These quotes show that HEINRICH HERTZ has indeed found in his
experiments that ELECTROSTATIC effects propagate INSTANTANOUSLY
and NOT at c as generally assumed.
Interference effects between waves in a wire orginating from the
center of a brass disc and the electrostatic effects of the disc
can be measured. If oscillations of 35.7 Megahertz are used and
the speed of the wire waves is 200 000 km/s, we get a wave length
of 5.6 meters. If electrostatic effects propagate instantanously,
after 5.6 m, 11.2 m, 16.8 m, ... the wire must be in phase with
the electrostatic effect of the disc.
 > Heinrich Hertz, Gesammelte Werke, Band 2, Leibzig, 1894:
 I'll reproduce the translation by D. E. Jones from

 "Electric Waves being Researches on the Propagation of Electric Action
 with Finite Velocity through Space", Dover, 1962:
 Introduction, p. 8:

 "Nor was there any greater difficulty in producing interference between
 the action which had travelled along the wire and that which had
 travelled through the air, and thus in comparing their phases.
 Now if both actions were propagated, as I expected, with one and the
 same finite velocity, they must at all distances interfere with the same
 phase. ... But when I had carefully set up the apparatus and carried out
 the experiment, I found that the phase of the interference was obviously
 different at different distances, and that the alternation was such as
 would correspond to an infinite rate of propagation in air.
 Disheartened, I gave up experimenting."
 On the finite velocity of propagation of electromagnetic actions
 p. 110:

 "The total force may be split up into the electrostatic part
 and the electromagnetic part; there is no doubt that at shorter
 distances the former, at greater distances the latter,
 preponderates and settles the direction of the total force."
 p. 118:

 "In the second place, we notice that the retardation of phase
 proceeds more rapidly in the neighborhood of the origin than
 at a distance from it. All the rows agree in showing this.
 An alteration of the speed of propagation is not probable.
 We can with much better reason attribute this phenomenon to the fact
 that we are making use of the total force [...] which can be split up
 into the electrostatic force and the electromagnetic.
 Now, according to theory, it is probable that the former,
 which preponderates in the neighborhood of the primary oscillation,
 is propagated more rapidly than the latter, which is almost the only
 factor of importance at a distance."
 p. 120:

 "The interference does not change sign every 2.8 m. Therefore the
 electromagnetic interactions are not propagated with infinite velocity."
 p. 121:

 "Since the interferences undoubtedly change sign after 2.8 m in the
 neighborhood of the primary oscillation, we might conclude that the
 electrostatic force which here predominates is propagated with
 infinite velocity."
 I have a few questions on those procedures and results:

 Based on which requirements/measurements did Hertz decide whether or not
 "the apparatus was set up carefully"?, and
Apart from the constancy of the primary oscillation and the
possibility to compare at different distances the phase of the
wire wave with the phase of the action propagating through
the air, there is nothing which must be set up carefully.
 How did he determine pairwise distance relations such as "2.8 m" or
 "neighborhood" (if not by employing Einstein's procedures, based on
 the exchange of light signals)?
A simple tape measure is enough to determine at which distances
from the emitter the interference changes sign.
In any case, it would make sense to repeat this crucial experiment.
Wolfgang Gottfried G.
Previous relevant posts:
http://www.deja.com/=dnc/getdoc.xp?AN=530719931
http://www.deja.com/=dnc/getdoc.xp?AN=531225851
http://www.deja.com/=dnc/getdoc.xp?AN=531506436
http://www.deja.com/=dnc/getdoc.xp?AN=531614175
> In any case, it would make sense to repeat this crucial experiment.
Sure, experimental results may be obtained in any number of trials.
If the experimental procedure being conducted is _reproducible_
then it is even meaningful to _compare_ and relate the individual
results of the various trials to each other, quite irrespective of
whether or not they happen to be "surprising" or to "make sense".
Of course, in order to determine which (if any) aspects of Hertz'
experiments are unambiguously reproducible, one must reach
clarification and agreement on ...
> [Frank Wappler wrote:]
> > a few questions on those procedures and results:
> > Based on which requirements/measurements did Hertz decide whether or not
> > "the apparatus was set up carefully"?
> Apart from the constancy of the primary oscillation
How did Hertz/how would you determine whether or not
the primary oscillation is "constant"; and wrt. what else?
> and the possibility to compare at different distances the phase
> of the wire wave with the phase of the action propagating through
> the air, there is nothing which must be set up carefully.
How did Hertz/how would you determine which "actions" are associated
with "the wire" and/or "the air"; whether exclusively so, or at all?
> > How did he determine pairwise distance relations such as "2.8 m" or
> > "neighborhood" (if not by employing Einstein's procedures, based on
> > the exchange of light signals)?]
> A simple tape measure is enough [...]
How did Hertz/how would you determine the calibration relation
between various distinct "simple tape measures", or
between the same "simple tape measure" being used in various distinct trials
(if not by employing Einstein's procedures, based on
the exchange of light signals)?
Regards, Frank W ~@) R
p.s.
> Frank Wappler, thank you very much for the translated quotes.
You're welcome.
> > [(Btw., the copy of Band 1 that I found in the library
> > was published by J. A. Barth in _Leipzig_ ...)]
Would you kindly doublecheck where the copy of
"Heinrich Hertz, Gesammelte Werke, Band 2" was published,
to which _you_'ve been referring?
Please read what you wrote. In a _static_ situation, _nothing_
propagates, because everything is static and nothing changes.
Note, please, that it is well known that the Maxwell's equations
plus the Lorentz force law predict no aberration for the
electromagnetic force due to a charge which is in inertial motion
wrt the observer (see, e.g. Feynman volume 2).
It appears that you make the same mistake for E&M as does Van
Flandern for gravity  absence of aberration is _not_ a measurement
of any "propagation speed".
The situation is different for E&M and gravitation  for
E&M this holds only for inertial motion, but holds exactly;
for gravitation it holds only approximately (except for a
universe containing only a single mass), but it holds for
any motion whatsoever (assuming a small and isolated mass,
and the observer is of negligible mass). For both theories
this is a really statement about the symmetry group of the
theory and the boundary conditions applied (no radiation
zooming in from infinity).
[I could not decipher the physical situation of Hertz's
experiment(s) from your remarks, so I respond _only_ to
the words quoted above.]
Statd differently: It does not look to me like Hertz's experiments
refute classical electrodynamics, in which E&M effects _do_ propagate
at c (not instantaneously). But I have not performed an analysis, I
merely note that nobody else claims they refute classical E&M.
Tom Roberts tjro...@lucent.com
Please, can you tell me what abberation is in this context? And how does
it affect measurement of propagation speeds?
Thanks in advance,
George
Where gravity is concerned, propagation speed should
not be an issue, but those who have the olden ideas that
something leaves mass A and travels to Mass B, and pulls
on Mass B, in just the right amount, to cause any amount
of Mass B to accelerate at the same rate..... They think
whatever travels has to have a speed, like the speed of
light.
Abberation is the same as always, the difference
in the angle between incoming signals and a straight
line to the source of the incoming signals.
Gravity does not propagate, "gravitational
radiation" may, if it exists.
Mostly everyone expects gravity to propagate,
if no other reason than, that they cannot see how
gravity could work without those little invisible
somethings pull on whatever is "attracted" by gravity.
For 85 years the theory and the math has
existed that deals with gravity _not_ pulling,
but because discussing a physical mechanism for
gravity that works by geometry rather than "pulling"
would be embarrassing, not much is said about how
the mechanism of gravitation works, or what it is.
This theory specifies that inertia causes
"falling" or "freefall". Nothing about inertia
would need or produce abberation, assuming inertia
arises from mass itself.
Nothing in General Relativity requires any
propagation of anything, As Far As I Know, for
the mechanism of gravitation.
If gravitational radiation (gravity waves)
exists, then it would propagate at the speed of
light, not because it is photons, but because
nothing can propagate faster than light, that
would mean arriving before starting the trip.
The motions of objects affected by gravity
are known very precisely, but the mechanism of
gravitation is still a complete mystery.
What seems to me to be certain, is that
gravity and inertia are closely related (a physics
teachers organization has a publication called
"Gravity and Inertia".
There may well be a fundamental process
that causes both gravity and inertia, and if there
is, present day physics is lagging far behind where
it should be.
Yet people go to school for twenty years, and
still learn and write about "propagation", "attraction",
"force", and other things which may not have any bearing
on the phenomenon of gravitation.
Joe Fischer
 How did Hertz/how would you determine whether or not
 the primary oscillation is "constant"; and wrt. what else?
 How did Hertz/how would you determine which "actions" are associated
 with "the wire" and/or "the air"; whether exclusively so, or at all?
 How did Hertz/how would you determine the calibration relation
 between various distinct "simple tape measures", or between
 the same "simple tape measure" being used in various distinct
 trials (if not by employing Einstein's procedures, based on
 the exchange of light signals)?
Hertz was an excellent experimenter and I do not doubt that
it was rather easy for him to produce a constant oscillation
resulting in wire waves with a wavelength of 5.6 m. I'm neither
knowledgeable about nor especially interested in experimental
physics and it was some years ago when I studied some of Hertz'
texts. Now I deal only with passages I marked then. But Hertz
has well decribed his experimental techniques and I'm sure that
for an experimental physicist it should not be very difficult
to compare the phases of the oscillation propagating in a wire
with the actions propagating through the air.
Once again Heinrich Hertz about his first attempt:
"But when I had carefully set up the apparatus and carried out
the experiment, I found that the phase of the interference
was obviously different at different distances, and that the
alternation was such as would correspond to an infinite rate
of propagation in air. Disheartened, I gave up experimenting."
Nevertheless, also after having learned to (produce and) detect
an 'electrodynamic force' propagating at c, the actions at a
distance from the 'electrostatic force' did not disappear:
"Since the interferences undoubtedly change sign after 2.8 m in the
neighborhood of the primary oscillation, we might conclude that the
electrostatic force which here predominates is propagated with
infinite velocity."
 > > [(Btw., the copy of Band 1 that I found in the library
 > > was published by J. A. Barth in _Leipzig_ ...)]

