Questioning the cosmological principle

[Tanelorn]

Thanks for reply Chalnoth. If it is as complicated as you say and we can't right down a model that can be verified by simulation then it must be very difficult to beleve that we even have a standard model? However perhaps I am wrong when you consider that we can't even accurately predect the weather or the climate yet we think we understand these.

 

[inflector]

I've been thinking about the OP again:

The cosmological principle essentially says that there are no preferred locations and directions in the universe (homogenity and isotropy). We know that strictly speaking this principle is violated at the accessible scales (filaments, galaxy clusters and supercluster, voids, CMB). So one could try to save this principle by assuming that beyond the accessible scales these inhomogenities will be smoothed out.

This strikes me as a very insightful indictment of the cosmological principle itself as it is interpreted by most.

The principle states that there are no preferred locations and directions in the universe yet the principle is violated at every scale we can measure. As someone who is only recently studying cosmology and the issues involved, this strikes me as the wrong conclusion. This could simply be because I misunderstand the meaning of the principle. Or perhaps it is indicative of where we are in the progression of our evolution of theory.

As a neophyte, I ask myself: how can one of the major assumptions of standard cosmology be contradicted at every level of our empirical observation yet it still remains "valid?"

I found marcus's quote interesting and perhaps illustrative of the issue raised in the OP:

So all we can know or estimate is the effective density over the largest scale we can observe and then we assume that whatever is out beyond that is either too far away to have an effect or enough like what we see that we can get good answers by assuming uniformity.

[second bold section is mine, the bold of "effective" was in marcus's post]

So it seems to me that the assumption that whatever is out beyond what we can see is "enough like what we can see" seems intuitively obvious as a decent starting place for assumptions about what we cannot see. It sure makes a lot more sense than assuming that it's filled with dragons or turtles or something we have never seen. It might not hold, but it sure seems like a good starting assumption.

Interestingly, the real question is what this idea of "enough like what we can see" really means. I can see two different directions that one can take this:

1) Statistical Similarity - Assuming that the density and other characteristics that we can measure represent the space outside our horizon. This is the approach that I believe cosmology takes with the cosmological principle.

2) Fractal Continuation - Assuming that the progression of fractal structure that we see starting with quarks in nucleons, to nucleons in atoms, to atoms in molecules, to atoms and molecules in plants and animals and planets and stars, to planets in solar systems, to stars in galaxies, to galaxies in clusters, to clusters in superclusters and filaments, ... assuming that this progression continues, which means that we are very likely to find continued fractal structure as we look out further and therefore it is unlikely that we'll just happen to be in the center of some uniform density system even at the largest scales.

So in both cases, similarity is being projected out beyond what we can see. In the first case, we assume that things smooth out despite our not seeing smoothness in our observations except at very narrow bands of observation. In the second case, we assume that we will see more fractal clumping and bunching of matter which matches the character of what we see.

I don't understand why the second perspective is not more widely held.

The first perspective seems tenable only in the realm where our instruments just happen to detect similarity at the boundaries, but how often has that been true even in the history of cosmology? To me, it doesn't even seem to be true now?

At first, when we only saw the stars, we were in the center of our universe. Then a bit later we once saw only the milky way as our universe, and we were on the bare edge of it. Later we came to see the galaxies as the dominant feature of the universe. But how do we really know that we are not missing as much as we once did when we only saw the Milky Way and didn't realize that some of the stars we saw were actual galaxies?

I think that one could equally well assume that instead the (infinite) universe has a kind of "fractal structure" extending on all scales. That would mean that the universe is filled by scale-free clusters, superclusters, ... and voids, super-voids etc.

That sure seems to me to be a better assumption. It fits the facts better.

 

[TrickyDicky]

I've been thinking about the OP again:


This strikes me as a very insightful indictment of the cosmological principle itself as it is interpreted by most.

The principle states that there are no preferred locations and directions in the universe yet the principle is violated at every scale we can measure. As someone who is only recently studying cosmology and the issues involved, this strikes me as the wrong conclusion. This could simply be because I misunderstand the meaning of the principle. Or perhaps it is indicative of where we are in the progression of our evolution of theory.

As a neophyte, I ask myself: how can one of the major assumptions of standard cosmology be contradicted at every level of our empirical observation yet it still remains "valid?"

I found marcus quote interesting and perhaps illustrative of the issue raised in the OP:

[second bold section is mine, the bold of "effective" was in marcus' post]

So it seems to me that the assumption that whatever is out beyond what we can see is "enough like what we can see" seems intuitively obvious as a decent starting place for assumptions about what we cannot see. It sure makes a lot more sense than assuming that it's filled with dragons or turtles or something we have never seen. It might not hold, but it sure seems like a good starting assumption.

