Critique of Mainstream Cosmology

  1. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    From the Review of Mainstream Cosmology thread.
    Here I have started just such a thread.

    This thread is not the place to discuss a multitude of speculative ideas, rather it is intended as a discussion of observations that may raise questions about the consensus [itex]\Lambda[/itex]CDM model.

    I start with a discussion of Lieu's eprint on today's physics ArXive. LCDM cosmology: how much suppression of credible evidence, and does the model really lead its competitors, using all evidence?
    Can this claim be substantiated?

    Garth
     
    Last edited: May 18, 2007
  2. jcsd
  3. Wallace

    Wallace 1,253
    Science Advisor

    Looks like we overlapped posts Garth, your thread is a better approach, I'll edit the other one to send people here.

    This should be a good discussion, looking forward to it!
     
  4. Wallace

    Wallace 1,253
    Science Advisor

    To get the discussion going a bit, I'd have to say the Lieu paper is over the top and goes to far. However, in amongst the rant there are some important and in my opinion very valid and interesting points. There are observations that are troublesome to LCDM (and to wCDM more generally) and these clearly don't get the attention they deserve.

    Lieu suggests two alternative models are almost as supported by the data as LCDM, on this point he is talking rubbish. The table in his own paper makes in pretty clear to start with that LCDM is far better supported. However science can be much more precise than his crude binary table. It is not enough to say that a model is or is not supported by the data, we need to know how certain we are about it. It's a small field that we call statistics, and when we confront the models to the data and run the number LCDM wins by a huge margin.

    The reason the LCDM is the dominant theory is that if you run it by all the evidence it gives a good (but not perfect) fit to (almost) all the available data. Compared to any other suggested model LCDM wins hands down.

    The most important point Lieu makes though is that just because it is the best model, does not mean it is right. I would have to agree that cosmologists do frequently go too far in their hubris over the wonderful achievement of the concordant model.

    There is clearly far more evidence in favour of LCDM than against it, but if the evidence against is conveniently ignored then the danger is that it will drive an over-reaction against the standard model, which is fundamentally what I think this paper is. LCDM, warts and all is still impressively robust to the data, and trying to dress it up to be even better is probably a mistake.

    One final comment is that the purpose of Lieu's paper fundamentally seems to be about money. He wants some of the big cash that is heading towards 'dark energy' projects to come his way instead (okay and to some others as well). However what I think he doesn't realize, or at least doesn't acknowledge, is that almost all so called 'dark energy' missions will generate a wealth of data that is highly relevant whatever the theoretical basis of cosmology is. So if someone came up with a great theory tomorrow that seemed to explain the data better than LCDM and maybe did away with the nasty unknowns of dark matter and or dark energy, then the big budget dark energy missions like Plank and SNAP would still be just as relevant to probing the details of this new theory.

    If LCDM is wrong then we are even more in the dark about our Universe than if it is right. To me, uncertainty over the validity of LCDM is more of a reason to go out and test it rather than a reason not to bother! We have a theory of our universe that contains at least 3 very uncertain elements (inflation,DM & DE/CC). However, despite the fact that the theory is so bizarre, it is what the majority of the evidence points to! Therefore we need to do two things, one is as Lieu suggests and look more carefully at the minority of contrary evidence to see how significant we can push the disagreement between theory and observation, but the second thing is that contrary to Lieu's suggestion we really do need to push on with the 'mainstream' big surveys to see how far we can push the agreement, and in the case of DE, nail down it's properties to find what true micro-physical theories for it might be reasonable.
     
    Last edited: May 18, 2007
  5. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    Wallace I cannot agree more! :smile:

    Lieu's paper makes three claims, each of which will bear some discussion.

    1. Observations have been 'cherry-picked' to support the mainstream model. (Section 4 - page 7)

    2. Other observations exist that question that model yet are ignored and even suppressed by the community. (Abstract- page 1)

    3. Funding is channelled into mainstream projects, thus alternative models remain untested and 'on the shelf'. (Conclusion - page 13)

    We may wish to deal with these one by one. Certainly the nub of the matter is 2, that is whether there are, actually, observations that contradict the mainstream model.

