Re: [Phys-l] Death of Time Dilation (or not)

[John Denker <jsd@av8n.com>]

On 04/13/2010 02:36 PM, chuck britton wrote:
> http://www.physorg.com/news190027752.html
>
> The article isn't very clear - just sensationalism??

1) The physorg report is complete bunk.  It is a travesty
of the New Scientist report.

2) The New Scientist report is itself a caricature of 
the actual scientific paper.

3) The actual paper has some strong points and some 
weak points.

  M. R. S. Hawkins
  On time dilation in quasar light curves
  http://arxiv.org/abs/1004.1824


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

1) Discussing the physorg article would be a complete
waste of time.

2) Discussing the New Scientist report would be mostly
a waste of time.

3) We can perhaps take a few moments to discuss the
physics of the actual paper. 

Executive summary:  The first part of the paper is 
valuable because it identifies some peculiarities in 
the observed data.  These peculiarities demand an 
explanation.  Alas the explanation suggested in the 
last part of the paper is dead on arrival.

Analysis:

By way of analogy, consider the following typesetting
project:

   1) Mary had a little lamb.

   2) M a r y   h a d   a   l i t t l e   l a m b . 

Case (2) is uniformly expanded relative to case (1).  
Obviously there is an increase in the distance from one 
letter to the next, and just as obviously there is an 
increase in the distance from the beginning of one _word_ 
to the next.  All corresponding lengths increase in the 
same proportion.  This should be obvious from the geometry 
of the situation, and the uniformity of the expansion.

We apply this analogy as follows:   In a supernova light 
curve, the overall timescale of the curve is the "words"
and the wavelength of the light is the "letters".  The 
central claim of the Hawkins paper is that the photons 
are expanded (redshifted) whereas the light curve is not 
expanded ("time dilated").  This is impossible.  This 
should be obvious from the geometry of the situation.  
We can safely assume that the rate of expansion of the 
universe does not change significantly over the timescale 
of one supernova event.

  Remember the rule for working almost any relativity
  problem:  Start by drawing the spacetime diagram.

  Start by drawing the spacetime diagram!

  If you do that, it should be obvious that my typographic
  analogy is a good model for the physics.

I don't mean to argue against the data.  Data is data.
But I can argue against the interpretation.  The paper
reaches a (somewhat tentative) conclusion about the 
"conventional cosmological framework" ... but the data
does not and simply cannot support any such conclusion.

I prefer the following interpretation instead:
 *) We observe that the photons are expanded (red shifted).
  That's OK.
 *) Therefore the light curves are expanded ("time dilated").
  They simply must be, by the geometry of the situation.
 *) We observe that the light curves, as received in their
  expanded state, line up when we compare light curves from
  different events.
 *) Therefore if we compute the "proper" (i.e. intrinsic)
  light curves by factoring out the expansion, they don't
  line up.  
   -- Maybe they shouldn't line up.  Maybe we have been
    comparing apples and oranges all along.
   -- Maybe they should line up, in which case the fact
    that they don't remains a mystery.