Interestingly, the real question is what this idea of "enough like what we can see" really means. I can see two different directions that one can take this:

1) Statistical Similarity - Assuming that the density and other characteristics that we can measure represent the space outside our horizon. This is the approach that I believe cosmology takes with the cosmological principle.

2) Fractal Continuation - Assuming that the progression of fractal structure that we see starting with quarks in nucleons, to nucleons in atoms, to atoms in molecules, to atoms and molecules in plants and animals and planets and stars, to planets in solar systems, to stars in galaxies, to galaxies in clusters, to clusters in superclusters and filaments, ... assuming that this progression continues, which means that we are very likely to find continued fractal structure as we look out further and therefore it is unlikely that we'll just happen to be in the center of some uniform density system even at the largest scales.

So in both cases, similarity is being projected out beyond what we can see. In the first case, we assume that things smooth out despite our not seeing smoothness in our observations except at very narrow bands of observation. In the second case, we assume that we will see more fractal clumping and bunching of matter which matches the character of what we see.

I don't understand why the second perspective is not more widely held.

The first perspective seems tenable only in the realm where our instruments just happen to detect similarity at the boundaries, but how often has that been true even in the history of cosmology? To me, it doesn't even seem to be true now?

At first, when we only saw the stars, we were in the center of our universe. Then a bit later we once saw only the milky way as our universe, and we were on the bare edge of it. Later we came to see the galaxies as the dominant feature of the universe. But how do we really know that we are not missing as much as we once did when we only saw the Milky Way and didn't realize that some of the stars we saw were actual galaxies?

That sure seems to me to be a better assumption. It fits the facts better.

But once again you have to understand that the assumption that seems better to you for the reasons you explain and that seems pretty logical is not compatible with standard cosmology and here enter "extra-scientific" factors that pertain to the sphere of human psychology. So you need huge evidences to change a paradigm that is believed to be true by most of the comunity, because the emotional anchor with the current paradigm is very strong.
On the other hand when you interpret observations in a way different than the standard model, you better bring with you a new model that is internally consistent, and also consistent with the laws and theories of physics and that explains at least as much and predicts at least as much as the model you're questioning, otherwise the practical thing for mainstream science is to basically ignore the alternative interpretation, and I think that is fair enough even if sometimes in forums it gets a little overacted.
The situation changes a little when you actually find overwhelming observations that contradict the paradigm, but that is not the case yet, and still if that were to happen you still need a viable alternative model that fits the new observations and the old ones.

 

[tom.stoer]

The principle states that there are no preferred locations and directions in the universe yet the principle is violated at every scale we can measure.

...

I can see two different directions that one can take this:

1) Statistical Similarity - Assuming that the density and other characteristics that we can measure represent the space outside our horizon. This is the approach that I believe cosmology takes with the cosmological principle.

2) Fractal Continuation - Assuming that the progression of fractal structure that we see ... assuming that this progression continues, which means that we are very likely to find continued fractal structure as we look out further ...

So in both cases, similarity is being projected out beyond what we can see. ...

I don't understand why the second perspective is not more widely held.

Inflector, thanks for the excellent summary of my thoughts; seems that you understand better than myself :-)

That sure seems to me to be a better assumption. It fits the facts better.

What I learned during this discussion is that I am certainly not the first one who had these ideas. Perhaps my idea that large-scale fractal-like structures have not been formed during expansion by gravitational attraction (which seems to be impossible) but may be relicts from a kind of phase transition (which might be a rather hasty conclusion) is new. But that was not my original intention. Instead it was mainly about asking how structures could continue beyond the observable universe and if and how structures on an these scales can cause physical effect within the visible universe.

From what I learned this discussion is already a few years old - and the interpretation of the data is by no means undisputed!

I skimmed through some papers trying to explain accelerated expansion via large inhomogenities and a violation of the cosmological principle in the sense that the solar system is in a rather special location. Btw.: I don't think that this idea has already been disproven (nor has it been proven):
a) both proponents and opponents do have not enough data and facts available to settle the discussion conclusively
b) the discussion regarding the cosmological principle strikes me as the same facts seem to allow two contradictory interpretations

The problem seems to be that beyond a certain scale the universe we see is too young to allow for bright structures like galaxies to be observed (as they have not been formed since the big bang). So instead using facts to derive conclusions regarding the large scale structure it may very well be that the only option we have is to find a natural principle w/o being able to give a sound proof. My claim is simply that - given that the two options just discussed cannot be decided experimentally - we have to discuss whether such a principle is natural.

From what I know and from what I see it is by no means clear that the cosmological principle as we know it today is the only natural principle one can imagine.

 

[tom.stoer]

@TrickyDicky: most of your arguments are based on "what is believed to be true by most of the comunity" and on an "emotional anchor". I think this is not a very good reason to believe in a scientific theory.

