The Ivie conundrum
Van Snyder
conundrum: A black hole, by definition, prohibits the escape of absolutely
everything. So how do the gravitons get out?
He got email from Hawking's assistant that his conundrum had caused quite
a stir there. And Thorne still isn't really satisfied with his response.
Anybody have ideas?
--
vsn...@jato.Jpl.Nasa.Gov
ames!elroy!jato!vsnyder
vsn...@jato.uucp
Lee Samuel Finn
Gravitons don't get out; rather, only what is already outside the
horizon makes its way to infinity.
If you analyze the emission of gravitational radiation from a
perturbed black hole, you will find that you need regard only the
perturbation outside the hole. That perturbation behaves much as a
travelling wave in an external potential (the potential owing to the
field of the unperturbed black hole). Some of it travels down the
horizon, and some of it travels out to infinity. Nothing travels out
from the horizon; that is in fact a boundary condition imposed by the
causal structure of the spacetime there.
There are circumstances where the energy of a rotating black hole can
be reduced owing to the phenomena of super-radiance (and related to
the Penrose process for energy extraction from a rotating black hole);
however, these are understood in much the same way.
Are you sure that you have faithfully related the central question in
Ivie's conundrum? I don't see any paradox in the description above.
Graeme Williams
>conundrum: A black hole, by definition, prohibits the escape of absolutely
>everything. So how do the gravitons get out?
Well I'm not too hot on Quantum stuff, but don't virtual particles
engage in all manner of spacelike journeys, hoping over an event horizon
shouldn't be much of a problem. That leaves the question of how real
(as opposed to virtual) gravitons behave.
Graeme Williams
gcwi...@daisy.waterloo.edu
Loren Petrich
>In article <1991Apr12.1...@jato.jpl.nasa.gov> vsn...@jato.Jpl.Nasa.Gov (Van Snyder) writes:
>>One of my colleagues, Chuck Ivie, recently posed approximately the following
>>conundrum: A black hole, by definition, prohibits the escape of absolutely
>>everything. So how do the gravitons get out?
I presume the question is: if gravitons cause the
gravitational field, how can there be a black hole, when not even
gravitons can escape?
The answer is: They don't. They are always there, as it were.
The gravitational field of a black hole does not change over time. The
no-escape theorem applies to perturbations of this field, however.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Loren Petrich, the Master Blaster: lo...@sunlight.llnl.gov
Since this nodename is not widely known, you may have to try:
gen...@husky1.stmarys.ca
Maybe some mechanism like "Hawking Radiation?"
Also, Gravitons are virtual particles (I belive) and therefor should not
be subject to gravity.
--
"What is love? 'tis not hereafter; present mirth hath present laughter
- Shakespeare.
Phil Laird | E-mail: GEN...@Husky1.stmarys.ca
Dartmouth. |
Nova Scotia|
Canada |
Mark William Hopkins
> One of my colleagues, Chuck Ivie, recently posed approximately the following
> conundrum: A black hole, by definition, prohibits the escape of absolutely
> everything. So how do the gravitons get out?
>
> He got email from Hawking's assistant that his conundrum had caused quite
> a stir there. And Thorne still isn't really satisfied with his response.
>
> Anybody have ideas?
>
In article <1991Apr12...@husky1.stmarys.ca> gen...@husky1.stmarys.ca writes:
>Maybe some mechanism like "Hawking Radiation?"
>Also, Gravitons are virtual particles (I belive) and therefor should not
>be subject to gravity.
Two other answers:
(1) The whole graviton idea is way off base, as is shown by minor
inconsistencies like this. Gravity's not a field. It look nothing like a
field, it acts nothing like it. There are no gravitons. It's warped spacetime
and that's all it is!
or
(2) Particles under the Planck mass do not gravitate (!) because their
gravitational radii are lie under their Planck wavelength. Gravitation
is a statistical property of ensembles much like temperature is. The spacetime
continuum is likewise a 'statistical' approximation of an underlying
discrete structure. Points don't exist. Field infinities likewise have no
reality. Particles are not delta functions.
Schroedinger's Cat can never exist, because gravitation of the animal provides
a continuous indication of its state. But Schroedinger's cat exists on the
quantum level for small particles, so I think (2) is the explanation (and
also that (1) is true).
Marc Roussel
fi...@theory.tn.cornell.edu (Lee Samuel Finn) writes:
>In article <1991Apr12.1...@jato.jpl.nasa.gov> vsn...@jato.Jpl.Nasa.Gov (Van Snyder) writes:
I don't keep up with this area of theoretical physics very much but I
was very much surprised to see people seriously discussing a problem
of graviton dynamics. Do many physicists believe in gravitons? Is
there any experimental evidence of their existence? Is there
a coherent theory of them? Who are its main proponents? (I noticed Hawking's
name being tossed about in the above-mentioned discussion.)
Just curious.
Marc R. Roussel
mrou...@alchemy.chem.utoronto.ca
Franklin Antonio
>conundrum: A black hole, by definition, prohibits the escape of absolutely
>everything. So how do the gravitons get out?
Black hole, by definition, has an escape velocity > speed of light.
>Anybody have ideas?
Sure. Gravitons travel faster than the speed of light.
This would also explain why the gravitational attraction of the earth toward
the sun is toward the actual rather than the apparent direction of the sun.
Mutant for Hire
> I don't keep up with this area of theoretical physics very much but I
>was very much surprised to see people seriously discussing a problem
>of graviton dynamics. Do many physicists believe in gravitons? Is
>there any experimental evidence of their existence? Is there
>a coherent theory of them? Who are its main proponents? (I noticed Hawking's
>name being tossed about in the above-mentioned discussion.)