 Would you kindly doublecheck where the copy of
 "Heinrich Hertz, Gesammelte Werke, Band 2" was published,
 to which _you_'ve been referring?
My copy:
Band 2, Leibzig, 1894, Johann Ambrosius Barth, (Arthur Meiner)
Unchanged reprint of the edition of 1895, #3112
SÃ¤ndig Reprint Verlag, 1988, Vaduz / Liechtenstein
Interestingly I live in Vaduz, a capital with a population of
around six thousand residents.
Cheers, Wolfgang
>  How did Hertz/how would you determine whether or not
>  the primary oscillation is "constant"; and wrt. what else?
>  How did he determine pairwise distance relations such as "2.8 m" or
>  "neighborhood" (if not by employing Einstein's procedures, based on
>  the exchange of light signals)?
> Hertz was an excellent experimenter and I do not doubt that
> it was rather easy for him to produce a constant oscillation
> resulting in wire waves with a wavelength of 5.6 m.
Easy or not, I'd like to be sure how he or you determined
(i.e. what he or you meant by)
"constant oscillation" and "meter" _at all_, trial by trial.
IMHO, a prerequisit for being an excellent experimenter
is to select experimental procedures and to define notions
that can be unambiguously communicated and reproduced.
> I'm neither knowledgeable about nor especially interested
> in experimental physics [...]
> I'm sure that for an experimental physicist it should not be
> very difficult to compare [...]
As an experimental physicist I can assure you that it is meaningless
to compare results of distinct experimental trials if they were
not obtained by the same measurement procedure.
This of course requires that the selected experimental and
measurement procedures be reproducible in the first place;
hence the conventional choice of the Einstein procedures.
SkÃ³l, Frank W ~@) R
p.s.
>  Would you kindly doublecheck where the copy of
>  "Heinrich Hertz, Gesammelte Werke, Band 2" was published,
>  to which _you_'ve been referring?
> My copy:
> Band 2, Leibzig, 1894, Johann Ambrosius Barth, (Arthur Meiner)
> Unchanged reprint of the edition of 1895, #3112
> SÃ¤ndig Reprint Verlag, 1988, Vaduz / Liechtenstein
> Interestingly I live in Vaduz, a capital with a population of
> around six thousand residents.
Well  I rather refer to the copy available through the Library of Congress
Author: Hertz, Heinrich, 18571894.
Title: Gesammelte Werke, von Heinrich Hertz.
Published: Leipzig, J. A. Barth, 189495 [v. 1, '95]
Description: 3 v. front. (port.) diagrs. (1 fold.) 22 cm.
LC Call No.: QC3.H49
Dewey No.: 530.81
Notes: Each volume has also special t.p.
Includes index.
Bd. I. Schriften vermischten Inhalts,
herausgegeben von Ph. Lenard.Bd. II. Untersuchungen Ã¼ber
die Ausbreitung der elektrischen Kraft. 2. Aufl.Bd. III.
Die Prinzipien der Mechanik.
Subjects: Physics.
Other authors: Lenard, Philipp Eduard Anton, 18621947, ed.
Control No.: 7361082
(Btw., the volume two in question seems to have been a second edition of
Author: Hertz, Heinrich, 18571894.
Title: Untersuchungen ueber die Ausbreitung der
elektrischen Kraft.
Published: Leipzig, J. A. Barth, 1892.
LC Call No.: QC661.H59)
Equally interesting is that only few (if any) of the about six thousand
cases annually in which "Liechtenstein" is misspelled in various ways
are being considered international incidents.
 > Hertz was an excellent experimenter and I do not doubt that
 > it was rather easy for him to produce a constant oscillation
 > resulting in wire waves with a wavelength of 5.6 m.

 Easy or not, I'd like to be sure how he or you determined
 (i.e. what he or you meant by)
 "constant oscillation" and "meter" _at all_, trial by trial.
 As an experimental physicist I can assure you that it is meaningless
 to compare results of distinct experimental trials if they were
 not obtained by the same measurement procedure.

 This of course requires that the selected experimental and
 measurement procedures be reproducible in the first place;
 hence the conventional choice of the Einstein procedures.
I agree with you in principle, but in the case we are
discussing, there is absolutely no need to include SR
reasonings. The experiments of Hertz precede SR by several
years and in addition to that they are not sensitive to
time dilation, length contraction and so on, because they
are performed in a frame at rest wrt the earth's surface.
So even if SR is assumed, clocks, rulers and simultaneity
are well enough defined to decide whether the measured
actions propagate at around 200'000 km/s (longitudinal wire
waves), at around 300'000 km/s (photons in the air) or
rather instantanously ('electrostatic force').
In any case, it is generally admitted that the situation
nearby an emitting dipole antenna does not agree with the
normal explanation and the drawings of waves peeling off,
which can be found in any textbook. So if we take seriously
logic we must conclude that this explanation is in principle
wrong.
If the electrostatic effects of an oscillating disc
propagated indeed at finite speed, longitudinal waves
similar to the ones propagating in wires would be a
consequence.
There is a fundamental difference between on the one
hand electrostatic and magnetic interactions and on the
other hand electromagnetic radiation:
After radiation having separated from a dipol, the dipol
is no longer affected by the radiation. Whether it is
absorbed by an antenna or not has no retroaction on the
emitting dipole. However, the induction of a current in
a neighbouring conductor has a retroaction on the dipole.
( translated from: http://members.lol.li/twostone/a3.html )
Gruss, Wolfgang
My previous two posts of this thread:
http://www.deja.com/=dnc/getdoc.xp?AN=532021977
http://www.deja.com/=dnc/getdoc.xp?AN=532263367
... if not by the conventional measurement procedures, i.e. by
the measurement procedures of SR: Einstein's calibration procedure
(which can be based on the exchange of roundtrip light signals),
and the associated distance definition (which is in turn
based on the results of sucessful calibration procedures, as
"c/2 * light_signal_roundtrip_interval;
if the beginning and end states of that interval, and the state
of reflecting the light signal, have been calibrated between
those two observers through Einstein's calibration procedure").
> So even if SR is assumed, clocks, rulers and simultaneity
> are well enough defined to decide whether the measured
> actions propagate at around 200'000 km/s (longitudinal wire
> waves), at around 300'000 km/s (photons in the air) or
> rather instantanously ('electrostatic force').
Yes, using the SR procedures one can measure relations between
various clocks ("simultaneity", i.e. pairwise calibration of
states/readings/proper_times; and subsequent calibration of intervals),
as well as relations between various ruler ends (measurement of
pairwise "distances");
and subsequent derivation of pairwise velocity, acceleration etc.
Note that measurements of instantaneous exchange of light signals,
i.e. zero light signal roundtrip intervals, imply _zero distance_
between those particular pairs of sources/emitters/clocks/ruler_ends
who exchange those particular signals, in those particular trials;
by the conventional procedure for determining "pairwise distance", above.
> In any case, it is generally admitted that the situation
> nearby an emitting dipole antenna does not agree with the
> normal explanation and the drawings of waves peeling off,
> which can be found in any textbook.
I'd be the first to admit that it can be difficult to reproduce in a
drawing the electromagnetic and other gauge fields nearby any particular
collection of about 10^24 or more charges, in any particular trial.
However, I don't know of disagreements with the conventional
explanation/preselected measurement procedure by which those fields
are conventionally determined in the first place, at least in principle:
namely via the Einstein procedures to measure coordinate relations
of charges/observers/clocks/ruler_ends wrt. each other,
and, taking those measurements as constraints, the subsequent
derivation of "the most probable" potential and fields
via the principle of stationary action.
Of course anyone is free to suggest other, preferebly reproducible,
measurement procedures; whose results might disagree with those
obtained by the conventional procedure, in any paricular trial,
given the same collected observations/data.
But _which_ "nonnormal" measurement procedures, specificly?
And _are they_ as unambiguously reproducible as the Einstein procedures
and the principle of stationary action?, i.e. requiring no a priori
assumptions about any charges/observers/clocks/ruler_ends other than
_that_ they can exchange light signals and compare/count/do_math,
at least in principle?
> There is a fundamental difference between on the one
> hand electrostatic and magnetic interactions and on the
> other hand electromagnetic radiation:
> After radiation having separated from a dipol, the dipol
> is no longer affected by the radiation. Whether it is
> absorbed by an antenna or not has no retroaction on the
> emitting dipole. However, the induction of a current in
> a neighbouring conductor has a retroaction on the dipole.
What do you mean by "radiation having separated from"? 
how do you suggest to measure coordinates of "radiation" wrt. any
of the charges/observers/clocks/ruler_ends who constitute a "dipol"?
Also: if a pair of charges/observers/clocks/ruler_ends exchange a light
signal, is this not observable by all charges/observers/clocks/ruler_ends,
at least in principle, incl. by those two themselves?
Salut, Frank W ~@) R
 > There is a fundamental difference between on the one
 > hand electrostatic and magnetic interactions and on the
 > other hand electromagnetic radiation:

 > After radiation having separated from a dipol, the dipol
 > is no longer affected by the radiation. Whether it is
 > absorbed by an antenna or not has no retroaction on the
 > emitting dipole. However, the induction of a current in
 > a neighbouring conductor has a retroaction on the dipole.