The standard model of cosmology has some severe shortcomings:

• it is to a very large extend based on invisible and poorly understood entities like dark matter and dark energy
• it seems to be the case that interpretation of data is by no means unambiguous
• it uses a principle that is simply wrong on the accessible scales and can perhaps be saved on scales that may never be accessible to observations

The starting point for scientific revolutions was never a new, internally consistent model; it was quite often "only" reasonable doubt about common belief; only years later these new models emerged and were widely accepted (quantummechnaics, general relativity)

My conclusion is that it is much too early to scrap the cosmological standard model - but it's never too early to challange it!

 

[inflector]

What I learned during this discussion is that I am certainly not the first one who had these ideas. Perhaps my idea that large-scale fractal-like structures have not been formed during expansion by gravitational attraction (which seems to be impossible) but may be relicts from a kind of phase transition (which might be a rather hasty conclusion) is new. But that was not my original intention. Instead it was mainly about asking how structures could continue beyond the observable universe and if and how structures on an these scales can cause physical effect within the visible universe.

[SNIP]

The problem seems to be that beyond a certain scale the universe we see is too young to allow for bright structures like galaxies to be observed (as they have not been formed since the big bang). So instead using facts to derive conclusions regarding the large scale structure it may very well be that the only option we have is to find a natural principle w/o being able to give a sound proof. My claim is simply that - given that the two options just discussed cannot be decided experimentally - we have to discuss whether such a principle is natural.

From what I know and from what I see it is by no means clear that the cosmological principle as we know it today is the only natural principle one can imagine.

You bring up an interesting point. One of the reasons that this thread caught my interest was because of the tie-in between quantum gravity and cosmology. My personal research in this area is exploring the nature of spacetime structure and gravity at the extremes: both the cosmological and quantum scales, and less-popular theories for these domains.

I believe that a valid quantum gravity theory, if found, should shed some light on this issue of whether or not fractal structure or homogeneity dominates cosmology outside the horizon. If, for example, it could one day be shown that spacetime itself has a fractal structure on the quantum scale, and it could be demonstrated that this fractal nature also was the basis for the fractal structure at larger scales, then it would certainly be reasonable to conclude that fractal structure continued out beyond the cosmological horizon.

 

[Chalnoth]

Actually, we won't "know" what the case is until we gather more observational info, that's the point here, and fortunately it seems it might be a relatively short time until we can tell, we are already reaching the threshold of the 100 Mpcs, and the limit of 200 Mpcs proposed in the paper I linked seems reasonable and it might not take as many years as it has taken to get near the 100 Mpcs.
It feels great when empirical observations needed to confirm or falsify a theoretical model seem so near and are not subject to different interpretations (provided a big enough sample of galaxies), that's a situation so infrequent in cosmology!

We already have measurements that confirm the standard model at length scales much, much greater than 100Mpc, such as CMB studies, baryon acoustic oscillation studies, and supernova studies.

 

[Chalnoth]

Thanks for reply Chalnoth. If it is as complicated as you say and we can't right down a model that can be verified by simulation then it must be very difficult to beleve that we even have a standard model? However perhaps I am wrong when you consider that we can't even accurately predect the weather or the climate yet we think we understand these.

The key is focusing primarily on areas where we can do the calculations. Basically, on large length scales, we can make certain approximations to the behavior of gravity that allows us to calculate things to high accuracy. In this regime, where we can be most sure of systematic errors, our simplest models match reality excellently.

 

[TrickyDicky]

@TrickyDicky: most of your arguments are based on "what is believed to be true by most of the comunity" and on an "emotional anchor". I think this is not a very good reason to believe in a scientific theory.

That is exactly my point, that majority belief and emotional factors are not scientific reasons to support a theory.

The standard model of cosmology has some severe shortcomings:

• it is to a very large extend based on invisible and poorly understood entities like dark matter and dark energy
• it seems to be the case that interpretation of data is by no means unambiguous
• it uses a principle that is simply wrong on the accessible scales and can perhaps be saved on scales that may never be accessible to observations

The starting point for scientific revolutions was never a new, internally consistent model; it was quite often "only" reasonable doubt about common belief; only years later these new models emerged and were widely accepted (quantummechnaics, general relativity)

My conclusion is that it is much too early to scrap the cosmological standard model - but it's never too early to challange it!

I basically agree with the shortcomings you list and your conclusion, just a couple of remarks, GR in its November 1915 form was new and internally consistent, not only that, it made accurate predictions and had a tremendous explanatory power based in the geometrical concept of curvature, it fulfilled all the demands I asked of new revolutionary theories to overcome resistance to change, it actually became accepted almost inmediately, even if it corrected no other than Newton.
In the case of QM in its 1926 version was also pretty much giving predictions and explaining results. So certainly in the two cases you cite the starting point was indeed a new and internaly consistent theory.
So I was just warning challengers about the difficulty of the task.

