So this is a bit of a duplicate: Is the universe fully transparent to gravitons?

[.Scott]

TL;DR Summary

Is the universe fully transparent to gravitons??
That seems to be presumed in a phys.org article.

In a phys.org article, the investigators make these statements:

Vagnozzi and Loeb say we can go even further back, however, by tracing gravitons, particles which mediate the force of gravity.

"The universe was transparent to gravitons all the way back to the earliest instant traced by known physics, the Planck time: 10 to the power of -43 seconds, when the temperature was the highest conceivable: 10 to the power of 32 degrees," says Loeb. "A proper understanding of what came before that requires a predictive theory of quantum gravity, which we do not possess."

If gravitons can be detected at all, that implies to me that the universe is not as transparent to them as is being asserted in the article.
How can you detect something without changing it - if nothing else, wouldn't that violate HUP?
In the contemporary universe, if the gravitons are changed, that doesn't sound to me like fully "transparent".
If the contemporary universe is not fully transparent to them, I would think that all bets are off at t0 + Planck time.

Is the universe fully transparent to gravitons??

----------

Edited to include link to the original article:
As noted by @vanhees71 , the article is now open to non-subscribers:
https://doi.org/10.3847/2041-8213/ac9b0e

 

[Vanadium 50]

Well, we don't have a theory of gravitons, so I don't know what "transparent to gravitons" really means.

However, I would suppose that the universe is not transparent to gravitons. Consider a graviton emitter (operating by some unknown principle) and a graviton detector (operating by some unknown principle, maybe the same one, maybe not). The detector sees the emitter.

Now place a black hole between them. I would expect -= and again, there's no theory so no way to do the calculation - the detector to stop seeing the emitter.

 

[PeterDonis]

In a phys.org article,

Which is already a red flag that you should expect to see claims that are questionable, or at least not stated very clearly.

Is the universe fully transparent to gravitons??

First, we need to restate the question. Instead of "gravitons", we need to say "gravitational waves". We have detected gravitational waves, so your argument that the universe can't be fully transparent to them is valid for that case. But we have not detected gravitons, since we have not detected any quantum aspects of gravitational waves, and do not expect to for the foreseeable future.

However, what the researchers described in the article are actually saying is not that the universe is "fully transparent" to gravitational waves. What they are saying is something much simpler: that our universe should have a cosmic gravitational wave background for much the same reason that it has a cosmic electromagnetic wave background: that at some point in the past the universe underwent a transition that created such a background. And that we should be able to detect such a cosmic gravitational wave background today, if it was indeed produced in the very early universe.

Such a transition does not require that the universe be "fully transparent" to the background radiation after it happens. The universe is not "fully transparent" to EM radiation, but that does not prevent the CMBR from existing. All that is required is that the universe is "transparent enough" to the radiation for the patterns we see in it if we detect it today to convey useful information about what the universe was like when it was emitted. And since it is more difficult for gravitational waves to interact with the rest of the matter and energy in the universe than it is for EM waves (which is why it's much more difficult to detect gravitational waves than it is to detect EM waves), if our universe is "transparent enough" for the CMBR to convey useful information about the universe when it was emitted, it should be "transparent enough" for a cosmic gravitational wave background, if we can only figure out how to detect one and if it does turn out to exist, to convey useful information about the universe when it was emitted. That is the basic argument I understand the researchers to be making.

 

[.Scott]

Would you agree that the fact that something is detectable means that it can be changed?
If I can tell that light is shining through a window by only looking at the glass, is that not prima facie evidence that the glass is not completely transparent?
Would there be any way for the graviton to escape that logic?

 

[.Scott]

[phys.org] Which is already a red flag that you should expect to see claims that are questionable, or at least not stated very clearly.

The claim is that this is based on an article in yesterdays' "Astrophysical Journal". At the moment, I can only search articles from Nov 2 and earlier. ... I'll keep checking.

