Speed of Gravity Revealed
Uncle Al
c_g = c within 20%.
Fundamental questions are the really fun ones. They tell us which
theories are empirically disqualified from serious consideration -
it's the only way to make a theoretician shut up "8^>).
--
Uncle Al
http://www.mazepath.com/uncleal/eotvos.htm
(Do something naughty to physics)
John Baez
In article <avo7ru$216$1...@panther.uwo.ca>,
Uncle Al <Uncl...@hate.spam.net> wrote:
>http://www.newscientist.com/news/news.jsp?id=ns99993232
>
>c_g = c within 20%.
>
>Fundamental questions are the really fun ones. They tell us which
>theories are empirically disqualified from serious consideration -
>it's the only way to make a theoretician shut up "8^>).
But it doesn't always work. Robert Irion of Science magazine
pointed out to me that Clifford Will has written a paper
disputing Kopeikin's interpretation of the experimental results:
http://arxiv.org/abs/astro-ph/0301145
Will says that the observed effects have "*nothing* to do
with the speed of propagation of gravity, insofar as it
affects the retardation of gravitational interactions".
This matters, since Clifford Will is generally regarded as a
real expert on experimental tests of general relativity -
see his book "Theory and Experiment in Gravitational Physics".
The New Scientist article you cite contains a quote by me, which
I find rather embarrassing. The reporter, Hazel Muir, asked
me how physicists would react if the experiment measured the
speed of gravity to equal the speed of light. I wrote something
like this: "IF the experiment is confirmed and IF Kopeikin's
interpretation of the experiment is correct, the reaction will be:
Einstein wins yet again."
I stuck in these caveats because I knew from this newsgroup
that the experiment was controversial. And I distinctly remember
putting those IF's in all capital letters.
But the New Scientist article quotes me as saying: "Einstein wins yet again."
Argh!
Of course it's my fault for not remembering the First Law of
Science Journalism: "Go for the snappy sound-bite". I probably
shouldn't have commented at all without first carefully checking
Kopeikin's interpretation and Asada's objections. And I *definitely*
shouldn't have embedded a juicy quote like "Einstein wins yet again"
in an otherwise cautious sentence.
In case anyone gets confused, I'm NOT saying that Einstein DIDN'T
win yet again. I'm just saying that I personally haven't gone
through the details of this experiment, so don't take me as supporting
it. If you want to understand what's going on, read the papers by
Kopeikin, Asada and Will.
Uncle Al
It is a remarkable disagreement.
1) Kopeikin has assembled and published a conservative theoretical
model accepted by folks skilled in the art.
2) Empirical measurements were made, the handle was cranked,
credible results were obtained.
3) Other folks skilled in the art have disputed Kopeikin's model as
being a sciolism, again backed by conservative theory and published.
4) If a matter of pure theory cannot be resolved to both sides'
complete satisfaction - empirical inputs totally ignored - physics as
an endeavor has been falsified. I don't want to hear about Godel's
this and that either, because the Liar's Paradox is irrelevant here.
A model is mathematically sound or it isn't. Ambiguity does not exist
in mathematics except in identified and contrived cases.
Gravitational physics is tightly coupled to its math, certainly
outside of quantum mechanics that makes no claim to any mathematical
validity at all. There cannot be any basis for unresolved dispute
here. Either Kopeikin and Ed Fomalont are correct or Asada and Will
are correct. It is a 0% or 100% thing.
How can there even be a sustained argument? You put them in a room
with a lot of chalkboards and you see who runs out of scribble first
while bulding up from first principles. Somebody should do the parity
Eotvos experiment and perhaps overturn the entire apple cart at the
postulate level. Metric theories of gravitation say "no output,"
teleparallel theories with a gravitational pseudotensor demand an
output. Qualified apparatus exits, the math has been explicitly
calculated to a 123 trillion atom test mass - a 20 micron diameter
sphere - with astoundingly perfect scalability. Somebody should look.
--
Uncle Al
http://www.mazepath.com/uncleal/eotvos.htm
(Do something naughty to physics)
--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
"Quis custodiet ipsos custodes?" The Net!
Steve Carlip
> It is a remarkable disagreement.
> 1) Kopeikin has assembled and published a conservative theoretical
> model accepted by folks skilled in the art.
> 2) Empirical measurements were made, the handle was cranked,
> credible results were obtained.
