The Fate of the Universe: Expansion, Matter Decay, and the Ultimate End

[MathematicalPhysicist]

I don't quite understand this.

Some models say that if the universe keeps expanding indefinitely eventually matter will disolve into radiation, I also read that the particles of matter such as protons wil decay (I am not sure how reliable is this if no one ever deteceted proton decaying). Now it's not as if one day the universe will be empty, right?

 

[Ryan_m_b]

Not empty but pretty much dead. Some models predict an era where all that is left are photons, neutrinos and electrons. Take a look at this timeline:
http://en.wikipedia.org/wiki/Future_of_an_expanding_universe#Timeline

Doesn't look fun from the Degenerate era onwards.

 

[MathematicalPhysicist]

But what of the quarks, will they decay as well?

Well doesn't this resemble the start of the universe with soup of particles?

It always leads to the question how can particles be assembled to planets and stars etc...

 

[Ryan_m_b]

But what of the quarks, will they decay as well?

Well doesn't this resemble the start of the universe with soup of particles?

It always leads to the question how can particles be assembled to planets and stars etc...

I'm not sure about quarks but it's not like the start of the universe because everything is vastly more spread out and entropy has increased to near maximum. There simply isn't that much energy in the entire universe to do work anymore.

Stellar and planetary formation is a product of gravity. Whilst gravity will still remain the sheer size of the universe and the rate of its expansion compared to the star formation era means that you won't be getting anymore stars or planets. Just a vast, cold, mostly empty universe where once an eon two leptons might fly past each other.

 

[bcrowell]

Some models say that if the universe keeps expanding indefinitely eventually matter will disolve into radiation,[...]

Not true. Although Hawking radiation will convert some matter into radiation, the universe is predicted to have matter in it at all future times. See Adams and Laughlin, http://arxiv.org/abs/astro-ph/9701131 , §VD. More recently, Penrose has had a lot of motivation to poke around and look at mechanisms for complete conversion to photons, and at one time was pushing nonstandard particle-physics mechanisms for this as a prediction of his conformal cyclic cosmology (CCC). The fact that he couldn't find standard mechanisms for it shows that the current state of the art does *not* predict it to happen.

But what of the quarks, will they decay as well?

If a particular quark ends up in a black hole, then the black hole will eventually evaporate, and the evaporation will produce mostly photons. Proton decay is also a possibility: http://en.wikipedia.org/wiki/Proton_decay

 

[MathematicalPhysicist]

How can you disprove proton decay?

I mean I understand how you can prove it exists, you just watch for such an occurence (though I am not sure what do one need to detect in order to be witnessing proton decay), but to disprove it looks tough empirically, isn't it?

 

[Nabeshin]

How can you disprove proton decay?

I mean I understand how you can prove it exists, you just watch for such an occurence (though I am not sure what do one need to detect in order to be witnessing proton decay), but to disprove it looks tough empirically, isn't it?

That's the difficulty with disproving ANYTHING in science.

 

[George Jones]

Have a look at these articles:

http://math.ucr.edu/home/baez/end.html;

http://arxiv.org/abs/astro-ph/9902189;

http://arxiv.org/abs/0704.0221.

 

[bcrowell]

How can you disprove proton decay?

I mean I understand how you can prove it exists, you just watch for such an occurence (though I am not sure what do one need to detect in order to be witnessing proton decay), but to disprove it looks tough empirically, isn't it?

All you can do is put a lower limit on the lifetime. But in any case it doesn't seem relevant to the current discussion. The proton would decay into leptons, not radiation.

 

[Chronos]

There is no evidence of proton decay in any experiments conducted to date.

 

[Chalnoth]

There is no evidence of proton decay in any experiments conducted to date.

Except for the fact that protons had to be produced in the early universe. That fact alone implies that proton decay must be possible.

 

[bcrowell]

Except for the fact that protons had to be produced in the early universe. That fact alone implies that proton decay must be possible.

Huh? No, that's wrong. There is no proton decay in the standard model. Therefore you seem to be claiming that the existence of protons disproves the standard model.

 

[Chalnoth]

Huh? No, that's wrong. There is no proton decay in the standard model. Therefore you seem to be claiming that the existence of protons disproves the standard model.

There's also no baryon asymmetry in the standard model. So yes, the existence of protons does disprove the standard model.

