Torbjorn_L said:
That MOND is ad hoc makes it even more problematic than physically motivated "modified gravity". It can predict galaxy behavior superficially naturally since it is a curve fit.
Why?
It necessarily follows that if MOND is a good fit over the entire range of dark matter phenomena up to elliptical galaxies with a single parameter, that any actually correct theory must closely reproduce this result and have approximately the same relationships of the variables and same numbers of degrees of freedom for all dark matter phenomena at these scales. If a dark matter or modified gravity theory doesn't reproduce MOND at galactic scales, it is necessarily wrong.
Any additional degrees of freedom and dependency on other factors than those found in the MOND formula must necessarily have negligible phenomenological impact at galactic scales. And, since MOND doesn't work at cluster scales, any correct theory must have significant scale dependence that becomes significant at some point between the ellipical galaxy scale and the small galactic cluster scale.
These are pretty strong hints regarding what kind of theory scientists should look for to explain dark matter phenomena.
Also keep in mind - MOND was developed with a curve fit to one narrow subtype of galaxies and only much later found to be accurately in other circumstances. And, there is nothing at all intuitive about the fact that it is possible to fit the entire range of dark matter phenomena in all galaxies with just one experimentally set parameter. No dark matter theories out there do that, and certainly no theory explaining dark matter at the time that MOND was proposed did that.
But it - like all other modified gravity theories - doesn't make sense, since then it can't predict galaxy collisions (or other gravity effects).
Hence my careful statement that it works in its domain of applicability. Newtonian gravity is just great for all sorts of purposes, even though it doesn't predict black holes or frame dragging or that light can be bent by gravity. MOND works everywhere that Newtonian gravity does and also accurately reproduces observation at the galactic scale where Newtonian gravity does not.
Similarly, the fact that the proton-neutron-electron model of the atom omits many of the aspects of the Standard Model that have been experimentally confirmed, doesn't mean that it isn't perfectly useful in the domain of doing chemistry. The perturbative QCD approximates the Standard Model and experimental results well in the ultra-violet, but fails dismally in the infrared, but it is still used every day at the LHC.
The fact that the initial version of MOND from he early 1980s is a non-relativistic toy model with a limited domain of applicability, doesn't mean that it is not a worthwhile theory to understand and apply in the appropriate circumstances.
There is no physics continuity, except an extraordinary claim without evidence - "just because". Morally, they fail before they start.
MOND has all sorts of evidence in the form of galactic behavior in systems where there was no observational evidence before it was proposed that have matched its predictions and which were not predicted by dark matter models.
Your complaint is not really that it is an extraordinary claim without evidence. Your really beef is that a
mechanism by which the proposed law comes into being is not provided. But, this isn't any different than other aspects of physics, like non-local entanglement effects, which are empirically demonstrated to happen and are accurately described by equations, but which our primate brains evolved to have different intuitions about the physical world derived from the problems we faces as hunting and gathering primates have a hard time accepting as real.
Lots of physics is "just because" notwithstanding perfectly good reasons why Nature (the jerk, not the journal) shouldn't behave as it does. The strong force CP violation parameter should "naturally" be 1 and not 0. The mass constants and mixing angles of the Standard Model take the values they do (in the Standard Model) just because. There are three kinds of quark color charges and eight color charge variants of gluons because that is what we observe, even though it is perfectly possible to imagine physics (and indeed Lattice QCD theorists routinely do imagine physics to extrapolate to the physical parameters) with a different number. It is "natural" for there to be SUSY sparticles at the TeV scale, but that isn't what we see.
Nature has no obligation to act in a manner we find to be "moral" and a theory that efficiently reproduces observations in its domain of applicability is perfectly respectable and adds insight, even when it doesn't work outside of its domain of applicability.
With latest standard cosmology models, they do all that and predict galaxy emergence and detailed structure to boot, from first principles. [
http://www.illustris-project.org/ ]
The problem, as noted above, is that the standard cosmology model, lamda CDM, does no such thing. It doesn't reproduce the correct number of satellite galaxies (it predicts too many). It does not correctly reproduce the shape of dark matter halos inferred from observations of galaxies. It predicts more inferred dark matter subhalo structure than is observed. It does not correctly predict the ansiotropic distribution of satellite galaxies around central galaxies. It does not get the timing of galaxies with modern levels of metallacity right. It predicts that dark matter should be possible to detect directly in circumstances where it has been experimentally excluded by direct detection experiments. It predicts a cross-section of interaction between dark matter particles at odds with what was observed in the bullet cluster (the lamda CDM cross-section would be lower).
At scales larger than central galaxies with satellite galaxies, lamda CDM does a good job. But, this doesn't mean that this model isn't seriously flawed. Indeed, the domain of applicability of lamda CDM and the domain of applicability of MOND are almost completely disjoint and between the two cover the entire realm of observed phenomena. The fact that the mechanisms in each case are completely different strongly suggests that there is a major conceptual leap that has been missed somewhere and that both theories are wrong, not just MOND.
MOND is competed away long since, it is fringe. The OT article shows that, it is attempting to tear down current observations with the erroneous idea that 'hence MOND'. Cue pseudosciences, like 'design' creationism.
I'm not sure failed ideas fits on PhysicsForum as such. Maybe if there is a historical discussion?
Lots of MOND papers are produced every year. Probably as many as Loop Quantum Gravity or any particular inflation theory. It is not merely a matter of historical interest, because the idea of explaining dark matter phenomena by tweaking the laws of gravity rather than inventing new particles and force carrying bosons continues to be a viable approach to investigate. All serious MOND proponents admit that the original model is flawed, but that doesn't mean that the generally theoretical approach of looking at modifications of gravity laws that could reproduce observational evidence, as opposed to the dark matter paradigm is discredited.
UPDATE: Who added the function that censors "bad words"?
Or maybe not.
