Dark Matter: A debate – afterwards while on safari
Fleeing the European continent to go back to Australia on safari for some satellite-galaxy hunting in Canberra with my friend Dr. Helmut Jerjen, I had a little time on my Quantas flight and in Singapore and Perth to reflect upon the debate, and I note the following:
Simon White gave an excellent presentation of the impressive agreement of standard cosmology, i.e. the LCDM model, showing some of the available data on large scales and the cosmic background radiation map (his slides are available on his website). Somebody in the audience during or after the debate was over, mentioned an interesting observation (unfortunately I do not recall who this was):
In order to get Einstein's theory of general relativity to fit the data one needs to postulate unknown physics, namely inflation, dark matter, dark energy.
(This I had indeed stressed in my presentation, therewith putting the LCDM success story implied by Simon on a different if not dis-satisfying footing.)
But, the unknown person continued:
Can one then, after introducing these unknowns to make the theory fit, argue that the LCDM model is correct because it fits the data? Is this not a circular argument?
Perhaps the LCDM model finds support in that various different lines of argument lead to similar values for the numbers which define the precise model. But, Mr. Unknown has raised a point central to how science advances:
Rather than demonstrating how excellently LCDM does on large scales, a cleaner argument that the LCDM model describes physical reality is as follows:
Accept that the LCDM model is adjusted to fit the data on large scales. Once it is fixed, it can be used to make predictions in a different regime. This different regime is on scales smaller than 8 Mpc, where the model makes very precise predictions how the cold dark matter must be distributed for it to be a valid description of nature and where we have truly exquisite observational data. This distribution is seen in the form of galaxies and how they cluster. And this is where the observational data, unfortunately, are in highly significant conflict with the model such that they exclude the LCDM model. This holds true despite the often invoked uncertainties and complexities in dealing with the physics of normal matter (for example, observations clearly tell us that galaxies are simple objects obeying simple scaling laws such that true physics describing their structure must be simple as stressed by Disney et la., 2008, Nature).
For example, in the debate after the two presentations, Simon White attempted to address the small scale problem by showing an excellent fit of the LCDM model to one of our satellite galaxies, namely Fornax. Fornax is far away, at 140 kpc, so far in fact, that it must be in dynamical equilibrium. Dynamical equilibrium means that the stars are orbiting within the galaxy such that the whole galaxy is not changing its appearance. Thus, when a star moves to the right, another one moves to the left such that they compensate each other statistically. So getting a good description of it with the LCDM model would indeed, as Simon White stresses, be a great success.
However, this is wrong, and it was surprising that Simon White did not note this. Indeed it is also surprising that I did not jump at this logical inconsistency, perhaps because it was such a self-evident failure that I dared not point this out in fear of causing an unpleasant situation. The LCDM fit is unphysical because Fornax has a complex inner structure: As is evident on slide 49 of my presentation, Fornax has a twisted and dislocated inner structure, such that it simply cannot be in dynamical equilibrium. Dynamical equilibrium is, however, one of the fundamentally important assumptions that go into modeling the data via the LCDM model. In the LCDM model, any complex structure would disappear on a short time-scale of about a hundred million years, as I indeed had put much emphasis on during my presentation. That is, the appearance of the little galaxy would be changing significantly on this astronomically short time-scale. Simon White must have missed this point, or simply ignored it.
Unfortunately, Simon did not address all the other failures I had put up, nor did I return to them during the debate - well, they had been stated in my presentation already. But, given the astrophysical literature, it is evident that there are no remedies to save the LCDM model, given its current ingredients.
Now, one attempt to advance from here is to add an additional Dark Unknown, a Dark Force which acts only between cold dark matter particles, as discussed by Peebles & Nusser (2010, Nature), or an additional Dark Force which acts only between dark matter and normal matter, as discussed by Kroupa et al. (2010, Astron. & Astrophys). These speculative forces, about which we know absolutely nothing, are none of the other three already known forces (electromagnetic, weak and strong) nor Einsteinian gravity.