> 3) Other folks skilled in the art have disputed Kopeikin's model as
> being a sciolism, again backed by conservative theory and published.
> 4) If a matter of pure theory cannot be resolved to both sides'
> complete satisfaction - empirical inputs totally ignored - physics as
> an endeavor has been falsified.
Note that no one is disagreeing about the prediction. Kopeikin,
Asada, and Will all agree about what the observation should see;
in fact, they end up with identical equations. The disagreement
is over the *interpetation*---does the term in question come from
the finite propagation speed of gravity, or is it a different velocity-
dependent effect of the sort that appears routinely in general relativity?
> A model is mathematically sound or it isn't. Ambiguity does
> not exist in mathematics except in identified and contrived cases.
Nobody disagrees about the math, only about the language that
should be used to describe it.
> Either Kopeikin and Ed Fomalont are correct or Asada and Will
> are correct. It is a 0% or 100% thing.
Maybe. The problem with interpreting the result is that in order to
say something about the speed of gravity, you need comparison
theories in which the speed of gravity is different. Those can't be
general relativity---there's no way to adjust the speed of gravity
in GR without simultaneously changing the speed of light. I can
imagine a situation in which different comparison theories give
different interpretations of the observations. (``The GR prediction
is different from that of theory X because theory X has a different
speed of gravity.'' ``OK, but it's different from theory Y in a very
similar way, because theory Y has a different speed of light.''
``Fine, but it's different from theory Z in almost exactly the same
way because theory Z has some extra velocity-dependent terms.'')
In this particular case, my guess is that the dispute will eventually
be resolved, but I expect it will be pretty tricky.
Steve Carlip
John Baez
>The New Scientist article you cite contains a quote by me, which
>I find rather embarrassing. The reporter, Hazel Muir, asked
>me how physicists would react if the experiment measured the
>speed of gravity to equal the speed of light. I wrote something
>like this: "IF the experiment is confirmed and IF Kopeikin's
>interpretation of the experiment is correct, the reaction will be:
>Einstein wins yet again."
>
>I stuck in these caveats because I knew from this newsgroup
>that the experiment was controversial. And I distinctly remember
>putting those IF's in all capital letters.
Actually, rereading my email, I see my memory was wrong.
In fact, I only put those IF's in a separate paragraph, where
I described what might happen if Kopeikin's experiment measured
the speed of gravity to be *different* from the speed of light.
(I was asked about this before the result of Kopeikin's experiment
was known, so I had to cover two hypothetical cases.)
Asked what would the reaction would be if he measured the speed to
*equal* the speed of light, I left out those big IF's. I just said
"Einstein wins yet again".
So, I guess I'm the only one to blame for my own silly quote.
But, at least this sordid incident may help some people see how
one has to be careful when reading popular science articles!
The "experts" being quoted do not always expect that they will
be quoted. They may not really be "experts" on the subject
at hand. And they may have been asked to comment on a development
before it has actually happened, since journalists love to be
first with a story. As a result, they may say dumb things.
And, maybe this incident will help me remember to say fewer
silly things in the future. But I doubt it.
Eugen Winkler
"Steve Carlip" <sjca...@ucdavis.edu> wrote
news:avv4tl$mnn$2...@woodrow.ucdavis.edu...
[unnecessary quoted text deleted by angry gods]
> Uncle Al <Uncl...@hate.spam.net> wrote:
> > It is a remarkable disagreement.
> Note that no one is disagreeing about the prediction. Kopeikin,
> Asada, and Will all agree about what the observation should see;
> in fact, they end up with identical equations. The disagreement
> is over the *interpetation*---does the term in question come from
> the finite propagation speed of gravity, or is it a different velocity-
> dependent effect of the sort that appears routinely in general relativity?
> In this particular case, my guess is that the dispute will eventually
> be resolved, but I expect it will be pretty tricky.
I can only see 0%:100%.
If you measure a light delay that can be assigned to the Shapiro effect of a
moving planet you have indirectly measured the speed of the moving planet
via comoving stationary metric (i.e. Schwarzschild, no spin effects). The
stationary ``deformation'' of this metric with respect to the observer frame
on earth can be estimated.
But gravitational speed should be assigned to travelling ``deformations''
within a metric and not to a stationary metric that is travelling with
respect to an observer.
