying general relativity. First of all, let me use the term "equivalence principle" without the redundant adjective because operationally speaking, all the "versions" of the equivalence principle are really equivalent to each other. The authors of the paper in Nature are really "mostly" astronomers and by their addition of adjectives to the "equivalence principle", they only reveal one thing – their personal problems with general relativity. It reminds me of Lee Smolin's talk about "strong AdS/CFT conjectures" and similar bizarre linguistic constructs. (Incidentally, I was offered to translate Smolin's latest book, Time Reborn, to Czech today. I needed a few seconds to refuse the offer LOL. I've done some low-brow physical work in my life but to translate a book by an idiot who can't reach my ankles would be somewhat too much of modesty and masochism for me.)

The point is the following: Even if the trinary system detected some deviations from the predictions of general relativity, there would exist more sensible, more well-motivated, more natural explanations of the deviations from general relativity. General relativity could be modified in much more plausible, more likely, milder ways than by the liquidation of the equivalence principle that is beating in its heart. Their implicit claim that they may be close to the falsification of the equivalence principle is a silly wishful thinking. After all, the star system doesn't even test the principle "directly" in any sense because the three objects are located at different points and are therefore probing different gravitational fields, not the same field that could cause acceleration that depends on the composition.

Similarly, all the verbal exercises in the media designed to talk general relativity down are dumb, too. For example, we read in many sources that GR is incompatible with quantum mechanics which is supposed to make it more likely that they will detect violations of the equivalence principle. This is just plain idiotic. The observations of the orbits in this star system (and other systems of celestial bodies) have nothing whatever to do with quantum mechanics or its reconciliation with general relativity.

Moreover, the equivalence principle is *needed* for quantum mechanics and general relativity to be compatible. The equivalence principle is linked to the diffeomorphism symmetry which is necessary for the decoupling of the unphysical polarizations of the gravitons that would sometimes lead to predictions of negative probabilities in a quantum theory. So if you decided that the equivalence principle has to be violated, you would actually make the tension between general relativity and quantum mechanics worse!

General relativity (and especially the equivalence principle underlying it) and quantum mechanics (and its postulates) are perfectly compatible as long as they are being unified in a careful and wise way rather than a naive or straightforward one. The full reconciliation of these two paradigms is known as string/M-theory but for many sufficiently "down to Earth" purposes, we don't really need too specific stringy features of our TOE. A semiclassical approximation is good enough. And the equivalence principle is one of the features of general relativity that make it possible to reconcile this Einstein's theory of gravity with quantum mechanics!

So these astronomers have contaminated their nice observations of this unusual system by their laypersons' misconceptions about modern physics – which is a bit unfortunate. And of course that much like in virtually all other situations, the stupid journalists have mindlessly repeated – and amplified – these misconceptions. But I am still happy that people will be able to observe this bizarre triple animal.


  1. Wikipedia and other authors make a distinction between "strong" and "weak" equivalence in that the strictly geometrical character of the former supposedly excludes Brans-Dicke and other "fifth force" theories. If I understand correctly, that would also mean incompatibility of "strong equivalence" with string theory. Is that right?

    Is that what you mean when you say there is only one equivalence principle? I didn't follow your reasoning there.

  2. Dear Pig,

    scalar-mediated forces are perfectly compatible with string theory. But these extra "fifth forces" shouldn't really be called gravitational - for the same reason why the quantum of the scalar field shouldn't be called a graviton.

    Whether such "fifth [long-range] forces" are viewed as a violation of the equivalence principle is a matter of the definition. People in HEP would usually count them as a violation of the equivalence principle even though they may be considered as "non-gravitational" as electromagnetism.

    But it doesn't matter whether we would view such "fifth forces" as a violation of the equivalence principle. We really know that these forces don't exist in this Universe from other experiments and considerations, anyway.

    Such "fifth forces" would be mediated by strictly massless scalars and they would lead not only to the violation of the equivalence principle in the sense above but they would also lead to time-dependent values of dimensionless constants such as the fine-structure constant.