We already have measurements that confirm the standard model at length scales much, much greater than 100Mpc, such as CMB studies, baryon acoustic oscillation studies, and supernova studies.

Those are very much open to different interpretations(meaning their interpretation is heavily model dependent), and BAO measurements are being questioned even in the mainstream journals.
Whilst the statistical treatment of the data recovered from redshift surveys has been pretty much agreed upon by the main competitor teams (orthodox cosmology-Peebles, Hogg etc- and Fractal cosmology team-Pietronero,Joyce, Baryshev,etc..). And therefore is less subject to model dependent interpretatons.

 

[Chalnoth]

Those are very much open to different interpretations(meaning their interpretation is heavily model dependent), and BAO measurements are being questioned even in the mainstream journals.
Whilst the statistical treatment of the data recovered from redshift surveys has been pretty much agreed upon by the main competitor teams (orthodox cosmology-Peebles, Hogg etc- and Fractal cosmology team-Pietronero,Joyce, Baryshev,etc..). And therefore is less subject to model dependent interpretatons.

The disputes of these measurements are all about the details, not the overall results. And it's really not reasonable for so many such extremely different measures to have all converged on the same cosmology if that cosmology wasn't, at least to a decent approximation, largely accurate.

 

[tom.stoer]

... just a couple of remarks, GR in its November 1915 form was new and internally consistent, not only that, it made accurate predictions and had a tremendous explanatory power based in the geometrical concept of curvature, it fulfilled all the demands I asked of new revolutionary theories to overcome resistance to change, it actually became accepted almost inmediately, even if it corrected no other than Newton.
In the case of QM in its 1926 version was also pretty much giving predictions and explaining results. So certainly in the two cases you cite the starting point was indeed a new and internaly consistent theory.

No, that is not what I wanted to say:
The story of GR did not start in 1915, it started in 1905 or soon after. It was clear for Einstein - and for the whole communty - that SR is a challenge for Newtonian gravity and that not both can be "true" but it took approx. ten years to present the final result.
The same applies to QM: the story started approx. in 1905 (or even earlier) and it took approx. 20 years to find a consistent andpredictive theory. So over 20 years the communty new that what was available (Bohr, Bohr-Sommerfeld, ...) was essentitally "wrong".

The development of QM was deeply rooted in reasonable doubts. Therefore it's absolutely legitimate to provide and discuss good reasons why certain assumtions or theories may be "wrong" w/o being able to provide a theory that is "right". It's this discourse that is required in order to make progress; w/o questioning the principles of theories science becomes religion.

My observation is that b/c of lack of sufficient experimental support in certain regimes (quantum gravity / string theory, cosmology) these discussions are more importent than ever.

 

[Chalnoth]

My observation is that b/c of lack of sufficient experimental support in certain regimes (quantum gravity / string theory, cosmology) these discussions are more importent than ever.

Well, one of the problems is that without the experimental support in regimes beyond the standard model, there really hasn't been any clear direction on where to go, so theorists have just been branching out more or less blindly in a tremendous variety of directions. Sadly, it's going to be nigh impossible to distinguish the correct theories from the incorrect ones until we start producing experiments that clearly and unambiguously deviate from our current theories. The problem is that those places where we may see some deviation all lie in regimes where systematic uncertainties are still quite high, so unfortunately we can't be sure just yet that we have seen any deviation.

 

[tom.stoer]

In principle you are right, but my claim is that regarding the cosmological principle these deviations are there (!) and can only be explained away by referring to scales beyond the observable universe. Of course systematic and statistical uncertainties exist, but they exist regardless if you want to prove the standard model or if you want to disprove it. The problems are equally severe for all reserach directions.

 

[Chalnoth]

In principle you are right, but my claim is that regarding the cosmological principle these deviations are there (!) and can only be explained away by referring to scales beyond the observable universe. Of course systematic and statistical uncertainties exist, but they exist regardless if you want to prove the standard model or if you want to disprove it. The problems are equally severe for all reserach directions.

What deviations? Where the cosmological principle is concerned, the large-scale experiments confirm expectations to a very high degree of accuracy. There are some interesting hints in some places that maybe there might be something interesting going on, but so far we don't know for sure whether it's an experimental/observational problem or something we don't understand about fundamental physics that is causing the discrepancy.

 

[tom.stoer]

What deviations?

The deviations of matter distribution (clusters, voids, ...), the deviations of the CMB from a perfect monopole profile.

Where the cosmological principle is concerned, the large-scale experiments confirm expectations to a very high degree of accuracy.

You can interpret the data in that way if you like; but you always refer to the idea that homogeneity will hold on scales beyond the observed scales. So you may use homogeneity as an element of your model, but it's certainly an input which cannot be confirmed experimentally. All measurements taken so far show direct evidence that visible matter is not distributed homogeniously (clusters, voids, ...). The CMB is not homogenious, either (higher multipoles) [there are even indications that there might be a small anisotropy in both matter flow and CMB. ]

The claim that homogeneity holds must always refer to scales larger than the observed ones and will therefore always make assumptions how matter distribution will continue outside the directly visible domain.

