Re: [Phys-l] Gravitons

[John Denker <jsd@av8n.com>]

On 11/09/2011 12:55 AM, Savinainen Antti wrote:
> My point is that *perhaps* we are extrapolating empirical evidence
> too far if we say that the binary star observation supports
> *gravitons*. OTOH I have no doubt that gravitational waves are real
> because of the binary star observation.

That's the sort of question that should be asked.  One should
always consider /all/ the plausible scenarios.

So the two scenarios on the table are:
  A) Gravitational waves exist and can be described as gravitons,
   in accordance with the usual rules of quantum mechanics, and
  B) Gravitational waves exist but are not subject to the usual
   rules of quantum mechanics.

We need to consider both scenarios ... but scenario A is a lot
simpler than scenario B. for the following reason (among others):

Recall that in statistical mechanics, area in phase 
space has to be measured by counting the number of "cells" ... and
as of 1898 the cell-size was an arbitrary parameter in the theory.
Let's call it "h".  Changing the value of h corresponds to shifting
the zero of entropy by an additive constant;  smaller h means smaller 
cells and therefore more entropy.  The cell-size cannot be determined 
from classical thermodynamics, because an additive constant drops out
of all the classical observables such as heat capacity.

On the other hand, the value of h is now known to high accuracy. and
has been for more than 100 years.  One line of evidence comes from
superfluid helium, which is a zero-entropy state, and there is no way 
that this zero could be shifted by an additive constant.
   http://physics.nist.gov/cgi-bin/cuu/Value?h

I reckon that gravitational waves must interact with superfluid helium
the same way they interact with anything else, by setting up tidal
stresses.

In scenario B, the phase-space of the gravitational wave is allegedly
/not/ quantized in units of h ... which means there is a huuuuge problem 
with this scenario.  The phase-space of the helium is quantized one way,
while the phase-space of the gravitational wave is (allegedly) otherwise,
and when you put them together all of thermodynamics becomes inconsistent
and invalid.

Maybe you can find a solution to this problem, but until then scenario 
A must be considered strongly preferable, since it does not suffer from 
this problem.

=======================================

Let's be clear:  I have not disproved scenario B ... in the same way
that I could not go back to 1491 and disprove the existence of the
New World by sailing a short distance west from Spain.  I can prove
that the trip will not be short or easy, but I have not proved (or
even attempted to prove) that it is impossible.

All I can say is that scenario B seems to have bugs, and until it is
debugged scenario A must be considered strongly preferable.