Natural Science Forum / Physics / Optics / August 2005

One Comment I have read, in Feinman's book, "QED", is that for short
distances light can and does go both faster and slower than C.  for
any appreciable distance those components cancel and light appears to
travel at C.

He's smarter than I, so I accept him at his word.

 In some optical fibers and waveguides, you have "phase velocity" and
"group velocity".

 Ever see a wake of a passing boat while on the shore of a calm lake or
calm slow river?  The wake often has a bunch of wavelets.  The "phase
velocity" is that of the wavelets, and the "group velocity" is that of the
cluster of them, as in the wake.  Often you can see wavelets materializing
at the trailing edge of the group, moving to the leading edge of the
group, then dying out.
 Another thing you can usually see in this case is the group getting
wider and more blurred as it travels.

 When the waves are electromagnetic waves, the group velocity does not
exceed C even when the phase velocity does.

 Another thing:  When phase velocity and group velocity differ, it
appears to me that usually the phase velocity varies with frequency.

- Don Klipstein (don@misty.com)

Some of my colleagues -- quite a few in fact -- like to analyse certain
dispersive linear systems, or even do experiments on them, in which the
envelope or the peak of a light pulse appears to travel forward at
faster than c -- an effect which they refer to as "superluminal" light
propagation.  

This effect occurs because the different frequency components that make
up the pulse travel at different phase velocities in the dispersive
system, and the pulse envelope is as a consequence reshaped such that
its peak, or some central portion of the pulse, appears to move forward
at greater than c.

The very frontmost edge of the pulse, however, never travels faster than
c, and I would bet that there is some general theorem about the energy
in any finite section of the pulse never moving forward at great than c.

In addition, I am not aware of any such systems, much less any
experimental results, in which the predicted or observed forward shift
of the pulse peak is anything more than a very small fraction of the
pulse width.  In one highly touted result the peak of the output pulse
was shifted forward relative to its ordinary velocity of light position
by, as I recall, some 10s of nanoseconds, in a pulse that was tens of
microseconds long.

I'd be very glad to hear of any systems in which the predicted forward
shift of the pulse peak through the system, relative to free space
propagation over the same physical distance, is even as much as one full
pulse width.

The whole of physics is one huge misconception bristling with misnomas.
Take for example "bent space". How can a vacuum be bent? Well of course
there has to be something physical in it to "bend" but no, there's no
proof for space having substance ( of course neglecting the stuff that
bends trajectories and stops alleged photons going too fast!) Then
there's magic "fields", magnetic and gravitational, again made of, yup
you guessed it, nothing.
Fizzics is a f.cking joke. They don't have a clue what gravity is, what
photons are, what fields are made of, even less what consciousness is.

I explain it to you, how it works. Inhomogene waves have a komplex
propagation vector.
Because the propagation vector is complex, the dispersion relation is
complex.
Those waves have been traditionally named inhomogenous waves. Their
difference to homogeneous
waves is, that the amplitude decreases exponentially in a certain direction.
You find inhomogeneous
waves in all media inclusive vacuum. While homogeneous vacuum waves travel
exactly at c, inhomogeneous vacuum waves travel with superluminal
velocities. When two inhomogeneous waves overlap, even nearly
infinite velocities can be reached --> Nimtz experiment.

There also exists a special type of wave, often referred to as bonded
surface plasmons.
It is nothing else than a inhomogenous wave with a propagation direction
parallel to the surface and the exponential decrease perpendicular away from
the surface. A photon in the case of the Nimtz double
prism experiment just stays a short time in this wave and this causes the
difference between nearly infinite
and infinite signal group velocity. The infinite part comes from an overlap
of an ingoing an reflected
inhomogene waves in the vacuum between two prisms.

Even though in a fiber we have other wave forms than inhomogene plane waves
which is difficult math,
the princile can be taken to illustrate this. The modulation of light speed
in vacuum is caused by different
real and imaginary parts of the propagation vector. There is a degree of
freedom for it.
In media, you have always inhomogene waves due to absorption. Here you have
also a degree of freedom
in the angle between the direction vectors of real and imaginary part of the
propagation vector.
You can control them for eample via a variation of the incident beam angle
or other possibilities depending
on certain equipment. But the key behind the propagation mechanism is always
the same. Its the complex dispersion relation, which always have one degree
of freedom in a certain material.

Josef

It hasn't, yet, as far as the received opinion of most scientists is
concerned.

*However*, there is one result that has some scientists wondering.

If a process generates a particle and an antiparticle such that their
combined angular momentum is zero, but each one has a spin of its own
(they being equal and opposite)... and the two particles go off in
opposite directions...

and then you measure the polarization of both particles at the same
time, as they are some distance from each other,

both will _always_ have spin up, or spin down - if that's what you
measure at both ends.

What one would have expected is that the "real" spin of the two
particles might have been, say, 37 degrees off the vertical, and the act
of measuring if they have spin up or down would then have led to odds of
the sine of 37 degrees squared for each particle *individually*.

Of course, if the particles cheated - and were carrying along two
identical books explaining exactly what to do if any conceivable
measurement were made on them - this is called "hidden variables" - this
result could be obtained without faster-than-light communications
between the particles.

Then, in that case, if you gradually change from measuring the spins in
the same direction to measuring them in the opposite direction, you have
a probability of crossing the sudden jump from "point to the top of the
apparatus" to "point to the bottom of the apparatus". So the probability
of a mismatch should go linearly as the discrepancy in angle increases.
(Bell's inequality.)

Not as the square of the discrepancy, as would be expected if particle 2
"really" had the (exact opposite of the) spin measured for particle 1.

Well, the experiment's been done. And it *is* the square of the
discrepancy.

But the official word is still that, NO, the particles aren't
communicating faster than light, reality is just "nonlocal", and we're
just too stupid to understand or imagine what that really means in a
Universe which is still 100% relative, and where being able to go faster
than light _therefore_ means being able to go back in time and kill your
maternal grandmother.

The term "popular misconception" is the problem here. This implies folk
origins, when in fact, if 'nothing can ever go faster than light' _is_ a
misconception (which is indeed possible in view of the above) it is a
misconception with *elite* origins among real scientists. If scientists
want to admit they were wrong, well, that's part of the scientific
process. But trying to shift the blame onto someone else, well, that's
just too much!

John Savard
http://www.quadibloc.com/index.html