From general relativity one can calculate gravitational radiation. From this
work you can also show that gravity has helicity 2, that is, it behaves under
rotations as a spin 2 particle. (see Weinberg for an excellent derivation)
From quantum field theory, if we want a tensor particle, we have to make it a
spin 2 particle. But the problem is that there currently is no way to
renormalize spin 2 particles. (references escape me off-hand) Renormalization
means that we can actually make calculations and predictions using only a
finite number of experimentally observed parameters. So right now no one has
a way to use gravity in current field theory.
Of course, even if it did work, we'd have the problem that there would be no
way to do experiments on quantum gravity. The force is so insanely weak that
we'd be stuck trying to get a manifestation of it on earth.
There are various experimental new theories out there, string theory, twistor
theory and so on, which try to come up with a smooth way to incorporate gravity
into quantum mechanics, usually to the heavy modification of both, but the
jury is out on that one.
Martin Terman, Mutant for Hire, Physicist from Hell, Bug in the Cosmic Program
Disclaimer: This posting was produced by an infinite numbers of monkeys.
"Its an extremely inferior person who doesn't have several major inherent
contradictions in their view of the universe and themselves." --MFT
David Scott McAnally
|
|Sure. Gravitons travel faster than the speed of light.
|
|This would also explain why the gravitational attraction of the earth toward
|the sun is toward the actual rather than the apparent direction of the sun.
But the electric field of a uniformly moving electrically charged
particle is is towards (away from) the actual postion, not the apparent
position, and light (photons) travel at the speed of light.
David McAnally
kurims.kurims.kyoto-u.ac.jp
``All the other king said I was mad to build on the swamp."
King of Swamp Castle: Monty Python and the Holy Grail
Mark William Hopkins
> But the electric field of a uniformly moving electrically charged
>particle is is towards (away from) the actual postion, not the apparent
>position, and light (photons) travel at the speed of light.
Actually, you use the retarted (and/or advanced) potentials in calculating
fields under relativistic conditions from their sources. And they include
within them the shift due to the finiteness of the speed of light.
Van Snyder
|
|Sure. Gravitons travel faster than the speed of light.
|
|This would also explain why the gravitational attraction of the earth toward
|the sun is toward the actual rather than the apparent direction of the sun.
Actually, the gravitational attraction is toward the apparent direction. We've
got stuff in our navigation programs called the "General Relativity Correction"
that account for this (and other things). Isn't it amazing that GR has effects
that affect spacecraft navigation even out to Jupiter's orbit?
--
vsn...@jato.Jpl.Nasa.Gov
ames!elroy!jato!vsnyder
vsn...@jato.uucp
Han de Bruijn
> This would also explain why the gravitational attraction of the earth toward
> the sun is toward the actual rather than the apparent direction of the sun.
Is this so? Isn't gravity propagated with the speed of light too?
Can somebody confirm the above statement with experimental evidence?
Tom Van Flandern
Earlier, Franklin Antonio <ant...@qualcom.qualcomm.com> wrote:
>> Sure. Gravitons travel faster than the speed of light. This would also
>> explain why the gravitational attraction of the earth toward the sun is
>> toward the actual rather than the apparent direction of the sun.
Van Snyder <vsn...@jato.jpl.nasa.gov> writes:
> Actually, the gravitational attraction is toward the apparent direction.
> We've got stuff in our navigation programs called the "General Relativity
> Correction" that account for this (and other things).
Franklin was correct, Van. The gravitational force of the Sun points
toward the true, instantaneous position of the Sun, not toward its apparent
position (the direction the light comes from). I assume those program
corrections are for aberration, or light time delays. They certainly could
not be for gravity delays, or you couldn't get the right answers. It's
been known for the past two centuries that you must calculate as though the
effects of gravity happen everywhere instantly, even though the visible
light suffers light time transit delays (8.3 minutes from Sun to Earth).
If you were to assume that the force of gravity suffered the same
transit delay as light, integrations of the solar system's planets with
that assumption fly apart in a few hundred revolutions. The differences
are quite large, and operate to continually accelerate the orbital motion
of the planets.
So how do relativists deal with this apparent superluminal effect?
They argue that the Sun's field curves spacetime at the Earth's orbit, and
that curved spacetime is already there waiting to affect the Earth when it
arrives; so the transit time is irrelevant.
If you find that reasoning a little sloppy, I certainly do. The Sun
constantly moves along its galactic orbit, and it accelerates around the
solar system's center of mass. It could not maintain its effects at the
Earth's orbit instantaneously without something propagating out from the
Sun.
Even assuming a stationary Sun, the direction of the force caused by
curved spacetime should suffer aberration, as light does. Aberration of
light, as you know, simply means that light arrives from the direction
which is the vector sum of the light's true source direction & velocity
with the Earth's direction & velocity. Since the Earth's velocity, 1E-4 of
lightspeed, influences the direction in which light acts (for example,
light pressure has aberration), that should also true for gravity. Yet the
direction of the Sun's gravitational force and the direction of its
arriving photons are not parallel. Logically, that seems to require that
they do not act at the same effective speed.
In my opinion, this is one of those "skeletons in the closet" which
dynamicists know about, but don't like to talk about. -|Tom|-
--
Tom Van Flandern / Washington, DC / met...@well.sf.ca.us
Meta Research was founded to foster research into ideas not otherwise
supported because they conflict with mainstream theories in Astronomy.
Steinn Sigurdsson
>Earlier, Franklin Antonio <ant...@qualcom.qualcomm.com> wrote:
>>> Sure. Gravitons travel faster than the speed of light. This would also
>>> explain why the gravitational attraction of the earth toward the sun is
>>> toward the actual rather than the apparent direction of the sun.