    Garth
     
    Last edited: May 18, 2007
  6. marcus

    marcus 24,908
    Science Advisor
    Gold Member
    2014 Award

    I like this statement a lot. It has the right balance.
    GR must be wrong (since it has singularities and is at best only an effective classical theory) but historically the alternatives have tended to be worse.

    LCDM based on GR must be wrong as well (behold its great warts! as Wallace points out) but it is a beautiful fit in many many ways and all the alternatives so far seem to be worse.

    It is an exciting moment in history to be watching because an improved model cosmology MUST emerge (with LCDM being both so wrong and at the same time so right) and the most strenuous efforts to check LCDM can only help this process happen.
     
    Last edited: May 18, 2007
  7. Wallace

    Wallace 1,253
    Science Advisor

    Is Newtonian gravity wrong? No, just valid in certain regimes. The same is almost certainly true of GR. The LCDM is based on GR, but only in the regime in which we think GR is valid. We don't encounter singularities in the use of GR to define the LCDM model with a FRW metric (Since we only describe t>0 not attempt to talk about t=0 in the LCDM model).

    Before you dismiss LCDM as being 'so wrong' consider the evidence against. Lieu mentions two things only, soft X-ray emission from clusters and the lack of SZ effect in WMAP. The first of these is highly dependent on how clusters are modeled (in terms of density profile etc) , and the second is very hard to do observationally and again requires a lot of modeling of clusters that is uncertain.

    You can, as Lieu points out, find papers in the literature that do not find any conflict with LCDM for every piece of evidence he claims refutes LCDM. All he s saying is that there are also papers (mainly his) that arrive at different conclusions about these effects.

    So before you gleefully dismiss the mountain of surprisingly concordant data, realize that their is no uncontroversial evidence against LCDM. The only thing I get from Lieu, as expressed in my previous post, is that given these uncertainties in areas that look like they may be troublesome to LCDM, we should look at them more closely and give greater acknowledgment to the disagreement between different research groups.

    There is only one reason the reasonably abandon the LCDM model to the scrap heap at this time. Theoretical prejudice. The data, such that it is, does not support this move. Remember that Lieu has been writing anti-LCDM papers for years and as is guilty of cherry picking results to support his preconceived position as anyone. Is it just as likely the Lieu would suddenly find some new data or method of his supports LCDM as it is that Dave Spergel would do the reverse (in both cases somewhat slim one would think).
     
  8. marcus

    marcus 24,908
    Science Advisor
    Gold Member
    2014 Award

    my post was intended as approval and corroboration of the one of yours that preceded it.

    I am consistently a proponent and expounder of LCDM, which I contemplate with delight and with amazement that it fits so much data so well.

    Indeed I suspect there is little differerence between our positions on that score.

    I think your analogy with Newtonian gravity is very apt. Newton gravity is elegant and amazingly accurate in a certain regime. What is exciting is when one comes to a point in history where it is realized that such a successful theory is not fundamental and is only an effective theory that works in a certain domain of applicability. Then one can hope to see a more fundamental theory emerge.

    the warts or flaws in LCDM that I referred to are the ones YOU mentioned.
    the flaws in GR which I mentioned are wellknown, and do it no dishonor.
    thank goodness GR is wrong, so we shall have the privilege to see a more fundamental theory of gravity appear (probably duplicating GR to a large extent except at extreme scale or near classical singularities)

    Let neither of us accuse the other of "gleefully dismissing" either GR or LCDM:smile:

    Pax

    ==============
    BTW I found Lieu's paper yesterday and had a look at it, but didn't think it so good. I liked Simon White's paper of a few weeks ago but Lieu paper
    seemed more like sorehead to me----potentially serious issues but aggravated by an exaggerated presentation. So I decided NOT to post a link here about the paper, or start a thread about it. The next morning I saw that both you and Garth chose to start threads. You may be right---the paper may merit discussion. I'm not yet convinced.
     
    Last edited: May 18, 2007
  9. marcus

    marcus 24,908
    Science Advisor
    Gold Member
    2014 Award

    I doubt it can be substantiated. LCDM is the best model going and testing its limits with more and more observation is the traditional way to do science.