So this "confirmation" of the cosmological principle always assumes that it may hold outside the visible universe - and is therefore not a confirmation.

 

[Chalnoth]

The deviations of matter distribution (clusters, voids, ...), the deviations of the CMB from a perfect monopole profile.

Ah, i see what you meant. But my point is that the results of our observations to date are fully consistent with the universe being approximately homogeneous and isotropic. How far beyond our horizon this holds is unknown, of course, but ultimately it's irrelevant to the fact that making these assumptions has so far proven to give an accurate picture of how our observable patch behaves.

 

[tom.stoer]

OK, no we have a common basis ;-)

Of course the universe is approximately homogeneous, but recovering homogeneity in the sense of the cosmological principle is not possible w/o making assumptions about invisible regions.

Here's my main point: in order to do that one has to make several assumptions
- continuation of matter distribution outside the visible universe
- data fitting requires the additional assumption regarding existence and properties of DM and DE

Instead one could make different assumptions like
- different continuation of matter distribution outside the visible universe
- no DM (there seems to be something wrong with DM in explaining globular clusters)
- no DE (using inhomogeneity / voids / backreaction instead)

I don't think that one can savely prove or disprove one of these options or models on the data basis that is available today. I don't think that one can safely rule out one of the above mentioned assumptions. And I don't think that the assumptions made in the standard model are "easier" in the sens of Ockhams razor.

Therefore one should take different assumptions and competing models seriously.

 

[Chalnoth]

Instead one could make different assumptions like
- different continuation of matter distribution outside the visible universe
- no DM (there seems to be something wrong with DM in explaining globular clusters)
- no DE (using inhomogeneity / voids / backreaction instead)

I don't think that one can savely prove or disprove one of these options or models on the data basis that is available today. I don't think that one can safely rule out one of the above mentioned assumptions. And I don't think that the assumptions made in the standard model are "easier" in the sens of Ockhams razor.

The first point is unlikely to ever be demonstrated exactly, except to state that due to the homogeneity of our own observable patch, we can be pretty darned sure that homogeneity holds for a significant distance outside it. Exactly how far we can't say.

As for dark matter and dark energy, however, those are pretty effectively confirmed. There still remains some slight possibility that dark energy might turn out to be modified gravity, but all other explanations have utterly failed.

 

[tom.stoer]

The first point is unlikely to ever be demonstrated exactly, except to state that due to the homogeneity of our own observable patch, we can be pretty darned sure that homogeneity holds for a significant distance outside it.

Seems that we don't agree here. What about the following:

Your first point (homogeneity) is unlikely to ever be demonstrated exactly b/c due to the small inhomogeneity of our own observable patch, we always have to refer to assumptions that homogeneity holds for a significant distance outside it.
(I am not saying you are wrong; I am only saying that you can't prove that you are right)

As for dark matter and dark energy, however, those are pretty effectively confirmed.

One has introduced invisible entities in order to fit the data. I wouldn't call that a confirmation. (Again I am not saying you are wrong; I am only saying that you can't prove that you are right)

There still remains some slight possibility that dark energy might turn out to be modified gravity, but all other explanations have utterly failed.

I don't want to insist on specific models (I don't like MOND, for example). But it has become clear that
a) there is some potential in explanations based on inhomogeneity (which I don't think has been ruled out)
b) some problems based on DM due to incompatibilities with observations for globular clusters)
c) missing experimental conformation of DM (OK, let's wait for the LHC :-)

I don't think that the position of the standard model of cosmology is as strong as it is claimed ...

 

[Chalnoth]

Seems that we don't agree here. What about the following:

Your first point (homogeneity) is unlikely to ever be demonstrated exactly b/c due to the small inhomogeneity of our own observable patch, we always have to refer to assumptions that homogeneity holds for a significant distance outside it.
(I am not saying you are wrong; I am only saying that you can't prove that you are right)

This just goes down to default assumptions, though. We do expect that there are likely significant inhomogeneities somewhere. But for there to be one near the edge of our observable universe without seeing any sign of it within our observable universe is very unlikely.

One has introduced invisible entities in order to fit the data. I wouldn't call that a confirmation. (Again I am not saying you are wrong; I am only saying that you can't prove that you are right)

Now you're getting about as absurd as saying that you can't prove neutrinos exist.