>Van Snyder <vsn...@jato.jpl.nasa.gov> writes:
>> Actually, the gravitational attraction is toward the apparent direction.
>> We've got stuff in our navigation programs called the "General Relativity
>> Correction" that account for this (and other things).
> Franklin was correct, Van. The gravitational force of the Sun points
>toward the true, instantaneous position of the Sun, not toward its apparent
>position (the direction the light comes from). I assume those program
>corrections are for aberration, or light time delays. They certainly could
>not be for gravity delays, or you couldn't get the right answers. It's
>been known for the past two centuries that you must calculate as though the
>effects of gravity happen everywhere instantly, even though the visible
>light suffers light time transit delays (8.3 minutes from Sun to Earth).
> If you were to assume that the force of gravity suffered the same
>transit delay as light, integrations of the solar system's planets with
>that assumption fly apart in a few hundred revolutions. The differences
>are quite large, and operate to continually accelerate the orbital motion
>of the planets.
Ah, no. Van Snyder is right and you and Franklin Antonio are completely
wrong. Gravity is propagated at the speed of light, and is well described
by a retarded potential, as a first approximation. I believe JPL's
calculations usually go to third order post-Newtonian? A full relativistic
treatment is possible but decidedly non-trivial!
If your planet integrations fail with retarded
potentials I would suspect your integration scheme, the planets
certainly seem to manage in practise.
A thorough perusal of "Gravitation" bu Misner, Thorne and Wheeler
might be in order...
>Tom Van Flandern / Washington, DC / met...@well.sf.ca.us
>Meta Research was founded to foster research into ideas not otherwise
>supported because they conflict with mainstream theories in Astronomy.
Just try to make sure they do not conflict with reality!
For, "Nature will not be fooled..."
--
| Steinn Sigurdsson |I saw two shooting stars last night |
| Physics, Caltech |I wished on them but they were only satellites |
| ste...@tybalt.Caltech|Is it wrong to wish on space hardware? |
| "standard disclaimer" |I wish, I wish, I wish you'd care - B.B. 1983 |
Mcirvin
>
> If you were to assume that the force of gravity suffered the same
>transit delay as light, integrations of the solar system's planets with
>that assumption fly apart in a few hundred revolutions. The differences
>are quite large, and operate to continually accelerate the orbital motion
>of the planets.
Somebody pointed out many messages ago that the electrostatic force from
a uniformly moving charged particle points toward the actual, not the
past, position of the particle. Someone else said that one does in fact
make corrections for the delayed effect due to finite propagation speed.
Unfortunately the difference between these two statements got lost
somewhere in the discussion, and it's illuminating. You do indeed calculate
electrostatic effects based on a "retarded potential" which depends on the
position of the charged particle at a previous time; but if you do the
calculation for a particle in uniform motion, the field you get is
in fact a flat "pancake" of electric and magnetic fields, with the
electric field pointing toward or away from the current position of the
particle.
No information is propagating faster than light; since the particle is
moving uniformly, you could have deduced its current position from its
old position and velocity at any time in the past. If the particle
ACCELERATES, on the other hand, the effect of the acceleration propagates
at c. The electromagnetic force is mediated by photons, but that doesn't
mean that the static force has to lag behind a uniformly moving particle.
Gravity is much the same. The proper comparison is not between solar
gravity and light, but between solar gravity and electrostatic force,
and between gravity WAVES and light. Solar gravity doesn't have to lag
behind the sun as it moves uniformly.
>
> If you find that reasoning a little sloppy, I certainly do. The Sun
>constantly moves along its galactic orbit, and it accelerates around the
>solar system's center of mass. It could not maintain its effects at the
>Earth's orbit instantaneously without something propagating out from the
>Sun.
>--
>Tom Van Flandern / Washington, DC / met...@well.sf.ca.us
>Meta Research was founded to foster research into ideas not otherwise
>supported because they conflict with mainstream theories in Astronomy.
The Sun does indeed accelerate, but if you calculate v^2/r with v equal to
galactic orbital speed and r equal to galactic radius, I think you will
find that the galactic-orbit component is rather small. The
accelerations about the solar system's center of mass, are, I believe,
those whose delayed effects are corrected for in the computer program referred
to..
(This is off the subject, but I apologize for the misinformation about
silly terminology in supersymmetry... )
Matt McIrvin
Tom Van Flandern
Matt McIrvin <mci...@husc9.harvard.edu> writes:
> No information is propagating faster than light; since the particle is
> moving uniformly, you could have deduced its current position from its
> old position and velocity at any time in the past. If the particle
> ACCELERATES, on the other hand, the effect of the acceleration propagates
> at c. ... Solar gravity doesn't have to lag behind the sun as it moves
> uniformly.
Completely understood. Unfortunately for ease of understanding within
the context of relativity, it is still the case with gravity that even
ACCELERATIONS appear to be felt without observable delay. For example,
numerical integrations of massive contact binary stars which use any delay,
even the slight delay for the difference between a star's extrapolated
linear position and its actual curved-orbit position, will not preserve
angular momentum and will cause the system to fly apart on a time scale of
centuries (which is of course not observed to happen with real contact
binaries).
> The accelerations about the solar system's center of mass, are, I
> believe, those whose delayed effects are corrected for in the computer
> program referred to..
Not so. These still would be proportional to v/c, using the smaller v
appropriate to the barycenter. But the relativistic corrections have no
terms larger than v^2/c^2.
***************************************************************************
Han de Bruijn <rct...@dutrun2.tudelft.nl>, refers to my statement that the
Sun's gravitational force is toward its true, not its apparent, position,
and writes:
> Can somebody confirm the above statement with experimental evidence?
