Sunday, February 27, 2011 ... Deutsch/Español/Related posts from blogosphere

MOND isn't beating dark matter

The BBC application on my iPod Touch led me to the following article:

Dark matter theory challenged by gassy galaxies result
This stuff is based on a preprint by McCaugh. This paper chooses some "gassy galaxies" which are consistent with MOND - Modified Newtonian Dynamics - while requiring a smaller number of parameters than the standard cosmological model with dark matter.

When I looked at the graphs that are claimed to contain all the support, I was terribly unimpressed. It's just some slope in a noisy graph that MOND is argued to predict pretty well - a linear curve. There are many other things, to be enumerated below, where MOND doesn't work and they contain many more details than a slope of a would-be linear function for a subclass of galaxies.

I think it's obvious that the author has been cherry-picking in order to support a particular viewpoint. And there's no real evidence against the standard model of cosmology in the paper - except for the observation that this model doesn't offer a trivial way to calculate some power law relating the masses of stars and gas in a galaxy.

So I fully subscribe to every sentence written by Sean Carroll. It's not a paper that can revolutionize cosmology and the journalists didn't do a good job in reporting it.

Carroll enumerates a couple of huge reasons why MOND is not a good explanation of the data that are otherwise explained by dark matter:
  • MOND predicts a very bad WMAP spectral curve of the microwave background; the agreement of the dark matter model is spectacular, and includes many perfectly reproduced "bumps" which clearly carry much more information than the noisy graphs in McCaugh's paper; with MOND, the agreement becomes very weak
  • dark matter and visible matter don't always share the center; the Bullet Cluster shows that these two chunks of mass may get totally separated; at least all the prevailing versions of MOND make such a separation of "body and soul" impossible, so they're really falsified by this and similar observations
  • MOND doesn't fit clusters; the scaling laws are OK for most galaxies but at a different distance scales, the relationship just fails; most MOND proponents accept that dark matter is needed to explain the motion of clusters - which makes MOND's importance highly questionable because dark matter may explain the galaxies' behavior, too
  • even galaxies may fail; certain small and hard-to-observe galaxies, "dwarf spheroidals", are predicted incorrectly
  • MOND realizations are given by unnatural equations; the addition of vectors and scalar fields, and their bizarre non-polynomial interactions, surely looks awkward. From the particle physics perspectives, the interactions are totally unnatural, and the very fact that one needs to add new field content whose only goal is to modify gravity is very contrived or "man-made"
I would add one more point, and that is the criticism of the MOND proponents' main motivation:
  • it is not natural to assume that all forms of matter and energy must be visible by our eyes, or by electromagnetic waves that we learned to love
The MOND proponents primarily build on the basic philosophy that there's something wrong with a theory that contains elements that are invisible by electromagnetic waves - the main gadget we got used to when observing the real world. I think that this basic philosophical notion - which is the true driving force behind all their model building - is just deeply flawed.

God or Nature didn't guarantee that our eyes He or She created have to be omni-potent. Quite on the contrary, they're almost certainly not. There is no reason - no theorem and no logical argument - that would imply that everything must be visible by the waves we like to use when watching things on the Earth. 

After all, our eyes only see a very small interval of possible frequencies of the electromagnetic waves. But even if we add additional frequencies that may be observed by our telescopes, there's no reason why we should see "everything". We only see the objects that emit the light, and there's absolutely no reason to think that all objects should emit light, or even the same amount of light.

The people who dislike "dark things" seem to think that they're returning us to the era of angels, ghosts, and superstitions. But they don't. Dark matter and dark energy are standard parts of physics because they do have observable consequences. The word "observable" doesn't have to be associated with light or electromagnetic waves.

During its cosmological evolution, Nature just chose what is the right percentage of the things that will be observable 13.7 billion years after the Big Bang. And the answer is that only the gay 4% visible matter is observable. You don't like that the heterosexual 96% majority of the mass in the Universe prefers to hide its intimate places from the lights of your camera? Well, move into another, less puritan Universe. ;-)

Add to Digg this Add to reddit

snail feedback (5) :

reader Anonymous said...

We need to get you an iPad 1.5

reader Andrew Oh-Willeke said...

A narrower way to look at MOND is that it is a simple mathematical relationship that summarizes a very large share of the empirical data on dark matter distributions at the galaxy scale, and has been more successful at predicting new discoveries at that scale than several versions of dark matter theory, all of which, generally, have more moving parts.

If one abandons the original possibility behind MOND that the law of gravity is just plain wrong and see it instead as an empirical approximation of how dark matter acts (a bit like classical thermodynamics), the existence of this empirical relationship doesn't have to mean that dark matter theory is wrong, any more than the great empirical accuracy of Newtonian gravity over a wide range of phenomena means that General Relativity is wrong. The Bullet Cluster evidence is pretty convincing evidence that MOND is not correct at a fundamental level as a true modification of gravity per se. But, the success of this far simpler MOND formula with fewer degrees of freedom at predicting a great deal of data in a great many different kinds of galaxy level phenomea and generating accurate predictions of future experimental results on multiple occassions, means that dark matter distributions in galaxy sized systems have more fewer degrees of freedom in practice than the bare bones structure of a CDM theory permits. In other words, some of the CDM variables related to the distributions of dark matter that we observe in the universe are related or constrained in ways that the existing CDM theory does not reveal in an obvious way.