But we have not detected gravitons, since we have not detected any quantum aspects of gravitational waves, and do not expect to for the foreseeable future.

Of course, if they cannot be detected, then for the purposes of their article, it wouldn't matter if they hold any secrets.

However, what the researchers described in the article are actually saying is not that the universe is "fully transparent" to gravitational waves. What they are saying is something much simpler: that our universe should have a cosmic gravitational wave background for much the same reason that it has a cosmic electromagnetic wave background: that at some point in the past the universe underwent a transition that created such a background. And that we should be able to detect such a cosmic gravitational wave background today, if it was indeed produced in the very early universe.

But I'm guessing that you wouldn't expect such a CGB to reflect too much information from t0 + Planck time (assuming such a moment exists).

 

[PeterDonis]

I'm guessing that you wouldn't expect such a CGB to reflect too much information from t0 + Planck time (assuming such a moment exists).

Why not?

 

[Vanadium 50]

Would you agree that the fact that something is detectable means that it can be changed?

We can detect the CMBR. Good luck changing it.

However, I think this is more a topic for philosophy than physics, as we can go around and around asking what it measn to be "detectable", "to change" and maybe even "something".

 

[.Scott]

Why not?

Actually, that's a good question. It's not only an issue of what would be left of the information that could still be detected, but how much information existed at t0+Planck time to begin with.

The amount of information would be limited by the Bekenstein Bound - or perhaps it's more like the diameter of the universe at Planks time is not so tiny - limited by that bound and how much information there is to be discovered.

 

[.Scott]

We can detect the CMBR. Good luck changing it.

CMBR is photons. They change when they are detected. You can even shield against photons.

 

[PeterDonis]

It's not only an issue of what would be left of the information that could still be detected

I have already addressed this, in the very paragraph that you quoted from me in post #5.

how much information existed at t0+Planck time to begin with.

The amount of information would be limited by the Bekenstein Bound

True, but the argument of the researchers described in the article is simpler: if a CGB is detected at all, it would amount to a falsification of inflation, because inflation, if it happened, should have diluted any CGB from the Planck time to the point where it would be undetectable.

I have not looked at the actual paper so I don't know what, if any, mathematical arguments the researchers used along these lines.

 

[Vanadium 50]

CMBR is photons. They change when they are detected. You can even shield against photons.

And that's exactly the kind of problem we are going to get into if we go down the quasi-philosophical rabbit hole. Does changing one photon change the CMBR? (If I replace a hammer's handle and later replace its head, is it still the same hammer?)

 

[Vanadium 50]

because inflation, if it happened, should have diluted any CGB from the Planck time to the point where it would be undetectable.

I did not read the article in question, but it sounds - at a minimum - like a model dependent statement.

 

[.Scott]

And that's exactly the kind of problem we are going to get into if we go down the quasi-philosophical rabbit hole. Does changing one photon change the CMBR? (If I replace a hammer's handle and later replace its head, is it still the same hammer?)

That was not the question I was asking.

 

[PeterDonis]

I did not read the article in question, but it sounds - at a minimum - like a model dependent statement.

Only in the sense that you need an inflation model; but the property of diluting whatever existed before inflation started is a general property of all inflation models.

 

[Vanadium 50]

Q. Why don't we see monopoles? Why don't we see domain walls?
A. Inflation swept them away.

Q. Why didn't inflation sweep away all the matter, leaving behidn a dull universe of only dark energy?
A. Inflation didn't last long enough to do it?

This is what I mean by model dependent. Nobody will propose a model that doesn't match what we see, but these are functions of the parameters of the model, not the model itself. And I don't think we know enough about quantum gravity to make anything but model dependent statements. (Especially when you include things like reheating, the surface of last scattering, etc.)

 

[PeterDonis]

Why didn't inflation sweep away all the matter, leaving behidn a dull universe of only dark energy?