Every numerical integration of the planets is an experiment of sorts.
But the only thing I know of which qualifies as an "experiment" in the
strictest sense is one I performed at the U.S. Naval Observatory about 15
years ago. I considered an Earth-centered reference frame in which we may
think of the Moon and Sun as in orbit around us. Using analytic theories
(which agree with numerical integrations), I then isolated all of the
perturbing effects of the planets, oblateness of the Earth, etc., to
simplify the problem to a 3-body one of Sun-Moon-Earth as point masses.
Finally, I compared to observations to solve for the *direction* of the
perturbing force on the Moon's otherwise elliptical orbit which was due to
the presence of the Sun. The result was that the observed direction in
which the Sun's force acts to perturb the Moon's orbit from elliptical
agreed with the Sun's true, instantaneous position, with an uncertainty of
+/- 1 arc second. The Sun's apparent position is 20 arc seconds away from
the its true position.
-|Tom|-
Tom Van Flandern
Steinn Sigurdsson <ste...@nntp-server.caltech.edu> writes:
> Ah, no. Van Snyder is right and you and Franklin Antonio are completely
> wrong. Gravity is propagated at the speed of light, and is well described
> by a retarded potential, as a first approximation. I believe JPL's
> calculations usually go to third order post-Newtonian? A full
> relativistic treatment is possible but decidedly non-trivial! If your
> planet integrations fail with retarded potentials I would suspect your
> integration scheme, the planets certainly seem to manage in practise.
>
> A thorough perusal of "Gravitation" bu Misner, Thorne and Wheeler might
> be in order...
It is amazing how widespread this mis-impression is. The textbooks
just do not like to deal with it. Gravity *waves* (if they exist) would
propagate at the speed of light. But the force of gravitation itself acts
instantaneously (to the accuracy of observations). *All* numerical
integrations of the solar system, including those done at JPL by their top
man Standish, those done at MIT by Shapiro & co., and those done at the
U.S. Naval Observatory [where I was Chief of the Celestial Mechanics Branch
in the Nautical Almanac Office for many years], all calculate the forces of
each gravitating body on each other body at their true, instantaneous
positions while integrating their orbits. No delay, big or small, is
applied until one wishes to convert an integrated position to an observed
one; and then one allows for the 8.3 minutes per astronomical unit it takes
a planet's light to travel from there to here.
When doing a relativistic integration, the largest terms are all of
order v^2/c^2 (see Misner, Thorne and Wheeler, who largely gloss over the
issue we are discussing). But any "retarded potential" terms would be of
order v/c, and there are no such terms. See also the discussion in
Eddington's "Space, Time, and Gravitation" (1920; reprinted 1987, Cambridge
U. Press), p. 94, to see that this is not a new problem, but has been of
special concern since General Relativity was first confirmed at the solar
eclipse of 1918 by Eddington.
I realize that in conventional theory, nothing can go faster than
light. I explained how that is dealt with in my original message,
criticizing it as "sloppy reasoning". But whether sloppy or valid, that
curved space-time argument *is* the reasoning to explain why gravitation
appears to act instantly.
If you still doubt, ask one of the people who do this for a living.
Seek out someone at CalTech or JPL who knows dynamics well, and quiz them
on this point.
Referring to the word "ideas" in my signature, you said:
> Just try to make sure they do not conflict with reality! For, "Nature
> will not be fooled..."
Interpretations of reality can vary widely, but observed facts are
facts. The gravitational force of the Sun is either toward its true,
instantaneous position (as I have explained), or toward its apparent one
(as you claim). We cannot both be correct. I've given my arguments are
references. And you? -|Tom|-
--
Kenneth Arromdee
>facts. The gravitational force of the Sun is either toward its true,
>instantaneous position (as I have explained), or toward its apparent one
>(as you claim). We cannot both be correct. I've given my arguments are
>references. And you? -|Tom|-
OK, question then. My friend Lar Gand has taken a trip to the sun and
thinks the long distance telephone cost from there is sky high, so he decides
to send me a message by shaking the sun. I check to see what direction the
gravitational force of the sun is towards in order to receive the message,
thus receiving the message at faster than speed-of-light propagation.
What's wrong with this?
--
"If God can do anything, can he float a loan even he can't repay?"
--Blair Houghton, cross-posting
Kenneth Arromdee (UUCP: ....!jhunix!arromdee; BITNET: arromdee@jhuvm;
INTERNET: arro...@cs.jhu.edu)
Steinn Sigurdsson
>Matt McIrvin <mci...@husc9.harvard.edu> writes:
>> No information is propagating faster than light; since the particle is
>> moving uniformly, you could have deduced its current position from its
>> old position and velocity at any time in the past. If the particle
>> ACCELERATES, on the other hand, the effect of the acceleration propagates
>> at c. ... Solar gravity doesn't have to lag behind the sun as it moves
>> uniformly.
> Completely understood. Unfortunately for ease of understanding within
>the context of relativity, it is still the case with gravity that even
>ACCELERATIONS appear to be felt without observable delay. For example,
>numerical integrations of massive contact binary stars which use any delay,
>even the slight delay for the difference between a star's extrapolated
>linear position and its actual curved-orbit position, will not preserve
>angular momentum and will cause the system to fly apart on a time scale of
>centuries (which is of course not observed to happen with real contact
>binaries).
This is in accord with my experience integrating hard binaries,
(and should have casued me to pause before challenging Tom's
original post on this). The binaries are described to O(v^2/c^2)
by Newtonian "action at a distance" potential.