 

[SHISHKABOB]

There's also no baryon asymmetry in the standard model. So yes, the existence of protons does disprove the standard model.

wait, what? I thought the Standard Model was... the standard model. How can it be disproved by something as fundamental as the existence of protons? Are you saying that the Standard Model doesn't explain the existence of protons? And so that would mean that the SM is not complete?

 

[Chalnoth]

wait, what? I thought the Standard Model was... the standard model. How can it be disproved by something as fundamental as the existence of protons? Are you saying that the Standard Model doesn't explain the existence of protons? And so that would mean that the SM is not complete?

Well, to put it more correctly, there is no way within the standard model to produce the asymmetry of matter and anti-matter in the early universe. So far we don't have enough experimental data to show us which model is the correct one, however (all experimental tests of the standard model have been quite consistent with it).

 

[bcrowell]

There's also no baryon asymmetry in the standard model. So yes, the existence of protons does disprove the standard model.

Oh, please.

First off, the existence of protons does not require baryon asymmetry. If (a) protons exist, and (b) an equal number of antiprotons does not exist, and (c) the initial conditions of the universe had zero baryon number, then (d) baryon asymmetry is required. The logic here is a & b & c -> d, not a -> d.

Second, this is a distraction from your mistake in claiming that the existence of protons implies proton decay. It does not.

When someone points out to you that you've made a mistake, please just admit it and move on rather than trying to make some new claim that distracts attention from the mistake.

 

[Chalnoth]

Second, this is a distraction from your mistake in claiming that the existence of protons implies proton decay. It does not.

When combined with the fact that protons had to be produced in the early universe, it does.

 

[PAllen]

When combined with the fact that protons had to be produced in the early universe, it does.

Even if you assume (reasonably) that the existing universe almost all matter, how on Earth do you favor proton decay over differential productions rates? I'll buy that, with reasonable cosmological assumptions, SM cannot explain existence of matter in current quantity, but I find differential productions rates via unknown CP violations more plausible than proton decay.

 

[Chalnoth]

Even if you assume (reasonably) that the existing universe almost all matter, how on Earth do you favor proton decay over differential productions rates? I'll buy that, with reasonable cosmological assumptions, SM cannot explain existence of matter in current quantity, but I find differential productions rates via unknown CP violations more plausible than proton decay.

Such CP violations imply non-conservation of baryon number. And if baryon number is not conserved, then protons can decay.

 

[twofish-quant]

How can you disprove proton decay?

You can't prove that protons decay at all, but you can set limits.

http://hep.bu.edu/~kearns/pub/kearns-pdk-snowmass.pdf

I mean I understand how you can prove it exists, you just watch for such an occurence (though I am not sure what do one need to detect in order to be witnessing proton decay), but to disprove it looks tough empirically, isn't it?

You disprove it within certain limits. The limits that we have right now have already falsified a number of theories.

 

[twofish-quant]

Such CP violations imply non-conservation of baryon number. And if baryon number is not conserved, then protons can decay.

If CP is violated then so is T, and if T is violated then you can have processes that create baryons violate baryon number conservation while requiring that the reverse processes conserve baryon number.

You can also have a situation in which you create X/anti-X bosons through pair production. The decay of X/anti-X bosons into protons would not conserve baryon number, but any processes involving protons decaying into lighter particles could.

If you have any literature that says that those scenarios are impossible, I'd be interested in seeing them.

 

[Chalnoth]

If CP is violated then so is T, and if T is violated then you can have processes that create baryons violate baryon number conservation while requiring that the reverse processes conserve baryon number.

This line of reasoning presumes that CPT is a perfect symmetry, which means that you can perform the time-reversal of all processes, just with the positions of matter and anti-matter switched. So I don't think this gets you out of forcing proton decay to exist.

You can also have a situation in which you create X/anti-X bosons through pair production. The decay of X/anti-X bosons into protons would not conserve baryon number, but any processes involving protons decaying into lighter particles could.

But then you'd still have decays with those heavier particles as intermediate steps.

 

[PAllen]

This line of reasoning presumes that CPT is a perfect symmetry, which means that you can perform the time-reversal of all processes, just with the positions of matter and anti-matter switched. So I don't think this gets you out of forcing proton decay to exist.

.

What if anti-protons can decay (slowly) but not protons? That seems to meet the requirements.