Will: "...the altered propagation speed of the gravitational signal has no
effect whatsoever on the time delay to first order in v/c beyond the leading
term, ..."
The conclusion of Will is plausible to me even without PPN. The linear
motion can be boosted away, orbital spin or planet spin effects can be
neglected in this case.
Eugen Winkler
John Baez
Uncle Al <Uncl...@hate.spam.net> wrote:
>It is a remarkable disagreement.
Hmm. Really?
> 1) Kopeikin has assembled and published a conservative theoretical
>model accepted by folks skilled in the art.
Not exactly: two papers by respectable physicists (Asada and Will)
have come out within about half a year after he came out with his,
both disputing his reasoning.
> 2) Empirical measurements were made, the handle was cranked,
>credible results were obtained.
Yes, he claimed the speed of gravity was within 20% of the speed
of light. But Asada and Will basically claim he got mixed up and
is measuring the speed of light, not the speed of gravity. That
would certainly explain his results.
> 3) Other folks skilled in the art have disputed Kopeikin's model as
>being a sciolism, again backed by conservative theory and published.
A sciolism, eh? Okay, "a superficial show of learning". Yeah,
they're claiming it's a sciolism.
(Gee, it's almost a sciolism to use the word "sciolism"!)
> 4) If a matter of pure theory cannot be resolved to both sides'
>complete satisfaction - empirical inputs totally ignored - physics as
>an endeavor has been falsified.
Right: and bridges will totter, buildings will collapse, and a bolt of
lightning will come from on high to announce THE END OF PHYSICS!!!
Ahem.
Let's not get too melodramatic about this, now!
Suppose, as seems quite possible, that Will's paper is correct.
Then whether the dispute can be resolved "to both sides'
satisfaction" mainly depends on Kopeikin's personality:
whether he has what it takes to admit he's wrong after being
lauded worldwide for having made a marvelous discovery.
If he does, I'll be truly impressed. If he doesn't, so what?
It proves he's human.
What matters in science is not whether the parties to a dispute
come to agreement, but whether *most of the other experts* do.
>How can there even be a sustained argument?
Because people are human and don't like to back down.
Weber never admitted his gravitational wave detector couldn't
possibly be seeing waves. He went to his deathbed an embittered
man, arguing all the way. That wasn't a crisis in *science*.
It was just a personal tragedy.
Or think of Pons and Fleischman. I don't think these people
were incompetents, or crackpots. They just got out on too long
of a limb to back down.
I guarantee you, in a couple months at most - or weeks, if
people care enough - this will be sorted out to the satisfaction
of most physicists. General relativity is tricky, but it's not
*that* tricky.
>Somebody should do the parity
>Eotvos experiment and perhaps overturn the entire apple cart at the
>postulate level.
Yeah, yeah - I should have known that was what this was all
leading up to. :-)
Gordon D. Pusch
[big wads of unnecessary quoted text deleted by moderator]
> If you measure a light delay that can be assigned to the Shapiro effect
> of a moving planet you have indirectly measured the speed of the moving
> planet via comoving stationary metric (i.e. Schwarzschild, no spin
> effects). The stationary ``deformation'' of this metric with respect to
> the observer frame on earth can be estimated.
>
> But gravitational speed should be assigned to travelling ``deformations''
> within a metric and not to a stationary metric that is travelling with
> respect to an observer.
>
> Will: "...the altered propagation speed of the gravitational signal has
> no effect whatsoever on the time delay to first order in v/c beyond the
> leading term, ..."
In hindsight, this result should have been obvious to anyone who followed
the debate surrounding van Flandern's false claims that the "speed of
gravity" was much larger than that of light,
<http://math.ucr.edu/home/baez/RelWWW/wrong.html#speed>.
The first-order effects in (c_g/c) always canceled out, leaving second
order as the leading correction.
> The conclusion of Will is plausible to me even without PPN. The linear
> motion can be boosted away, orbital spin or planet spin effects can be
> neglected in this case.
Exactly.
-- Gordon D. Pusch
perl -e '$_ = "gdpusch\@NO.xnet.SPAM.com\n"; s/NO\.//; s/SPAM\.//; print;'
Al Jackson
> http://www.newscientist.com/news/news.jsp?id=ns99993232
>
> c_g = c within 20%.
>
> Fundamental questions are the really fun ones. They tell us which
> theories are empirically disqualified from serious consideration -
> it's the only way to make a theoretician shut up "8^>).