    Those processes don't occur which is why all string theorists I know feel pretty much certain that there can't be any strictly massless scalars in Nature - and that's why string phenomenologists always consider stabilized vacua only (all scalars have to get positive squared masses).

    One may create lots of fog but any violation of "any" equivalence principle is pretty much excluded and impossible.


  3. It may be possible to make a meaningful distinction between different equivalence principles, but Wikipedia certainly fails to do so.

    1.The gravitational motion of a small test body depends only on its

    initial position in spacetime and velocity, and not on its constitution.

    2.The trajectory of a point mass in a gravitational field depends only on
    its initial position and velocity, and is independent of its composition
    and structure.

  4. 6-7 orders of magnitude more sensitive checks of some aspects of GR is still pretty fantastic!

  5. Yup, it is cool, but even 6-7 orders of magnitude improvement of the accuracy of a quantity can't replace the falsification of a grand principle.

  6. @Lubos:
    Hmm, if you are still considering translating Smolin's book, I could send you some valium, or maybe an anti-psychotic :)
    ( A favorite quote from George Carlin--
    "I don’t have pet peeves. I have major psychotic fucking hatreds." )

  7. All measurable observables are EP violation-inert. Here's a yummy failure: arXiv:1304.6875;
    Science 340(6131) 448 (2013); Physics
    Today 66(7) 14 (2013). Violating the weak or strong EP requires finessing Einstein's inertial elevator. EP violation through a preferred direction in space is awful. EP violation through a variable vacuum refractive index toward matter is trivially encoded within Einstein-Cartan-Kibble-Sciama
    gravitation. Dear Reader is invited to list candidate unmeasurable observables. HINT: It's a screwy footnote, so get torqued.

  8. LOL, thanks, no, I was really not considering it for a second and I am not considering it. The chemicals couldn't possibly be enough.

    Brian Greene is a reasonable honest guy of a sort but due to the mystifications about quantum mechanics in particular, The Hidden Reality was already near the edge of I can handle. ;-) There are still lots of books with various attitudes and perspectives but books that are essentialy correct and insightful so it still makes sense to choose.

  9. kashyap vasavadaJan 7, 2014, 1:48:00 PM

    Hi Lubos: Is it easy (!) to see why scalar (dilaton or some other) particle would break equivalence, when in fact tensor graviton enforces it?

  10. kashyap vasavadaJan 7, 2014, 3:55:00 PM

    Great answer. Thanks.

  11. Dear Dilaton, all the best wishes and I hope you get well fast. Good to hear from you.

  12. It's not an order, it's "the" order. Because they invented the concept. It's about their concept that Sabine says that it don't exist.
    The free will is about our ability to decouple from our environment, represented by the laws of nature, our genetic inheritance or our culture. We have no free will when we talk about decoupling from the laws of nature. We have the ilusion of free will only because we are ignorant about them. We have the ilusion of free will because we do not have the complete causal dependencies of our decisions. We are in totality a construction of the environment, strictly speaking we are not independent from it.

  13. Dear Calin, if "the free will" (or anything else) is claimed to have existed or not before people began to think about the world, then one cannot possibly "invent it". At most, one may "discover it" but things that may be "discovered" cannot be "copyrighted".

    I will generously overlook your incoherent philosophizing twaddle.

  14. The concept was invented to "explain" things like bullshits from the bible.
    New concepts with the same name? How about the redefining of aether?

  15. Sorry, the meaning of every word is evolving, has always been evolving, and will be evolving.

    The Bible is irrelevant today in these discussions so almost no one uses some "Bible flavors" of this concept while discussing the free will - and many other things.

    Eat your feces if you have a trouble with that.

    The aether is no exception. The meaning of the word "aether" has evolved significantly over centuries. It was a soul-less element, one of the five hypothetical elements in some civilizations, but it became the luminiferous aether a particular compound composed of atoms of a sort whose mechanical properties were claimed to be identical with the electromagnetic phenomena. It was a long way from the aether of the ancient civilizations.