From: hal@finney.org
Date: Thu Jul 20 2000 - 11:03:00 MDT
Scerir writes:
> That superluminal light propagation (Princeton experiment) on Nature
> http://www.nature.com/nature/fow/
It's actually worse than superluminal; the pulse literally arrives before
it is sent (in their interpretation of events).
For those who can't read the article, I have made a temporary copy of
the most important graph at http://www.finney.org/~hal/nature_ftl.jpg.
It shows two waveforms, a solid one labelled A, and a dashed one
labelled B. They are almost superimposed, with B a little bit ahead of A.
These are the measurements of the shape of the wave coming out of the
chamber, in two different cases. A is when the laser is tuned to a
frequency which will not be affected by the gas in the chamber. In that
case it travels at the speed of light. B is when the laser is tuned to
the frequency which should be accelerated by the gas. We see that B
is earlier than A, by about 62 ns.
The chamber is only 6 cm long, which takes about 0.2 ns to traverse,
so if taken literally this measurement implies that the pulse began
leaving the chamber about 62 nanoseconds before it began arriving.
Given that this is enough time to travel six feet, which is likely to
be shorter than the optical path for the whole experiment, it means that
the pulse arrives at the detector before it is sent. (Note though that
the pulse itself takes several microseconds, so what it really means is
that the pulse recorded at the detector "leads" the state of the pulse
as it is being emitted from the laser.)
I think it's clear that this could not actually have happened; nothing
left the chamber until something entered it. At the very beginning of
the pulse, the first faint stirrings of the oncoming electromagnetic
winds triggered the chamber to emit a matching pulse which was 62 ns
ahead in time of the original pulse.
If this is true, it should be possible to see the effect at the very
beginning of the pulse; the 62 ns lead time should not be present at
sufficiently early points. There are insets in the graph above showing
enlargements of data near the beginning and ending of the pulse, and
no such phenomenon is evident. However I can explain this simply by
postulating that they did not look close enough to the beginning of
the pulse. If they were able to look at that data sufficiently closely,
I am confident that they would see the effect I am describing, which
would contradict their interpretation of backwards-in-time transmission.
Hal
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:30:04 MST
