gjphw@ihuxm.UUCP (05/31/84)
This brief item is a comment upon a comment made about thought experiments,
and an anecdote about the curvature of space.
E. Turner (astrovax!elt) seems to have made one unfortunate statement
concerning the value of thought experiments. The most famous proponent of
thought experiments was probably A. Einstein, but these experiments were not
used to demonstrate or prove anything. Instead, Einstein used thought
experiments to clarify concepts surrounding physical phenomena. It appears
dangerous to me to suggest that thought experiments can serve to prove
anything. They should only be used to clarify concepts (are you reading
this, Sarfatti?).
Also, you should be concerned about the caliber of the individual who is to
perform these thought experiments with you. Einstein used Max Born and Max
Planck as sounding boards for his thought experiments. Unfortunately, I
cannot claim to have acquaintances of similar stature or promise for my
thought experiments.
As has been discussed, one of the predictions of general relativity (now
often called Einstein theory) is that the curvature of space becomes
significant near massive objects. Since light follows the path of least
action (Fermat's principle), the light path should deflect with the curvature
of space. Einstein calculated this bending for light traveling very near the
Sun, and the (at that time) young astronomer Eddington said that he would
attempt to measure this bending.
At the next available solar eclipse, Eddington made his measurements for the
position of a star that he had previously selected as a good candidate. And,
he obtained the shift in the star's apparent position in complete agreement
with Einstein's prediction. Later people have commented that the errors of
Eddington's technique were greater than the effect he was attempting to
measure, but we will not dwell on this.
A newspaper reporter went to Einstein with the news that Eddington had
confirmed one of the predictions from his latest theory.
"Were you worried?" the reporter asked.
"No." answered Albert.
"But what if Eddington had not measured the effect you predicted?"
"Then he would have been wrong." replied Albert.
--
Patrick Wyant
AT&T Bell Laboratories (Naperville, IL)
*!ihuxm!gjphw