General relativity: a poor description of spacetime

[skylark53]

Everyone is congratulating the recent Nobel Prize winners, whose work uses observations of distant supernovae to attempt to show the existence of universal acceleration.

Let's not forget that supernova studies of cosmology are all based on a theory (General Relativity) that has huge problems - as demonstrated first by Zwicky in the 1930s, again by Rubin et al in the 1970s, and continually since. But instead of coming up with a new theory of gravity that fits the large-scale data, the community has continually propped up GR by introducing free parameters. These include exotic forms of matter and energy for which there is no theoretical understanding and no observational evidence, despite decades of searching.

GR works well within the solar system. But time and time again serious, intractable problems have emerged on large scales. Not only does the metric have no clothes, we do not even know its shape.

Ironically, were he still with us, Einstein would be first to roll up his sleeves.

 

[DrChinese]

Let's not forget that supernova studies of cosmology are all based on a theory (General Relativity) that has huge problems - as demonstrated first by Zwicky in the 1930s, again by Rubin et al in the 1970s, and continually since. But instead of coming up with a new theory of gravity that fits the large-scale data, the community has continually propped up GR by introducing free parameters. These include exotic forms of matter and energy for which there is no theoretical understanding and no observational evidence, despite decades of searching. ...

Welcome to PhysicsForums, skylark53!

Honestly, I think your first post is a bit extreme. I would say that GR is extremely useful, and I question your comment about "huge problems". As we gain more information about the large scale, which is where there are a lot of gaps, we should be able to see where there are further modifications might be necessary.

The Zwicky reference is a bit odd to me. He was known for alternating brilliant and somewhat off-kilter ideas ("nuclear goblins" ?). So citing his name really does nothing for your point.

If there is a specific current open issue with GR you wish to cite, I would be interested in hearing it along with why that makes GR problematic. After all, I don't think anyone imagines that GR is supposed to single handedly explain everything. You might want to consider that it is good within the scope of its domain. I consider that a good way to judge any theory.

 

[marcus]

Doctor, good to see you!
I think Fritz Zwicky showed that galaxy motion could predict Dark Matter----using either Newton or if you prefer Einstein.
http://en.wikipedia.org/wiki/Fritz_Zwicky#Dark_matter
That was 1933. He deduced there was "missing" matter needed to hold a cluster together.

Now DM has been observed (bullet cluster etc) and has become an important part of modeling structure formation, you could say that the prediction of DM was a great triumph of Gen Rel.

Except that Newton gravity serves just about as well to get the Zwicky prediction of DM.

Where I think Gen Rel plays essential role in this success story is it tells us about gravitational lensing. This is one of the ways we OBSERVE dark matter and map its density.

So Newtonian gravity is enough to predict DM but Gen Rel plays an essential role in actually mapping the density of the clouds of it which we see in clusters of galaxies.

 

[PAllen]

Dark Energy (rather than Dark Matter) can also be viewed as a success of GR. The cosmological constant simply falls out of the most general form of the field equations (and really falls right out if you derive from an action principle). Despite the history of Einstein's favoring non-zero value, then calling it a blunder, it always was part of the theory. It is really the only adjustable parameter of the theory, and it was there from day one.

A positive cosmological constant makes specific predictions about the nature of accelerated expansion. So far, observations are increasingly consistent with this, and a number of alternative models have been discarded.

 

[marcus]

EDIT: PALLEN YOU ARE RIGHT. I did not see your post, or I would not have bothered to write this.
Lambda has to be in the equation. Mankind's job is to find its value. New methods are in the works. It is an exciting time. And yes this is another success of GR!

...the community has continually propped up GR by introducing free parameters...
Ironically, were he still with us, Einstein would be first to roll up his sleeves.

No new free parameters have been introduced into GR.

The two parameters in GR are the Newton G and Lambda, whether 90 years ago or today. It is standard practice in physics to include whatever terms are consistent with the symmetries of the theory. Besides G the only other possible constant is Lambda. So Lambda HAS to be in the Einstein equation for GR.

If you took Freshman Calculus, it is analogous to introducing a constant of integration. The answer is incorrect if you omit it. The question is then what is the VALUE of Lambda.
Even before 1998 several physicists had figured out that the value was probably not zero.
Rafael Sorkin was one such. It did not require supernovas for them to arrive at this conjecture. But until 1998 there was no observational proof and the majority just assumed it was zero.

However Lambda is a natural feature of the GR equation just like G is.

There is nothing bizarre or exotic about it. (That is what you get from hyped up popular accounts.) It is just a constant which has been there essentially from the beginning.
It has as much right to be there as Newton's G.

It is simply untrue that "the community has continually introduced free parameters" into GR.

If anyone wants the straight story about Lambda, as an antidote to all the hyped up speculation, a good treatment to read is the paper by Bianchi and Rovelli.

