On Sun, 17 Sep 2006

In a perverse way, I like this chapter in Jackson:-) Here he spends
16 chapters to give you the sense of "hey, we know how to deal with
this stuff" (though not necessarily easily), then, within few pages,
this feeling gets completely demolished. Interestingly, most EM
courses never reach this chapter.

I like to drop it in about 1/3 - 1/2 of the way through the semester, just
for a change.

The next trauma is when I do scattering & diffraction. Scattering by an
infinite screen with an aperture is fun. With no aperture, the field on
the far side of the screen is E_total = E_incident + E_scattered. But this
is zero, so clearly E_scattered = -E_incident. Put an aperture in. The
scattered wave arises from where the screen is opaque, not from the
aperture. Best student response so far: "I'm not sure I'm comfortable with
that idea."

Eric Gisse wrote:

mmeron@cars3.uchicago.edu wrote:

Why *would* one expect an electron to exert a force on itself? If it
did, how could we tell the difference than if it didn't?

Imagine an electron with size. The field created by the left side

exert a force on the charge density in the right side, and vice versa.
Turns out that doesn't go away even as the diameter is brought to

In article <1158653398.319133.311870@d34g2000cwd.googlegroups.com>, "Eric Gisse" <jowr.pi@gmail.com> writes:

In article <200609190746.k8J7kQq7022076@ipp.mpg.de>, Bruce Scott TOK <Use-Author-Supplied-Address-Header@[127.1]> writes:

Timo N wrote:

On Sat, 16 Sep 2006, Bruce Scott TOK wrote:

Try googling with: self forces electrodynamics

also: look up articles by Steve Parrott in sci.physics.research a few
years ago when there were lively (and well informed) threads on the topic

I'm not sure how much this will reveal about what is clear and what is not
clear. Perhaps a better rule of thumb would be that if it isn't in (and
clear in) Griffiths and Jackson, it isn't clear.

That's a good point. Basically, the point of Parrott's book is that
this stuff has never been solved.

For me the best elementary treatment of the problems with the problem is
the one in Feynman's Lectures, Vol II, where he asks precisely that
question: what it the force of an electron on itself.

At rest the self-produced electric field is symmetric and the force is
zero. In motion, considering relativity, that is no longer true and you
get infinities. Feynman makes an insightful connection between that and
the infinities in QED (in commentary... QED itself is of course beyond
the scope of the lectures).

In sci.physics about 10 years ago the thing that started the whole
discussion was when someone asked if a charge at rest in a gravitational
field should radiate (massless charged particle near a gravitating
object with relative velocity of zero to be precise).

I got the impression from the discussion (long, with Parrott involved)
that there is no rigorous answer to that question in any reasonable
theory.

That's my impression as well. And, yes, the classical divergences and
the QED ones are not unrelated.

Thus my presense in this group is justified, because I would not hear
about this stuff for several more years. Mabey only a few if I asked
the right questions somewhere appropriate.

Why *would* one expect an electron to exert a force on itself? If it
did, how could we tell the difference than if it didn't?

Well, the best I can suggest is to read the last chapter in Jackson,
it deals with the issue. Classically, the electron generates a field
and since thsi field exists everywhere, including its own location,
the interaction of the electron with itself has to be considered.
This leads to very "interesting" equations of motion, with solutions
where an electron, initially at rest, can start spontaneously to
accelerate. This can be eliminated given some assumptions but then
you get an equation of motion when the acceleration of the electron is
proportional not to the force acting on it but to weighted time
average of the forces which will act on it in the *future*. Of
course, Jackson stresses that the time frames involved are so short
that we know that classical physics no longer applies, and a QM
treatment is needed. Ok, so we passed the buck to QM where same
issues reappear under different guise and are swepped under a new rug,
this of "renormalization". Bottom line, as long as the existance of
the electron (or any elementary particle) is just taken as a fact,
instead of following from some dynamical law (and we have no such at
present), these foundational issues keep reappearing.

Read the chapter I suggested. It is worthwhile.

What about gravitation?