Higher order terms account for the periastron advance and the
decrease in orbital period - consistent with GR for observed
neutron star binaries (although a paper at the Jan 1991 AAS claimed
two Main-sequence binaries in conflict with GR (Carroll \etal 1991),
I have not gone through their paper carefully enough to say if
their claim is valid)
.
>> The accelerations about the solar system's center of mass, are, I
>> believe, those whose delayed effects are corrected for in the computer
>> program referred to..
> Not so. These still would be proportional to v/c, using the smaller v
>appropriate to the barycenter. But the relativistic corrections have no
>terms larger than v^2/c^2.
--
Steinn Sigurdsson
>Steinn Sigurdsson <ste...@nntp-server.caltech.edu> writes:
Ooops, I retract that. But not that!
!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!
>> Ah, no. Van Snyder is right and you and Franklin Antonio are completely
>> wrong. Gravity is propagated at the speed of light, and is well described
>> by a retarded potential, as a first approximation. I believe JPL's
>> calculations usually go to third order post-Newtonian? A full
>> relativistic treatment is possible but decidedly non-trivial! If your
>> planet integrations fail with retarded potentials I would suspect your
>> integration scheme, the planets certainly seem to manage in practise.
>>
>> A thorough perusal of "Gravitation" bu Misner, Thorne and Wheeler might
>> be in order...
Ok, I was wrong, the gravitational force experienced by a mass
in orbit about the sun is towards the _actual_ not the _apparent_
position of the sun. (Ref Lightman, Press, Price and Teukolksy, 12.4)
Coulomb forces are aberred - hence the first order "action at a distance"
effect.
I do however insist that this is _not_ due to any instantaneous
propagation of gravitational influences, but a consequence of assuming
(and experiencing) a static source for the gravitational potential.
> It is amazing how widespread this mis-impression is. The textbooks
>just do not like to deal with it. Gravity *waves* (if they exist) would
>propagate at the speed of light.
Agreed.
>But the force of gravitation itself acts
>instantaneously (to the accuracy of observations). *All* numerical
>integrations of the solar system, including those done at JPL by their top
>man Standish, those done at MIT by Shapiro & co., and those done at the
>U.S. Naval Observatory [where I was Chief of the Celestial Mechanics Branch
>in the Nautical Almanac Office for many years], all calculate the forces of
>each gravitating body on each other body at their true, instantaneous
>positions while integrating their orbits. No delay, big or small, is
>applied until one wishes to convert an integrated position to an observed
>one; and then one allows for the 8.3 minutes per astronomical unit it takes
>a planet's light to travel from there to here.
> When doing a relativistic integration, the largest terms are all of
>order v^2/c^2 (see Misner, Thorne and Wheeler, who largely gloss over the
>issue we are discussing). But any "retarded potential" terms would be of
>order v/c, and there are no such terms. >
No, MTW deals with this in sections 18.4 and 39.9 onwards.
The eqn of motion to Post-Newtonian order if given by eqn 39.64,
The first order terms vanish when integrating over the source,
leaving v^2/c^2 as the dominant term.
> I realize that in conventional theory, nothing can go faster than
>light. I explained how that is dealt with in my original message,
>criticizing it as "sloppy reasoning". But whether sloppy or valid, that
>curved space-time argument *is* the reasoning to explain why gravitation
>appears to act instantly.
Agreed, I jumped on you too quickly, sorry, you were right. I knee-jerked
on the original posters "this proves gravity acts instantaneously" and
I was wrong about apparent aberration - should have known to check.
> If you still doubt, ask one of the people who do this for a living.
>Seek out someone at CalTech or JPL who knows dynamics well, and quiz them
>on this point.
Thanks, but I've embarrassed myself enough, I managed to figure
this one out by myself, I hope...
>Referring to the word "ideas" in my signature, you said:
>> Just try to make sure they do not conflict with reality! For, "Nature
>> will not be fooled..."
I stand by that, I never said I could not be fooled ;-)
> Interpretations of reality can vary widely, but observed facts are
>facts. The gravitational force of the Sun is either toward its true,
>instantaneous position (as I have explained), or toward its apparent one
>(as you claim). We cannot both be correct. I've given my arguments are
>references. And you? -|Tom|-
I concede defeat on this point. The gravitational force of the sun is
towards its true instantaneous position.
This is however in now way due to anything propagating faster
then light. The first orders terms cancel for a static source by
virtue of symmetry, nothing else needs be evoked (Ref MTW, p. 1080 )
This better be the last time I post to sci.physics before checking
my facts!
Jose Castejon-Amenedo
met...@well.sf.ca.us (Tom Van Flandern) writes:
> Gravity *waves* (if they exist) would propagate at the speed of
> light.
I'll be insufferably nitpicking in this one :-) and point out
that this is certainly true of general relativity, but not of all
gravity theories. Anyway, I think it is true of all currently viable
theories.
Graeme Williams
> Interpretations of reality can vary widely, but observed facts are
>facts. The gravitational force of the Sun is either toward its true,
>instantaneous position (as I have explained), or toward its apparent one
>(as you claim).
"Instantaneous" isn't a well defined concept, it's not Lorentz invariant.
(I seem to say this every week!)
So suppose we have a spacecraft on its way to Jupiter, and I calculate
the sun's "instantaneous" position. Whose version of "instantaneous"
do I use - that defined by the rest frame of the spacecraft?, the rest
frame of the sun? or some other rest frame? And are the differences
between these choices of the order v/c? Or perhaps (v/c)^2 ?