    Besides, a lot of research has been and is being devoted to checking how close w is to -1. That is, to ruling out some kind of dark energy that is not equivalent to a cosmo. constant.
    and a lot has been devoted to checking that Lambda is CONSTANT. so people are always probing around the margins, and checking for variants of standard LCDM

    My feeling is Lieu paper is not very appetizing and I have no urge to discuss it.

    In your thread you broach a more general topic. Basically what are the observations that show us where LCDM might be weak, and possibly subject to improvement?

    That's very interesting and well worth discussing. I hope you get knowledgeable people to contribute.

    About the MONEY question---I think the direction of funding has been great and that the LCDM has been an organizing principle in identifying good research and observational strategies.
    Lieu, when he complains about bad direction of the funding, sounds like a sorehead IMHO. I think they've done a great job and have launched and set up great space and groundbased instruments.
    And the signs are that observational astro is on a roll and this will continue.

    So Lieu is so off-base I dont even see a need to talk about his paper.
    Unless it starts being influential, which I don't expect.

    Garth, this is what I LIKE about your original post, and it doesnt need to have much to do with Lieu's paper. Good question to discuss on its own:

     
    Last edited: May 18, 2007
  10. Nereid

    Nereid 4,014
    Staff Emeritus
    Science Advisor
    Gold Member

    It seems to me that Lieu makes a mistake that can be seen rather often in 'agin the mainstream' public pronouncements - that effort devoted to testing a reasonably well-established theory could be better spent testing an alternative theory ... despite the fact that the alternative is demonstrably inferior, in terms of its match with relevant observational results*.

    As Wallace has already noted, a great many of the proposed tests of LCDM models should serve as equally good tests of many, if not most, alternatives.

    Further, many proposed tests, if properly conducted, should show flaws in LCDM models in ever harsher light.

    And the astonishingly rich trove of high quality astronomical observations, available to anyone with a broadband internet connection, surely presents any developer of alternative theories with fantastic opportunities.

    It gets worse, for Lieu and any others trying to make a similar case ... the very lack of control over what we observe is also a reason why effort will continue to be devoted to astronomy's many facets - soft x-ray emission from clusters and groups is of considerable (astrophysical) interest, irrespective of any cosmological implications it might (or might not) have; getting a good handle on the nature of irregular dwarf satellite galaxies is something many astronomers will want to do, no matter what role DM (or GR, or ...) may play on the cosmological stage.

    It's not a zero sum game.

    *I'm turning up the contrast of course.
     
  11. Chronos

    Chronos 10,127
    Science Advisor
    Gold Member

    Observables still trump alternative theories. It is the one thing LCDM has going for it that the others do not. Naysaying does not replace observational evidence.
     
  12. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    As far as claim 1. is concerned (Observations have been 'cherry-picked' to support the mainstream model), we will have to see what other observations there are that might have been ignored.

    As far as claim 3 is concerned (Funding is channelled into mainstream projects, thus alternative models remain untested and 'on the shelf') - a claim made by Arp even though in the 1970's he had been given an inordinate amount of telescope time to investigate his conjecture that quasar red shifts were not cosmological, as Nereid has pointed out many of the tests of GR and the standard model will also test the alternatives to them, as Gravity Probe B is doing at this moment.

    However my own work has thrown up a question about the equivalence principle that could be tested: the question, "Do photons 'fall at the same rate' as free-falling particles?" A question I think that is as equally important as the Eötvös-type experiments yet that has not been tested in the laboratory. (The deflection of light by the Sun is not a clean test as the free-falling component is convoluted with the space-curvature component.) Now where may I get funds to put this to the test.......

    Which leaves us with claim 2 - are there observations that contradict the mainstream model?

    Lieu includes a selection of seven out of what he calls a "long list of counter evidence", however, as Wallace pointed out, only two have any substance; soft X-ray emission from clusters and the lack of SZ effect in WMAP, and it is highly problematic whether they actually provide "counter evidence".

    I have my own list of questions for the [itex]\Lambda[/itex]CDM model, both cosmological and questions about the GR theory on which it is based.