One interesting question, who were the referees on the Kopeikin and
Will papers?
[Moderator's note: The odds that anyone reading this post will (a)
know the answer and (b) be willing to say are quite low. -TB]
John Baez
Al Jackson <aa...@flash.net> wrote:
>One interesting question, who were the referees on the Kopeikin and
>Will papers?
>
>[Moderator's note: The odds that anyone reading this post will (a)
> know the answer and (b) be willing to say are quite low. -TB]
I know that Will was a referee on Kopeikin's paper, and recommended
that it not be published. The paper was then sent to another referee.
How do I know this? Why am I willing to reveal such information,
which is normally kept secret?
Because it's in the news:
http://www.space.com/scienceastronomy/gravity_speed_030116.html
Interestingly, Will also says he's friends with Kopeikin!
Uncle Al
Referees' identities are confidential to lessen bias (and
retribution). The plot thickens,
http://www.nature.com/nsu/030113/030113-6.html
http://www.nature.com/nsu/030106/030106-8.html
astro-ph/0301145
gr-qc/0212121
gr-qc/0210015
gr-qc/0206022
gr-qc/0201001
Uncle Al's parity Eotvos experiment is much less ambiguous and much
more important if it gives net output.
Stephen Speicher
> Uncle Al <Uncl...@hate.spam.net> wrote in message news:<avo7ru$216$1...@panther.uwo.ca>...
> > http://www.newscientist.com/news/news.jsp?id=ns99993232
> >
> > c_g = c within 20%.
> >
> > Fundamental questions are the really fun ones. They tell us which
> > theories are empirically disqualified from serious consideration -
> > it's the only way to make a theoretician shut up "8^>).
>
> One interesting question, who were the referees on the Kopeikin and
> Will papers?
>
"were" is the proper tense for Kopeikin, since he had a
half-dozen or so papers on the theory and proposed experiment
published in journals such as "Physical Review D" and "The
Astrophysical Journal." So the theory and procedures have been
around for more than three years.
Kopeikin's most current paper (gr-qc/0212121) on the
post-Newtonian treatment, and Will's treatment of such
(astro-ph/0301145), have not yet been published, so who "are"[is]
the referee[s] would be the question to ask for those two papers.
--
Stephen
s...@speicher.com
Ignorance is just a placeholder for knowledge.
Printed using 100% recycled electrons.
-----------------------------------------------------------
Eugen Winkler
"Gordon D. Pusch" <gdp...@NO.xnet.SPAM.com> wrote
Nevertheless, they have made a great measurement. Take
the wonderful team(s) plus equipment(s) to a space-trip
with say two-three distant space ships (for optimum S/N,
like on earth). Everybody should agree, that they will
measure an effect depending on the relative position and
velocity of spaceships with respect to the source-Jupiter
axis. They will find delays induced by gravitation,
highly relevant for a cosmic GPS. I would say, measuring
a linear velocity effect would be highly remarkable,
determining the velocity-independent effects would be
extra. Probably they could even accelerate their
spacecrafts while making measurements and find more
``higher terms''.
As long as the Jupiter metric is stationary, why should
there be a gravitational signal travelling from Jupiter
towards the light ray that could allow to determine
the gravity speed? The relative speed of Jupiter with
respect to the light source could be interesting regarding
relativistic velocity addition, but thats another part.
All people on board should see after making some
manoeuvres that most of the delay was generated
by the _stationary_ Schwarzschild metric of Jupiter,
where the effects induced by a relative motion
can be boosted away (of course most or even all
scientists could stay on earth in the mission control
center).
Eugen Winkler
Steve Carlip
> In hindsight, this result should have been obvious to anyone who followed
> the debate surrounding van Flandern's false claims that the "speed of
> gravity" was much larger than that of light,
> <http://math.ucr.edu/home/baez/RelWWW/wrong.html#speed>.
> The first-order effects in (c_g/c) always canceled out, leaving second
> order as the leading correction.
It's not so obvious in this case. The reason, basically, is that
here you have a three-body system (quasar, Jupiter, Earth; four
bodies if you add in the Sun). This makes the situation enough
more complicated that a detailed analysis is needed; it is not
obvious that the same cancellation will be exact here.
(That isn't so easy, either, since such an analysis requires some
``comparison theories'' in which c_g differs from c.)