Graeme Williams
gcwi...@daisy.waterloo.edu
Toby Kelsey
> It is amazing how widespread this mis-impression is. The textbooks
>just do not like to deal with it. Gravity *waves* (if they exist) would
>propagate at the speed of light. But the force of gravitation itself acts
>instantaneously (to the accuracy of observations). *All* numerical
>integrations of the solar system, including those done at JPL by their top
>man Standish, those done at MIT by Shapiro & co., and those done at the
>U.S. Naval Observatory [where I was Chief of the Celestial Mechanics Branch
>in the Nautical Almanac Office for many years], all calculate the forces of
>each gravitating body on each other body at their true, instantaneous
>positions while integrating their orbits.
Forgive my naive question, but aren't gravity waves caused by
the lag in the response of space-time to the accelerations of
massive objects? The finite speed of gravity waves then
implies gravitational attraction is not to the "instantaneous"
position of an accelerating object. If that were really so
then gravity waves would have infinity velocity and we would
have faster-than-light communication.
In article <24...@well.sf.ca.us> you mention
>numerical integrations of massive contact binary stars which use any delay,
>even the slight delay for the difference between a star's extrapolated
>linear position and its actual curved-orbit position, will not preserve
>angular momentum and will cause the system to fly apart on a time scale of
>centuries (which is of course not observed to happen with real contact
>binaries).
I was under the impression that the slowing of the period of
binary star systems due to gravity wave emission had been
experimentally seen, and agreed with the calculated values.
>Tom Van Flandern / Washington, DC / met...@well.sf.ca.us
--
Toby Kelsey (chp...@uk.ac.bath.gdr)
Tom Van Flandern
Kenneth Arromdee <arro...@cs.jhu.edu> writes:
> My friend Lar Gand has taken a trip to the sun and thinks the long
> distance telephone cost from there is sky high, so he decides to send me
> a message by shaking the sun. I check to see what direction the
> gravitational force of the sun is towards in order to receive the
> message, thus receiving the message at faster than speed-of-light
> propagation. What's wrong with this?
In fact, the Sun is actually being "shaken" in a manner of speaking by
the planets, which cause it to constantly accelerate around the center of
mass (barycenter) of the solar system. At times, that barycenter lies
entirely outside the body of the Sun. In order to successfully integrate
the motions of the planets, one calculates as if each planet instantly
feels where the true Sun is. If one were to calculate with the assumption
that the planets only knew where the Sun apparently was based on a linear
extrapolation of the motion of its visible disk (information which arrives
with a lighttime delay), then the solar system integration will go badly,
and the results will not match observations by a significant amount.
* * * * * * * * * * * * * * * * * * * * * * * * *
and Steve Robiner <rob...@mizar.usc.edu> writes:
> Whoa, whoa. Now then that's FTL communication! If I can move a massive
> body about at the other end of the solar system (now this is a thought
> experiment since it's not likely to be possible in the near future...)
> like say, a black hole or brown dwarf, or whatever, then its gravitional
> force on other bodies will be instantaneous???!?!?
It is true that the Earth gravitationally senses and responds to the
acceleration of the Sun about the solar system's barycenter *before* the
Sun's light arrives to show us what acceleration occurred. However the Sun
is just responding to the gravitation of the planets in a perfectly
predictable way, as is evident from the fact that we do predict its motion
years in advance. So no actual faster-than-light information is
necessarily communicated.
But I admit that I am personally very uneasy with the argument that
the true dynamical path of the Sun is imitated by the curved spacetime
continuum surrounding it, so that the Earth can respond to it instantly.
This is part of what I alluded to earlier as "sloppy reasoning". I now
strongly suspect that the conventional explanation of this is contrived,
and may contain a logical fallacy. Does anyone here have a really clear
way to explain the answer to the question of either respondent? In my
reading, the explanations in MTW and other books are tautological, and
assume the conclusion they are trying to prove. -|Tom|-
Bob Lodenkamper
attraction combined with gravity waves in any way analogous to the
situation in electrodynamics where the total field can be split up
into the instantaneous Coulomb term plus a radiation field - in
several different ways, if I remember right?
My - uninformed, I admit - guess is that in certain nearly linear
approximations, General Relativity may admit this division of terms.
If so, the Coulomb terms and the radiation terms separately are
unphysical in certain respects, but the combined field is
well-behaved. In cases where, in addition, the radiative corrections
are quantitatively negligible, one would predict the position of stars
or planets using an apparently unphysical model, but this
unphysicality is not really a problem.
- Bob
Todd A. Brun
>
>> My friend Lar Gand has taken a trip to the sun and thinks the long
>> distance telephone cost from there is sky high, so he decides to send me
>> a message by shaking the sun. I check to see what direction the
>> gravitational force of the sun is towards in order to receive the
>> message, thus receiving the message at faster than speed-of-light
>> propagation. What's wrong with this?
>
> In fact, the Sun is actually being "shaken" in a manner of speaking by
>the planets, which cause it to constantly accelerate around the center of
>mass (barycenter) of the solar system. At times, that barycenter lies
>entirely outside the body of the Sun. In order to successfully integrate
>the motions of the planets, one calculates as if each planet instantly
>feels where the true Sun is. If one were to calculate with the assumption
>
>and Steve Robiner <rob...@mizar.usc.edu> writes:
>
>> Whoa, whoa. Now then that's FTL communication! If I can move a massive
>> body about at the other end of the solar system (now this is a thought
>> experiment since it's not likely to be possible in the near future...)
>> like say, a black hole or brown dwarf, or whatever, then its gravitional
>> force on other bodies will be instantaneous???!?!?
>
> It is true that the Earth gravitationally senses and responds to the
>acceleration of the Sun about the solar system's barycenter *before* the
>Sun's light arrives to show us what acceleration occurred. However the Sun
>is just responding to the gravitation of the planets in a perfectly
>predictable way, as is evident from the fact that we do predict its motion
>years in advance. So no actual faster-than-light information is
>necessarily communicated.