    The overall question of course is the the fact that the inflaton particle required by Inflation, the DM particle and DE itself have not yet been discovered in the 'laboratory' even after three decades of research. This together with the apparent difficulty in producing a quantum gravity theory leads one to understand we are missing something.

    Cosmologically I ask the following questions:
    1. Are SNe Ia Standard Candles?
    2. Is there an Age Problem in the Mainstream Model?
    3. Are the Cosmological Coincidences just coincidences?
    4. The Axis of Evil, is there a low-l mode deficiency in the WMAP power spectrum?

    Of local questions I list a few, which question whether we fully understand GR, even in the solar system:
    1.The Pioneer Anomaly.
    2. The Fly-by anomaly.
    3. The fact that the residual anomaly in Uranus' orbit, after Neptune is accounted for, is not explained by Pluto (mass 2 OOM too small).
    4. Why does the residual increase in the Earth's rotation rate, after other factors are removed, equal Hubble's constant? (h=0.67) See Leslie Morrison and Richard Stephenson, 1998, Astronomy & Geophysics Vol. 39 October, The Sands of Time and the Earth’s Rotation and again by Stephenson, 2003, Astronomy & Geophysics Vol. 44 April, Historical eclipses and Earth’s rotation.
    They obtain △T/day/cy = −6 x 10−4 sec/day/cy, which is equal to H if H = 67km.sec−1Mpc−1.


    We wait to the end of the year to see if the hint of a non-GR residual in the GP-B results is substantiated.

    Just food for thought...

    Garth
     
    Last edited: May 20, 2007
  13. marcus

    marcus 24,908
    Science Advisor
    Gold Member
    2014 Award

    Well Garth, you started the thread for this purpose
    and now you have offered for discussion some sample observations that raise questions about the standard cosmo model.

    I will try to follow suit---maybe in a different style but with parallel intention

    I think LCDM looks like a KLUDGE, tinkered manyways to fit, and that any cosmology has to be a kludge as long as it is based on GR. Until we have a theory of space time and matter that is more fundamental than GR, all the cosmology models are going to be somewhat funnylooking.

    No disrespect to either LCDM or GR, both are great!

    But the more tinkered and contrived our best fit looks the more it is begging for a more fundamental theory. The situation is very hopeful because QG replacements for GR are in the works and because the consensus Kludge cosmology shows what the more fundamental theory must aim to accomodate and make natural (instead of contrived)
     
    Last edited: May 19, 2007
  14. turbo

    turbo 7,366
    Gold Member

    Marcus, I agree entirely. Penrose keeps saying that both GR and quantum theory will have to change before they can be unified, but I suspect that GR will take the bigger hit by far. Consider: If Einstein had known that the rotation curves of spiral galaxies are too flat to be explained by GR, and that clusters exhibit far more binding force and stronger lensing than expected, would he have blamed that on the existence of a mysterious entity that cannot be detected by any means, or would he have considered that gravitational attraction and inertial effects might vary with location, based on the local matter-density? After reading his 1920 Leyden address and his 1924 essay "On the Ether" I know where I'd put my money. GR's failure on galactic and cluster scales should not be surprising, since so little was known about our universe in 1917. It is a wonderful model within the domain of its applicability, but now we need something more fundamental and more general than GR.
     
  15. Nereid

    Nereid 4,014
    Staff Emeritus
    Science Advisor
    Gold Member

    Nice list Garth (I'll comment on it later).

    Taking advantage of the informality of this forum, here is a very much off-the-top-of-my-head list of possible topics and questions, relating to GR and cosmology:

    * tests of strong field GR, using SgrA* and binary pulsars: what regions of parameter space can we realistically hope to explore?

    * gravitational wave detectors - LIGO etc: non-detection sometime in the next decade would trigger a crisis, I think; detection would open a new window, with all that that implies

    * UHECR research - anisotropy, local sources, GZK limit, etc: another new window, maybe a crisis triggered too?

