I would like to know the main reasons why the push gravity concept is
not considered as a viable concept by mainstream science. I know it
gave rise to numerous published works, amongst which we have those of
Lorentz, H.Poincare, F.Brush, Secchi, Leray, V.Thomson, Schramm, Tait,
Isenkrahe, Preston, Jarolimek, Waachy, Rynsanek, Darwin, Majorana... so
it cannot be all wrong.
Please note, I am NOT asking about Le Sage ultramundane particles
theory (which also falls under the push gravity category), which I can
easiely discredit myself. I'm mostly interested in the concept of
electromagnetic radiation pressure of high frequency radiation acting
as the gravitational mechanism, and its shadowing creating the inverse
square law, low pressure areas.
Electromagnetism propagates as spin-1 vector bosons. If gravitation
is quantized it propagates as spin-2 tensor bosons. The selection
rules for allowed transitions are different. EM and gravitation do
not unify - not even if you are wearing Kaluza-Klein jeans. EM is
trivially shielded with alternating layers of grounded conductor
(Faraday cage) and lossy inductor (e.g., ferrite) and eventually by
electron scattering (nuclear shielding for beta-rays). Gravitation
cannot be shielded.
The source of monopole radiation is a changing monopole moment for a
charge q or for a mass m. Since charge and mass are conserved, there
can be neither monopole electromagnetic radiation nor monopole
The source of dipole radiation is a changing dipole moment.
(Punctiliously, you need a second time derivative of the dipole
moment.) For a pair of charges
d = qr + q'r'
and there's nothing special about the derivatives. For a pair of
masses, the gravitational dipole moment is
d = mr + m'r'
and its time derivative is
mv + m'v' = p + p'
By conservation of momentum the second time derivative of the
gravitational dipole moment is zero, and you can go to a center of
momentum frame and set the first derivative to zero as well. There
is no gravitational "electric dipole" radiation.
Consider the analog of "magnetic dipole" radiation. The gravitational
equivalent of the magnetic dipole moment for a pair of charges is
M = mv x r + m'v' x r'
("x" is the cross product, "mv" is the "mass current")
But M is the total angular momentum, which is also conserved. There
is no gravitational "magnetic dipole" radiation.
The next moment up is quadrupole, with no relevant conservation laws,
so gravitational quadrupole radiation is permitted.
(Toxic URL! Unsafe for children and most mammals)
Isn't this exactly what Brush wrote about? A search to see who cite
Brush's paper might be revealing. Perhaps it's just the observation that,
apart from rather speculative push-gravity effects, we don't seem to be
immersed in a bath of lots and lots of ultra-gamma rays?
Modern experiments on the falling of single cold atoms might be a
conclusive disproof, since radiation pressure due to ultra-high frequency
radiation tends to be in discrete jumps; E=hf and all that. This isn't
I see you list Majorana, although perhaps his suggestion should be called
"anti-push" (surely better than "suck"!) gravity.
My impression is that while push gravity, at least in certain limits, give
plausible results, doesn't offer any improvement over other theories of
gravitation, while introducing severe difficulties related to the exchange
of energy between the gravitational particle flux and conventional matter.
A relativistic treatment of gravitational particles (relevant if photons)
doesn't seem to improve matters such as galactic rotations curves (thanks
Rob for this).
Timo Nieminen - Home page: http://www.physics.uq.edu.au/people/nieminen/
Shrine to Spirits: http://www.users.bigpond.com/timo_nieminen/spirits.html
> I would like to know the main reasons why the push gravity concept is
> not considered as a viable concept by mainstream science.
There are a few generic objections, along with particular problems with
particular models. The main generic objections I know of are
1. Drag: As Feynman pointed out in the Feynman Lectures, anything
that's capable of "pushing" will also create drag on a moving object.
There are very strong observational limits on such drag, in the
Solar System and in binary pulsar systems.
2. Aberration: Suppose "pushing" particles move at a speed v, and
look at the effect on the Solar System. For a planet at distance d
from the Sun, the "push" will not be toward the instantaneous
position of the Sun, but towards its position at a time d/v in the
past. This is a drastic effect -- if v is the speed of light, the
Solar System would be drastically unstable over a thousand-year
(The effect of aberration is to increase the velocity of a planet,
and you might hope that drag would cancel it. But it's easy to
check that such cancellation can occur at, at most, one radial
distance from the Sun.)
3. Principle of equivalence: It is observed that gravity acts not
only on mass, but on all forms of energy. A "push gravity" theory
would have to come with an explanation of how the particles that do
the pushing manage to push against, for example, electrostatic binding
energy and the kinetic energy of electrons in an atom, and why that
"push" exactly matches the "push" against ordinary matter.