 What do you mean by "radiation having separated from"? 
 how do you suggest to measure coordinates of "radiation" wrt. any
 of the charges/observers/clocks/ruler_ends who constitute a "dipol"?
Energy and momentum conservation are empirical facts at least
in the case of highfrequency radiation. If for instance an
atom emits a photon, the atom suffers a recoil impulse in the
opposite direction. The atom also loses the mass corresponding
to the emerging photon.
In the same way as there is an interaction between the emitting
atom and the emerging photon, there is an interaction between
an emitting dipole and the emerging radiation. When radiating,
the dipole loses energy which normally is compensated by an
energy supply. (Unlike real photons, QED 'photons' have neither
mass nor impulse, therefore a charged body can maintain its
electrostatic effect even if it emits many more QED 'photons'
than it absorbes.)
So whereas in the beginning there is an undeniable interaction
between an emitter and its raditation, there is certainly no
interaction between emitter and emitted radiation after the
latter having separated from the former, e.g. when the
radiation is absorbed by a radio antenna.
Gruss, Wolfgang
http://members.lol.li/twostone/E/physics1.html
>So whereas in the beginning there is an undeniable interaction
>between an emitter and its raditation, there is certainly no
>interaction between emitter and emitted radiation after the
>latter having separated from the former, e.g. when the
>radiation is absorbed by a radio antenna.
>
Physically, the separation means the "exchanged" photon satisfies
the condition q**2 = 0, i.e., the photon is real, not virtual and
therefore, is free to propagate without being tied to its emitter,
the antenna. Any form of the em field is carried by the photon, it's
just that the static or quasi static electric and magnetic fields
are what one sees as a result of virtual photon exchange (not on
mass shell) where the photon does not live independent of its
emitter.
> Frank Wappler wrote:
>  What do you mean by "radiation having separated from"? 
>  how do you suggest to measure coordinates of "radiation" wrt. any
>  of the charges/observers/clocks/ruler_ends who constitute a "dipol"?
You're addressing my questions more directly below.
Untill then let's establish some context:
> Energy and momentum conservation are empirical facts at least
> in the case of highfrequency radiation.
"Empirical facts" must be based on measurements, and therefore require
the selection of reproducible measurement procedures to begin with.
Measurements of energy and momentum (pairwise, of observers wrt. each other)
can be used to _determine_ which pairs exchanged which quantities
in the first place, and whether or to which extent some region is
"homogenious", "isotropic" and "closed".
> If for instance an atom emits a photon, the atom suffers a recoil
> impulse [momentum] in the opposite direction.
Don't forget to specify _how and wrt. whom_ "momentum" and "direction"
were determined, and with whom this photon was exchanged.
> The atom also loses the mass corresponding to the emerging photon.
I'm not sure what you mean by "the photon emerging" without defining
how to measure "coordinates of the photon" wrt. that atom;
AFAIK _exchange of a photon_ by a pair can be defined as
"exchange of _no_thing/charge/observer", but merely as
a transition between two states of one
correlated to a transition between two states of the other,
without reference to "coordinates of the photon".
You have a point:
Given measurements of energy and momentum of some particular observer A
(wrt. any other, B), one can express an invariant of A (wrt. any B):
"(energy_B( A )/c^2)^2  (momentum_B( A )/c)^2", i.e. "mass_A^2".
Consequently a series of measurements of energy and momentum of A
(wrt. any other, B) can be used to determine the values of this invariant
throughout the course of those measurements.
Note that the measurements of "energy E" and "momentum p" are in turn
defined through measurements of pairwise coordinate relations "t" and "x",
E == hbar i d/dt(), and p == hbar/i d/dx.
> In the same way as there is an interaction between the emitting
> atom and the emerging photon, there is an interaction between
> an emitting dipole and the emerging radiation.
I consider "interaction" a relation between pairs of
atoms/dipoles/things/charges/observers (or systems thereof);
for instance _their_ exchange of photons, radiation, and/or light signals
_with each other_.
Atoms exchanging photons or dipoles exchanging radiation
are surely in some sense similar interactions.
> Unlike real photons, QED 'photons' have neither mass nor impulse,
AFAIK, QED distinguishes exchange of real, onmassshell photons;
and exchange of virtual, offmassshell photons.
Both notions are defined in terms of measured exchanged energy and momentum
(i.e. fourmomentum, and the corresponding invariant mass).
> a charged body can maintain its electrostatic effect
> even if it emits many more QED 'photons' than it absorbes.
That seems implied in the notion of exchange of a "photon" as
"exchange of no_thing/charge" but merely correlated transitions.
> So whereas in the beginning there is an undeniable interaction
> between an emitter and its raditation, there is certainly no
> interaction between emitter and emitted radiation after the
> latter having separated from the former, e.g. when the
> radiation is absorbed by a radio antenna.
O.k.  IIUC by "interaction at the beginning" and
"interaction after separation" apparently you're referring to the
two transitions, between two states of the "dipole sender",
and between two states of the "radio antenna",
which toghether constitute the exchange of radiation, a light signal,
one or many photons.
Still, I consider "interaction _with_ radiation" misleading; instead,
there's "interaction between observers, between sender and antenna, etc.".
Return to what I thought was the main point:
Do you understand that Einstein's calibration procedure and the
associated distance definition allow this sender and this antenna
to determine their coordinate relations wrt. each other,
such that everyone else can understand their results?
(given sufficiently interactions/exchanges of light signals
with each other, as well as with certain auxiliary observers.)
How else do you suggest that they should measure their pairwise
coordinate relations, "simultaneity", "distance", "velocity", etc. at all?
>
>> Unlike real photons, QED 'photons' have neither mass nor impulse,
>
>AFAIK, QED distinguishes exchange of real, onmassshell photons;
>and exchange of virtual, offmassshell photons.
>Both notions are defined in terms of measured exchanged energy and momentum
>(i.e. fourmomentum, and the corresponding invariant mass).
>
Virtual photons can not only have mass, they can have the longitudinal
polarization states that result. There is only one difference between
a real photon and a virtual one: a real one satisfies q^2 = 0 and
may freely propagate.
>
>> a charged body can maintain its electrostatic effect
>> even if it emits many more QED 'photons' than it absorbes.
>
This doesnt make sense. Photons carry momentum. The electrostatic
effect is the momentum carried by the photons due to the charge
of an object. The photons do not carry charge and so cannot change
any feature related to charge. All a photon can do is change the
momentum of another charged body upon absorption. Any charged body
may absorb a photon. If the absorber and emitter are different
bodies, the result looks like a force because the momentum change
in the emitter has to equal the momentum change in the absorber.
Virtual photons do not exist independent of the emitter. If a
charged object emits a photon, it must either reabsorb it to
satisfy heisenberg, or another charged object must absorb it to
satisfy heisenberg. Since the exchange could occur by swapping
the roles of emitter/absorber, it is symmetric and should be
interpereted that way. In that sense, an object absorbs the same
number of virtual photons it emits.
This will come as a terrible shock to those who have been involved
in the design of very successful electromagnetic wave (r.f) based
communications systems and equipment for the past 80+ years.
Also, James Clerk Maxwell must be turning over in his grave on
learning this rather remarkable fact! ;)
Harry C.
> Frank Wappler [wrote]:
> > AFAIK, QED distinguishes exchange of real, onmassshell photons;
> > and exchange of virtual, offmassshell photons.
> > Both notions are defined in terms of measured exchanged energy and
> > momentum (i.e. fourmomentum, and the corresponding invariant mass).
> There is only one difference between
> a real photon and a virtual one: a real one satisfies q^2 = 0
Precisely; where q denotes the fourmomentum that has been transferred
between the two charges/observers who have exchanged this photon.
> and may freely propagate.
What do you mean by "a photon propagating", "freely" ?
How would you measure "coordinates of a photon" wrt. other observers?
> If a charged object emits a photon
Note that this by itself is kinematically forbidden, at least for
real photons. Instead one considers _exchange_ of photons by pairs
(of not necessarily distinct charges/observers).
> Since the exchange could occur by swapping the roles of emitter/absorber,
... yes, surely emitter and absorber can observe each other, mutually ...
> it is symmetric and should be interpreted that way.
The conventional measurement procedures, which are based
on light_signal_roundtrips obviously use the fact that this symmetry
is not perfect at all:
if I see the deer, the deer has thereby not necessarily
seen me in turn already
(unless the roundtrip interval and corresponding distance
is zero, of course);
observer A's state "Ax: I've observed B in state B5" does _not require_
or imply that this state of observer B is in turn
"B5: I've observed A in state Ax, observing B5";
but _if_ this were found, then A and B would be able to conclude
zero light signal roundtrip interval wrt. each other,
in this particular trial and pair of states.
Otherwise there _may_ well occur state
"B9: I've observed A in state Ax, observing B5",
from which observer B can conclude that state B9 is _after_ state B5.
(That's a way by which an observer can _order_ her/his/its own
set of states in the first place.)
> In that sense, an object absorbs the same number of virtual photons
> it emits.
That's obvious for photons which _one_ particular object/charge/observer
emits/absorbs/exchanges only wrt. him/her/it_self_.
But otherwise I fail to find sense in your point.
Consider a pair of distinct objects/charges/observers who exchange
just one (virtual) photon.
Did either one emit and absorb the same number?
Or are you suggesting that this example is not possible or illdefined?
Except for one "little" thing  in a basis where individual photons
have welldefined 4momenta, the number operator does not have a well
defined value, and you cannot count them! In a basis where the number
operator has a welldefined value, individual photons do not have
welldefined 4momenta! This is intimitely related to the fact that
photons are indistinguishable Bosons, and to the necessity to
symmetrize the wavefunction over Bosons and antisymmetrize over
Fermions  in a perturbative approximation this intermixes all the
Bosons/Fermions in all of the different diagrams....
If one _really_ tries to take into account _all_ of the properties
of photons in QED, the discussion gets so convoluted and complicated
that it is essentially useless....
There seems to be a Heisenberg uncertainty relationship
between correctness and understandibility (:)).
Tom Roberts tjro...@lucent.com

No!, except in the most general Fourier measurment
sense there is no HUP associated with the
"correctness and understandibility" of photons. It
is JUST inadequate description like we(some of us)
see so often in SR. On the contrary, when the
description is done correctly the understanding
seems simple. Dennis has presented you with many
historical examples of this truth. In the photon
case we need, as you note, to treat the
indistinguishability explicitly [ when we talk
about combinations and averages as QM does ] but
even more importantly we must distinguish the
transient and the steady state description. This
means distinguishing what is happening to the
photon envelope or shape or antenna pattern
distinctly from what is happening to the carrier
phase and frequency under the envelope which in
steady state is not constrained by c. Good seeing.
JD