I don't want to insist on specific models (I don't like MOND, for example). But it has become clear that
a) there is some potential in explanations based on inhomogeneity (which I don't think has been ruled out)
b) some problems based on DM due to incompatibilities with observations for globular clusters)
c) missing experimental conformation of DM (OK, let's wait for the LHC :-)

I don't think that the position of the standard model of cosmology is as strong as it is claimed ...

a) Inhomogeneities have been ruled out as an explanation for the accelerated expansion, as I posted earlier:
http://arxiv.org/abs/1007.3725

b) I'm just not impressed at all at people claiming to have found such inconsistencies. Such observations are liable to help us nail down the precise nature of dark matter, but other observations have already confirmed beyond any reasonable doubt that it exists.

c) So? This isn't unexpected. The LHC is a very poor dark matter detector, by the way. I'd be extremely surprised if we saw it there.

 

[tom.stoer]

But for there to be one near the edge of our observable universe without seeing any sign of it within our observable universe is very unlikely.

The problem is that we are talking about scales that cannot be seen in principle, neither directly not indirectly. A structure just beyond the particle horizon will influence visible objects via gravity, but in an infinite universe we must talk about arbitrary large structures which are arbitrary far away! You can't ever see them. So you have to make an assumption. You will never be able to prove or disprove this assumption but you must honestly admit that you made an assumption.

Now you're getting about as absurd as saying that you can't prove neutrinos exist.

No, certainly not. Neutrinos have been detected, so there's no doubt about their existence. And of course I will change my mind as soon as one is able to demonstrate the existence of dark matter. But currently there is no proof, therefore it's allowed to be skeptical and to think about alternatives.

a) Inhomogeneities have been ruled out as an explanation for the accelerated expansion, as I posted earlier:
http://arxiv.org/abs/1007.3725

I know this paper; I would like to wait for some more responses and discussions before calling it a disproof.

b) I'm just not impressed at all at people claiming to have found such inconsistencies. Such observations are liable to help us nail down the precise nature of dark matter, but other observations have already confirmed beyond any reasonable doubt that it exists.

I think the real reason of disagreement between us is a different conception of science, here especially about existence. I absolutely agree that observations have confirmed the existence of an effect that cannot be explained via standard hadronic matter and standard GR. But that is not a confirmation of the existence of dark matter itself. (as an example: the observation of beta decay did not proof the existence of the neutrino; it simply revealed an effect that was not compatible with the models known at that time and that required new physics w/o any indication regarding violation of conservation of energy or the existence of a new particle; the proof of the existence of the neutrino was a second, independent experiment). In the same way DM will not be proven by using it as a parameter to fit the data.

c) ... The LHC is a very poor dark matter detector, by the way.

The LHC and the detectors are especially designed and constructed to detect light SUSY particles (e.g. the neutralino, depending on the specific model like MSSM; mainly by detecting missing energy) which are the best candidates for DM; is there any other experiment that could do the job?

Btw.: is there a preferred mass scale for light SUSY particles to explain DM? What happens of the LHC disproves the existence of SUSY particles below 14 TeV; is (C)DM compatible with much larger SUSY mass scales?

 

[Chalnoth]

The problem is that we are talking about scales that cannot be seen in principle, neither directly not indirectly. A structure just beyond the particle horizon will influence visible objects via gravity, but in an infinite universe we must talk about arbitrary large structures which are arbitrary far away! You can't ever see them. So you have to make an assumption. You will never be able to prove or disprove this assumption but you must honestly admit that you made an assumption.

There's no need to make any assumptions about what lies significantly beyond our cosmological horizon. That stuff can't effect our observable universe anyway.

No, certainly not. Neutrinos have been detected, so there's no doubt about their existence. And of course I will change my mind as soon as one is able to demonstrate the existence of dark matter. But currently there is no proof, therefore it's allowed to be skeptical and to think about alternatives.

There's plenty of proof. The CMB and a number of cluster studies (such as the bullet cluster) are quite conclusive that there is some form of at most weakly-interacting massive particle that is not in the standard model and makes up around 80% of the matter density in our universe.

I think the real reason of disagreement between us is a different conception of science, here especially about existence. I absolutely agree that observations have confirmed the existence of an effect that cannot be explained via standard hadronic matter and standard GR. But that is not a confirmation of the existence of dark matter itself. (as an example: the observation of beta decay did not proof the existence of the neutrino; it simply revealed an effect that was not compatible with the models known at that time and that required new physics w/o any indication regarding violation of conservation of energy or the existence of a new particle; the proof of the existence of the neutrino was a second, independent experiment). In the same way DM will not be proven by using it as a parameter to fit the data.

Dark matter isn't just a parameter fit to the data, however. The hypothesis of dark matter's existence makes a number of directly-testable claims that have been tested and found to be accurate. Yes, it was just a parameter fit back when Zwicky first proposed it some 75 years ago to explain his cluster observations, and later when Vera Rubin and others in the '60's used it to explain galaxy rotation curves. But since then our observations have advanced dramatically, and all of the other explanations have basically been ruled out.