Steve Carlip
Jim Graber
have been made to refute Kopeikin's claim
that the Jupiter-quasar observation
measures the speed of propagation of gravity.
Perhaps the most crucial one is the
detailed calculation from first principles
performed by all three published authors, i.e.
Kopeikin, Asada and Will.
The crucial point of difference is apparently isolated
in Will's appendix A.
This is indeed a deep and subtle calculation,
and I am not surprised that even Dr. Baez hesitates
to dive into these deep waters, and that Dr. Carlip
says that it is tricky and may take a while to clear up.
Kopeikin, Will and now Nordtvedt are still arguing this point.
So I will grant that this issue is *not* obvious.
However, there is another argument which is not nearly
so detailed or subtle.
This is the dimensional analysis argument.
In their (Arxiv) published papers,
both Asada and Will in effect say that they knew,
even before they began their detailed calculations,
that Kopeikin had to be wrong because the
dimensions of his result disagree with all that is
known about the gravitational field in GR.
This argument seems very close to much of the analysis
that Dr Baez and also Dr. Carlip have done on previous
topics on s.p.r and elsewhere. I wonder if they or others
would not like to comment on this argument?
TIA.
Jim Graber
Al Jackson
> In article <fa9de496.03011...@posting.google.com>,
> Al Jackson <aa...@flash.net> wrote:
>
> >One interesting question, who were the referees on the Kopeikin and
> >Will papers?
> >
> >[Moderator's note: The odds that anyone reading this post will (a)
> > know the answer and (b) be willing to say are quite low. -TB]
>
> I know that Will was a referee on Kopeikin's paper, and recommended
> that it not be published. The paper was then sent to another referee.
Notice Ken Nordtvedt has weighed in on this issue, agreeing with
Will.
Means Kopeikin is up against some GR heavy weights!
I wonder if Kip Thorne, John Wheeler, Stephen Hawking, Bob Wald or
Michael Turner had any comments on it?
Steve Carlip
Jim Graber <jgr...@mailaps.org> wrote:
> A number of different arguments
> have been made to refute Kopeikin's claim
> that the Jupiter-quasar observation
> measures the speed of propagation of gravity.
> However, there is another argument which is
> not nearly so detailed or subtle.
> This is the dimensional analysis argument.
> In their (Arxiv) published papers,
> both Asada and Will in effect say that they knew,
> even before they began their detailed calculations,
> that Kopeikin had to be wrong because the
> dimensions of his result disagree with all that is
> known about the gravitational field in GR.
Neither paper has a dimensional analysis argument.
They *do* have arguments that propagation effects
should first appear at a higher order in v/c than
Kopeikin claims. This may well be right, but it is
*not* obvious from previous work. For a two-body
problem, an argument of this sort is known to be
correct, but Kopeikin's analysis has to be considered
as at leat a three-body problem (quasar, Jupiter, and
Earth). So while this sort of argument is a good reason
to look at Kopeikin's analysis with care, it's not a proof
that he's wrong; it could just be a proof that he's very
clever.
(For the two-body problem, the simplest argument
is to approximate the motion as linear, and then go
to a frame in which one of the bodies is at rest. But
you can't do this with three bodies. A second approach
to the two-body problem is based on conservation of
energy and angular momentum, but it's certainly not
obvious how to apply that here. Will's introductory
argument does that, but at best only in the far zone.)
Steve Carlip
John Baez
>ba...@galaxy.ucr.edu (John Baez) wrote in message
>news:<b081j2$7dv$1...@glue.ucr.edu>...
>> I know that Will was a referee on Kopeikin's paper, and recommended
>> that it not be published. The paper was then sent to another referee.
>Notice Ben Nordtvedt has weighed in on this issue, agreeing with
>Will.
>Means Kopeikin is up against some GR heavy weights!
>I wonder if Kip Thorne, John Wheeler, Stephen Hawking, Bob Wald or
>Michael Turner had any comments on it?
Dunno. Amusingly, there is a post on sci.math about this issue,
which purports to be written by Roger Penrose, though it's not
Penrose's usual email address. Does anyone know if it's for real?
............................................................................