Sent via Deja.com http://www.deja.com/
Before you buy.
>: These quotes show that HEINRICH HERTZ has indeed found in his
>: experiments that ELECTROSTATIC effects propagate INSTANTANOUSLY
>: and NOT at c as generally assumed.
>
>This will come as a terrible shock to those who have been involved
>in the design of very successful electromagnetic wave (r.f) based
No pun intended, I presume?
>What do you mean by "a photon propagating", "freely" ?
>How would you measure "coordinates of a photon" wrt. other observers?
>
It isnt tied to the source from which it was emitted. One that
doesnt leave its emitter off massshell and other equivalent
statements.
Since the only means to determine its coordinates is by observing
it (and thus destroying it in the process), I'd have to define
its coordinates by the point of observation or by inference from
the point of origin and a density matrix containing whatever
information there is to be had wrt things that might render
the probability of finding it along some directions, less likely
than others (like a slit). However, I'm not going to suggest
it has coordinates beyond what a probabilty amplitude buys
you. Obviously, there's no frame to obtain the photon's idea
of where it is (I've always wondered what the world looks
like to a photon).
>Note that this by itself is kinematically forbidden, at least for
>real photons. Instead one considers _exchange_ of photons by pairs
>(of not necessarily distinct charges/observers).
>
Sure.
>if I see the deer, the deer has thereby not necessarily
>seen me in turn already
However, murphy's law states the deer then will see you
and you wont necessarily see that it moved before pulling
the trigger and missing.
>(unless the roundtrip interval and corresponding distance
>is zero, of course);
>That's obvious for photons which _one_ particular object/charge/observer
>emits/absorbs/exchanges only wrt. him/her/it_self_.
>
>But otherwise I fail to find sense in your point.
>Consider a pair of distinct objects/charges/observers who exchange
>just one (virtual) photon.
>Did either one emit and absorb the same number?
>Or are you suggesting that this example is not possible or illdefined?
I'm suggesting just that. The physical situation has to include
both possibilities since it isnt possible to distinguish between
the two. Actually, you have to include more. If you have identical
particles, you cant even say incoming particle line goes with
which outgoing line. I've always taken the concept of distinguih
ability to be literal fact: If something cannot be observed in
principle, it isnt real or physically meaningful to consider and
attempting to do so "as if it were to see what happens", from the
looks of some of the more creative posts I've seen, leads to
pleas to support research to extract free energy from little
rotating particles in the vacuum :)
I'm not sure it makes sense to enumerate them at all, since a
proper count requires the infinity of diagrams.
> I'd have to define its coordinates by the point of observation
How would you determine the coordinates of emission and reception
of some particular photon, wrt. each other?
> or by inference from the point of origin and a density matrix
> containing whatever information there is to be had wrt things that
> might render the probability of finding it along some directions, less
> likely than others (like a slit).
How is such density matrix information obtained in the first place?,
if not from measurements of coordinates of emission and reception alone;
perhaps together with the expectation that
"the most probable potential of some slit or any other thing/observer
to render the probability of finding a photon along some directions
doesn't change a lot, soon".
> However, I'm not going to suggest it has coordinates beyond
> what a probabilty amplitude buys you.
Thanks; I had been concerned that by "a photon propagating" you implied
that one could _measure_ coordinates of a photon other than
the coordinates of emission and reception.
Still, are you implying that exchange of a photon were not
_completely_ described by emission and reception?:
> the only means to determine its coordinates is by observing
> it (and thus destroying it in the process)
What "it" do you suggest is "destroyed in the process of observation"?
AFAIK, the exchange of a photon is only _established_ in the first place
by the receiver observing the transition between states of the emitter.
> [A photon that propagates "freely" ...] doesnt leave its emitter
> off massshell
"Off massshell" wrt. _whom_?
> > [bilge/serling wrote:
> > > Since the exchange could occur by swapping the roles of
> > > emitter/absorber, it is symmetric ...
> > The conventional measurement procedures, which are based
> > on light_signal_roundtrips obviously use the fact that
> > this symmetry is not perfect at all:]
> > if I see the deer, the deer has thereby not necessarily
> > seen me in turn already
> However, murphy's law states the deer then will see you
Please note the distinction (asymmetry) between "seen already", and
"then will see". This distinction vanishes (only) if
> > [the roundtrip interval and corresponding distance is zero]
(Also, this distinction is defined whether Murphy's law holds, or not,
in any particular trial.)
> > Consider a pair of distinct objects/charges/observers who exchange
> > just one (virtual) photon.
> > Did either one emit and absorb the same number?
> > Or are you suggesting that this example is not possible or illdefined?
> I'm suggesting just that. The physical situation has to include
> both possibilities since it isnt possible to distinguish between the two.
They or anyone else might use the distinction indicated above
(and in more detail in the preceding post).
But of course anyone is free to ignore distinctions, too.
> Actually, you have to include more. If you have identical particles,
Well  I had been asking about distinct objects/charges/observers.
I don't know how otherwise to determine coordinate relations,
through the Einstein procedures anyways.
> you cant even say [which] incoming particle line goes with
> which outgoing line.
Right  distinct Feynman diagrams which aren't being distinguished
by what's observed/exchanged (and which are therefore only
"potentially distinct") are described as "interfering with each other".
Though one can determine "the most probable contribution" of each,
along with the determination of "the most probable potential"
(i.e. coordinates of potential slits, walls, etc.) in the region
containing the given set of trials.
> I'm not sure it makes sense to enumerate them at all
I'm not sure why the exchange of _one_ (virtual) photon
shouldn't be described as interference of many diagrams.
Regards, Frank W ~@) R
p.s.
> I've always taken the concept of distinguishability to be literal fact:
> If something cannot be observed in principle, it isnt real or physically
> meaningful to consider
I've always found the inverse consideration more interesting:
An observation collected by an invidual observer is surely real
and may be meaningful to that individual;
but for statements to meaningfully describe reality as agreed upon
by _all_ observers, at least in principle, they must be derived
from individual observations by reproducible measurement procedures.
Measurements can (or ought to) be unambigously copied,
communicated and reproduced/understood;
individual observations, and individuals themselves, can not.
Could you describe the problematic Hertz experiment briefly. I still dont
see exactly what he did.
As to the possible reasons: Something more relevant than the usual
inappropriate mantras from the usual suspects is the Wheeler Feynman theory
of advanced and retarded potentials or the improved version as follows:
The source of the radiation produces a repeated oscillation sequence of
instantaneous forces at a distance on the receiving antenna wire. Each force
in the sequence acts on the free electrons and on the interior of the atomic
nuclei and of the free electrons.
A transverse polarization of charge is produced by each such force inside
the atomic nuclei at the same time the free electrons are made to move in
the direction of this applied force. This transverse polarization of charge
is what produces the so called magnetic force. The next in the oscillatory
sequence of instantaneous forces at a distance produces the same sort of
effect but now of a different magnitude.
This causes a change in the transverse force and this change in the
transverse force produces a longitudinal polarization of charge inside other
atomic nuclei and a force field in the opposite direction of the
instantanous force at a distance.
After many oscillations the field resulting form the longitudinal
polarization of charge inside the atomic nuclei becomes dominant.
The delay before this field becomes dominant is given by the speed of
light and the distance from the source. And the mechanism involves the
elasticity of charge polarization inside the atomic nuclei which is the true
significance of the speed of light.
see http://www.bestweb.net/~sansbury
>
>Interference effects between waves in a wire orginating from the
>center of a brass disc and the electrostatic effects of the disc
>can be measured. If oscillations of 35.7 Megahertz are used and
>the speed of the wire waves is 200 000 km/s, we get a wave length
>of 5.6 meters. If electrostatic effects propagate instantanously,
>after 5.6 m, 11.2 m, 16.8 m, ... the wire must be in phase with
>the electrostatic effect of the disc.
>
>
>A simple tape measure is enough to determine at which distances
>from the emitter the interference changes sign.
>
>In any case, it would make sense to repeat this crucial experiment.
>
>
A related scalar potential instantaneous propagation
curiosity is discussed nicely in:
O. L. Brill and B. Goodman, "Causality in the Coulomb Gauge."
Am. J. Phys. _35_, 832 (1967).
J. D. Jackson, Classical Electrodynamics, (John Wiley & Sons,
New York, 1975, 2nd edition) pp. 220223.
Result: Causality is maintained...
Dale Woodside
Macquarie Univ.Sydney
> There seems to be a Heisenberg uncertainty relationship
> between correctness and understandibility (:)).
Or as Bohr liked to say, between truth and clarity. Which is why
those science journalists have such a devil of a time, don't you know.
 > : These quotes show that HEINRICH HERTZ has indeed found in his
 > : experiments that ELECTROSTATIC effects propagate INSTANTANOUSLY
 > : and NOT at c as generally assumed.
 >
 > This will come as a terrible shock to those who have been involved
 > in the design of very successful electromagnetic wave (r.f) based
 > communications systems and equipment for the past 80+ years.
 >
 > Also, James Clerk Maxwell must be turning over in his grave on
 > learning this rather remarkable fact! ;)
 A related scalar potential instantaneous propagation
 curiosity is discussed nicely in:

 O. L. Brill and B. Goodman, "Causality in the Coulomb Gauge."
 Am. J. Phys. _35_, 832 (1967).

 J. D. Jackson, Classical Electrodynamics, (John Wiley & Sons,
 New York, 1975, 2nd edition) pp. 220223.

 Result: Causality is maintained...
A quote from the German translation of Jackson's 2nd edition:
"Am Rande sei auf eine Besonderheit der CoulombEichung
hingewiesen. Elektromagnetische Wellen breiten sich bekanntlich
mit endlicher Geschwindigkeit aus. Gleichung (6.45) besagt
jedoch, dass sich das skalare Potential momentan im ganzen Raum
"ausbreitet". Das Vektorpotential dagegen genÃ¼gt der Wellen
gleichung (6.52), die die endliche Ausbreitungsgeschwindigkeit
c enthÃ¤lt. Auf den ersten Blick erscheint es schwierig, dieses
offensichtlich unphysikalische Verhalten zu umgehen."
This "obviously unphysical behaviour" is yet inherent in Maxwell's
equations. Despite Maxwell's contradictory claim, the possibility
to derive e.m. radiation does not entail that electrostatic
effects (described by the first Maxwell equation) propagate at
the same speed as the radiation.
And this first Maxwell equation entails that it is possible to
transmit information instantanously. The change in charge of
a body affects its neighbourhood instantanously. In the same
way, the induction effect is an instantanous effect. If this
effect is used to transmit information over a distance of 3 m
no time delay of 10 nanoseconds is predicted by the most
obvious interpretation of Maxwell's equations.
Maxwell's 'displacement current' which is assumed to propagate
at c is also an erroneous concept. The effects currently
explained by displacement currents can be better explained by
the first Maxwell equation which states that the net charge
inside a closed surface can always be determined by its effect
on the surface. The movements of charges affect distant surfaces
instantanously, otherwise the first Maxwell equation would be
valid only in static situations and could not be used to derive
e.m. radiation.
In any case, Jackson's section on "scalar potential instantanous
propagation" shows that instantanous e.m. effects not only have
to be explained away on the experimental side but also on the
theoretical side.
Wolfgang Gottfried G.
http://members.lol.li/twostone/E/physics1.html
> A quote from the German translation of Jackson's 2nd edition:
> "Am Rande sei auf eine Besonderheit der CoulombEichung
> hingewiesen. Elektromagnetische Wellen breiten sich bekanntlich
> mit endlicher Geschwindigkeit aus. Gleichung (6.45) besagt
> jedoch, dass sich das skalare Potential momentan im ganzen Raum
> "ausbreitet". Das Vektorpotential dagegen genÃ¼gt der Wellen
> gleichung (6.52), die die endliche Ausbreitungsgeschwindigkeit
> c enthÃ¤lt. Auf den ersten Blick erscheint es schwierig, dieses
> offensichtlich unphysikalische Verhalten zu umgehen."
This corresponds to my copy of Jackson's 2nd edition (p. 222 and 223):
"In passing we note a peculiarity of the Coulomb gauge. It is well
known that electromagnetic disturbances propagate with finite speed.
Yet (6.45) indicates that the scalar potential "propagates"
instantaneously everywhere in space. The vector potential, on the
other hand, satisfies the wave equation (6.52), with its implied
finite speed of propagation c. At first glance it is puzzling how
this obviously unphysical behavior is avoided."
which is stated with the note
"See O. L. Brill and B. Goodman, Am. J. Phys. 35, 823 (1967)
for a detailed discussion of causality of the Coulomb gauge."
After some calculation, they in turn obtain the unsurprising result
that the E == d/dx( Phi )  1/c d/dt( A ) is _still_ the solution
of a wave equation, indicating "propagation with finite speed".
The scalar and the vector potentials themselves _don't_ have
direct physical relevance;
_that's why_ there is a gauge freedom in the first place when
expressing Maxwell's equations in terms of potentials A and Phi,
and that's why the mathematical consequences of the chosing
the Coulomb gauge don't constitute a physical puzzle.
(Note that even for the remarkable AharanovBohm effect
the physically relevant quantity is _magnetic flux_,
and not for instance "the vector potential itself".)
> This "obviously unphysical behaviour" is yet inherent in
> Maxwell's equations.
Obviously not so for the quantities E, D, B, and H in terms of which
Maxwell's equations (6.28) are stated;
but merely after rendering them in terms of quantities that are not
"obviously physical" themselves anyways, namely the potentials A and Phi.
(Not to be misunderstood:
the fact that Maxwell's equations _allow_ themselves to be rendered
with such gauge freedom is itself physically relevant, of course.)
> [The] first Maxwell equation entails that it is possible to
> transmit information instantanously
... _IF_ one could make or destroy charge.
But the notions of what constitutes/how to count "charge",
and directly related, what constitutes/how to recognize
"transmission of information" preclude this by _definition_.
Of course you're free to apply (the formalism of) Maxwell's equations
to describe relations between entities that/who are _not_ charges.
Though then it may be difficult or impossible for others to obtain
any unambiguous, reproducible, meaningful information;
IOW, it may be impossible to _measure_ what you might wish to describe.
Not true. You cannot create or destroy charge, because the
electromagnetic current is conserved (div J = 0)  that's a direct
consequence of Maxwell's equations.
> In the same
> way, the induction effect is an instantanous effect.
Ditto.
Tom Roberts tjro...@lucent.com
>How would you determine the coordinates of emission and reception
>of some particular photon, wrt. each other?
I suppose it depends upon the circumstances. If you look
at a radioactive decay for example which emits a gamma
and beta:

\ e
\

 gamma


I know where the radioactive material is located, I can see the
beta decay occur, I can detect a gamma with a detector and know
it's correlated with the stuff located in the target.
>How is such density matrix information obtained in the first place?,
>if not from measurements of coordinates of emission and reception alone;
>perhaps together with the expectation that
>"the most probable potential of some slit or any other thing/observer
>to render the probability of finding a photon along some directions
>doesn't change a lot, soon".
>
>> However, I'm not going to suggest it has coordinates beyond
>> what a probabilty amplitude buys you.
>
>Thanks; I had been concerned that by "a photon propagating" you implied
>that one could _measure_ coordinates of a photon other than
>the coordinates of emission and reception.
>
Not I. I believe it when quantum mechanics says you cant measure
certain things, even in principle. That's why I suggested a
density matrix. It contains all the information there is to know
for a given situation.
>Still, are you implying that exchange of a photon were not
>_completely_ described by emission and reception?:
>
No. I'm really not trying to imply too much about the process beyond
trying to emphasize any symmetry that, if absent,would suggest
fundamentally different roles played by the emitter and absorber in
the virtual case. The analogy shouldnt be taken too seriously as
a literal occurence. I was more interested in making the symmetry
explicit, not suggesting you can literally assume two photons do
precisely as I suggested. That's one of those things you cant measure,
cant ever know and only represents a small part of the whole
process.
>What "it" do you suggest is "destroyed in the process of observation"?
>AFAIK, the exchange of a photon is only _established_ in the first place
>by the receiver observing the transition between states of the emitter.
>
I dont undertand your statement. Consider a transition that
emits a photon, A>B + gamma. The photon doesnt have an
existence which depends upon B. The photon is on mass shell.
If I observe it, certainly no one else will. I can only observe
it if it creates some disturbance I can measure. In the process,
the photon gets absorbed by something along the way. I dont
see what your statement means in this context, for example.
Maybe you could use this or a similar context to clarify it for me.
>> [A photon that propagates "freely" ...] doesnt leave its emitter
>> off massshell
>
>"Off massshell" wrt. _whom_?
>
If you look at just one piece of the exchange process:
 
' 
'  
'   This photon is clearly not propagating freely
'  
'  
' 
 
e
\ /
' \ / Neither is the one at A.
' \ B /
' /''''''\
A' / \
/ \
e
e
' 
' 
'  If the momentum and energy at B are such that
'A the photon at A is kinematically able to meet
 B the condition q^2 = 0, then it isnt bound by
''''''''' any physical requirement to be reabsorbed.
e it prpopagates. If this isn't satisfacrory, I'm
afraid I'm out of ways in which I intend to make
an attempt to explain it. Further attempts to
play semantic games with quantum effects is bound
to add more confusuion than it eliminates and
goes against the quantum nature of an observable.
q^2 = 0, should be sufficient to describe a
propagating photon.
If there were no difference between a virtual photon and one
that propagates, you'd have no fermi sea to work with as a
vacuum state.
>
>They or anyone else might use the distinction indicated above
>(and in more detail in the preceding post).
>But of course anyone is free to ignore distinctions, too.
>
>"potentially distinct") are described as "interfering with each other".
>Though one can determine "the most probable contribution" of each,
>along with the determination of "the most probable potential"
>(i.e. coordinates of potential slits, walls, etc.) in the region
>containing the given set of trials.
>
The only parts I consider to bear a resemblece to reality are those
subsets that produce observables. There are two many problems
with suggesting it's possible to guess what is unobservable in
principle as a mechanism for anything.
>
>I'm not sure why the exchange of _one_ (virtual) photon
>shouldn't be described as interference of many diagrams.
>
I thought that was basically what a feynman diagram does. It provides
an intuitive way to dissect a complex interaction into tractable
calculations which have physical interperatations. One photon line
represents both moeller and bhabba scattering diagrams, right?
It doesnt represent just one though, it represents them all. I was
merely suggesting that the symmetry of the exchange be retained in
any conceptual description in stating both particles play both roles
in the exchange process. If you dont, the obvious question that
someone is bound to ask from your choice of a single photon going from
A to B as reality is "why doesnt it go from B to A". I dont think you
were really suggesting nature makes such distinctions.
> > [The] first Maxwell equation entails that it is possible to
> > transmit information instantanously
Let me address this  perhaps your main point  in a little more detail:
Note that Maxwell's first equation (Jackson 6.28) is a
differential equation which concerns values only _at one point_:
"divergence of vector field D at coordinates r and t
= 4 Pi charge density at coordinates r and t"
This equation by itself doesn't involve any "transmission".
If there's no charge (density) at given coordinates r, then divergence
of vector field D is defined and remains equal to zero, no matter what
else might be going on at other coordinates.
One can consider the integrated equation, since this involves different
coordinates which might allow to comtemplate some sort of "transmission"
between them at all. Using Gauss' theorem, the statement then is:
"vector field D integrated over a closed surface (noting direction)
= 4 Pi the charge contained within that surface"
There are two ways of changing the content of a closed surface,
in the attempt to somehow "affect the entire surface, perhaps at once":
Either content simply appears or vanishes without crossing the boundary 
but such content is not "charge", by definition; as discussed earlier.
Or the content moves through the boundary, as a current 
that's described as time dependence in the remaining Maxwell equations
and by the resulting wave equations.
>way, the induction effect is an instantanous effect. If this
>effect is used to transmit information over a distance of 3 m
>no time delay of 10 nanoseconds is predicted by the most
>obvious interpretation of Maxwell's equations.
>
Then it would serve as an excellent way to check. Especially since c
should really be associated with the propagation of information and
light received the honor by the grace of a massless photon and being
the only means of transmitting information. It did not have to be so.
Light could have consisted of photons with very small masses, leaving
the rate at which information is constrained to travel still at c,
photons a bit slower, a much harder time undertanding nature, and as
of the present, without a means of transmitting information at c.
The speed at which light travels isnt the constraint  it's the
ability for two observers to reconcile events using a description
that has the same form in both of their rest frames. Light just
happens to propagate at the same rate as information.
>inside a closed surface can always be determined by its effect
>on the surface. The movements of charges affect distant surfaces
It doesnt state that. You added the word "always" to mean "right now
in every frame of reference". Maxwell's eqns. predate such concepts.
You cant expect them to be infallible when it comes to describing
phenomena that wasnt known in terms of concepts that werent known.
>instantanously, otherwise the first Maxwell equation would be
>valid only in static situations and could not be used to derive
>e.m. radiation.
>
That's not true. A complete explanation of radiation from charges
lies outside of classical em, with or without considering relativity.
An ad hoc term, appended to maxwells eqns. that produces a result
that is approximately correct and has theoretical merit is a bonus,
not a liability. Since maxwell's eqns do not address the origin of
the radiation, you might expect some inconsistencies. However,
the inconsistency is in maxwell's eqn. It's ridiculous to try and
assert the equations are infallible and that a consistency argument
which uses as evidence, phenomena which relativity doesnt claim to
fully explain, invalidates sr. Radiation requires accelerated charges.
Acceleration is not feature built explicitly into either maxwell's
eqn. nor sr.
>In any case, Jackson's section on "scalar potential instantanous
>propagation" shows that instantanous e.m. effects not only have
>to be explained away on the experimental side but also on the
>theoretical side.
>
Sure. But they are related to localized effects that arent a part of
the theory. When you start asking about how the reaction of the
charge affects the radiation it emits, you're going to have a hard
time explaining things. The large scale effects of instantaneous
propagation that are testable should be evident if true. So far,
I've seen exactly one candidate that could potentially pose an
interesting problem for relativity and that's the "teleportation"
that quantum mechanics apparently permits, but it would only show
where sr was limited, not that it was totally invalid.
 > And this first Maxwell equation entails that it is possible to
 > transmit information instantanously. The change in charge of
 > a body affects its neighbourhood instantanously.

