We don't know what dark matter is, or if it even is a kind of matter rather than e.g. modified gravity (though modified gravity has many problems of its own), but I think it is a common mistake that people are so eager to associate it with aether on the grounds only of what they find 'elegant'.
Edit: Also, the notion of a kind of matter that can hardly be observed is not particularly weird or 'arcane'. For instance, neutrinos have many of the right properties. I think they are ruled out as dark matter candidates mainly because their mass is too low to fit with observations.
Is it? As I understand it the concept of an aether is not so much provably wrong as utterly meaningless; once you factor in various dilations there is no effect that would be caused by an aether. Nothing could possibly measure it, and it has no impact on predictions.
>Many aspects of Lorentz's theory were incorporated into special relativity (SR) with the works of Albert Einstein and Hermann Minkowski.
>The introduction of length contraction and time dilation for all phenomena in a "preferred" frame of reference, which plays the role of Lorentz's immobile aether, leads to the complete Lorentz transformation (see the Robertson–Mansouri–Sexl test theory as an example). Because the same mathematical formalism occurs in both, it is not possible to distinguish between LET and SR by experiment.
Are you thinking of the Michelson-Morley experiment? Interestingly, even after the development of special relativity, Lorentz and Einstein [1] continued to discuss the possibility of an ether. Dirac also discussed the topic in 1951 [2]. The fact that it continued to be discussed by these scientists has always made me wonder whether the idea would ever return.
[1]: http://en.m.wikisource.org/wiki/Ether_and_the_Theory_of_Rela...
[2]: http://www.researchgate.net/publictopics.PublicPostFileLoade...
That the Aether does not exist is only the simplest explanation (by a huge margin), but there are infinite possible theories that satisfy the experiments. Also, as the anisotropy is the only relevant characteristic of the Aether, thus most people prefer to consider the case settled.
As far as special relativity goes, the aether ends up being equivalent to a having preferred rest frame. It's not wrong, but it's unnecessary. You still have the light speed limit and length contraction and time dilation and all the same math (which is an unfortunately post-hoc insight), they just take on a different philosophical character. PhysicsForums has a good summary [1]:
> Both BU and LET use the Lorentz transform, etc., to make all of their experimental predictions, and therefore they are scientifically indistinguishable, making the same experimental predictions in all cases. Because of this experimental equivalence, there is little if any serious ongoing debate between the two in professional physics circles (although the philosophy literature does have ongoing debate).
For me the intuition pump for BU=LET was reading The Strange World of Classical Physics [2]. Notice that, by the end, as more things are considered, it's sounding more and more like relativity...
All that being said, I agree with the PhysicsForums policy. It's a pointless debate. Switching between BU and LET might give you some intuitive understanding, but day to day you end up using special relativity because it's a lot more mathematically elegant. Plus I have no idea how you'd make the aether work with general relativity.
1: https://www.physicsforums.com/threads/what-is-the-pfs-policy...
The whole thing seems ludicrous -- but for the problem that no better explanation has yet been put forth. One gets the feeling though that one day in the future 'dark matter' is gonna look a lot like 'epicycles'.
One might more generously say that dark matter is a proposed explanation for many measurements deviating from predictions based on visible matter in a way very consistent with the existence of large amounts of non-electromagnetically-interacting matter.
I don't think there's anything fundamentally ridiculous about this. For instance, while you characterise it as 'radically' different to any matter that we observe directly, the neutrino has most of the relevant properties, and e.g. supersymmetry has been noted to include dark matter candidate particles. While one may doubt that supersymmetry is actually correct, I think it demonstrates that the combination of properties (massive + electromagnetically inert + maybe weak force interacting) isn't necessarily a problem.
Also there's apparently information about the universe's composition to be gleaned from the CMB anisotropies; its all somewhat beyond me, but that curve certainty seems to fit extremely well, I'll give them that.
http://www.astro.ucla.edu/~wright/CMB-DT.html
The combination of all this stuff is apparently pretty convincing.
EDIT: I see that your comment is actually referring to the bullet cluster more than the rotation curves. It's all pretty tenuous in the grand scheme of things, I agree, but then so much of cosmology is. I guess they have the benefit of the string people next door making them look reasonable by comparison.
It wasn't until Kepler proposed elliptical orbits at the beginning of the 17th century that heliocentrism actually made good mathematical sense though - Copernicus' perfectly circular orbits still made use of epicycles to explain deviations between calculation an observation.
From the mid 15th century when Copernicus first proposed heliocentrism, to the mid 17th century when Kepler's theories became widely accepted is a long time though.
I want to believe society has advanced past needing centuries to advance from a possibly superior model that still has some issues, to an obviously superior model, but the way seemingly intelligent people cling dogmatically to things like Pluto being a planet, and defining a kilobyte as 1024 bytes because kibibytes 'looks dumb' gives me pause though.
I'm not surprised; humans always keep searching for more elegant models. There seems to be something in the brain that associates positive emotion with models that compress information well.
A mind completely incapable of dealing with the universe would have been selected against waaaaay back.
Dark Matter is the Phlogiston
of Contemporary Cosmology
I'm looking forward to new theories that can explain what we now observe as "dark matter" without having to invent a new kind of matter specifically for it!
Carver Mead is a very famous guy who also wants to shake things up[1]. Maybe he's turned into a crank in his old age, but, to steal from Ghostbusters: "the usual stuff isn't working".
[1] http://www.theregister.co.uk/2013/02/20/carver_mead_on_the_f...
So alternative theories like MOND may explain galaxies but have nothing to say about the big bang.
The potential existence of dark matter candidates, which some suspect to be sparticles predicted by Supersymmetry (maybe the LHC throws something up there or, as always, it's just the next collider away), is another reason to pursue dark matter ideas.