Milgrom's perspective on the Bullet Cluster

I have been asked by several people to comment on these claims regarding the "bullet" cluster. This short write up tries to answer all of you.

I think it is an over-dramatic statement.

They seem to insinuate that they have proven the DM doctrine. Dark matter is supposed to be an omnipresent entity in galaxies, groups, clusters, etc.. If they found something it is that there is some matter in clusters that we haven't yet seen directly; then they generalize to the universe at large, as if this is a proof of the general doctrine. Far from it. And, as I explain below, it doesn't undermine the motivation for MOND in any way.

I am not really clear why this new outburst. Exactly these claims using the same system have already been published almost three years ago in more then one paper (e.g., ). So we have had plenty of time to digest the matter, to discuss it at conferences, and to let the authors know what we think, but they don't seem to listen.

Anyway, the fact is that they don't add anything to what we knew already for years about MOND and DM. We have known for some fifteen years now that MOND does not fully explain away the mass discrepancy in galaxy clusters. (See e.g. the 1999 paper by Sanders:, but there have been quite a few others discussing this before and after, starting from 1988). Even after correcting with MOND you still need in the cluster some yet undetected matter in roughly the same amount as that of the visible matter. Call it dark matter if you wish, but we think it is simply some standard matter in some form that has not been detected. It could easily be in the form of dim stars or cold gas clouds (or, some people suggested neutrinos). The thing is that you do not need much of it, only about as much as the already visible matter in the cluster. In other words, for the cluster globally where the total mass discrepancy is about a factor 10 say, MOND would correct only by a factor 5 roughly, leaving still factor 2 discrepancy. The mass balance of the cluster is made up of stars in galaxies, say of total mass M*, about 5-10 times as much in the form of x-ray emitting gas, say of total mass Mg. If you believe in DM you need about 10 times Mg in the cluster at large. With MOND you need only about as much as Mg in a still undetected form. However, it is not distributed like the gas, but rather more like the galaxies so it is more centrally concentrated. Mind you, in galaxies (in general not inside clusters) the measured global discrepancies in the outskirts have reached a factor of 50-100 and is accounted in full by MOND. So we have to say that we are left with this corner of the universe (the cores of clusters) where we haven't yet detected everything. Now, this situation certainly does not undermine the cause for MOND. The cause for MOND is based on the fact that it has predicted with uncanny accuracy the full dynamics in over a hundred galaxies without DM, and even in cluster at large it removes a large part of the discrepancy. The fact there is still to be detected some normal matter in the universe is not really alarming. Anyway, this was the situation based on analysis of many isolated clusters to date. What these people find is exactly what is expected from the above. In fact, it could have been damaging or at least puzzling for MOND had they not found what they did: When two clusters collide head on the gas components of the two just stick together and stay in the middle, while the rest (galaxies plus this extra component I spoke of) just go through and stay together. So it's an interesting and informative result, but it is totally expected both in the DM picture and on the basis of what we know about MOND. I should also mention a recent paper claiming that MOND can actually explain these "bullet" observations without this extra matter (see; but I don't think this is necessary. We know we need some more matter in clusters than we have seen so far. Lastly, I think the authors labor under some misconception about MOND. They don't discuss their argument in detail but they say something to the following effect: "We found that most of the baryons (standard matter) are concentrated at the center but we see unaccounted-for mass concentrations flanking the center, where the galaxies are (which make up a smaller fraction of the observed matter). But MOND (or other modified gravity theories) predict that the "DM" should be found around the visible matter not elsewhere"; so they say. However, this last statement is incorrect. For example, in galaxies the baryons are concentrated at small radii but the putative DM according to MOND is far beyond that. But again, I don't think we have to resort to this. As I said we do know that there is some (still) dark matter in clusters and that's what they found.

All the best, Moti Milgrom

PS, The accepted picture has it that about 4 percent of the mass in the universe is in the form of baryons (standard matter). Of this we have so far seen (detected) only a tenth. So we know anyway that there is still much standard matter in the universe to be detected (beside the putative DM). What MOND still requires in clusters is a small fraction of that, so really it's no big deal.

Return to MOND home page