WTF is the speed of gravity

[Ethelred]

Originally posted by Ned
Ethelred, Then, as I understand what you are saying, in GR, gravity has no speed - it is not a wave that propagates.

In GR I think that is the case. Gravity waves would be an oscilation in the warp of space.

The problem as Rogan pointed out is the GR fails on the quantum level in some way. I don't know the way myself but if it is true than GR is in need of some improvement at the very least. If its just that it fails in calculations maybe it QM that has the problem. If however particles can be seen to behave in ways compatible with QM and incompatible with GR then its GR that is broken.

[KrazyHorse]

Eth, we're not sure if GR fails or the quantum theories do. We just know they're incompatible. We also know that each provides answers in their own realm (has predictive power).

Obviously there's some sor of quantum nature to the Universe. And just as obviously there's some sort of geometric nature to gravity.

[KrazyHorse]

BTW, I don't know QFT so I don't know why the two fail to mesh.

[Ned]

Frogger and Ethelred, I must admit that I don't have sufficient understanding of the topic to provide an intelligent reply to what you said. But, I gather from what you said,

1) there is no time delay between the optical eclipse maximum and the gavity eclipse maximum in contrast to what the article reports as fact; and

2) gravity is a space warp; it is not a wave; it propagates at the spead of light and it traverses an event horizon, which seemingly is impossible because, if I understand it correctly, the meaning of an event horizon is that space is so warped that it is discontinuous.

[IoT]

Originally posted by Ned
Frogger and Ethelred, I must admit that I don't have sufficient understanding of the topic to provide an intelligent reply to what you said. But, I gather from what you said,

1) there is no time delay between the optical eclipse maximum and the gavity eclipse maximum in contrast to what the article reports as fact; and

2) gravity is a space warp; it is not a wave; it propagates at the spead of light and it traverses an event horizon, which seemingly is impossible because, if I understand it correctly, the meaning of an event horizon is that space is so warped that it is discontinuous.

I can't answer 1) but for 2)

GR is a geometrical theory of space and does NOT involve waves or particles. Wave/particle theory is Quantum Gravity. These 2 models do not currently mesh and resolving this has been the focus of a large umber of theoretical physicists for a quite a few years.

In GR the only in space that is discontinuous is the singularity of a black hole. The event horizon is something completely different. The event horizon is simply a boundary (non physical) at which the gravitiational force of a black hole is just sufficiently strong to prevent photons escaping.

[Ned]

IoT, But, if photons cannot escape, how can gravitons that have the same speed?

But, I still think you confirmed what I was driving at - space/time between the Black Hole and the rest of the Universe is discontinous under GR, but we still feel the effect of the Black Hole's mass. How?

[Caesar the Great]

but isn't everything always moving at the speed of light?

[Felch]

I sure don't.

[KrazyHorse]

Originally posted by Ned
Frogger and Ethelred, I must admit that I don't have sufficient understanding of the topic to provide an intelligent reply to what you said. But, I gather from what you said,

1) there is no time delay between the optical eclipse maximum and the gavity eclipse maximum in contrast to what the article reports as fact

I've already explained that the reason why gravitational retardation is a 5th order effect is the same reason that the eclipse "problem" is not a problem. There is a time delay between "felt" eclipse and viewed eclipse because, in Newtonian gravity, taking the "speed" of gravitation as infinite is a better approximation of GR than taking it as c. All three reduce to equal in the low speed/weak field limit; infinite-speed Newt does so faster

2) gravity is a space warp; it is not a wave; it propagates at the spead of light and it traverses an event horizon, which seemingly is impossible because, if I understand it correctly, the meaning of an event horizon is that space is so warped that it is discontinuous

a) It depends what you mean by wave.

b) Space isn't discontinuous at an event horizon; it's discontinuous at the singularity itself. An event horizon is only important in that it is the point at which the escape velocity exceeds that of light...

c) Gravity does not have to "pass through" the event horizon if it makes you uncomfortable; you can think of it as originating at the event horizon and propagating from there.