 

[Chalnoth]

What if anti-protons can decay (slowly) but not protons? That seems to meet the requirements.

Honestly, I'm not entirely sure. It just seems incredibly unlikely.

As far as I am aware, all GUT's yet proposed require proton decay in order to have baryogenesis.

 

[PAllen]

Honestly, I'm not entirely sure. It just seems incredibly unlikely.

As far as I am aware, all GUT's yet proposed require proton decay in order to have baryogenesis.

That is also true of what I know, but I don't see that as equivalent to the statement that any modification of SM to allow for baryon asymmetry must entail proton decay. It is an assumption that a GUT is true of our universe. Further, I've certainly seen proposed mechanisms for the origin of baryon asymmetry that don't entail proton decay (irrespective whether the model contains proton decay as one of its other predictions).

 

[Haelfix]

Further, I've certainly seen proposed mechanisms for the origin of baryon asymmetry that don't entail proton decay (irrespective whether the model contains proton decay as one of its other predictions).

It is true in a sense, but it depends on the rates and details of the new physics to a certain degree.

In the standard model, proton decay proceeds via dimension 6 operators. This is a nonrenormalizable, baryon number violating interaction that is suppressed by 2 powers of a cutoff scale that is taken to be very high.

Postulating some sort of new physics between the electroweak scale and the Planck scale, will set this cutoff scale, and you must be very careful that these dimension 6 operators don't come dangerously close to violating the proton decay bounds (which are bounded by experiment to something like >10^33 years).

This is a typical and powerful constraint that phenomenologists use, but there are mechanisms or extra symmetries that you can introduce to tame the rates to a certain extent.

However if you violate baryon number, there will be proton decay at some level, it just might be small relative to some other new physics.

So for instance, in the standard model baryon number is an accidental symmetry and is actually violated by nonperturbative physics. Eg instanton physics will yield nonzero tunneling cross sections between degenerate weak SU(2) vacua that imply proton decay on the order of ~10^170 years or something enormous like that. So you see, it is a true statement, but it can be made essentially irrelevant for any sensible physics.

 

[twofish-quant]

This line of reasoning presumes that CPT is a perfect symmetry, which means that you can perform the time-reversal of all processes, just with the positions of matter and anti-matter switched.

CPT has to hold in order for their to be Lorenz covariance.

So I don't think this gets you out of forcing proton decay to exist.

I'm still not understanding how CP-violation *requires* proton decay. What I'm looking for is some mathematical statement that if CP-violation exists and protons don't decay then some fundamental physical constraint (i.e. Lorenz covariance) is violated.

But then you'd still have decays with those heavier particles as intermediate steps.

You can get around that. If the particles are heavy enough then any reactions that get you to the heavy particles would in the process produce at least one new proton for anyone that decays. The mass scales are large enough so that's possible.

 

[twofish-quant]

As far as I am aware, all GUT's yet proposed require proton decay in order to have baryogenesis.

And that's a limitation with GUT's rather than some fundamental constraint of the universe.

Basically the way that GUT's work is to put the SM into some larger group. If you want anything non-trivial then this involves mixing baryon and lepton groups. If you don't mix those groups, then you end up with just the SM and you have something that is non-publishable.

However this is "argument by aesthetics" or "argument by lack of imagination." I don't know of (and would be interesting to hear of) any arguments that say that CP-violation *requires* proton decay because of some fundamental physical constraint. I can think of some mechanisms that would supress proton decay. You'd end up with an ad-hoc and ugly theory, but argument by aesthetic or mathematical simplicity isn't something that i think counts for every much here.

 

[twofish-quant]

However if you violate baryon number, there will be proton decay at some level, it just might be small relative to some other new physics.

And that new physics could very well generate new protons at faster than the rate that protons decay. We are talking about energy scales much larger than the mass of the proton.

 

[Chalnoth]

You can get around that. If the particles are heavy enough then any reactions that get you to the heavy particles would in the process produce at least one new proton for anyone that decays. The mass scales are large enough so that's possible.

Then that wouldn't be a baryon-number violating decay.

Granted, maybe it's possible to come up with a theory which allows baryon number violation in one direction only. My knowledge of high-energy physics isn't sufficient to rule something like this out. But it naively seems massively unlikely.

 

[twofish-quant]

Then that wouldn't be a baryon-number violating decay.