In particular, we observe that gravitational binding energy itself
gravitates. This seems to require self-interaction among the
pushing particles. On the other hand, the accuracy of the inverse
square law over long distances requires that the self-interaction
be very small -- you certainly need a mean free path larger than
the size of the Solar System if you don't want to mess up Pluto's
4. Gravitational screening: There are very strong limits on the kind
of "gravitational screening" one would expect from a "push gravity"
model -- see, for example, Unnikrishnan et al., Phys. Rev. D 63 (2001)
> Please note, I am NOT asking about Le Sage ultramundane particles
> theory (which also falls under the push gravity category), which I can
> easiely discredit myself. I'm mostly interested in the concept of
> electromagnetic radiation pressure of high frequency radiation acting
> as the gravitational mechanism, and its shadowing creating the inverse
> square law, low pressure areas.
You immediately run into trouble with the principle of equivalence,
for one thing. Electromagnetic waves don't interact with other
electromagnetic waves (except by truly tiny quantum effects); but
gravity bends light. Nor do electromagnetic waves interact with
internal energy, not with neutrinos; but these *are* affected by
gravity. You also run into grave problems with aberration (see above),
and very probably with drag. You would *further* have to explain why
this high frequency radiation is not absorbed by the Earth enough to
lead to gravitational screening of the type ruled out by experiment.
Note that "high frequency [electromagnetic] radiation" is gamma radiation.
There are experimental measurements of very high energy gamma rays, and
a fair amount is known about their spectrum. I suspect you would have
a very hard time reconciling your model with these observations.
> Electromagnetic waves don't interact with other
> electromagnetic waves (except by truly tiny quantum effects);
> Steve Carlip
Could you please provide a reference to:
"truly tiny quantum effects"
"interacting Electromagnetic waves"
OTOH, de Broglie showed that treating a particle as a standing wave
would predict many effects which were subsequently found to be just so.
If a particle is a standing wave, then (as Wheeler and Feynman got
close to saying) it is a combination of both an in and out wave at
the Compton frequency of the particle. This is indeed ultra-gamma
rays, but it is not something that "happens to the particle" but
rather "what the particle is".
I highly recommend the web site of Gabriel LaFreniere at
which has many animated GIFs showing how standing waves look and
produce all the effects of de Broglie, including waves relating to
particles in motion and much more.
> My impression is that while push gravity, at least in certain limits, give
> plausible results, doesn't offer any improvement over other theories of
> gravitation, while introducing severe difficulties related to the exchange
> of energy between the gravitational particle flux and conventional matter.
If the particle as a standing wave idea is adopted, then LeSage gravity
does follow still.
> 1. Drag: As Feynman pointed out in the Feynman Lectures, anything
> that's capable of "pushing" will also create drag on a moving object.
> There are very strong observational limits on such drag, in the
> Solar System and in binary pulsar systems.
> 2. Aberration: Suppose "pushing" particles move at a speed v, and
> look at the effect on the Solar System. For a planet at distance d
> from the Sun, the "push" will not be toward the instantaneous
> position of the Sun, but towards its position at a time d/v in the
> past. This is a drastic effect -- if v is the speed of light, the
> Solar System would be drastically unstable over a thousand-year
> time scale.
When the in and out waves are considered, it seems to me that both the
drag and aberration problems are solved. That is because there is an
almost exactly equal and opposite effect from each of the two parts
of the wave.
I say almost equal and opposite because there does have to be a
difference of 1 part in 10^40 between the two fluxes in order to
explain why gravity is that must weaker than other forces.
That difference also leads to a correct prediction of the
cosmological redshift as being a side effect of the imbalance.
These relationships are deeply satisfying.
> 3. Principle of equivalence: It is observed that gravity acts not
> only on mass, but on all forms of energy. A "push gravity" theory
> would have to come with an explanation of how the particles that do
> the pushing manage to push against, for example, electrostatic binding
> energy and the kinetic energy of electrons in an atom, and why that
> "push" exactly matches the "push" against ordinary matter.
If particles are a type of e/m standing wave then this would of
course be so.
> 4. Gravitational screening: There are very strong limits on the kind
> of "gravitational screening" one would expect from a "push gravity"
> model -- see, for example, Unnikrishnan et al., Phys. Rev. D 63 (2001)
There are of course observations of effects of shadows from eclipses
on pendulums (Maurice Allais) and on gravitational acceleration
(Wang and Wang(?)) which do show that there is screening, although
it might better be described as a mixture of screening and scattering.