From: pen...@math.com (Roger Penrose)
Newsgroups: sci.math
Subject: Re: Speed of Gravity Revealed
Date: 14 Jan 2003 21:29:03 -0800
Organization: http://groups.google.com/
Lines: 22
Message-ID: <23be94f3.03011...@posting.google.com>
References: <avo7ru$216$1...@panther.uwo.ca> <avo9do$8j6$1...@glue.ucr.edu> <3E21C9E2...@hate.spam.net> <avv4tl$mnn$2...@woodrow.ucdavis.edu>
NNTP-Posting-Host: 202.9.180.36
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1042608544 9879 127.0.0.1 (15 Jan 2003 05:29:04 GMT)
X-Complaints-To: groups...@google.com
NNTP-Posting-Date: 15 Jan 2003 05:29:04 GMT
As I see, there is considerable excitement, over this rather
ingenious experiment to measure the "speed" of gravity.
I shall refrain from revealing the details, but I have
gone through the basic paper, and have written an analysis,
which at least, from the theoretical stand point, should
vindicate the claim.
Best,
Roger Penrose
University of Oxford
Mathematical Institute
----------------------------------------------------------------------------
24-29 St Giles'
Oxford
OX1 3LB
England
Phone (reception): +44 (0)1865 273525
Phone( direct): +44 (0)1865 273550
Fax: +44 (0)1865 273583
Email: pen...@math.com
Minor Crank
>
> It is a remarkable disagreement.
Disagreements of at least equal magnitude have occurred before, and
will continue to happen so long as people are fallible in their
judgement. You forget your history of science.
> 1) Kopeikin has assembled and published a conservative
theoretical
> model accepted by folks skilled in the art.
> 2) Empirical measurements were made, the handle was cranked,
> credible results were obtained.
> 3) Other folks skilled in the art have disputed Kopeikin's model
as
> being a sciolism, again backed by conservative theory and published.
> 4) If a matter of pure theory cannot be resolved to both sides'
> complete satisfaction - empirical inputs totally ignored - physics
as
> an endeavor has been falsified. I don't want to hear about Godel's
> this and that either, because the Liar's Paradox is irrelevant here.
Resolve to BOTH sides complete satisfaction? No, that's not necessary
for science to progress. Any science, not just physics. Think of
Einstein versus Bohr. Huxley versus Wilberforce. Duesberg versus
practically every other AIDS researcher.
Science is a self-correcting social endeavor. That doesn't mean that
SCIENTISTS are necessarily self-correcting. Individual scientists are
human, fallible, and often very stubborn.
> A model is mathematically sound or it isn't. Ambiguity does not
exist
> in mathematics except in identified and contrived cases.
> Gravitational physics is tightly coupled to its math, certainly
> outside of quantum mechanics that makes no claim to any mathematical
> validity at all. There cannot be any basis for unresolved dispute
> here. Either Kopeikin and Ed Fomalont are correct or Asada and Will
> are correct. It is a 0% or 100% thing.
>
> How can there even be a sustained argument?
The matter should be resolved quickly, yes, to the great majority of
physicists' satisfaction. Not necessarily to
Asada/Will/Kopeikin/Formalont's satisfaction. An all too human trait
is to hang on to your position long after the war is lost.
> You put them in a room
> with a lot of chalkboards and you see who runs out of scribble first
> while bulding up from first principles. Somebody should do the
parity
> Eotvos experiment and perhaps overturn the entire apple cart at the
> postulate level.
Aw, come on, Al. What in heck does your personal crusade have to do
with the issue at hand?
> Metric theories of gravitation say "no output,"
> teleparallel theories with a gravitational pseudotensor demand an
> output. Qualified apparatus exits, the math has been explicitly
> calculated to a 123 trillion atom test mass - a 20 micron diameter
> sphere - with astoundingly perfect scalability. Somebody should
look.
If you're going to go off-topic, let me repeat to you what I've
written to you before. If you want to have your experiment done, get
together with the guys with the torsion balances and the drop towers.
Write a grant proposal in joint with them. Help finance their
research, help them upgrade their equipment, or support another
postdoc, or fund a satellite experiment, try to do SOMETHING for them
to make it worthwhile to devote the considerable investment it would
take to carry out your experiment.
Minor Crank
[Moderator's note: The discussion of Al's "chiral Eotvos" experiment
is indeed getting a bit repetitive. We'd like to keep this civil,
though. -MM]
Barry
> Dunno. Amusingly, there is a post on sci.math about this issue,
> which purports to be written by Roger Penrose, though it's not
> Penrose's usual email address. Does anyone know if it's for real?
