[Phys-L] Re: accelerating charge

[jsd at AV8N.COM (John Denker)]

John Mallinckrodt wrote:
> > ... we know from special relativity that uniform motion cannot radiate
>
> Right, but don't we also suspect that uniform acceleration can't
> radiate?

I wouldn't have suspected that.

> As Feynman points out, it would seem to be inconsistent
> with the strong equivalence principle and the first law of
> thermodynamics.  Do we want to admit a universe in which a static,
> charged spherical body emits steady EM radiation?

I can't follow that argument.  There are too many steps
missing.  If we are supposed to get the details from
Feynman, please provide a specific citation.

I imagine that the scenario involves a spherical planet
with both mass and charge.

Every version I know of the equivalence principle is *local*,
i.e. the gravitational field is *locally* indistinguishable
from an acceleration.  If you have a version that is stronger
than that, it is easily refuted, e.g. using the non-uniform
centrifugal field of a playground merry-go-round.

I don't see how to apply a local principle to the putative
radiation of the planet, since the situation is non-local,
in that it is static (infinite timescale) and involves the
electric field (a long range, i.e. infinite-range force,
i.e. 1/r potential).

At each point near the surface of the planet, you need to
take into account not just the local bit of charged surface,
but the entire surface, because it is a long-range force.

The infinite range of the force is very special, and is
"just right" to give us a number of remarkable consequences.
Faraday cages, for instance.  It has been known since
Newton that a spherically-symmetric shell of mass is
externally indistinguishable from a point-mass at the
center (and the same goes for charge instead of mass).
So tell us, in which direction is this central point
accelerating?