[knzhou]

One should keep in mind that most dark matter "alternatives", including this one, actually include dark matter. It says so right on the 2nd page of their paper:

> Consider requirement (iii), that is, successful cosmology. In (2) we have a new d.o.f. φ [...] What should the expectation for a cosmological evolution of ϕ be? The MOND law for galaxies is silent regarding this matter. There is, however, another empirical law which concerns cosmology: the existence of sizable amounts of energy density scaling precisely as a^(−3).

In other words, they are saying that to get the cosmology right, they need to add stuff that behaves exactly like dark matter -- that is what they are alluding to with the "sizable amounts of energy". They make their φ field play this role. It's just like TeVeS, the other major relativistic MOND theory, where the scalar "S" field does the same thing.

The popular press likes to frame the debate as "dark matter vs. modified gravity", but it's really "dark matter vs. dark matter plus modified gravity", which is much less dramatic.

[dnautics]

is that right? You're referring to the fact that in general in modern physics a "field" implies a field-carrier particle. Lambda-CDM dark matter theories specifically posit that various astronomical discrepencies can be explained using particles that only interact gravitationally, with a huge number of degrees of freedom (e.g. the difference between the bullet cluster and the milky way alone represents a LARGE set of degrees of freedom).

This seems rather different from a "single" field which may have a particle (that may or may not itself interact gravitationally), with only one additional degree of freedom.

[throwaway894345]

> The popular press likes to frame the debate as "dark matter vs. modified gravity", but it's really "dark matter vs. dark matter plus modified gravity", which is much less dramatic.

Honestly for us lay folks there isn't a perceptible difference in the amount of drama between the two. :)

[raattgift]

> We remark that A_\mu also contains a pure vector mode perturbation which is expected to behave similarly as in the Einstein-Æther theory [90, 91]"

Their [91] is Jacobson & Mattingly https://arxiv.org/abs/gr-qc/0007031 whose §VII (DISCUSSION) contains this, which I struggle to see as helpful for them: "With the action adopted in this paper the aether vector generically develops gradient singularities even when the metric is perfectly regular. We take this as a sign that the theory is unphysical as an effective theory". (That doesn't stop Jacobson from investigating things like (time-independent) black hole solutions https://arxiv.org/abs/gr-qc/0604088 "It is a plausible conjecture that nonsingular spherically symmetric initial data will evolve to one of the regular black holes whose existence has been demonstrated here, but this has certainly not been shown", and worse they show that the aether does not obey the Raychaudhri equation, so the relativistic MOND authors seem to need more ghosts).

For the life of me, I can't figure out the relevance of their reference [90] which I believe is https://www.jstor.org/stable/2414316

I wonder who their Reviewer 2 was.

[dogma1138]

The question is how much dark matter is required and if it’s little enough to be filled by baryonic dark matter candidates.

[techdragon]

And given how hard it has been to find the "Dark Matter", theories that reduce the amount of it seem like valuable contributions to the overall understanding. Dark matter has so many "if its like ... then ..." scenarios that theories like this are effectively "working backwards" on the problem by giving us better constraints on the "then ... " part.