Also posted on s.p.relativity.
Definitely not for real for reasons discussed there, including the fact
that the post emanated from India - not Oxford.
Barry
Jim Graber
> Jim Graber <jgr...@mailaps.org> wrote:
> > A number of different arguments
> > have been made to refute Kopeikin's claim
> > that the Jupiter-quasar observation
> > measures the speed of propagation of gravity.
> [...]
>
> > However, there is another argument which is
> > not nearly so detailed or subtle.
> > This is the dimensional analysis argument.
> > In their (Arxiv) published papers,
> > both Asada and Will in effect say that they knew,
> > even before they began their detailed calculations,
> > that Kopeikin had to be wrong because the
> > dimensions of his result disagree with all that is
> > known about the gravitational field in GR.
>
> Neither paper has a dimensional analysis argument.
> They *do* have arguments that propagation effects
> should first appear at a higher order in v/c than
> Kopeikin claims.
You are right, an order of appearance argument is different than a
dimensional analysis. I stand corrected.
>This may well be right, but it is
> *not* obvious from previous work. For a two-body
> problem, an argument of this sort is known to be
> correct, but Kopeikin's analysis has to be considered
> as at leat a three-body problem (quasar, Jupiter, and
> Earth). So while this sort of argument is a good reason
> to look at Kopeikin's analysis with care, it's not a proof
> that he's wrong; it could just be a proof that he's very
> clever.
>
> (For the two-body problem, the simplest argument
> is to approximate the motion as linear, and then go
> to a frame in which one of the bodies is at rest.
Note that as discussed below, in the final result,
only two bodies are involved and
the motion has been reduced to its linear approximation.
In particular, no curvature, acceleration or other second
order terms appear.
>But
> you can't do this with three bodies. A second approach
> to the two-body problem is based on conservation of
> energy and angular momentum, but it's certainly not
> obvious how to apply that here. Will's introductory
> argument does that, but at best only in the far zone.)
>
> Steve Carlip
Hi Steve (aka Dr. Carlip),
(Its weird to mix Usenet style
(i.e. Steve instead of Dr. Carlip)
with academic style
(i.e. Dr. Kopeikin instead of Sergei)
(I hope I am not screwing up my netiquette, or
insulting anyone. If so, I apologize.)
I still think that basic arguments make the conclusion
clear even before the decisive detailed PPN analysis.
I think a key factor in understanding the controversy
over the Kopeikin Fomalont claim to have measured
the speed of propagation of gravity is to consider the
question:
How many bodies are effectively involved
in the Jupiter-quasar observation?
As you point out in your response to Gordon Pusch,
and as Dr. Kopeikin emphasizes in his papers,
actually at least four bodies are involved.
(Quasar, Sun, Earth, and Jupiter.)
However, in the final result, only the relative speed
of Earth and Jupiter appears in the expression for
the measured effect. Therefore, I think it
is permissible to think of this as a two body problem.
(Or even less as I discuss below)
The effects of the Sun are negligible.
So are the effects of the velocity of the Quasar.
In fact we don't even know the direction of
the velocity of the quasar, much less its magnitude.
Therefore, we are free to imagine that the
velocity of the quasar is identical to that of the Earth,
or of Jupiter, or even the Sun for that matter.
Alternatively, we can treat the quasar as merely
a direction in space, and not as a "body" at all.
Thus it is possible to reduce this problem to a problem
of two bodies and one direction, which implies that
the two body results about the order of the effect
which you refer to above are applicable to this calculation.
Going further, I suggest this is an acceptable way to
demonstrate that Kopeikin's claim is in error.
(For the record, Dr. Asada does suggest that the
problem be considered in the frame in which Jupiter
is at rest, and he also points out that the only velocity
explicitly entering the final formula is the relative velocity
of Earth and Jupiter.)
Since the key velocity involved is that of Earth and Jupiter, it
is not clear that this should be considered
the velocity of Earth's gravitational field or
the velocity of Jupiter's gravitational field.
Of course, the Shapiro effect differential that is measured
(to leading order) comes only from Jupiter's gravitational
field, not Earth's. In order to claim to have measured
the speed of propagation of Jupiter's gravitational field,
one would expect to find a velocity
that cannot be equally well considered to be
the velocity of Earth's gravitational field.