 Not true. You cannot create or destroy charge, because the
 electromagnetic current is conserved (div J = 0)  that's a direct
 consequence of Maxwell's equations.
It is possible to change the charge of a body without creating or
destroying charge. So your remark is not relevant.
The question is whether the effect of the displacement of charged
particles propagates at c or instantanously.
 > In the same
 > way, the induction effect is an instantanous effect.

 Ditto.
Ditto.
Gruss, Wolfgang
Relationality versus Relativity:
http://members.lol.li/twostone/E/physics1.html
Sure it is, otherewise to change the charge of a body requires the
passage of an electric current to supply or remove charge.
: The question is whether the effect of the displacement of charged
: particles propagates at c or instantanously.
Standing waves, the design of certain types of microwave
devices (magnetrons, klystrons, and planar triodes) along
with the design principles employed successfully on antennas all make
it very clear that the propagation rate is c (actually c modified by the
propagation factor of the dielectric medium).
Harry C.
> > How would you determine the coordinates of emission and reception
> > of some particular photon, wrt. each other?
> I suppose it depends upon the circumstances.
You're using not one, but different measurement procedures
to determine coordinate relations??
How would you then determine and characterize various "circumstances"
in order to select the corresponding measurement procedure to begin with?,
if not already through measured coordinate relations.
> a radioactive decay for example which emits a gamma and beta: [...]
> I know where the radioactive material is located, I can see
> the beta decay occur
My question was how you'd _determine "location"_ in the first place,
given what you see (of various charges, in various trials) ...
> I can detect a gamma with a detector
... or given what other detectors/observers see, in those trials.
> I suggested a density matrix. It contains all the information
> there is to know for a given situation.
And I asked already:
> > How is [the] density matrix [...] obtained in the first place?,
for a given situation;
and what exactly is "given" in any one trial, e.g. in the _next_ trial?
> > What "it" do you suggest is "destroyed in the process of observation"?
> > AFAIK, the exchange of a photon is only _established_ in the first place
> > by the receiver observing the transition between states of the emitter.
> I dont undertand your statement. Consider a transition that
> emits a photon, A>B + gamma.
It's not very helpful just to invent new names like "emitted photon"
or "+ gamma" in place of the presumed "it".
However, I can consider a transition "A>B" between two states
of the ordered set of one particular observer (or observer system) ...
> In the process, the photon gets absorbed by something along the way.
... and I can even consider someone observing this transition in turn, e.g.
"You_before_observing_A>B > You_after_observing_A>B", or
"r_before_observing_A>B > s_after_observing_A>B",
i.e. perhaps more concisely: (A>B ~ r>s);
and such a relation constitutes "exchange of a photon", AFAIK.
I can even consider "self observation":
either (simply, evidently):
"A_before_observing_A>B > B_after_observing_A>B", i.e. (A>B ~ A>B),
or (more involved):
"B_before_observing_A>B > C_after_observing_A>B", i.e. (A>B ~ B>C).
> If I observe it, certainly no one else will.
This seems to describe specificly
(A>B ~ You_before_observing_A>B > You_after_observing_A>B).
> I can only observe it if it creates some disturbance I can measure.
That makes sense  I wouldn't know either how to distinguish between
you collecting the observation that "A>B",
you undergoing the transition between two states of your ordered
set of states: "You_before_observing_A>B > You_after_observing_A>B", or
you "measuring a (this particular) disturbance".
> The photon doesnt have an existence which depends upon B.
Interesting, given that you introduced B (with the notion of
"transition") above. Surely you're not trying to explain existance
and measurements of "photons" independent of "transition"?
The question is still how to measure and unambiguously characterize any
particular "it/photon/gamma", given mutual observations of transitions.
Specificly, how to measure _whether or not_ or to which extent
> [...] the momentum and energy at B are such that the photon at A
> is kinematically able to meet the condition q^2 = 0 [...]
in the first place.
> q^2 = 0, should be sufficient to describe a propagating photon.
Well  if so, then at least "a propagating photon" seems described
entirely as a _relation_ between A, B, etc. alone (though I'd still
like to know more specificly how that relation is evaluated);
and "it/photon/gamma" is not "some_thing_it_self". That was my point.
> the obvious question that someone is bound to ask from your choice
> of a single photon going from A to [r] as reality is
> "why doesnt it go from [r] to A".
Well, _that_ would be a _different_ exchange of a photon,
as observed by A and r.
(A>B ~ r>s) is _not_ necessarily the same as (r>s ~ A>B).
(Btw., QM incorporates this principle in general;
cf. Sakurai, Modern Quantum Mechanics, eq. 1.2.12.)
> I dont think you were really suggesting nature makes such distinctions.
Surely observers who _describe_ each other/nature do.
Regards, Frank W ~@) R
p.s.
> I'm really not trying to imply too much about the process beyond
> trying to emphasize any symmetry that, if absent, would suggest
> fundamentally different roles played by the emitter and absorber in
> the virtual case. The analogy shouldnt be taken too seriously as
> a literal occurence. I was more interested in making the symmetry
> explicit, not suggesting you can literally assume two photons do
> precisely as I suggested. That's one of those things you cant measure,
Then this doesn't seem relevant to the question I had asked in
reply to Moataz H. Emam,
Frank Wappler [wrote]:
. By which measurement procedures are the extent of time, translation
. and/or rotation symmetries determined experimentally in the first place?
For "teleportation" to be a problem,
the determination of pairwise coordinate relations
("localization", "separation", "propagation", etc.)
would have to independent of the "teleportation" procedure
(and the procedure by which to assert what, if anything, has been
"teleported" at all).
However, in relativity
(i.e. when using Einstein's procedures for determining pairwise
coordinate relations, based on the exchange of light signals itself)
they are not independent;
therefore "teleportation" can't be a problem for relativity.
>Then this doesn't seem relevant to the question I had asked in
>reply to Moataz H. Emam,
>
>Frank Wappler [wrote]:
>. By which measurement procedures are the extent of time, translation
>. and/or rotation symmetries determined experimentally in the first place?
Then somewhere I missed a (some) post(s). There seemed to be two
distinct topics here. If that wasnt the case, then I'm contributing
to confusion by pursuing this apparent tangent.
Frank Wappler explains:
> Precisely; where q denotes the fourmomentum that has been transferred
> between the two charges/observers who have exchanged this photon.
bilge continues:
> Photons carry momentum. The electrostatic
> effect is the momentum carried by the photons due to the charge
> of an object. The photons do not carry charge and so cannot change
> any feature related to charge. All a photon can do is change the
> momentum of another charged body upon absorption. Any charged body
> may absorb a photon. If the absorber and emitter are different
> bodies, the result looks like a force because the momentum change
> in the emitter has to equal the momentum change in the absorber.
> Virtual photons do not exist independent of the emitter. If a
> charged object emits a photon, it must either reabsorb it to
> satisfy heisenberg, or another charged object must absorb it to
> satisfy heisenberg. Since the exchange could occur by swapping
> the roles of emitter/absorber, it is symmetric and should be
> interpereted that way. In that sense, an object absorbs the same
> number of virtual photons it emits.
Tom Roberts comments:

> [pretty good description of photon absorbtion/emission]

 Except for one "little" thing  in a basis where individual photons
 have welldefined 4momenta, the number operator does not have a well
 defined value, and you cannot count them! In a basis where the number
 operator has a welldefined value, individual photons do not have
 welldefined 4momenta! This is intimitely related to the fact that
 photons are indistinguishable Bosons, and to the necessity to
 symmetrize the wavefunction over Bosons and antisymmetrize over
 Fermions  in a perturbative approximation this intermixes all the
 Bosons/Fermions in all of the different diagrams....

 If one _really_ tries to take into account _all_ of the properties
 of photons in QED, the discussion gets so convoluted and complicated
 that it is essentially useless....
On the one hand we have the extremely simple and elegant
Coulomb law and on the other hand this obscure quasi
mechanistic explanation of the same relation.
How can somebody taking seriously Ockham's razor prefer such
an obscure and logically inconsistent explanation to the
simple and elegant Coulomb law?
The QED explanation contains several concepts (elements)
which are at least as complex as the Coulomb law itself.
How can somebody taking seriously Ockham's razor explain
one simple concept by a combination of several complicated
concepts?
Virtual particles are assumed on the one hand to have
the needed properties (e.g. mass and momentum) and on
the other hand, if necessary, not to have the same
properties.
Actions at a distance are a far better explanation:
1) Photons as postulated by Einstein are real entities with
concrete values for mass, momentum and frequency. QED
'photons' only share the name with the original photons.
2) QED 'photons' would be logically refuted if Heisenberg had
not invented his famous uncertainty relations which allow
to circumvent necessary logical conclusions.
3) Electrostatic attraction cannot even be explained
qualitatively by 'QED' photons because under momentum
conservation two objects can only drift apart by exchanging
particles. (Perhaps the Heisenberg uncertainty relations
allow the assumption of negative mass and momentum :)
4) In order to prevent isolated charged objects from radiating
more QED 'photons' than they absorb, one must assume that
the 'photons' are somehow tied to the objects. They fly away
at c and if they don't find another charge, they change
direction and fly back to the object. (It is certainly not
always easy for them to find back home :)
5) Explaining interactions between charged objects by mediating
particles leads to the even more complex problem of how
these mediating particles interact with the charged objects.
6) Many experiments could be interpreted in a simpler and more
transparent way as confirmation of actions at a distance
than as confirmation of the currently accepted theories.
Wouldn't it be almost a miracle if such a strange and complex
behaviour as the one assumed for 'QED' photons resulted in
exactly the Coulomb law in all the many situations where this
law is experimentally confirmed?
Wolfgang Gottfried G.
Instantanous propagation of the 'electrostatic force':
http://www.deja.com/=dnc/getdoc.xp?AN=532021977
http://www.deja.com/=dnc/getdoc.xp?AN=532263367
http://www.deja.com/=dnc/getdoc.xp?AN=532665217
http://www.deja.com/=dnc/getdoc.xp?AN=533126325
http://www.deja.com/=dnc/getdoc.xp?AN=534705489
But Coulomb's law doesn't fit experiment. It's not like we invent
these things just for our own amusement, after all.
>
>How can somebody taking seriously Ockham's razor explain
>one simple concept by a combination of several complicated
>concepts?
>
>Virtual particles are assumed on the one hand to have
>the needed properties (e.g. mass and momentum) and on
>the other hand, if necessary, not to have the same
>properties.
How many times does this have to be said? Virtual particles only
need to exist in a didactic sense. The full theory is a theory of
fields. You're attacking a caricature of the theory.
Aaron