Didn't even touch on this (and I doubt anybody did). In principle,
taking relativity into account, no particle can be truely "elementary"
if it has finite extent. But if it has zero extent with finite mass,
it represents an infinite mass density which runs afoul of gravity.
So, a quantum theory of gravity is needed. It is too late for us old
foggies, so we'll leave the job to you:-)

Mati Meron | "When you argue with a fool,
meron@cars.uchicago.edu | chances are he is doing just the same"

mmeron@cars3.uchicago.edu wrote:

In article <1158653398.319133.311870@d34g2000cwd.googlegroups.com>, "Eric Gisse" <jowr.pi@gmail.com> writes:

mmeron@cars3.uchicago.edu wrote:

In article <200609190746.k8J7kQq7022076@ipp.mpg.de>, Bruce Scott TOK <Use-Author-Supplied-Address-Header@[127.1]> writes:

Timo N wrote:

On Sat, 16 Sep 2006, Bruce Scott TOK wrote:

Try googling with: self forces electrodynamics

also: look up articles by Steve Parrott in sci.physics.research a few
years ago when there were lively (and well informed) threads on the topic

I'm not sure how much this will reveal about what is clear and what is not
clear. Perhaps a better rule of thumb would be that if it isn't in (and
clear in) Griffiths and Jackson, it isn't clear.

That's a good point. Basically, the point of Parrott's book is that
this stuff has never been solved.

For me the best elementary treatment of the problems with the problem is
the one in Feynman's Lectures, Vol II, where he asks precisely that
question: what it the force of an electron on itself.

At rest the self-produced electric field is symmetric and the force is
zero. In motion, considering relativity, that is no longer true and you
get infinities. Feynman makes an insightful connection between that and
the infinities in QED (in commentary... QED itself is of course beyond
the scope of the lectures).

In sci.physics about 10 years ago the thing that started the whole
discussion was when someone asked if a charge at rest in a gravitational
field should radiate (massless charged particle near a gravitating
object with relative velocity of zero to be precise).

I got the impression from the discussion (long, with Parrott involved)
that there is no rigorous answer to that question in any reasonable
theory.

That's my impression as well. And, yes, the classical divergences and
the QED ones are not unrelated.

Thus my presense in this group is justified, because I would not hear
about this stuff for several more years. Mabey only a few if I asked
the right questions somewhere appropriate.

Why *would* one expect an electron to exert a force on itself? If it
did, how could we tell the difference than if it didn't?

Well, the best I can suggest is to read the last chapter in Jackson,
it deals with the issue. Classically, the electron generates a field
and since thsi field exists everywhere, including its own location,
the interaction of the electron with itself has to be considered.
This leads to very "interesting" equations of motion, with solutions
where an electron, initially at rest, can start spontaneously to
accelerate. This can be eliminated given some assumptions but then
you get an equation of motion when the acceleration of the electron is
proportional not to the force acting on it but to weighted time
average of the forces which will act on it in the *future*. Of
course, Jackson stresses that the time frames involved are so short
that we know that classical physics no longer applies, and a QM
treatment is needed. Ok, so we passed the buck to QM where same
issues reappear under different guise and are swepped under a new rug,
this of "renormalization". Bottom line, as long as the existance of
the electron (or any elementary particle) is just taken as a fact,
instead of following from some dynamical law (and we have no such at
present), these foundational issues keep reappearing.

Read the chapter I suggested. It is worthwhile.

I would have thought any self forces would be symmetric and could not
possibly lead to what you described.

I have time today and if I remember I'll take a look. The concept is
whacky enough to merit at least a look. However, I have a feeling that
at best the concept will be filed away to my subconscious to re-appear
sometime down the road considering my experience with MTW's
Gravitation.

What about gravitation?

Didn't even touch on this (and I doubt anybody did). In principle,
taking relativity into account, no particle can be truely "elementary"
if it has finite extent. But if it has zero extent with finite mass,
it represents an infinite mass density which runs afoul of gravity.
So, a quantum theory of gravity is needed. It is too late for us old
foggies, so we'll leave the job to you:-)

I know I'm interested in gravity. That much is clear to me. What I
can't decide is whether I'm more interested in being an experimentalist
looking for a hole in gravity as we understand it, or as a researcher
building new theories.

To me it is abundantly clear which one is needed more, but I need
another oh, say, 5 years of education first before I can make a useful
crack at either. Hopefully someone will either observe something cute
that can't be explained, or technology will improve enough to probe
Just A Little Further as with particle physics. GR is wrong - it has to
be. The question is *where* is it wrong.