Your premise is false. Inflation did dilute away any "matter" (or "radiation") that might have existed before it started, and did leave a "dull universe of only dark energy"--but dark energy with a huge energy density that was in a "false vacuum" state. Inflation ended when that dark energy (the inflaton field) underwent a phase transition from the "false vacuum" state that transferred its huge energy density to the Standard Model fields.

 

[Vanadium 50]

Yes, but if it ran "longer" (and I know I am oversimplifying) there would be no visible matter left. Obviously nobody wioll propose this, because it's silly, but it's a perfectly valid outcome of a "long" inflationary period.

 

[PeterDonis]

if it ran "longer" (and I know I am oversimplifying) there would be no visible matter left.

Sure there would. The transfer of energy density to the Standard Model fields happens whenever inflation ends, and does not depend on how long it lasts.

 

[Vanadium 50]

Only if there is a baryon asymmetry at that energy scale. If not, you end up withmostly radiation. Again, model dependent.

Inflation predicts ∑Ω = 1. It doesn't say anything about the individual Ωs, although individual models might.

 

[PeterDonis]

Only if there is a baryon asymmetry at that energy scale.

What energy scale? The "energy scale" during inflation does not change since the energy density of the inflaton field does not change. The energy density changing significantly means inflation is ending.

If not, you end up withmostly radiation.

I have been saying "matter and radiation" in my posts, or using the more precise term "Standard Model fields", because the issue is not whether radiation counts as "matter" but which fields contain the significant energy density in the universe--the inflaton, or the SM fields. I don't think a distinction between "radiation" and "matter" is relevant here.

Inflation predicts ∑Ω = 1. It doesn't say anything about the individual Ωs,

Huh? It says that all of the energy densities except the inflaton field get diluted to the point of being negligible. AFAIK that's true of all inflation models. Where particular inflation models differ is on how inflation starts and how it ends, not on what it does to other fields besides the inflaton field while it is happening.

 

[vanhees71]

The claim is that this is based on an article in yesterdays' "Astrophysical Journal". At the moment, I can only search articles from Nov 2 and earlier. ... I'll keep checking.

The article is open access:

https://doi.org/10.3847/2041-8213/ac9b0e

 

[Vanadium 50]

You're practically making my point for me. If you have inflation and this and this and this you get a universe like what we see. To me, this sounds very model dependent.

 

[PeterDonis]

If you have inflation and this and this and this you get a universe like what we see. To me, this sounds very model dependent.

I think you're missing my point: my point is that, as far as I know, every inflation model includes "this and this and this", so as soon as you assume inflation at all, you are assuming all those other things. They aren't separate things that get added in one inflation model but not another.

 

[vanhees71]

That's an important point: The details of a specific inflation model are not that important, i.e., inflation independently of a specific model solves many problems of the cosmological models without inflation (horizon, flatness problem, etc).

 

[Vanadium 50]

every inflation model

I get that, but I would add "viable" to it.

Nobody is going to propose a model that leaves an empty universe, or one that lasts under a second, or was teeming with monopoles, or any of a zillion other features that conflict with observation. That would be silly.

However, to assume that the only possible models are ones that are presently popular does not seem to me to be wise. History is full of things that "everybody just knew" only to learn later that things were different. How could nature possibly treat left and right handed particles differently? (To pick an example from my own field)

I think this is particularly important in a situation where we don't have an underlying theory. I don't believe we understand gravity well enough to say for certain that it must act in a particular way in untested conditions. (To bring up an example from a recent thread: does ZPE gravitate? GR says it should, but that causes its own set of problems as well.) Assuming GR works at all relevent scales is, IMO, the best possibility, but I wouldn't say it is the only possibility.'

 

[PeterDonis]

to assume that the only possible models are ones that are presently popular does not seem to me to be wise

If you have an example of an inflation model that doesn't have the properties I described, please give a reference.