Therefore, the failure of LCDM on scales smaller than 8 Mpc is due to it being wrong. Note here that it is wrong even if one adds the above dark forces, since with these forces the LCDM model becomes a different one with different properties on large scales. This is where MOND or another alternative (e.g. MOG) comes in. MOND is not a dark force, but merely a simple modification of either gravity or inertial mass, depending on its interpretation.
It is remarkable how brilliantly MOND has been performing since its conception in 1983 by Mordehai Milgrom. In fact, in his most recent paper, Milgrom ("MD or DM? Modified dynamics at low accelerations vs dark matter", 2010, Proceedings of Science) writes in his abstract:
Some of the complaints leveled at MOND are: (i) "MOND was designed to fit rotation curves; so no wonder it is so successful in predicting them". This is both incorrect and quibbling: The first ever MOND rotation curve analysis was undertaken more then four years after the advent of MOND. And, even if MOND, epitomized by a very simple formula, could have been designed to predict hundreds of rotation curves, it would still be a great achievement. (ii) "MOND outperforms CDM only on small, galactic scales, where formation physics is anyhow very messy, but falls behind in accounting for `simpler', large-scale phenomena". Quite contrarily, all the salient MOND predictions on galactic scales follow as unavoidable, simple, and immediate corollaries of the theory - independent of any messy formation scenario - just as Kepler's laws, obeyed by all planetary systems, follow from an underlying theory, not from complex formation scenarios. To think, as dark-matter advocates say they do, that the universal MOND regularities exhibited by galaxies will one day be shown to somehow follow from complex formation processes, is, to my mind, a delusion. What is left for MOND to explain on large scales is a little in comparison, and has to await a full fledged relativistic MOND theory. (iii) "The `bullet cluster' shows that MOND still requires some matter that is dark". Yes, it has long been known that MOND does not fully remove the mass discrepancy in the cores of galaxy clusters. Some additional still-dark matter is needed. But this need not be THE "dark matter"; a small amount of the still-missing baryons, in some dark form (dead stars? cold gas clouds?), or perhaps (sterile?) neutrinos, could fit the bill.
Finally it serves to be useful to note the following statement from the paper by Peebles & Nusser ("Nearby galaxies as pointers to a better theory of cosmic evolution", 2010, Nature, p.568):
The variety of problems we have considered in the interpretation of the present baseline motivates serious consideration of adjustments of the fundamental theory.
Prof. Jim Peebles at Princeton University is one of the leading cosmologists who had actively worked in developing the LCDM model.
The 18th November was very special. Not only because we had such a debate in Germany, which is otherwise overall well on-track with the LCDM model with most major professorships having been filled with its adherents. More importanly, the generally well educated public in Germany is interested, and for me having so many cameras around was a new experience which clearly had an effect on how the scientific debate proceeded. I would like to sincerely thank Prof. Gerhardt Hensler from Vienna, Prof. Robert Sanders from Groningen and Prof. Tom Shanks from Durham for following my call to join-in with the debate. Prof. Hensler is an expert on star-formation and gas-dynamical processes in galaxies. Prof. Sanders is an expert on gravitational dynamics and the astrophyics of galaxies. Prof. Shanks is an expert on observtional cosmology andextragalactic astronomy. The presence at the debate of my long-term supporting colleague Prof. Klaas S. de Boer from Bonn was also central. Their expertise was essential during the debate, and their active participation also demonstrates that here are a substanial number of scientists who see major problems with the LCDM model, such that, with adequate funding support, significant progress in cosmology can be hoped for.
by Pavel Kroupa and Marcel Pawlowski (27.11.2010): "Dark Matter: a debate - afterwards while on safari" in "The Dark Matter Crisis - the rise and fall of a cosmological hypothesis" on SciLogs. Written in Perth. See the overview of topics in The Dark Matter Crisis.