From: George Murphy (gmurphy@raex.com)
Date: Sun Oct 19 2003 - 19:26:21 EDT
Don Winterstein wrote:
Keith Miller wrote:
"Any accepted scientific theory is simply the best existing
unfalsified explanation for the observations already made. This is as
true for physics as it is for evolutionary biology."
Keith, I liked your post and mostly agree with it, but for my
edification I'd like to hear your (or anyone else's)responses to two
sets of questions:
(1) Given what we know about Mercury's orbit today, wouldn't the
inability of Newton's theory of gravity to account for the motion
constitute falsification? That is, Newton would predict one thing,
and we'd observe another. Yet, if we didn't have an alternative such
as general relativity or something similar, wouldn't physicists still
accept Newton even though he'd been so falsified?
.............................
The anomalous precession of Mercury's orbit was known by at least the mid 19th
century yet physicists & astronomers did not consider Newton's theory to be falsified.
You could, e.g., postulate sufficient mass between the sun & Mercury, in the form of a
planet (Vulcan), or dust, or meteorites, to cause this precession. Leverrier, who had
predicted Neptune & solved a similar problem for the orbit of Uranus, devoted a good
deal of effort to the search for Vulcan. If you wanted to be a bit more radical you
could keep Newton's laws of motion & his idea of a universal gravitational but change
the dependence of the latter on distance from 1/r^2 to 1/r^(2 + s) with s a small
number.
All of these ideas illustrate Lakatos' approach that I sketched earlier -
maintaining the hard core of a theory by modifying the protective belt of auxiliary
hypotheses. But in doing this Newtonian theory wasn't predicting any "novel facts".
Einstein's theory did that by explaining the anomalous precession of 43''/century in a
natural way. & this counts as a "novel fact" for Einstein's theory even though the
observational result had been known for quite awhile because it wasn't used in the
construction of his theory.
In Keith's statement, "best" would need to be understood as "most progressive"
(in the sense of being better able to predict novel facts) or perhaps in some cases
"least regressive" (in the sense of having to do the least adjustment of the protective
belt). In the case of Mercury's orbit both Newton's & Einstein's theory were
"unfalsified" - though of course the failure to detect Vulcan or other necessary masses
counted more & more against Newton as time went on.
& for all that Boltzmann said that elegance should be left to tailors, it does
play a role in evaluating theories. The proposal to replace 1/r^2 by 1/r^(2 + s) is
just ugly & would be acceptable only under duress.
Shalom,
George
George L. Murphy
gmurphy@raex.com
http://web.raex.com/~gmurphy/
-- George L. Murphy gmurphy@raex.com http://web.raex.com/~gmurphy/
This archive was generated by hypermail 2.1.4 : Sun Oct 19 2003 - 19:27:38 EDT