If it's two different processes it would be.

Something like that happens in atomic nuclei. Free neutrons will decay to protons, but when you bind neutrons and protons then any n->p decay is balanced by p->n.

Granted, maybe it's possible to come up with a theory which allows baryon number violation in one direction only. My knowledge of high-energy physics isn't sufficient to rule something like this out. But it naively seems massively unlikely.

I think it's trivial to allow one way baryon number violation. You go into your favorite GUT, and say *God says so* and it is done. You end up with an ugly ad-hoc theory, but one that doesn't contradict any observations or settled physical principles.

Also, what I'm looking for is something "clean" like the Sahkarov arguments. You can easily show that any matter/anti-matter imbalance starting from a symmetry situation requires CP-imbalance, and that's a clean argument based on very firm physical principles.

If there is a similar argument saying CP-imbalance -> proton decay, I'd be interested. Proton decay is a generic feature of GUT's, but you can just argue "GUT's are wrong."

 

[PAllen]

You know, there is just as much an asymmetry in leptons. Why not claim electrons exist implies electrons decay? Whatever mechanism could allow for lepton asymmetry could, in some not yet described theory, allow for baryon asymmetry.

Again, there is a whole chain of assumptions here:

- Universe began with a big bang scenario (of course I don't dispute this, but it is separate from particle physics; in the past, eternal cyclic theories were considered, in which case baryon asymmetry is just a boundary condition of the infinite past).

- Any extension to SM to account for evolution of baryons asymmetry from assumed symmetric big bang state must have the character of known GUTs.

Please note: in no way do I dispute that big bang is highly likely, and GUT type phenomenology is likely in a successful BSM theory, however I prefer to be clear on chain of certainty and reasoning:

- we are confident there is baryon asymmetry (but not certain)
- the most successful cosmology models start with zero baryon number
- thus, there must be a mechanism to explain this

The above are not, IMO ironclad, and further assumptions requiring proton decay are somewhat less certain.

 

[Chronos]

I remain skeptical on the basis of lack of experimental evidence favoring proton decay.

 

[twofish-quant]

You know, there is just as much an asymmetry in leptons. Why not claim electrons exist implies electrons decay?

Not that I disagree with your main point, but for completeness...

One general result of GUT's is that B and L are not conserved but B-L is. The reason that people don't talk about electron decay is that there aren't any particles lighter than the electron that it can decay to without violating some conservation rule.

- Any extension to SM to account for evolution of baryons asymmetry from assumed symmetric big bang state must have the character of known GUTs.

I think that's the big one that I find difficult to swallow. Things could change quickly if we see some strong evidence that GUT's are right, but right now GUT's have enough problems that if someone argues that we are just wrong, that I can't counterargue.

- the most successful cosmology models start with zero baryon number

I wonder about this one. Starting with zero baryon number makes the universe nice and symmetric, but is there any physical reason way we can't start with non-zero baryon number?

 

[Chalnoth]

You know, there is just as much an asymmetry in leptons. Why not claim electrons exist implies electrons decay? Whatever mechanism could allow for lepton asymmetry could, in some not yet described theory, allow for baryon asymmetry.

Electrons are the lightest charged particle. So they can't decay unless there is non-conservation of charge, which we have no reason to believe.

Again, there is a whole chain of assumptions here:

- Universe began with a big bang scenario (of course I don't dispute this, but it is separate from particle physics; in the past, eternal cyclic theories were considered, in which case baryon asymmetry is just a boundary condition of the infinite past).

- Any extension to SM to account for evolution of baryons asymmetry from assumed symmetric big bang state must have the character of known GUTs.

Please note: in no way do I dispute that big bang is highly likely, and GUT type phenomenology is likely in a successful BSM theory, however I prefer to be clear on chain of certainty and reasoning:

- we are confident there is baryon asymmetry (but not certain)
- the most successful cosmology models start with zero baryon number
- thus, there must be a mechanism to explain this

The above are not, IMO ironclad, and further assumptions requiring proton decay are somewhat less certain.

I would say requiring baryogenesis from a hot dense symmetric state is ironclad at present given the evidence. And I have a really, really hard time buying that you can produce a theory which explains baryogenesis from such a state without proton decay that also respects CPT symmetry.

 

[PAllen]

Electrons are the lightest charged particle. So they can't decay unless there is non-conservation of charge, which we have no reason to believe.

ok, I agree this argument by analogy has flaws.

