The award yesterday of the Nobel prize to Weiss, Thorne and Barish (and the LIGO team) for gravitational waves is interesting because they have discovered a new phenomenon. It could be gravitational waves or something else, but it should be treated as an interesting observation for further open-minded study. What bothers me about it is the assumption that the anomaly was caused by the merger of a couple of black holes in a far off galaxy, which is unfalsifiable and apparently unquestioned (as I also said in a previous blog entry).
Imagine you're sat on a beach and a particular waveform rolls in from the deep ocean. You have a supercomputer at hand and a love of dolphins and because the computer is so powerful you manage to compute the exact action a nice old dolphin out to sea must take to produce that pattern of waves. Maybe he jumped out of the sea whistling "I'm a little teapot" (have you heard of Russell's Teapot?) and plunged in with a twist of its tail. The fact that you can post-dict a scenario that leads to that pattern of waves is not surprising these days because we have such powerful computers. It does not mean you have proved it was the dolphin. It is not a direct proof: other things could have caused it, since in the case of physics the prevailing framework is not as sure as people suppose (see below). The ability of computers to invent unfalsifiable facts to support a comforting conclusion is one of the great problems of 21st century physics. To explain here's a quote from Douglas Adams (Dirk Gently's.., p55):
"..there have been several programs written that help you to arrive at decisions by properly ordering and analysing all the relevent facts so that they point naturally toward the right decision, but the decision that all the properly ordered facts point to is not necessarily the one you want. .. Gordon's great insight was to design a program which allowed you to specify in advance what decision you wished it to reach, and only then to give it the facts. The program's task, which it was able to do with consummate ease, was simple to construct a plausible series of logical-sounding steps to connect the premises with the conclusion. Gordon was able to buy a Porsche almost immediately."
You may say that black holes are the only entities that can produce the chirp that was seen by LIGO, but in saying this you are relying on a theoretical framework (general relativity, GR) that has been falsified in thousands and thousands of cases (at very low accelerations, not at high ones where it is supported by Gravity Probe B). This may come as a surprise since GR is supposed to be the highest creation of the human intellect, but it was falsified in the 1930s and then again in the 1970s by galaxy rotations - low acceleration phenomena very far from our normal experience. GR failed to predict any galaxy's rotation speed, badly, and galaxies are a pretty huge chunk of the universe not to predict (almost all of its matter!). The old theoretical framework has been patched up by the addition, where needed, of a huge amount of invisible (dark) matter, but this is an arbitrary addition, a fudge. It means that GR still cannot predict any galaxy's rotation from directly-observed quantities. You have to observe both the visible matter, the lit stars, then the rotation of the stars (the answer), and then add the dark matter distribution by computer so that GR can predict the right answer - Gordon's program is post-dicting the facts that are needed to make GR right. This goes unchallenged because it is an unfalsifiable prediction because dark matter is invisible, so Gordon is still buying Porsches.
Please note that, forgetting far off black holes for a minute the LIGO team still have discovered a real and very interesting effect, but it is the connection of that to a specific unfalsifiable scenario (two merging black holes) that I believe is unscientific and stops healthy debate. This is due to an unfortunate blind spot that the mainstream has, caused by the over-use of computer programs and it is serious perverting the progress of science. Note that in quantised inertia, a new framework that I am proposing, all the inputs are observed parameters, so Gordon's program is powerless.
9 comments:
Mike - much the same applies to the Nobel for discovery of the Higgs boson, where the theory is somewhat uncertain about what the mass ought to be and, after much searching, a particle was found in one of the ranges that were possibilities. There are various fudges in accepted theory, such as inflation of the early universe to overcome the limitation of the speed of light and thus explain the relative uniformity of the background radiation, and at times it would seem better to admit we can't explain it and leave it as "unknown" rather than produce an explanation that doesn't really make sense.
With gravity, we currently don't know the actual velocity of propagation, and it's only assumed to be light-speed. Possibly if a third LIGO comes on-line then the velocity of the disturbance can be measured and the source triangulated, but it's only an assumption at the moment that it's actually gravity-waves we're seeing. When we're talking about a change in distance of approximately 1 atomic separation in the diameter of the Earth, though, there are maybe quite a few other sources of confusion. With only two LIGOs running, and the signals ignored unless they are within a time-span corresponding to the distance between them, an earth-movement close to the central point could provide that close timing. Putting a triangle of sensors out in space seems a lot more certain to avoid such problems, and I suspect that when that goes up we'll probably get a null result and a lot of head-scratching as they try to explain it.
Excellent post. After the first announcement when the NY Times and others showed a hi-res animation of two colliding black holes I was unhappy. To extrapolate from a chirp of a signal of unknown origin to such a pile of fairy dust seemed so irresponsible.
Mike,
See Erik Verlinde's presentation at the Perimeter Institute on October 4th:
https://www.theguardian.com/science/life-and-physics/2017/oct/03/could-the-theory-which-predicted-gravitational-waves-be-wrong
He is coming at the problem of galaxy rotation from a different direction to you (this is a public lecture so fairly top-level, but you might want to look at his published papers) but it is interesting that he sees gravitation as an emergent phenomenon.
Laurence: Verlinde's theory disagrees with the data, and is an complex mess besides, see: http://physicsfromtheedge.blogspot.co.uk/2016/11/critique-of-verlindes-gravity-1.html
/* the LIGO team still have discovered a real and very interesting effect */
The gravitational wave chirps found were isolated from myriads of very similar noise events, so it's approach similar to searching of shapes in the clouds.
Mike,
Thanks. I hadn't spotted that blog post before.
Best wishes, Laurence
Mike,
FYI, an article at New Scientist details a new method of detecting hot gas streams in between galaxies that fall just short of x ray emission. Long story short, we now need about half as much "dark matter" in the traditional calculations to explain other observations.
So that's half the dark matter gone. Now to clean up the other half.
https://www.newscientist.com/article/2149742-half-the-universes-missing-matter-has-just-been-finally-found/
RichD
Wow. What would the QI modelling of these galaxies look like now with half of the DM gone?
RichD, joesixpack: They have found half of the missing 'baryonic' matter, so it won't have much affect on the galaxy rotation problem.
Post a Comment