Aaron Bergman
<http://www.princeton.edu/~abergman/>
> How can somebody taking seriously Ockham's razor prefer such
> an obscure and logically inconsistent explanation to the
> simple and elegant Coulomb law?
Ockham's razor only applies when two or
more explanations equally match the observations.
Opinions expressed herein are my own and may not represent those of my employer.
>Except for one "little" thing  in a basis where individual photons
>have welldefined 4momenta, the number operator does not have a well
>defined value, and you cannot count them! In a basis where the number
>operator has a welldefined value, individual photons do not have
>welldefined 4momenta! This is intimitely related to the fact that
>photons are indistinguishable Bosons, and to the necessity to
>symmetrize the wavefunction over Bosons and antisymmetrize over
>Fermions  in a perturbative approximation this intermixes all the
>Bosons/Fermions in all of the different diagrams....
>
>If one _really_ tries to take into account _all_ of the properties
>of photons in QED, the discussion gets so convoluted and complicated
>that it is essentially useless....
>
> There seems to be a Heisenberg uncertainty relationship
> between correctness and understandibility (:)).
>
>
I'm going to have to sidestep the issue by asserting the the concepts
and physics cannot depend upon a particular representation and it's
not my intent to try and substitute the concept literally as a novel
computational scheme. I recognize the principle your principle is a
limitation on extending analogies, (but never in quite the excellent
and concise reperesentation you've chosen to coney it) :)
> Frank Wappler [wrote]:
> > [at least "a propagating photon" seems described entirely
> > as a _relation_ between A, B, etc. alone]
Therefore
> > "it/photon/gamma" is not "some_thing_it_self". That was my point.
> Unfortunately, the language doesnt make it easy
> to talk about them not as things.
Fortunately, language allows us to talk instead about
"the relations with each other of particles/charges, systems of those,
things, etc. who observe each other, at least in principle".
Most generally therefore: relations between observers;
as opposed to observers themselves.
> I dont consider particles as "things" either.
Really? Unfortunately, the language doesn't make it easy
to talk about them not as things.
Perhaps one could more precisely refer to particles as
"those who can determine coordinate relations wrt. each other,
at least in principle; and who can count, and be counted"?
> > [bilge/serling wrote:
> > > ... That's one of those things you cant measure]
> > Then this doesn't seem relevant to the question I had asked [...]
> Frank Wappler [wrote]:
> . By which measurement procedures [...]
Specificly: through which procedures should observers determine
and agree upon
> . the extent of time, translation and/or rotation symmetries
of the region in which they are contained, in any particular trial,
from their mutual observations collected in that trial.
Best regards, Frank W ~@) R
p.s.
> There seemed to be two distinct topics here.
(But there's only _one_ way to find out. :)
>
>On the one hand we have the extremely simple and elegant
>Coulomb law and on the other hand this obscure quasi
>mechanistic explanation of the same relation.
>
>How can somebody taking seriously Ockham's razor prefer such
>an obscure and logically inconsistent explanation to the
>simple and elegant Coulomb law?
>
For one thing, Coulombs law provides no mechanism that tells
you how a charge exerts a force on another charge. You have to
assert some new physics and the idea of a special force, due
to charges. qed reduces the problem to conservation of momentum.
Most people consider reducing new physics to old physics with
a nice physical picture a reduction in complexity. You dont
get to insist on a simple disguise for the old physics to
use, however.
>The QED explanation contains several concepts (elements)
>which are at least as complex as the Coulomb law itself.
>
But, you can eliminate all of those concepts if you want
to give up what you give up accepting coulombs law  an
explanation of the underlying mechanism. qed allows you
to use coulomb's law. qed seeks explanations beyond what
coulomb's law can tell you. You just have to accept new
physics without hope of a mechanism if you want coulomb's
law to be the end of the line. You wont build quantum
computer's though.
>How can somebody taking seriously Ockham's razor explain
>one simple concept by a combination of several complicated
>concepts?
>
Nature chooses the implementation. Physics attempts to reverse
engineer it by second guessing the design as a whole. The pieces
dont fit into a working model without qed. A pile of screws
is simple, but you cant build a watch from them.
>Virtual particles are assumed on the one hand to have
>the needed properties (e.g. mass and momentum) and on
>the other hand, if necessary, not to have the same
>properties.
>
The standard vocabulary doesnt really have the words to convey
those concepts. We're much bigger than any effect quantum mechanics
predicts differently than classical mechanics and much smaller than
any effects you need relativistic mechanics to explain. So, an
everyday description vocabulary to evoke a picturesque and correct
discription of virtual particles just hasnt been important in
developing a vocabulary. On the other hand, the existing vocabulary
reinforces concepts which you have to abandon to have any intuition
for relativity or qm. Like, "now", or "solid object".
>Actions at a distance are a far better explanation:
>
>1) Photons as postulated by Einstein are real entities with
> concrete values for mass, momentum and frequency. QED
> 'photons' only share the name with the original photons.
>
Even here, your "classical" photon fails and must be rescued
by concepts of field theory. A monochromatic plane wave is
the only photon with definite values for these things. Classically
the energy is infinite if the frequency is definite. So you
must construct a real photon by superposition. Where do they
originate to perform the classical construction? Quantum
mechanics rescues you here. A plane wave has an infinite
energy and it must fill all of space. They may exist, but if
they fill all of space, they arent observable and so may
be redefined into the vacuum. Observing any effect of a plane
wave isnt obsevable, but you arent precluded from summing an
infinite number of them to produce something that is.
>2) QED 'photons' would be logically refuted if Heisenberg had
> not invented his famous uncertainty relations which allow
> to circumvent necessary logical conclusions.
>
Sure. All of quantum mechanics and the semiconductors it made
possible rest on the uncertainty principle. The very existence
of stable matter rests upon the uncertainty principle.
>3) Electrostatic attraction cannot even be explained
> qualitatively by 'QED' photons because under momentum
> conservation two objects can only drift apart by exchanging
> particles. (Perhaps the Heisenberg uncertainty relations
> allow the assumption of negative mass and momentum :)
>
No. Here is where heisenbergs uncertainty principle really
earns its reputation. The location of the exchanged photon
is uncertain to the extent allowed by the uncertainty principle
for any given uncertainty in the momentum. Nothing precludes
the photon from having a momentum in the opposite direction
than you would expect classically from its position. It may
then, for example, be in the vicinity of a particle to the left
of the one that emitted it AND have a momentum which points
to the right, TOWORD the particle that emitted it.
>4) In order to prevent isolated charged objects from radiating
> more QED 'photons' than they absorb, one must assume that
> the 'photons' are somehow tied to the objects. They fly away
> at c and if they don't find another charge, they change
> direction and fly back to the object. (It is certainly not
> always easy for them to find back home :)
>
No. By the same argument as above, it's location may be anywhere
that the uncertainty principle allows. The uncertainty principle
allows it to live no longer than the uncertainty in energy permits.
The photon's position always has a probability of being right where
it started, so when it's time is up, no inconsistency results
because it's always where it needs to be.
>5) Explaining interactions between charged objects by mediating
> particles leads to the even more complex problem of how
> these mediating particles interact with the charged objects.
>
You cant win them all. Unfortunately, having to figure this out
doesnt go away classically, either. Because only the fields matter
and not the potentials, you are allowed a choice of gauge to
simplify your life for any particular problem. The gauge freedom
forces you to have photons.
>6) Many experiments could be interpreted in a simpler and more
> transparent way as confirmation of actions at a distance
> than as confirmation of the currently accepted theories.
>
Only if you wish to skip understanding the mechanism. Trying
to recast an explanation of the electron magnetic moment
into ad hoc terms appended to maxwell's eqns would be worse
than hideous and more opaque than a political speech.
>Wouldn't it be almost a miracle if such a strange and complex
>behaviour as the one assumed for 'QED' photons resulted in
>exactly the Coulomb law in all the many situations where this
>law is experimentally confirmed?
>
It does, its wonderful, but not a miracle or particularly strange.
It would be strange if it didnt and anyone considered it wonderful
enough to call a miracle.
 > Virtual particles are assumed on the one hand to have
 > the needed properties (e.g. mass and momentum) and on
 > the other hand, if necessary, not to have the same
 > properties.