>
forth, one is using a force that is *not* gravitational in nature. One is
not solving simple orbital equations (if I may mis-use the term "simple") but
rather equations with a forcing term which *isn't* gravitational (it would
presumably be electromagnetic, since nuclear forces aren't too good at
moving things as big as the sun). In these forced equations, there *would*
be something like a retarded potential effect. All that this really
means is that if one can move the gravitating bodies about somehow, their
positions stop being predictable; not, when you think about it, a very deep
insight, and certainly nothing to do with FTL communication (there's enough
rubbish about that floating around the net).
Of course, if you are moving the sun gravitationally, then you are just
another body moving around according to a set of unforced equations, and your
influence will be felt in the direction of your instantaneous position. Since
that motion is (in principle) exactly predictable, it's hard to see how you
can use it to transmit a message. As soon as some conscious person adjusts
things to send a signal, you're back to the forced equation case.
Hope this clarifies things.
-- Todd
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Department of Physics
California Institute of Technology
"Feezeeks? Ve don't need no stinking feezeeks!"
Tom Van Flandern
Toby Kelsey <chp...@gdr.bath.ac.uk> writes:
> Forgive my naive question, but aren't gravity waves caused by the lag in
> the response of space-time to the accelerations of massive objects? The
> finite speed of gravity waves then implies gravitational attraction is
> not to the "instantaneous" position of an accelerating object. If that
> were really so then gravity waves would have infinity velocity and we
> would have faster-than-light communication.
Toby, you ask an excellent question, and one that I cannot really
answer. But in telling you what I can, let me first distinguish between
questions about WHAT is true (the observations); and questions about WHY it
is that way (the theory). It is observed that gravitational attraction is
(always, in our experience) toward the true, instantaneous position of the
source body. This holds true even if the source is accelerating. The
direction of the gravitational force is in general NOT parallel to the
direction from which the photons arrive. I believe it to be true that no
one who has looked into this question disputes these statements as factual
about WHAT is true. It is only in answer to the WHY part that explanations
differ.
When speaking about the WHY of the behavior of gravitation, physicists
draw a distinction between the gravitational POTENTIAL (related to local
spacetime), the gravitational FORCE (gradient of the potential, or
curvature of spacetime), and gravitational WAVES (= ?).
For a long time, I thought as you did that gravity waves were caused
by the lag in the response of space-time to the accelerations of massive
objects. But then physicists began talking about detecting a gravitational
wave from the 1987 supernova in the LMC. By my calculations, assume that a
20-solar-mass supernova at 50 kpc instantly ceased to exist (leaving not
even pure energy behind), and the resultant "wave" traveled to the solar
system. Then if we had a detector 1 astronomical unit long, the change in
acceleration between its two ends when that "wave" arrived would be about
1/30 of the classical radius of the electron per year per year! If a
detector resided on Earth, every passing auto on distant freeways, the
movements of every bird, the dropping of a leaf to the ground from a nearby
tree, indeed the rebound of every colliding air molecule in a room, would
all produce orders of magnitude stronger gravitational waves than an LMC
supernova.
So I have concluded that I don't, after all, know what physicists are
talking about when they talk about gravitational waves. Can anyone else
help? -|Tom|-
--
Tom Van Flandern / Washington, DC / met...@well.sf.ca.us
Steve Robiner
>In article <24...@well.sf.ca.us> met...@well.sf.ca.us (Tom Van Flandern) writes:
>>
>>Kenneth Arromdee <arro...@cs.jhu.edu> writes:
>>
>>and Steve Robiner <rob...@mizar.usc.edu> writes:
>>
>>> Whoa, whoa. Now then that's FTL communication! If I can move a massive
>>> body about at the other end of the solar system (now this is a thought
>>> experiment since it's not likely to be possible in the near future...)
>>> like say, a black hole or brown dwarf, or whatever, then its gravitional
>>> force on other bodies will be instantaneous???!?!?
>>
>forth, one is using a force that is *not* gravitational in nature. One is
>not solving simple orbital equations (if I may mis-use the term "simple") but
>rather equations with a forcing term which *isn't* gravitational (it would
>presumably be electromagnetic, since nuclear forces aren't too good at
>moving things as big as the sun). In these forced equations, there *would*
>be something like a retarded potential effect. All that this really
>means is that if one can move the gravitating bodies about somehow, their
>positions stop being predictable; not, when you think about it, a very deep
Yes, exactly, not predictable by a standard set of equations, but
yet if FTL gravity is true, readable at a distance *instantly* -
so therefore it would imply FTL communication!
>insight, and certainly nothing to do with FTL communication (there's enough
>rubbish about that floating around the net).
I think the confusion results from a predicted result based on this solar
system where no external forces are being applied. In the solar system no
FTL commincation is happening because everything is just moving along a
path prescribed by the space-time it's in. The question is, what if
an external (non-gravitional) force is applied (a coding of a message,
perhaps) to one of the gravitional bodies in the system. If that
body's change in position is felt by the other bodies in the solar
system instantly (or for that matter - on the other side of the universe)
then you have FTL communication.
=steve=
=steve=
joefish
Tom>
I would have to call what Weber and others have looked for, and
apparently not found, is more like tidal pulses, than gravity waves.
The gravitational potential at the distance to a supernova is so
small that I don't even care to think about it. But tides are
real, and maybe a sensor can be built to detect a tidal pulse
of fairly small amplitude. I am taking bets that it is a waste
of resources, because it is still not certain that gravity works
by gravity waves or gravitons or any other mediating particle.
