Speed of Gravity Waves: An Unmeasured Assumption?

[Mark Harder]

Many popular accounts claim that gravity waves move at the speed of light. Now, I know 2 things: Special relativity says their speed cannot be greater than the speed of light in a vacuum. Gravity is a different fundamental force than the electromagnetic force. The same goes for their fields, which are what their respective waves perturb. Why, then, should gravitational waves be expected to behave like electromagnetic waves? We couldn't have based the speed-of-light assertion on empirical observation either, since we haven't had them in our laboratories until now. In fact, even now it doesn't appear that we have measured their speed. So, why are all these people saying gravity waves travel at the speed of light?

 

[Jorrie]

Careful with the terminology - gravity waves are actually a certain type of pressure wave in our atmosphere and travel at the speed of sound.

What you are referring to are gravitational waves, which Einstein's theory says travel at the same speed as light. Since all else in his theory has been found to be correct, why would you doubt this particular one?

In any case, LIGO has measured the speed of the detected waves roughly (by various means) and they have traveled close to the speed of light. We will have to wait for more detectors to come on line to do much better. Or until we find an event both optically and gravitationally in order to compare their arrival times.

 

[Orodruin]

In addition, the modern viewpoint on SR is not that the speed of light is the same in all frames, but that there exists an invariant speed. It then falls out from theory that massless fields propagate at this speed and both the EM field and the disturbances in space-time which are gravitational waves satisfy this. It is not because of some underlying assumption on a connection between gravity and electromagnetism.

 

[Mark Harder]

Careful with the terminology - gravity waves are actually a certain type of pressure wave in our atmosphere and travel at the speed of sound.

What you are referring to are gravitational waves, which Einstein's theory says travel at the same speed as light. Since all else in his theory has been found to be correct, why would you doubt this particular one?

In any case, LIGO has measured the speed of the detected waves roughly (by various means) and they have traveled close to the speed of light. We will have to wait for more detectors to come on line to do much better. Or until we find an event both optically and gravitationally in order to compare their arrival times.

I like your answer. I didn't know that LIGO has measured their speed (how?), even approximately. That's reassuring. Your other point is more philosophical. Let's back up a year. Since the rest of GR had been verified by then, why would you doubt that gravitational waves exist? And if you have no doubt about it, why spend billions verifying what you already know? In any event, I would say that it's a double accomplishment to both detect the waves and measure their velocity, thus demonstrating the correctness of two aspects of GR, one of which is quantitative, which impresses me even more.

 

[fresh_42]

In addition, the modern viewpoint on SR is not that the speed of light is the same in all frames, but that there exists an invariant speed. It then falls out from theory that massless fields propagate at this speed and both the EM field and the disturbances in space-time which are gravitational waves satisfy this. It is not because of some underlying assumption on a connection between gravity and electromagnetism.

But what if a) there are gravitons and b) they have a mass. I've read it would be below $10^{-55}kg$, but it still could be positive. Would this change the picture?

 

[Orodruin]

But what if a) there are gravitons and b) they have a mass. I've read it would be below $10^{-55}kg$, but it still could be positive. Would this change the picture?

Yes, if gravitons were massive then gravitational waves would not travel at the speed of light. Also, GR would not be the correct classical limit of the quantum gravity theory where those gravitons arise.

The same can be said for photons though. To out best knowledge, they are massless and light propagates at the speed of light. This might seem like a tautology, but in the modern view it is not.

 

[Jorrie]

I didn't know that LIGO has measured their speed (how?), even approximately.

They found no (or very little) dispersion in the waves that arrived. Since the wave packet consisted out of a changing frequency, no dispersion essentially means waves propagating at c. Secondly, the signals did not arrive simultaneously at Hanford and Livingston, so assuming that the two parts of the wave front propagated at the same speed, we can find the rough direction of the source and the rough propagation speed of the wave packet..

And if you have no doubt about it, why spend billions verifying what you already know?

Gravitational waves have been indirectly confirmed by means of binary pulsars that slowly spirals inward, losing energy exactly at the rate that GR predicted for gravitational wave emission. LIGO is the pathfinder for a whole new branch of astronomy - that's why its name included the 'O' for Observatory.

 

[giulio_hep]

Sorry, I hope my comment is not off topic here. I assume we all agree that a wave describing a massless particle travels at c. Now I want to add that only the planar waves travel at c, but that's not true for a generic signal, not formed by planar waves, (even though it is produced by the same massless field): that signal may travel at a speed less than c.
e.g. this post: "Strictly speaking, photons only travel at c if they are plane waves in free space."

Another example http://www.colorado.edu/philosophy/vstenger/Briefs/Is%20the%20Speed%20of%20Light%20Variable%3F.pdf : ... when we have a localized pulse of light that contains photons of different energies or frequencies, the pulse will not move at c but at some "group velocity." This, too, is usually less than c.
In fact, for circular wavefront propagation, the spatial shape determines the velocity of the wavefronts, so that a convex traveling wave propagates slower than a planar one (see the pdf p.679).

P.S. Also in absence of a perfect vacuum, we will have an effective speed c/n, (for instance the dispersion relation for electrons in graphene)

 

[PeterDonis]

why spend billions verifying what you already know?

You might as well ask why we spend billions on radio and optical and X-ray telescopes when we already know that electromagnetic waves exist.

 

[sanman]

Can Casimir effect and certain geometries alter the speed of Gravitational Waves the same way that it can alter the speed of light?

 

[PeterDonis]

Can Casimir effect and certain geometries alter the speed of Gravitational Waves the same way that it can alter the speed of light?

Why do you think the Casimir effect alters the speed of light?