I would say requiring baryogenesis from a hot dense symmetric state is ironclad at present given the evidence. And I have a really, really hard time buying that you can produce a theory which explains baryogenesis from such a state without proton decay that also respects CPT symmetry.

Still waiting for the tight argument from CPT. Independent of whether you judge evidence for " a hot dense symmetric origin" ironclad (you) or highly likely (me), it is still another assumption. Let me rephrase what seems the minimal chain of reasoning leading to proton decay:

1) There is baryon asymmetry in the current universe.
2) At some earlier time there was no baryon asymmetry. Therefore baryon number is not exactly conserved.
3) The principle that a decay that is not prohibited, must occur, is an ironclad principle.
4) Without baryon conservation, nothing prevents the decay of a proton to positron and neutrino. [edit: corrected to neutrino]

To me:

(1) is extremely likely, based on observation + theory, but is not 100%.
(2) is very likely, not quite as strongly as (1)
(3) is very plausible, but what I would like to see is a tight argument that this is required by CPT.
(4) Given (1)-(3), this is ironclad.

 

[Chalnoth]

Still waiting for the tight argument from CPT.

See post #30.

 

[Haelfix]

1) There is baryon asymmetry in the current universe.
2) At some earlier time there was no baryon asymmetry. Therefore baryon number is not exactly conserved.
3) The principle that a decay that is not prohibited, must occur, is an ironclad principle.
4) Without baryon conservation, nothing prevents the decay of a proton to positron and neutrino. [edit: corrected to neutrino]

I would say that 1 is an empirical fact (modulo the caveat that we could be missing all those unobserved anti galaxies) and 3 is a statement about quantum mechanics as we currently understand it.

2 is the most problematic here, b/c you could imagine a scenario where initial conditions simply give a large excess of baryons. The scenario is highly contrived and tuned of course, but well its not logically impossible. Baryogenesis would then not be needed.

4 is also somewhat wishy washy, b/c as I said you can always create new physics that suppresses the rate to values which are so tiny as to be effectively zero for all intents and purposes. But yes, we know that some proton decay must occur in the universe by nonperturbative physics, so its never identically zero.

I think the best evidence for proton decay really comes from experience trying to model new physics. A what else can it be type of argument. Very generically, (and not just from GUTs/SuSY/etc) almost anything you can write down is going to start catalyzing proton decay, and the model builder invariably has to suppress it somehow. In general, this is very much done by hand and while there are some natural mechanisms (for instance imposing R symmetry and various flavor structures in supersymmetric extensions) to make proton decay long, at some point making it too long starts to look contrived and unnatural.

In fact, I only know of one semi natural possibility that kind of avoids this. Namely assuming only the standard model (except for a heavy right handed neutrino) all the way up to the Planck scale and postulating a bout of leptogenesis followed by first order Sphaleron induced baryogenesis. In this case you will get the proton lifetime to be ~10^170 yrs or so. But this model is problematic for different reasons in about 20 different ways.

Anyway...

 

[twofish-quant]

I would say requiring baryogenesis from a hot dense symmetric state is ironclad at present given the evidence.

I wouldn't it's anywhere near ironclad. The strongest argument that I know of against primordial baryon asymmetry is that any pre-inflationary asymmetry would get washed out, but since we don't know much about inflation, that's not a strong argument.

 

[twofish-quant]

I would say that 1 is an empirical fact (modulo the caveat that we could be missing all those unobserved anti galaxies) and 3 is a statement about quantum mechanics as we currently understand it.

One question that does come up is how difficult would it be in an inflationary scenario to come up with a universe that is "randomly" baryon asymmetric. People are messing with anthropic type arguments for other things, so it's worth looking into if someone hasn't done this already. I mean, if we are talking about different universes with different fine structure constants...

2 is the most problematic here, b/c you could imagine a scenario where initial conditions simply give a large excess of baryons. The scenario is highly contrived and tuned of course, but well its not logically impossible.

And if you embed that universe in a multiverse that gets rid of symmetry problems. Now you run into the standard problems with anthropic models (i.e. how the heck do you falsify?), but it's not logically impossible, and I don't know of any observational constraint (although I'd be open to people bring one up).