As I understand it, the properties I described are general features of all inflation models precisely because they are inflation models--they come from the fact that inflation is happening at all, not the details of how it starts or how it ends. So I don't think the assumption that all inflation models will have such properties is unwise.

 

[Vanadium 50]

Sure. Reduce the duration of inflation. Now the universe looks less homogeneous and might have spme monopoles left for us to see. So inflation accommodates a solution to the horizon problem (which is quite an accomplishment) and it accommodates a lack of monopoles, but doesn't actually predict either.

"Plausible inflation models" and "inflation models" are not synonyms. and (this is my opinion, which people can disagree with) the correct model of inflation may not be one that we consider plausible today. We may not have considered it at all.

 

[PeterDonis]

Reduce the duration of inflation.

So when you say "model dependent", you basically mean "duration dependent"? And by "viable" you basically mean "has a long enough duration that it solves the horizon and monopole problems, i.e., is consistent with observations"?

inflation accommodates a solution to the horizon problem (which is quite an accomplishment) and it accommodates a lack of monopoles, but doesn't actually predict either.

Ok.

correct model of inflation may not be one that we consider plausible today.

But whatever it is, it won't be one that doesn't solve the horizon problem or the monopole problem, correct?

 

[Paul Colby]

Well, we don't have a theory of gravitons, so I don't know what "transparent to gravitons" really means.

your statement lead me to what looks to be a high quality answer on Physics Stackexchange as to why we have no theory of gravitons. My takeaway is that at low energies we do have a workable theory of gravitons. So, could one reframe the OPs question in terms of low energy gravitons?

 

[PeterDonis]

My takeaway is that at low energies we do have a workable theory of gravitons.

Where are you taking that away from?

 

[PeterDonis]

could one reframe the OPs question in terms of low energy gravitons?

As I already pointed out in post #3, the OP question can and should be reframed in terms of classical gravitational waves, not gravitons. Nothing in the OP's question actually depends on any putative quantum aspects of gravity.

 

[Paul Colby]

Where are you taking that away from?

the final statement of the reply.

“But it's not all lost. As I tried to explain in the last paragraph, the problem of non renormalizability is actually an issue of high energies. The theory remains predictive at the energies we can reach in the collider. However, at energies larger than $M_p$ we have no clue.”

 

[PeterDonis]

the final statement of the reply.

Should not be taken as a statement about an actual quantum theory of gravity, since that would imply that we have a collider that can run experiments at or near the Planck energy--which we don't, by many orders of magnitude. We certainly do not have collider experiments involving gravitons, and we don't expect to. Where gravity is concerned, the "low energy limit" is classical GR; to the extent the non-renormalizable spin-2 quantum field theory of the "graviton" plays any role, it is purely an abstract one, that the field equation for this quantum field turns out to be the Einstein Field Equation, so the theory is consistent with classical GR in the low energy limit.

 

[Paul Colby]

Should not be taken as a statement about an actual quantum theory of gravity, since that would imply that we have a collider that can run experiments at or near the Planck energy--

Fermi’s theory of weak interactions was not renormalizable yet it yielded answers that could and were, tested. If there are low energy gravitational quanta, they may well be beyond detection, as all the cross sections end up being super small.

 

[PeterDonis]

Fermi’s theory of weak interactions was not renormalizable yet it yielded answers that could and were, tested.

So what? I am not saying that no non-renormalizable theory can ever be tested.

If there are low energy gravitational quanta, they may well be beyond detection

You're missing the point. We can detect low energy classical gravitational waves already, so obviously there are processes that, if we insist on modeling them using the spin-2 quantum field theory at low energy, do not have cross sections that are too low to detect. It's just that, for those processes, the predictions of this low energy quantum field theory are exactly the same as those of the classical theory of gravitational waves; so there is no way to test the quantum theory by experiment in this way.

What we do not have any prospect of testing, now or in the foreseeable future, is any scenario in which the quantum theory of gravity would make different predictions from the classical theory. That is what we would need a Planck energy accelerator for.