    * DM searches, both terrestrial and astronomical (e.g. XENON, gamma-ray astronomy as a means of detecting SUSY DM particle decays)

    * AMANDA and other neutrino 'telescopes': mostly (cosmological) serendipity

    * hi-res, deep x-ray studies of cluster environments: homogeneity of the ICM, cluster to cluster variation, structure within the ICM, ... I feel at least part of the CMB/WMAP/SZ 'problems' may be due to over-simplifications about the ICM

    * 6Li and 7Li abundance: IIRC, this is the only significant mismatch between theory and observation, for primordial light nuclide abundances

    * hi-res DM profiles of Local Group objects - MW, M31, M33, ... right down to the smallest dwarfs, GCs, ... and stellar streams: the more precisely the distribution of DM can be characterised, the sharper the tests that can be done of competing theories about its nature.

    (to be continued)
     
  16. Chronos

    Chronos 10,127
    Science Advisor
    Gold Member

    DM clustering is very poorly understood. DM constitutes the vast majority of mass in the universe, but is almost unnoticeable, save on large scales. And even then, mapping it is exquisitely difficult. It is also capriciously distributed, not smooth. It could account for the 'axis of evil' as well as other apparent artifacts in the CMB data. I find it more interesting than distressing for the LCDM model. A DM detector would be fascinating.
     
  17. Garth

    Garth 3,512
    Science Advisor
    Gold Member

  18. turbo

    turbo 7,366
    Gold Member

    Thank you for that link, Garth. Interesting reading - a nice summary by an author willing and able to present observational evidence without couching it in terms of the assumptions that so frequently accompany their description.
     
  19. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    The residual Pioneer Anomaly, after allowing for 'normal physics' from on-board systematics, and the residual spinning up of the Earth, after allowing for tidal braking, both can be explained by a clock drift between ephemeris and atomic clocks equal to the Hubble parameter.

    But we may ask, "Is the PA to be seen in the orbits of the outer planets?"

    Generally the answer given is no, but as I have pointed out we cannot explain Uranus' orbit, after allowing for Neptune, by the perturbations of Pluto and other 'Plutons', the masses of the trans-Neptunian planets are 2 OOM too small. So, what about Neptune itself?

    It appears there is an unexplained residual in its orbit as well, consider Rawlins' 1970 paper: The Great Unexplained Residual in the Orbit of Neptune.
    Now that Lalande prediscovery observation of Neptune (he recorded it but did not recognise he had discovered a new planet! :rolleyes:) was made in 1795 and the -7" arc corresponds to a 7/15 seconds of time discrepancy which equals 0.467 secs. over 1968 - 1795 = 173 years.

    This corresponds to a time discrepancy, or annual clock drift in which the atomic clock is speeding up relative to the ephemeris clock, of 0.467/(365.25x24x3600) per 173 years, which equals 8.55 x 10-11 yr-1 and this is equal to Hubble's parameter if H = 84 km.sec−1Mpc−1!

    So within the bounds of error of the observation it seems that the Hubble parameter turns up again!

    These 'Hubble Parameter' clock drifts are getting to be a bit of a habit, perhaps the universe is trying to tell us something!

    Just more food for thought. :wink:

    Garth
     
    Last edited: May 23, 2007
  20. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    As that observation of Neptune seems to be significant, especially as a confirmation that the PA may apply to the Outer Planets after all, it would be important to study it further.

    There is a Letter to the Editor of Nature from Charles T. Kowal & Stillman Drake: Galileo's observations of Neptune.
    This was answered by Rawlins Galileo's Observation of Neptune; and Reply but I don't have access to the Abstract.

    So it appears the pre-discovery observations of Neptune are unclear as to whether there is an anomaly or not.

    Garth
     
    Last edited: May 24, 2007
  21. Garth

    Garth 3,512
    Science Advisor
    Gold Member

    Arp is still at it; from today's physics arXiv: A concentration of quasars around the jet galaxy NGC1097.

    His thesis is that quasars are ejected from active galaxy nuclei. And although I am not convinced at all, it makes an interesting case to examine Lieu's claims.

    The question is simply one of a statistical analysis of the data, is Arp making a false correlation or not? And if not are his observations being deliberately ignored because it does not fit the mainstream model?

    Garth
     
Know someone interested in this topic? Share a link to this question via email, Google+, Twitter, or Facebook

Have something to add?