The strong point of McGaugh's article is that MOND made a prediction about gassy galaxies that other dark matter theories (and there is not just one at this point) got wrong. Similarly, it previously predicted the amount of dark matter effects seen in ellipical galaxies relative to spiral galaxies that the dark matter theories at the time failed to predict.

Any correct theory of dark matter will necessarily closely approximate MOND at the scale of galaxies and should be capable of explaining why the MOND equation is predictive at that scale from first principles. Moreover, any correct theory of dark matter should also explain why the MOND equation that is so powerful at the galaxy level in explaining dark matter distributions with very few empirically determined constants is not good at explaining phenomena at the galactic cluster level. The fact that the dark matter distributions in a CDM theory are so closely tied to the scale of the system in a very particular way is itself an interesting insight. These kinds of insights can guide a theorist when trying to answer questions like "what properties should a non-baryonic WIMP have?", if it is to fit the data.

reader Luboš Motl said...

Dear Andrew, I think that your remarks are pure hype. Could you please be more specific about the fascinating relationship that MOND explains so well?

It's a very narrow set of questions - essentially equivalent questions - that may be parameterized in a simple way, and MOND was (awkwardly) fabricated to match this subset. But it disagrees with other things.

Gassy galaxies happens to belong to this group, and what? A valid theory must predict the whole class correctly. If you need to modify the MOND predictions - e.g. by adding old good dark matter - it is just a falsification of the MOND theory as a replacement of dark matter.

It doesn't matter how much you hype some accidental agreements, by talking - in a Woit-like crackpotish fashion - about someone making predictions. One disagreement and a theory is dead. This is how science works. Whether someone is bold enough and makes predictions, that are sometimes just wrong, can't help you to "forget" a falsification. Please learn it.

This is really not a debate about MOND. This is a debate about what the scientific method is, and you badly misunderstand what it is.


reader Gustav said...

Beauty is in the eye of the beholder.

The issue of MOND versus Dark Matter is not a new one. When anomalous Mercury precession was discovered, the first solution people reached for was to postulate a yet another planet. Einstein resolved the problem differently, by changing the theory of gravity instead--how natural or aesthetically pleasing was that? Perhaps there is a dark matter planet orbiting between Mercury and the Sun after all?

It's a similar story with Dark Matter on the galactic and inter-galactic scales. Unless it can be seen other than gravitationally only, not necessarily by electromagnetic means, but at least in the accelerators... is it really "matter"?

How natural or unnatural are the various terms of MOND? I've seen MOND's derivation from the entropic principle of Verlinde. This looks pretty natural to me. But then, it depends also on how natural you find Verlinde's approach to begin with? I personally find it attractive, because I always thought gravity to be quite different from other interactions that could be explained quantumly--different in so many ways.

Science often works by postulating something phenomenologically first, then seeking to explain it in more depth. MOND is phenomenology. Dark Matter is phenomenology. In both cases we merely add "further terms" in our fits to the observed phenomena. If we find Dark Matter at CERN, great! We'll have a solution to the conundrum, but... not until then.

reader Luboš Motl said...

Dear Gustav,
this is not about beauty, or at least not "just beauty", it's about physics argument.

The difference between the transition from Newton's mechanics to GR, which was needed to explain the Mercury perihelion precession, and the hypothetical transition from GR
to MOND, which some people claim is the right step to explain the galactic rotation curves, is that the first transition ended with a vastly superior, technically more natural theory (GR) that was (much more) compatible with the other pillars of physics (such as special relativity) than its predecessor, while MOND is a technically totally unnatural, man-made, awkward pile of mud that is inconsistent with various stringent enough principles of physics, unlike its natural "predecessor", GR.

Moreover, MOND may be falsified by its disagreement with many other things.

Also, historically, your analogy is absurd because Einstein didn't construct GR in order to explain the Mercury's perihelion precession. He constructed the theory to get rid of inconsistencies between Newton's gravity and special relativity. While doing so, the Mercury perihelion precession turned out to be one of the numerous tiny observational consequences of his newer, superior theory.

Nothing like that is true for MOND. MOND was fabricated and adjusted to explain away the galactic rotation anomalies and it has absolutely no other positive evidence that it is true. There is no analogy of the "compatibility of GR with SR" and other things.

If you don't see those totally fundamental differences between the two cases, then you simply have no clue about physics. You don't understand its logic. You don't understand what it means to have an argument or not to have an argument. This is not in the eye of the beholder. Physics is an objective science, not a postmodern demagogy that depends on eyes of the beholders.


(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');