 How many times does this have to be said? Virtual particles only
 need to exist in a didactic sense. The full theory is a theory of
 fields. You're attacking a caricature of the theory.
Since when do didactic particles carry momentum? :(
Huge forces of electostatic repulsion and attraction do occur
in nature. QED explains these forces by assuming QED 'photons'
carrying momentum, isn't it? If you tried to create the same
forces using real photons, you would recognize how many high
energy photons would be necessary for that. But the higher
the energy of photons, the less relevant are Heisenberg's
uncertainty relations. So we must assume that huge (but
uncertain) numbers of low energy 'photons' must be involved.
It makes no sense to reduce electrostatic forces superficially
to 'photons' and conservation of momentum, if conservation of
momentum in principle can only lead to the opposite of what is
observed in the case of attraction. I think it is better to
explain a force by an action at a distance than by a local
action which according to correct logical reasoning could only
lead to a force in the opposite direction.
Feynman or who else is responsible for this strange idea may
have overlooked the fact that not all mechanical situations are
time reversible:
Two ships can drift apart if the passengers throw objects from
one ship to the other. The opposite however, is not possible.
QED explains this by the interaction of various fields. In the
first order, this can be approximated by something that looks
like the expression for the exchange of a particle. This
"particle" doesn't obey all the same rules as a normal particle,
so it's called a virtual particle. But, it's only an
interpretation of an approximation.
Because QED explains many more phenomena to far better accuracy than
does Coulomb's law.
> 1) Photons as postulated by Einstein are real entities with
> concrete values for mass, momentum and frequency. QED
> 'photons' only share the name with the original photons.
And the problem with this is?
> 2) QED 'photons' would be logically refuted if Heisenberg had
> not invented his famous uncertainty relations which allow
> to circumvent necessary logical conclusions.
Nonsense. You just do not understand QED, that's all. For a simple
and entertaining introduction:
Feynman, _QED_.
> 3) Electrostatic attraction cannot even be explained
> qualitatively by 'QED' photons because under momentum
> conservation two objects can only drift apart by exchanging
> particles. (Perhaps the Heisenberg uncertainty relations
> allow the assumption of negative mass and momentum :)
You _really_ do not understand QED or how virtual particles work.
Attraction is easily explained  the virtual photons are off the
mass shell, and their _interference_ causes net momentum transfer.
> 4) In order to prevent isolated charged objects from radiating
> more QED 'photons' than they absorb, one must assume that
> the 'photons' are somehow tied to the objects.
You _really_ do not understand QED. One cannot "count" photons
except in exceptional circumstances. But no matter  energy and
momentum are conserved, at each vertex, for each diagram, and
overall for the entire summation of diagrams.
This is, of course, in a perturbative approximation to QED.
That's really the only context in which photons arise.
> 5) Explaining interactions between charged objects by mediating
> particles leads to the even more complex problem of how
> these mediating particles interact with the charged objects.
You _really_ do not understand QED. Actually, the vertex functions
in QED are quite simple.
> 6) Many experiments could be interpreted in a simpler and more
> transparent way as confirmation of actions at a distance
> than as confirmation of the currently accepted theories.
Try accounting for the gyromagnetic ratio of the electron. QED agrees
with experiment to 11 significant digits.
> Wouldn't it be almost a miracle if such a strange and complex
> behaviour as the one assumed for 'QED' photons resulted in
> exactly the Coulomb law in all the many situations where this
> law is experimentally confirmed?
As Arthur C. Clarke said, "Sufficiently advanced technology is
indistinguishable from magic."
Here the "advanced technology" is the simple fact that in the
appropriate limit QED is accurately approximated by Maxwell's
equations, including Coulomb's law. But Maxwell's equations cannot
explain the gyromagnetic ratio of the electron....
Tom Roberts tjro...@lucent.com
Where do you think that it is assumed anywhere that *static*
(unchanging in time) fields have to propagate (travel) at all?
Certainly not in the theory of Maxwell that Hertz was testing.
Note followups.

James A. Carr <j...@scri.fsu.edu>  Commercial email is _NOT_
http://www.scri.fsu.edu/~jac/  desired to this or any address
Supercomputer Computations Res. Inst.  that resolves to my account
Florida State, Tallahassee FL 32306  for any reason at any time.
QED is better understood as an exact theory of particle interactions in
pregeometric space, than an approximate theory of fields.
http://xxx.lanl.gov/abs/physics/9905058
A Theory of Quantum Spacetime
http://xxx.lanl.gov/abs/physics/9909051
A PreGeometric Model Exhibiting Physical Law

Charles Francis
cha...@clef.demon.co.uk
Precisely infinite, or finite but indefinite? Infinite just means a
finite number larger than the number you first thought of. The number of
photons is finite, but whatever number you think of, there is a
possibility that there may be more. Only changes in the number of
photons actually have a physical effect, so the actual number is
completely unknown, and partly explains gauge invariance.
>
> >Feynman or who else is responsible for this strange idea may
> >have overlooked the fact that not all mechanical situations are
> >time reversible:
> >
Only on the basis that time symmetry is broken by statistical laws.

Charles Francis
cha...@clef.demon.co.uk
Charles Francis wrote:
>
> Precisely infinite, or finite but indefinite? Infinite just means a
> finite number larger than the number you first thought of.
Nonsense. A set has infinite cardinality if it can be put into
a one to one onto mapping of itself into a proper subset of
itself.
Bob Kolker
Mathematics has several definitions of infinity. I was using the
definition in analysis, which is the most practical and the most useful.
The definition, or rather axiom, of set theory to which you refer is
fine in so far as mathematics is only concerned with consistent systems
of thought, but cannot be applied to any physical situation.

Charles Francis
cha...@clef.demon.co.uk
>Precisely infinite, or finite but indefinite? Infinite just means a
>finite number larger than the number you first thought of. The number of
>photons is finite, but whatever number you think of, there is a
You are being pedantic. If you perform an explicit calculation and
identify photons with photon lines in the diagrams, you have an
infinite (countably, too) number of calculations to perform. The
electrons dont sit and calculate this and any inconsistency in
bookkeeping is from trying to approximate the interaction rather
than a real description. You can only talk about which diagrams are
important. At some point the exact semantics are important to obtaining
a mathematical answer, but it usually detracts from a physically
intuitive picture. Since every text I've seen feels free discuss
photons as single entities when convenient and to for physical appeal,
use "infinite" in ways that might make mathematicians shudder in a
mthematical context requiring rigor, I certainly dont feel compelled
to reconcile every physical description with a mathematically rigorous
one. I dont see that the doing so adds any appeal for the person that
wanted to know why qed was more appealing than using coulombs law,
since the lack of a physically intuitive process was a primary
objection.
I accept the charge, but I would not be so pedantic if I did not think
that there are significant misconceptions wrapped up in the question of
the infinite. In this instance the question is of direct relevance to
gauge invariance, and in such a way that leads me to question the
relevance, or meaning, of so called "gauge theories".
>If you perform an explicit calculation and
> identify photons with photon lines in the diagrams, you have an
> infinite (countably, too) number of calculations to perform. The
> electrons dont sit and calculate this and any inconsistency in
> bookkeeping is from trying to approximate the interaction rather
> than a real description.
The diagrams do not individually state what happens in a given
interaction, but describe the possibilities of what may happen. There
are an infinite number of possibilities, but that is not the same as an
infinite number of photons.
>You can only talk about which diagrams are
> important. At some point the exact semantics are important to obtaining
> a mathematical answer, but it usually detracts from a physically
> intuitive picture. Since every text I've seen feels free discuss
> photons as single entities when convenient and to for physical appeal,
> use "infinite" in ways that might make mathematicians shudder in a
> mthematical context requiring rigor, I certainly dont feel compelled
> to reconcile every physical description with a mathematically rigorous
> one. I dont see that the doing so adds any appeal for the person that
> wanted to know why qed was more appealing than using coulombs law,
> since the lack of a physically intuitive process was a primary
> objection.
In this instance the question is of fundamental importance to the
interpretation of qed, and quantum mechanics in general. I believe that
is the main interest of the person wanting a nonrigorous description of
physical theory. To give such a person a nonrigorous description, and
be sure that it is right, someone has to work on the rigorous
description. At the moment it seems that academics and lecturers refuse
to acknowledge the need for such treatments in field theory. I regard
that as a matter of supreme incompetence, as well as professional
negligence in view of what the founders of the subject such as Dirac and
Feynman had to say on the subject.

Charles Francis
cha...@clef.demon.co.uk
>In this instance the question is of fundamental importance to the
>interpretation of qed, and quantum mechanics in general. I believe that
>is the main interest of the person wanting a nonrigorous description of
>physical theory. To give such a person a nonrigorous description, and
>be sure that it is right, someone has to work on the rigorous
>description. At the moment it seems that academics and lecturers refuse
>to acknowledge the need for such treatments in field theory. I regard
>that as a matter of supreme incompetence, as well as professional
>negligence in view of what the founders of the subject such as Dirac and
>Feynman had to say on the subject.
I've read through your paper and when I get the chance, I'll
make comments if I have any. There is an irony though in your
support of for the concept of discrete space (and which has
numerous other adherents). The most enthusiastic support is
probably from people you would view in the same light of
negligence as the two you mention. Most notaably, john wheeler
supports the idea because it's a requirement to communicate
information.
I do not think so. In my paper qed is derived from first principles,
what I read into it is based on those first principles, not on qed
itself. All I am saying is that the relative contribution of a diagram
is a contribution to a probability calculation, given that we do not
know what happened. In any given case there is a diagram which describes
exactly what happened, but we do not know which one, just like an
ordinary probability. The reason we do not get the familiar probability
relationship, and get the quantum relationship, is that, in addition to
the diagram itself, we also do not know what happened to the individual
particles in the environment, and apparatus. In his published papers,
Mark Hadley also derives quantum law from "continuous timelike curves"
in gr. My core argument is very similar to his, but I think mine is more
general.
>What "really" happens has
> no meaning where you want it to. No observable can tell you
> what's under the hood.
That is true, but the requirement for logical consistency of
interpretation is very powerful.
>Charge renormalization in qed is perfectly
> legitimate. If you were questioning renormalization in qcd, I
> wouldnt be able to agree or disagree. I find it somewhat suspect,
> but havent spent enough time looking at that to say much.
>
I do not question the need for renormalisation, but I do not accept
infinite renormalisation. I
>
> >In this instance the question is of fundamental importance to the
> >interpretation of qed, and quantum mechanics in general. I believe that
> >is the main interest of the person wanting a nonrigorous description of
> >physical theory. To give such a person a nonrigorous description, and
> >be sure that it is right, someone has to work on the rigorous
> >description. At the moment it seems that academics and lecturers refuse
> >to acknowledge the need for such treatments in field theory. I regard
> >that as a matter of supreme incompetence, as well as professional
> >negligence in view of what the founders of the subject such as Dirac and
> >Feynman had to say on the subject.
>
>
> I've read through your paper and when I get the chance, I'll
> make comments if I have any.
I look forward to it.
> There is an irony though in your
> support of for the concept of discrete space (and which has
> numerous other adherents). The most enthusiastic support is
> probably from people you would view in the same light of
> negligence as the two you mention. Most notaably, john wheeler
> supports the idea because it's a requirement to communicate
> information.
>
I do not include Wheeler in my tirade against attitudes which I have
found in my personal contact with certain professors, but these
attitudes seems to reflect the "orthodox" view. For example, although
Feynman says the maths of qed is dodgy, and thinks that the divergences
arise from a technical inaccuracy, when I set about demonstrating that
this is in fact the case, I was told that Feynman, Schwinger, et al were
very clever people and could not possibly have made a mistake. When I
set about reconciling field theory with the foundations of quantum
mechanics and started to show that it describes pointlike particles, (
no wave structure) in the absence of a background of spacetime, I was
told that if Dirac and the founding fathers could not resolve such
questions, no one should try.

Charles Francis
cha...@clef.demon.co.uk