The LHC and the detectors are especially designed and constructed to detect light SUSY particles (e.g. the neutralino, depending on the specific model like MSSM; mainly by detecting missing energy) which are the best candidates for DM; is there any other experiment that could do the job?

The LHC is good at detecting charged particles. It isn't so good at detecting missing mass (dark matter would simply fly through the detector and not be counted). Basically, the proton-proton interactions it relies upon are too dirty for this kind of analysis. What we need is an electron-positron or electron-electron collider in the same energy range, but those are much more difficult to build.

Btw.: is there a preferred mass scale for light SUSY particles to explain DM? What happens of the LHC disproves the existence of SUSY particles below 14 TeV; is (C)DM compatible with much larger SUSY mass scales?

It's been a while since I've looked at the allowed parameter space for dark matter particles. However, whatever the dark matter particle is, it must be stable. So the only possible dark matter particle is the lightest neutral supersymmetric particle (anything more massive would decay). So I'm pretty sure that the LHC will, at the very least, place some nice limits on the allowable mass range of the dark matter particle, if supersymmetry is true, which will narrow the parameter space for dedicated dark matter searches (such as DAMA/Libra and CDMS, to name a couple).

 

[Tanelorn]

It is very difficult for neophytes and especially laymen to distinguish which aspects of the standard model are conjecture and tentative and which ones are widely accepted as a part of the theory. Obviously not everyone is going to agree on what is and what isn't accepted, myself I am forced to keep an open mind for now on anything so complex I can't even begin to comprehend it!

Regarding homogeneity of matter distribution, when something is quasi infinite like our universe appears to be we must specify over what scales we are talking about. The larger the scale the more important that homogeneity is to the standard model, however even if we can prove a little homogeneity due to dark flow at the limit of the observable universe, in a quasi infinite universe it may still be insignificant.

 

[tom.stoer]

There's no need to make any assumptions about what lies significantly beyond our cosmological horizon. That stuff can't effect our observable universe anyway.

The topic was about the cosmological principle. As it is not valid on the visible scales you have to make an assumption what will happen beyond the horizon if you still want to believe in it.

There's plenty of proof. The CMB and a number of cluster studies (such as the bullet cluster) are quite conclusive that there is some form of at most weakly-interacting massive particle that is not in the standard model and makes up around 80% of the matter density in our universe.

That's only indirect. If you want to prove the existence of a particle you must detect the particle. I am sorry, but that's my opinion.

It's been a while since I've looked at the allowed parameter space for dark matter particles. However, whatever the dark matter particle is, it must be stable. So the only possible dark matter particle is the lightest neutral supersymmetric particle (anything more massive would decay). So I'm pretty sure that the LHC will, at the very least, place some nice limits on the allowable mass range of the dark matter particle, if supersymmetry is true, which will narrow the parameter space for dedicated dark matter searches (such as DAMA/Libra and CDMS, to name a couple).

I am asking b/c the LHC is expected to say something about SUSY and therefore perhaps about string theory. If SUSY is not found at the LHC this is no problem for string theory as SUSY at a higher energy scale would be OK as well. Regarding the MSSM the LHC should find something, otherwise the simplest MSSM is ruled out.
My question is about DM, so we should check the allowed parameter space for the lightest SUSY particle to be required by CDM.

 

[Tanelorn]

Does the distribution of the hypothetical dark matter violate the homogeneity principle? Is it found wherever baryonic matter is found?

 

[Chalnoth]

That's only indirect. If you want to prove the existence of a particle you must detect the particle. I am sorry, but that's my opinion.

There's no such thing as direct evidence by the most strict definition. All that we do have are models and experiments/observations that either confirm or falsify those models. When a specific model holds up under a variety of conditions, and alternative models do not, we gain confidence that the model is accurate. The model for dark matter has held up. Other models have not. By the standard, tested practices of science, this is sufficient to consider a WIMP to be highly likely to explain these observations.

I am asking b/c the LHC is expected to say something about SUSY and therefore perhaps about string theory. If SUSY is not found at the LHC this is no problem for string theory as SUSY at a higher energy scale would be OK as well. Regarding the MSSM the LHC should find something, otherwise the simplest MSSM is ruled out.
My question is about DM, so we should check the allowed parameter space for the lightest SUSY particle to be required by CDM.

Well, again, you're probably not going to see it at the LHC no matter the mass.

 

[tom.stoer]

Does the distribution of the hypothetical dark matter violate the homogeneity principle? Is it found wherever baryonic matter is found?

DM will form lumps as well, but probably smoother than ordinary matter. So galaxies would swim in a halo of DM which is much larger than the visbily galaxy. That means that with DM you can come closer to homogeneity, but certainly not exactly.

 

[tom.stoer]

Looking at the discussion at http://cosmocoffee.info/viewtopic.php?p=4737 my conclusion is that the affair regarding large voids and a possible violation of the cosmological principle is by no means resolved.