Whether you consider this to be the same argument, or
another closely related argument, it seems to me to be
another way to see clearly that Kopeikin's claim
can not possibly be true.
In fact, you can essentially consider this problem to be
a one body problem: The sun is irrelevant,
the quasar is merely a direction in space, and
even the earth can be considered to be merely
a moving observation point (or "test particle")
with no gravitational field of its own.
(You will still get the correct answer for
the additional delay observed by Fomalont and Kopeikin.)
But in this view, the gravitational field does not move
or change at all. There is therefore no possibility that
its propagation speed could possibly be measured by
this observation.
Jim Graber
Stephen Speicher
> Steve Carlip <sjca...@ucdavis.edu> wrote in message news:<b0hn3r$bhn$1...@woodrow.ucdavis.edu>...
> > This may well be right, but it is *not* obvious from previous
> > work. For a two-body problem, an argument of this sort is
> > known to be correct, but Kopeikin's analysis has to be
> > considered as at leat a three-body problem (quasar, Jupiter,
> > and Earth).
[...]
>
> As you point out in your response to Gordon Pusch, and as Dr.
> Kopeikin emphasizes in his papers, actually at least four
> bodies are involved. (Quasar, Sun, Earth, and Jupiter.)
>
> However, in the final result, only the relative speed of Earth
> and Jupiter appears in the expression for the measured effect.
> Therefore, I think it is permissible to think of this as a two
> body problem. (Or even less as I discuss below)
No. This conflates the original solution with its approximation.
Kopeikin's original paper [1] gives an exact solution (at the
linear approximation) for the light propagation equations for an
N-body system. The fact that the equations are simplified, for
calculational and experimental reasons, does not mean it was not
based on the more general N-body solution.
[1] S.M. Kopeikin and G. Schaefer, "Lorentz covariant theory of
light in gravitational fields of arbitrary-moving bodies,"
_Physical Review D_, Vol. 60, 124002, 5 Nov. 1999.
Eugen Winkler
news:Pine.LNX.4.33.03012...@localhost.localdomain...
The quasar light couples via gravitation exclusively with Jupiter
and not with the passive observer.
Lorentz velocity terms that multiply the Shapiro velocity terms can
be boosted away. The gravitational lense can be assumed to be
time-invariant. The Shapiro delay does not depend on the observers
relative velocity directly, it's the position. A constant relative
velocity can be helpful to look through a delaying lense from
different optical positions, but is not necessary.
So it is hard to identify low-order Jupiter velocity terms that could
be somehow related to the speed of gravity.
Steve Carlip already mentioned that you need comparison
theories in which the speed of gravity is different.
Eugen Winkler
Eugen Winkler
The quasar light couples via gravitation exclusively with Jupiter and not
with the observer. But using only one relative velocity term excludes the
possibility of a Newtonian effect. In this case different observer measuring
at the same time and location with different velocities can be synchronized
by a simple Lorentz-transformation _independent_ of the Shapiro effect and
Jupiter speed.
Regarding the first order terms in this context, the Shapiro delay does not
depend on the observers relative velocity directly, it's the position. An
observer needs of course time and relative velocity to look through a
delaying lense at different optical positions. The lense can be assumed to
be time-invariant.
The fact that the N-body equations have been simplified to a velocity
necessary to change position while observing the Shapiro effect indicates
that the analysis was not done with the most general assumptions according
to the initial questions.
Eugen Winkler
Stephen Speicher
[Snip comments unrelated to the subject at hand.]
>
> Steve Carlip already mentioned that you need comparison
> theories in which the speed of gravity is different.
>
As I understand the comment you refer to by Steve Carlip, namely:
"(That isn't so easy, either, since such an analysis
requires some ``comparison theories'' in which c_g
differs from c.)"
it was made in reference to another poster's concern about
first-order effects in c_g/c cancelling out, and Carlip prefaced
his response with:
"It's not so obvious in this case. The reason,
basically, is that here you have a three-body system
(quasar, Jupiter, Earth; four bodies if you add in the
Sun). This makes the situation enough more complicated
that a detailed analysis is needed; it is not obvious
that the same cancellation will be exact here."
I do not see Steve Carlip here disputing the multi-body system
which is apparent here, nor is he saying that Kopeikin's analysis
does not apply simply because, as you imply up above, comparison
theories are needed in which c_g is not identically equal to c.