 

[marcus]

Tamara Davis et al have recently proposed a new "standard candle" method to determine distance which goes out beyond where you can use supernovae. They think it will work out to about redshift z = 4.
If this is right and the method works well it will be like another 1998. Or so I think.

I will get a link to the antihype Bianchi Rovelli paper.

http://arxiv.org/abs/1002.3966
Why all these prejudices against a constant?
Eugenio Bianchi, Carlo Rovelli
(Submitted on 21 Feb 2010)
The expansion of the observed universe appears to be accelerating. A simple explanation of this phenomenon is provided by the non-vanishing of the cosmological constant in the Einstein equations. Arguments are commonly presented to the effect that this simple explanation is not viable or not sufficient, and therefore we are facing the "great mystery" of the "nature of a dark energy". We argue that these arguments are unconvincing, or ill-founded.
9 pages, 4 figures

Basically they say there's needless elaboration and attention-getting hyperbole that just needs to take a cold shower. In this they try to help by providing a splash of cold water.

Here is the Tamara Davis AGN distance measure paper:
http://arxiv.org/abs/1109.4632
A new cosmological distance measure using AGN
I hope their new measure works!

==================
EDIT TO MATT-O: Your pointed-to paper about DM in galaxy clusters is from Copenhagen's Dark Cosmology Center. That also where the Tamara Davis paper I mentioned here is from. It looks like the Dark Cosmology Center is way up the list of places to watch for new results these days.

 

[matt.o]

GR has recently passed tests on galaxy cluster scales, too:

http://arxiv.org/abs/1109.6571

 

[Imax]

Let's not forget that supernova studies of cosmology are all based on a theory (General Relativity) that has huge problems - as demonstrated first by Zwicky in the 1930s, again by Rubin et al in the 1970s, and continually since.

General relativity is a very good theory. The huge problem is in understanding what it's saying.

Not only does the metric have no clothes, we do not even know its shape.

GR can't tell you the shape of the Universe (i.e. its topology).

 

[Chronos]

Railing against the current accepted model is a 'flavor of the day' exercise. Put up some math that is logically consistent with whatever 'idea' you have in mind.

 

[skylark53]

Thanks all for your thoughts. I think your reaction reveals how deeply entrenched we are in the current paradigm - and that there is (as yet) no safe haven if we’re to leave this one. Thomas Kuhn had this to say about making the leap: "The scientist who embraces a new paradigm is like [a] man wearing inverting lenses".

The failure of GR on large scales is so clear and unambiguous that it's hard to know where to begin. Sadly Zwicky's paper (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1933AcHPh...6..110Z) is in German but here's a paper that examines the observed problems of GR as they are known about in 1963 (and in addition attempts some MoND): http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1963MNRAS.127...21F&data_type=PDF_HIGH&whole_paper=YES .

Many more problems have emerged since then. Rubin et al's spiral galaxy studies in 1978-80 (http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1980ApJ...238..471R&data_type=PDF_HIGH&whole_paper=YES&filetype=.pdf ) demonstrate just as clearly that there's a problem with the way the gravitational force falls off with distance in GR. The authors (thinking within the standard paradigm) conclude there must be a lot of ‘dark matter’ outside the visible parts of spiral galaxies - but of course it has never been found. The discrepancy between model and observations is worse still for dwarf galaxies.

The failure of GR to explain the motions of test particles grows steadily larger the larger the scale you look on. Yes, GR predicts well the gravitational lensing observed in clusters of galaxies, but this might be true of many other metric tensor theories.

The problem with all supernova cosmology is that the data are fitted to a model that is assumed to be correct; all workers do is establish the best-fit parameters. This approach clearly cannot address the question of whether we have the right model.

If we jump away from GR, where are we to go? This is a huge part of the problem, for the community believes in GR with such faith that almost no-one is doing what Einstein was doing 100 years ago: coming up with a theory that explains the observations better than the incumbent model does.

Though it is not a theory, MoND has given us a description of how & where GR fails, and therefore acts as a 'filter' for any new theories. Here's a recently-proposed model that apparently can reproduce the observations of high-z supernovae without either dark matter or dark energy: http://arxiv.org/PS_cache/arxiv/pdf/0903/0903.4096v2.pdf. (I haven't read it.) The Weyl gravity models of Kazanas and Mannheim (and others) also need no DM or DE.

Alternative models of gravity are speculative and a huge amount of work is needed to know which -if any - is/are the most promising.. But at the moment the community is so fixated on GR that all our energies are going into the bottomless pit of trying to make it work.

Believe me, Chronos, this is not a rail, or even a rant; it is a plea for some of us, at least, to put on a pair of Kuhn’s glasses.

 

[Imax]

The interpretation of some observables may be subjective, but, absent a paradigm shift, the Universe is expanding.