[KrazyHorse]

Originally posted by Ned
IoT, But, if photons cannot escape, how can gravitons that have the same speed?

But, I still think you confirmed what I was driving at - space/time between the Black Hole and the rest of the Universe is discontinous under GR, but we still feel the effect of the Black Hole's mass. How?

We also feel the "temperature" and charge of a black hole. I'm unsure as to the details (the basics seem relatively straightforward) of Hawking radiation, but QED at least requires the exchange of virtual photons, same as QGD would require the exchange of virtual gravitons. So your question applies to EM as well...

[Rogan Josh]

OK. I give in. I have to make a few points before my fingers become unbearably itchy.....

1. As Ethelred already pointed out, in classical gravity (ie. GR) nothing crosses the event horizon because of time dilation. (This is not true in QFT, but they are really not understood in detail. Hawking for example uses QFT on a classical gravitional field background, which to me always seems like cheating.)

2. A pseudotensor is like a tensor but transforms with an extra minus sign. In physics this transformation is usually CP. Then one describes spin 0 particles as being scalars or pseudoscalars, spin 1 particles as being vector or axial-vectors, and spin-2 as being tensor or pseudotensor, depending on how their field change under CP transformations.

3. Gravity does propagate. You can actually build a completely classical version of GR on a flat space time in terms of propagating waves (like electromagnetism). The curvature of space-time is not needed - it is just a nice way of thinking about it. Of course, even with the geometrical picture, it still propagates because the space-time distortions spread out at a finite speed.

4. As already pointed out, GR breaks down at a quantum level.

When one 'quantizes' a theory like electromagnetism turning it into quantum-electrodynamics (QED) one finds that lots of quantities are formally infinite. At first sight it predicits that certain observable quantities are infinite, in contradiction with experiment. Looking closer however, one notices that although the quantities are infinite, the difference between any 2 such observable quantities is finite. Since any measurement we make is actually a comprison between the new observable and things we already know or have measured, all predictions are actually finite and in remarkable agreement with experiment. It is only ceratin 'bare' parameters in the theory, which we cannot measure, which are infinite. In technical jargon QED is 'renormalizable'.

The same 'first sight' applies the the quantum version of GR. Infinities crop up all over the place. Unfortunately for GR, it also predicts infinites for 'observable' quantities in contradiction with experiment. These cannot just be brushed away. Quantum gravity is 'non-renormalizable'.

[KrazyHorse]

4. As already pointed out, GR breaks down at a quantum level

Grr. Or Quantum breaks down at macro-level.

[Ethelred]

That is the first time I have seen a hint of the reasoning behind renormalization. I knew there were infinities and there was a process called renormalization which sounded to me like dividing one infinity with another and then filling in the resulting unknown with experimental results. What you wrote gives a reason for why the technique can be used for some things and not for others.

This is what I get for not having the math. Two semesters of engineering calculus thirty years ago just doesn't cut it.

OK. I give in. I have to make a few points before my fingers become unbearably itchy.....

Now that one I fully understand.

Its why I start writing up a reply to the longer posts while I am still reading them. Kind of like taking notes of my thoughts as I read and then trying to clean up the mess afterwards.

[Rogan Josh]

Originally posted by Frogger

4. As already pointed out, GR breaks down at a quantum level

Grr. Or Quantum breaks down at macro-level.

No. Not true. The theories which work at the quantum level (like QED) explain physics at large scales completely correctly. They are just rather hard to work with.

The only thing we can't do is describe large scale gravity with a quantum theory. And since we can't describe it at a quantum scale either, this is hardly surprising.

And there is a hell of a lot more evidence for QED being correct than there is for GR. In fact, QED is the best tested physical theory ever.

[KrazyHorse]

You can't describe gravity on a quantum scale? I'm pretty sure you can. What would the problem be? Just use QED, slap on coupling constant. Only problem is that when you hit large scales it breaks down, no?