I also see no way that any natural event at that distance would
cause any pulse at all, because a star should have the same
gravity field at a distance regardless of whether it explodes
or condenses or collapses.
I would rather see at least part of the effort going into testing
simple things like placing a small body inside a plastic tube
with massive end plugs. According to classical theory, the
body should stay against one end if placed there, but by extending
the principle of equivalence to this special case, I think the
small body will move toward the center, moving about half the
distance in less than an hour, and half again, an so forth,
never quite getting there.
This would also be an excellent test to see if the present popular
concept of gravitational attraction should be examined closer,
rather than doing experiments which are expected to confirm the
present thought. I thought students should try to disprove
the accepted thought.
There is much about Newtonian gravitation that is bothersome. Any
force carrier must exhibit a flux density through given cross sections.
The strongest steel cables could not even begin to pull down on a
mountain with the force that gravity does, and gravity must do it
within the same cross sections as the steel cables. That means
that gravity supposedly can generate flux densities exceeding
hundreds of thousands of pounds through a cross section of one square
inch.
Since both gravity and inertia are able to generate these apparent
forces, and in fact are the only process that can generate forces
in these flux densities, there has to be something going on that
we are not aware of. Popular thought credits all the mass in
the universe for inertia, but only particular bodies for gravity.
This is a very serious conflict, because gravitation and inertia
are essentially one and the same thing.
Would a test body move toward the ends of the tube I mentioned
if heavy masses were in each end, or would it move towards the center.
Joe Fischer joe...@disk.UUCP
David Scott McAnally
Yes, but moving a massive body around would mean producing gravity waves, and
these would travel at speeds less than the speed of light. If some violent
moved the sun, the earth would not be affected in its orbit until at least 8
minutes later, and Neptune would not be affected until at least 6 hours later
(Is this value right?). In short, the affects would not propogate any faster
than light. We would expect to _see_ the change before (or at least at the
same time as) the earth feels it.
David McAnally
kurims.kurims.kyoto-u.ac.jp
``All the other kings said I was mad to build on the swamp."
Andrew Ferencz
>>
>>> Whoa, whoa. Now then that's FTL communication! If I can move a
massive
>>> body about at the other end of the solar system (now this is a
thought
>>> experiment since it's not likely to be possible in the near
future...)
>>> like say, a black hole or brown dwarf, or whatever, then its
gravitional
>>> force on other bodies will be instantaneous???!?!?
I think one must remember that when one object moves another, their
center of mass will remain in the same place. That is, as the planets
move around the sun, their (and the sun's) center of mass will be at the
same place.
Therefore, If this Lar Gand is big enough to push the sun away from him,
for example, he will move in the opposite direction, an thier center of
mass will remain in the same way. Since the gravitrational atraction to
those two objects will remain in the same place, their movement could not
be detected by using gravitational fields!!!!
Andrew
amix!underground!aferencz
----------
Andrew Ferencz
UUCP: uunet!cbmvax!amix!undrground!aferencz
Internet: undrground!afer...@amix.commodore.com
Christopher Neufeld
>
>I think one must remember that when one object moves another, their
>center of mass will remain in the same place. That is, as the planets
>move around the sun, their (and the sun's) center of mass will be at the
>same place.
>
>Therefore, If this Lar Gand is big enough to push the sun away from him,
>for example, he will move in the opposite direction, an thier center of
>mass will remain in the same way. Since the gravitrational atraction to
>those two objects will remain in the same place, their movement could not
>be detected by using gravitational fields!!!!
>
gravitational quadrupole is affected by this distortion. An observer
close enough not to be swamped by the 1/r^4 force falloff might notice
this, and would be able to distinguish between forces directed at the
apparent positions of the masses, or at the instantaneous "true"
position outside the light cone.
As one poster has already remarked, instantaneous in whose reference
frame? How can the force be directed at a point which is not itself
Lorenz invariant?
--
Christopher Neufeld....Just a graduate student | Flash: morning star seen
neu...@aurora.physics.utoronto.ca Ad astra! | in evening! Baffled
cneufeld@{pnet91,pro-cco}.cts.com | astronomers: "could mean
"Don't edit reality for the sake of simplicity" | second coming of Elvis!"
Steve Robiner
>center of mass will remain in the same place. That is, as the planets
>move around the sun, their (and the sun's) center of mass will be at the
>same place.
>
>Therefore, If this Lar Gand is big enough to push the sun away from him,
>for example, he will move in the opposite direction, an thier center of
>mass will remain in the same way. Since the gravitrational atraction to
>those two objects will remain in the same place, their movement could not
>be detected by using gravitational fields!!!!
>
>Andrew
Sorry, wrong.
If the system is far enough away from the detector, then the center of
mass might be important, but each of the bodies has it's own center of
mass as well which can independantly affect another body, especially if
it is close.
Also, I never said the force had to be gravitational anyway - I just said
'shake' or 'move' the massive body. You don't necessarily have to do it
by attracting it to another mass.
=steve=
David Scott McAnally
>propagate at the speed of light. But the force of gravitation itself acts
>instantaneously (to the accuracy of observations). *All* numerical
Could this be because the sun and planets travelling in an inertial
fashion have well-determined trajectories, and so the positions at all
future times are predetermined. Each part of the metric propogates itself
at less than the speed of light, so if this is followed up, the above
quote might be shown to square with causality.
David McAnally
kurims.kurims.kyoto-u.ac.jp
``Other kings said I was daft to build a castle on a swamp."
David Scott McAnally
>the planets, which cause it to constantly accelerate around the center of
>mass (barycenter) of the solar system. At times, that barycenter lies
But according to GR, the sun is still undergoing inertial motion.