re: article on Gen. Relativity

From: hal@finney.org
Date: Mon Mar 05 2001 - 16:38:37 MST


Neal asks about:
> http://www.linguafranca.com/print/0103/hypoth_eclipse.html
>
> . . . which left me scratching my head as to the author's intent. It
> seems like an interesting bit of information, but the article as a whole
> is not very substantive. Maybe the author came across this info while
> researching something else, and thought he had an article.

I think I've heard this before, that the Eddington confirmation of General
Relativity was not as strong as the publicity implied. It's certainly
true that observations are informed and guided by theory. If you do an
experiment twice and get data conforming to the theory in one run and
contradicting it on the other, you'll probably accept the conforming data,
especially if the experiment is tricky and hard to do.

This does mean that it takes longer to find out that we are on the
wrong foot, but after all, the proof is in the pudding. Science has
done a pretty good job over the centuries and this bias has existed
all that time. In a sense, we can say that a process of evolution
has been operating on the institution of science, and it has built in
a conservative bias. Scientists are more successful who assume that
guiding theories are correct than to track down every flaky result with
the assumption that they are about to turn the world upside down.

Keep in mind that we're getting spin from what I assume is a book
skeptical of Einstein, "Was Einstein Right?". I suppose that Eddington
fully described his data reduction methods in his report, else the
skeptics would not have had these juicy details to chew on. So he was
honest and open about it, and it was his professional judgement that these
results constituted a confirmation of Einstein's theories over Newton's.
If you disagree, well, he gave you the data you need to make your own
determination.

Hal



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:39 MDT