 

[Chalnoth]

Looking at the discussion at http://cosmocoffee.info/viewtopic.php?p=4737 my conclusion is that the affair regarding large voids and a possible violation of the cosmological principle is by no means resolved.

The existence of some larger-than-expected voids is perhaps somewhat reasonable, and should be examined further. However, having us at the center of a large void being the explanation for the accelerated expansion is highly unreasonable and should be considered massively unlikely.

 

[tom.stoer]

The existence of some larger-than-expected voids is perhaps somewhat reasonable, and should be examined further. However, having us at the center of a large void being the explanation for the accelerated expansion is highly unreasonable and should be considered massively unlikely.

I agree that according to the standard model this seems to be unreasonable. But I want to entertain the idea that introducing unobservable quantities may be as unreasonable as the violation of the cosmological principle as long as it is not a proven fact.

So we should stop to insist in claiming something to be "(un)reasonable", "(un)physical", "(un)natural" etc. Currently there is neither a way to proof the standard model nor to rule it out. In addition there is no consensus that the ideas of large voids have been ruled out. So my conclusion is that we have competing ideas which must undergo a detailed examination. The conclusion is not that we have only one model subject to minor corrections.

 

[Chalnoth]

I agree that according to the standard model this seems to be unreasonable.

Even if you assume that such voids are likely, it's a completely unreasonable model, because it requires we be situated pretty much exactly at the center of one such void, and that the void be almost perfectly spherical.

But I want to entertain the idea that introducing unobservable quantities may be as unreasonable as the violation of the cosmological principle as long as it is not a proven fact.

There are no unobservable quantities in the standard model of cosmology. Every single one of them has an effect on observations.

 

[tom.stoer]

There are no unobservable quantities in the standard model of cosmology. Every single one of them has an effect on observations.

You should accept that up to now both dark matter and dark energy are known only phenomenologically as physical effects and that they haven't been detected experimentally as physical entities. You can't claim to have proven the existence of DM as long as no single DM particle has been detected. You can't claim to have proven the existence of DE - and you will never be! You have hints based on cosmic acceleration - but you will never be able to proof its existence in the same way as you have e.g. detected the neutrino as as you may hopefully be able to detect the Higgs and perhaps SUSY particles. You are not able to say what the DE really is. Comsomological constant, effect of f(R) or ScVeTe theories, quantum gravity relict, scaling limit for one parameter in asymptotic safety, ...

Both DE and DM are unvisible as physical entities, they are only observable as phenomena - but these phenomena do not allow us to deduce what they really are; alternative candidates are known, but the currebtly available facts do not allow us to rule out all alternatives.

Nevertheless: I think we should stop exchanging arguments as we are spinning in a circle. We are neither able to prove nor to refute our positions and should therefore be patient and wait for future experiments to shed more light on these affairs.

 

[Chalnoth]

You should accept that up to now both dark matter and dark energy are known only phenomenologically as physical effects and that they haven't been detected experimentally as physical entities.

Look, it is completely and utterly unreasonable to make arbitrary statements about what sorts of evidence you need before believing a model is at least approximately accurate. This is the exact same tactic that many creationists use, for instance, by claiming they won't believe in evolution until they see every single step of every transition between species. This attitude is fundamentally anti-science.

When approaching a scientific model, the only reasonable thing to do is to approach the model on its own grounds, and test it on those grounds. We have no expectation whatsoever that reality should bend to our whims, and so we should be flexible about what sorts of evidence we accept. The evidence in favor of dark matter is, today, so varied and robust that it really cannot be denied. The evidence in favor of dark energy is substantially less, but by basic inductive logic, the most reasonable explanation is the one that requires the fewest new parameters. Since the cosmological constant fully explains all observations with just one parameter, it is, by far, the most likely explanation.

You have hints based on cosmic acceleration - but you will never be able to proof its existence

This is the problem of induction and applies to all science. Your distinctions between this and neutrino physics are arbitrary and irrelevant.

 

[Chronos]

Interesting question. The fractal explanation is not well supported. Irregulaties in the CMB are inconsistent with a fractal distribution on scales less than the size of the observable universe. This may or may not be significant. They appear more gaussian, but, not decidedly so. It is a puzzle. I think part of the problem is WMAP has unresolved bias errors. The next generation of IR space probes should help.

 

[Calimero]

The existence of some larger-than-expected voids is perhaps somewhat reasonable, and should be examined further. However, having us at the center of a large void being the explanation for the accelerated expansion is highly unreasonable and should be considered massively unlikely.

I agree that large inhomogeneities are probably poor attempt to explain acceleration, but if you want to be completely scientific you can't exclude it easily. Your argument that it is highly unlikely to have us at center of such void does not hold if, somehow (don't ask me why), that is the most favorable place for life to emerge.