GR Point Masses: Seeking Help from Relativists

[humanino]

Dear relativists,

I have problems trying to understand the following statement in Forces from Connes' geometry

Einstein’s equation is nonlinear and therefore does not allow point masses as source

I would appreciate if somebody with a better understanding of GR could elaborate. I know there is no proper general local definition of gravitational energy but I always had difficulties on this aspect.

Thanks in advance for your comments.

 

[shoehorn]

The answer to your question is subtle, and it's too late here for me to come up with a convincing explanation. Nonetheless, this claim

I know there is no proper general local definition of gravitational energy

does deserve comment. There's no difficulty with defining energy at a point in general relativity, just as there's no difficulty in defining global energy; it's the notion of a quasi-local definition of energy which GR seems to lack, i.e., energy in an extended but finite region of spacetime.

Laszlo Szabados has many good papers on the ArXiv on this subject.

 

[atyy]

Apparently it's somewhat possible: Poisson, http://relativity.livingreviews.org/Articles/lrr-2004-6/ .

 

[humanino]

energy in an extended but finite region of spacetime.

Yes, that seems more accurate even to me
Thanks for the comment

Apparently it's somewhat possible: Poisson, http://relativity.livingreviews.org/Articles/lrr-2004-6/ .

I am not too sure, here he deals with point masses on a background fixed spacetime. Intuitively, I would guess, if one does not fix the background, then point masses will be black holes in GR. Maybe that is what the author meant. But black holes are "allowed", at least several authors in the past have tried to describe fundamental particles as "sort of" black holes.

 

[HallsofIvy]

It is not that you can't have a point source for gravity (theoretically) but that with non-linear equations you can't "add" solutions. That is you cannot treat an extended mass as being a "bunch of point sources" as you could with Newton's theory.

 

[humanino]

It is not that you can't have a point source for gravity (theoretically) but that with non-linear equations you can't "add" solutions. That is you cannot treat an extended mass as being a "bunch of point sources" as you could with Newton's theory.

Ah, yes, sure, that definitely makes sense of the sentence. Thank you very much.

 

[George Jones]

Read Stingray's comments in

https://www.physicsforums.com/showthread.php?t=111148

and maybe the first page of the reference that Stingray gives in post #10.

 

[humanino]

Read Stingray's comments in

https://www.physicsforums.com/showthread.php?t=111148

and maybe the first page of the reference that Stingray gives in post #10.

Thanks a lot (and to Stingray too !) for this reference.
Even the MTW does not go into those considerations.

 

[atyy]

I am not too sure, here he deals with point masses on a background fixed spacetime. Intuitively, I would guess, if one does not fix the background, then point masses will be black holes in GR. Maybe that is what the author meant. But black holes are "allowed", at least several authors in the past have tried to describe fundamental particles as "sort of" black holes.

Yes, he deals with a point mass as a black hole, and the point mass does perturb the background. However, reading Stingray's comments, there is no "source" here, since everything is a vacuum solution.

 

[humanino]

Yes, he deals with a point mass as a black hole, and the point mass does perturb the background. However, reading Stingray's comments, there is no "source" here, since everything is a vacuum solution.

I understand. Stingray's comment address my question exactly. I downloaded the reference he provided Phys. Rev. D 36, 1017 (1987).

 

[shoehorn]

Read Stingray's comments in

https://www.physicsforums.com/showthread.php?t=111148

and maybe the first page of the reference that Stingray gives in post #10.

God, I'd forgotten what an insufferable idiot pmb was.