I think the best evidence for proton decay really comes from experience trying to model new physics. A what else can it be type of argument. Very generically, (and not just from GUTs/SuSY/etc) almost anything you can write down is going to start catalyzing proton decay, and the model builder invariably has to suppress it somehow. In general, this is very much done by hand and while there are some natural mechanisms (for instance imposing R symmetry and various flavor structures in supersymmetric extensions) to make proton decay long, at some point making it too long starts to look contrived and unnatural.

Yup. However the problem with this argument is that the universe itself is looking pretty contrived and unnatural. Trying to reason out the universe by symmetry arguments seems to have hit something of a brick wall.

I think this is something that people can have pretty strong and legitimate disagreements over because a lot has to do with how one "weights" different arguments. I don't put too much weight on arguments by symmetry, but if someone does and we don't have "smoking gun" observations then we are just going to have to agree to disagree.

One thing that is interesting is to step back and ask the question *why* proton decay is a generic part of these models. It looks to me that the basic reason is that you have baryons and leptons in the same big matrix, and any thing that cases the matrix elements to "leak" slightly between the two is going to change quarks into leptons.

 

[PAllen]

I would say that 1 is an empirical fact (modulo the caveat that we could be missing all those unobserved anti galaxies) and 3 is a statement about quantum mechanics as we currently understand it.

I deliberately put this as a stated assumption because way back in high school I read Hannes Alfven's book on his cyclic theory that involved an equal number of galaxies and anti-galaxies. An implausible theory by a serious, Nobel winning scientist. Still, it is purely indirect observational evidence that there is baryon asymmetry - a distant anti-galaxy cluster would be indistinguishable from a galaxy cluster.

 

[Chalnoth]

I wouldn't it's anywhere near ironclad. The strongest argument that I know of against primordial baryon asymmetry is that any pre-inflationary asymmetry would get washed out, but since we don't know much about inflation, that's not a strong argument.

I don't see how it is remotely reasonable for the baryon asymmetry to be inherited from the initial conditions.

 

[Chalnoth]

I deliberately put this as a stated assumption because way back in high school I read Hannes Alfven's book on his cyclic theory that involved an equal number of galaxies and anti-galaxies. An implausible theory by a serious, Nobel winning scientist. Still, it is purely indirect observational evidence that there is baryon asymmetry - a distant anti-galaxy cluster would be indistinguishable from a galaxy cluster.

I would say that the existence of the CMB is pretty direct observational evidence that anti-matter doesn't make up a substantial fraction of the normal matter mass.

 

[PAllen]

I would say that the existence of the CMB is pretty direct observational evidence that anti-matter doesn't make up a substantial fraction of the normal matter mass.

Actually, Alfven's theory purported to explain CMB radiation. However, Peebles disputed the explanation. Point is, the interpretation of CMB radiation as proof of no anti-galaxies is theory dependent - it is not direct observational evidence. On the other hand, a galaxy - antigalaxy annihilation would provide definitive evidence of anti-galaxies.

Don't get me wrong - I find Alfven's theory ridiculous, but this gets at the issue that interpretation of observations is theory dependent, making it harder to make statements about observations ruling out 'all theories except ...'.

Penrose has written good essays on the problems of the Popper criterion of falsifiability:

- there are theories that are verifiable but not falsifiable
- there are theories that are falsifiable but not verifiable
- and there are theories that are both.

 

[Chalnoth]

Actually, Alfven's theory purported to explain CMB radiation. However, Peebles disputed the explanation. Point is, the interpretation of CMB radiation as proof of no anti-galaxies is theory dependent - it is not direct observational evidence. On the other hand, a galaxy - antigalaxy annihilation would provide definitive evidence of anti-galaxies.

Well, if you want to get technical, any piece of evidence you gather is theory-dependent. But as our understanding of the CMB now rests upon such a wide body of mutually-consistent evidence, it is overwhelmingly unlikely that any radically-different explanation can be correct.

I would say the most direct evidence of the CMB's veracity are our baryon acoustic oscillations, which relate the typical sizes of the hot and cold spots on the CMB to the typical distances between galaxies. There would be no reason for those two to be related to one another if the CMB were extremely wrong.

So yes, I think the CMB is quite direct evidence.

 

[PAllen]

Well, if you want to get technical, any piece of evidence you gather is theory-dependent. But as our understanding of the CMB now rests upon such a wide body of mutually-consistent evidence, it is overwhelmingly unlikely that any radically-different explanation can be correct.