I assume (perhaps incorrectly) that you are referring to the paragraph
in Vol. I, pages 7-9 to 7-10, in which Feynman commented on the theory
of a mechanism of gravitation. I was thinking that if these
"push-particles" are traveling at the speed of light, c, something like
the following might hold. Let F be the flux of these particles thoughout
space (i.e., the number of particles passing through unit area in unit
time.) Also, assume the flux is isotropic in direction. Consider a thin
sheet of matter traveling at speed u in the +X direction (traveling
broadside so you see the full area when looking along X.) To simplify,
consider only those particles going either in the +X or -X direction.
(Nothing is lost, in principle, by doing this, as you could integrate
over velocity components for other directions.) When the object is at
rest, it sees the same particle flux, F,coming from both the front side
and the hind side. But in motion, the flux it meets is increased to
F(c+u)/c and the flux from behind is decreased to F(c-u)/c. If Feynman's
anology with running in the rain applies, the thing would certainly
absorb more particles from the front than from the back per unit time,
and would feel a resistance to the motion. (With raindrops, if they hit,
they are absorbed.) However, the sheet of matter is composed of
individual absorber particles, say "atoms". Looking at a single atom,
the number of encounters per second it has with a push-particle is
proportional to the particle flux in the vicinity of the atom. The
number absorbed per second by that atom is equal to the number of
encounters per second times the probability, p, of absorption per
encounter.So, for push-particles coming from the front, an atom in the
sheet of material would absorb
N(1) = ApF(c+u)/c particles per second (1)
where A is the proportionality constant mentioned above for encounters,
and p is the probability of absorption per encounter.
This same atom would absorb from behind,
N(2) = ApF(c-u)/c particles per second. (2)
If the probability were the same in each case, the atom would certainly
absorb more per second from the front than from behind. However, the
atom (or whatever absorbing "particle") may be assumed to have an
effective absorbing diameter,d. A particle can be absorbed by it only
when it is traversing this distance through, or close by, the atom. It
takes a time t(1) = d/(c+u) for the particles meeting the atom to
traverse its sphere of influence. And for those coming from the rear, it
takes a time t(2) = d/(c-u) for them to get away from its influence. The
probability of absorption per encounter should also be proportional to
the time lapse of the encounter. (if it stays in the vicinity of the
atom longer, it should have a higher probability of absorption.)
Therefore, the probability of absorption in each case would be p(1) =
Bd/(c+u) for particles meeting it, and p(2) = Bd/(c-u) for particles
coming from behind, where B is the proportionality constant.
Replacing the probability p in equations (1) and (2) above with these
probabilities as a function of the time lapse of encounter, gives:
the number absorbed from the front per second by a given atom as
N(1) = A[Bd/(c+u)]F[(c+u)/c] = (ABdF)/c
and the number absorbed from behind per second by the same atom
N(2) = A[Bd/(c-u)]F[(c-u)/c] = (ABdF)/c
The result is the same, which shows that a moving object will absorb the
same number per second of push-particles from the front as from the
back. Therefore the object will feel no net force due to motion in this
isotropic flux of particles. (If one worries about the energy build-up,
we may assume that the particles, once absorbed, are very quickly
Whether I'm right or not,
Have one on me!
>> Electromagnetic waves don't interact with other
>> electromagnetic waves (except by truly tiny quantum effects);
> Could you please provide a reference to:
> "truly tiny quantum effects"
> "interacting Electromagnetic waves"
One place to look is www.hep.ucl.ac.uk/opal/gammagamma/gg-tutorial.html.
For observations involving real (not virtual) photons, see, for example,
Burke et al., Phys. Rev. Lett. 79 (1997) 1626 and Bamber et al., Phys.
Rev. D 60 (1999) 092004. There is even a proposal to build a photon-
photon linear collider -- see, for example, www.desy.de/~telnov/ggtesla/
For a description of the process in QED, you can look at most quantum
field theory textbooks, under "photon-photon scattering." For example,
see section 7-3-1 of Itzykson and Zuber.
Richard Saam <rds...@att.net> asked for references for this.
The usual phrase for this is "photon-photon scattering". A brief
bout of googling this phrase found (among others) the following pages
which look quite informative:
The last of these is an M.Sc thesis on the possible observability of this.
Cheng and Wu,
Phys Rev D 1, 3414 (12 June 1970),
give a detailed calculation of photon-photon scattering cross sections.
gives an experimental observation, abeit in a dilute gas rather than
in a vacuum (which would be a "purer" situation).
-- "Jonathan Thornburg -- remove -animal to reply" <jth...@aei.mpg-zebra.de>
Max-Planck-Institut fuer Gravitationsphysik (Albert-Einstein-Institut),
Golm, Germany, "Old Europe" http://www.aei.mpg.de/~jthorn/home.html
"Washing one's hands of the conflict between the powerful and the
powerless means to side with the powerful, not to be neutral."
-- quote by Freire / poster by Oxfam