On first reading Hawking's talk

[marcus]

somebody got a transcript of Hawking's 21 July talk at dublin
to the blogger/cosmologist and generally cool guy Sean Carroll
and Peter Woit, the mathematician blogger of Columbia U and
Not Even Wrong fame, passed the link on to us

http://pancake.uchicago.edu/%7Ecarroll/hawkingdublin.txt

So I printed out these words of Hawking
and sit reading them
they are no-formulas, readable, and fairly much
for general audience

He says among other arresting things:

"There is a problem describing what happens, because Strictly speaking,
the only observables in quantum gravity, are the values of the field
at infinity. One can not define the field at some point in the middle,
because there is quantum uncertainty in where the measurement is
done."

I dont know if I buy that. One observes correlations of measurments made in this life in one's own backyard---I would have thought----not at infinity.
the entertainment of this talk is high, and it is very well written I'd say, but there is something

whole-hog-graphic

about it (pun chuckle chuckle) that I mistrust.

[marcus]

Is there anyone here who found hawking's talk, which we now have a transcript of, at all convincing?

Please tell us about it, what did you find persuasive
or not persuasive.

A science reporter who was at the conference
was just being interviewed on the radio
and said that after Hawking's lecture he talked to a lot of scientists attending the conference
and his impression that they mostly remained unconvinced.
And that particular science journalist could be wrong or
have talked to an unrepresentative sample of the audience etc.

Ultimately it comes down to what you think.
Were you persuaded?

[pnjabiloafer]

Well, im undecided still, im waiting to see the detailed paper.

[wolram]

Take a large bucket and go wash that multi teated quadruped

[wolram]

Well im sorry but it seems hawkings just had to publish something, why i don't
know, but this is as lame as it gets, i also know im a nobody whose opinion is
negligible, but who can give any credence to this.

[selfAdjoint]

Is there anyone here who found hawking's talk, which we now have a transcript of, at all convincing?

Please tell us about it, what did you find persuasive
or not persuasive.

A science reporter who was at the conference
was just being interviewed on the radio
and said that after Hawking's lecture he talked to a lot of scientists attending the conference
and his impression that they mostly remained unconvinced.
And that particular science journalist could be wrong or
have talked to an unrepresentative sample of the audience etc.

Ultimately it comes down to what you think.
Were you persuaded?


I'm at the level of "I see why he says that, but I don't see how it settles the case". The unmentioned 4 ton elephant in the room is who says these euclidean path integrals from infinitely far away have anything to do with the problem? Yes I know about shooting beams from infinity to near the black hole and then seeing what has been added/subtracted when it gets back to infinity again. But there are too many questions around the path integral methods and Wick rotation in curved space time to convince me.

[Hurkyl]

Well, there is the holographic principle; the boundary of a region contains enough information to reconstruct the interior. "At infinity" is the "boundary" of the universe, so given this principle, measurements at infinity should be sufficient to describe anything in the universe. I'm not sure how much I believe that uncertainties go away at infinity, though.

[selfAdjoint]

Yes he does seem to be using holography and your satz of "at infinity" is the boundary of the universe, supposed to be the CFT that expresses all the physics of the bulk. I could see infinity as shorthand for far away and asymptotically flat, but this more literal interpretation as the infinitely far off boundary of spacetime gives me the willies. I don't care who does it, it just seems manifestly unsound to me.

[Hurkyl]

Maybe we live in projective 4-space? :biggrin:

I don't know much about CFT, but I've seen similar things in other fields. For instance, there's this classic technique from complex analysis:

Problem: Find the contour integral of some function along the real axis.
Solution: analytically extend the function to the upper half plane, and consider contour integrals along the boundary of semicircles with the center at the origin. Using residues, you can find the values of these integrals. Then, you look at just the line integral along the curve of the semicircle and find the limit as the radius goes to infinity. Then, you can subtract to get the value of the integral along the real axis. (of course, this is all under suitable conditions) The net result is as if you did an integral around the boundary of the entire upper half-plane!


And in Algebraic Geometry, a lot of affine varieties can be extended to projective varieties (which have these points "at infinity") and back, so "at infinity" is always fair game.