I would say the most direct evidence of the CMB's veracity are our baryon acoustic oscillations, which relate the typical sizes of the hot and cold spots on the CMB to the typical distances between galaxies. There would be no reason for those two to be related to one another if the CMB were extremely wrong.

So yes, I think the CMB is quite direct evidence.

But that doesn't address at all my point. Alfven's theory didn't dispute CMB - it explained it as originating from a different source. In particular, it explained as due to primordial annihilation of a portion of matter and anti-matter.

I actually find most convincing the absence of evidence of gamma rays consistent with annihilation. Even with separation into clusters you would expect more than is seen. Also, there is the point that GR pretty much rules out an Alfven type theory due to the inverse singularity theorems.

 

[twofish-quant]

I don't see how it is remotely reasonable for the baryon asymmetry to be inherited from the initial conditions.

What I hear from the quantum gravity people is that there is no reason for baryon number or any other quantum number to have a specific value. So from quantum gravity you can get any initial baryon number that you like.

http://scipp.ucsc.edu/papers/06_07.pdf

The basic argument seems to be that if you take something with a random baryon number and then collapse it into a black hole. By no-hair, you end up with the same state regardless of what the initial baryon number is. Now if you move time in reverse so that things start popping out of a singularity (i.e. the BB), there's no reason why you should get a particular baryon number whether it's zero or anything else.

Put in another way. You have a ball of matter and a ball of antimatter. You dump the ball of antimatter into the black hole. The universe now has positive baryon number. If the black hole had any measurable baryon number, you would have violated no-hair.

A lot of arguments about baryongenesis seem to assume that "zero" is the natural baryon number for quantum gravity to produce, and that's just not so. You can wave your magic wand and force QG to give you zero baryon number ab inito, but zero is as good as any other number.

Now the a stronger argument is that this doesn't matter. QG can give you any random number for baryon number and inflation pushes that to zero. That's a good argument, but that depends on the details of the inflation process. I can't think of a way of avoiding washing out asymmetry, but I'm seeing a few papers that argue that pre-inflation can affect the CMB, and I haven't looked very closely to see how that would impact baryon number.

 

[Chalnoth]

But that doesn't address at all my point. Alfven's theory didn't dispute CMB - it explained it as originating from a different source. In particular, it explained as due to primordial annihilation of a portion of matter and anti-matter.

Right, but that would have had a completely different spectrum.

That said, what I was attempting to say is that we have more than enough evidence to say that our current understanding is highly, highly unlikely to be wildly wrong. I don't see how pulling out a theory that was obviously wrong from the start overturns this.

 

[Chalnoth]

Now the a stronger argument is that this doesn't matter. QG can give you any random number for baryon number and inflation pushes that to zero. That's a good argument, but that depends on the details of the inflation process. I can't think of a way of avoiding washing out asymmetry, but I'm seeing a few papers that argue that pre-inflation can affect the CMB, and I haven't looked very closely to see how that would impact baryon number.

It really doesn't depend upon the details of inflation at all. The key point of inflation that causes it to dilute the universe so much is the early exponential expansion. This early exponential expansion is constrained to have been at least around 70 e-foldings in order for inflation to explain the horizon problem and the flatness of our universe. I don't believe for a moment that you could get that many e-foldings and still have a baryon number worth measuring, because if you did, then the baryons would have been so dense early-on that they would have absolutely overwhelmed the inflaton field and you would have had no inflation at all.

 

[twofish-quant]

That said, what I was attempting to say is that we have more than enough evidence to say that our current understanding is highly, highly unlikely to be wildly wrong.

I think the issue is that we need to be careful in describing what we know and what are strong conclusions. I wouldn't classify large regions of anti-matter existing as an "empirical fact" but rather as a strong conclusion based on the lack of gamma rays and the baryon acoustic oscillation evidence.

One reason to be clear about these things is that I'm wary of evidence by lack of imagination. I don't think that anyone has tried very hard to try to fit the BAO data to a world with antimatter (because there really is no good reason to) so I can't say with any certainty that it can't be done.

One reason to be careful is that the odds that you've overlooked a particular thing is very low. However cosmological models are complex enough so that I think it's highly likely that one assumption of the dozens that we are making is seriously wrong is high.