File spoon-archives/bhaskar.archive/bhaskar_1997/bhaskar.9710, message 24


Date: Tue, 7 Oct 1997 12:31:16 -0400 (EDT)
To: bhaskar-AT-jefferson.village.Virginia.EDU
From: Howie Chodos <howie-AT-magi.com>
Subject: Re: BHA: But can you prove it?


Alan Norrie raises some important questions relating to a recurring
question, how do we go about preferring one explanation (or theory) to
another? He wrote:

"Explanation A is better than B when it explains everything that B explains
plus some significant aspect of the object under investigation which B can't."

I have never been entirely satisfied with this formulation, though I'm not
sure how to do better. There are two limitations to it that seem to me to be
significant. In the first place, I can't help feeling that it begs a number
of questions. A is better than B because it explains more than B does. But
what counts as an explanation? Is there a neutral, unproblematic way to
tally up which explanation explains more phenomena? Is it simply a
quantitative matter? Can we explain more things and still be wrong about
certain key ones? Who does the adjudicating and from what vantage point? Can
it be rational to adhere to a theory which explains less?

The second limitation is that explanatory power seems to me to be just one
aspect that enters into the evaluative equation. The degree of explanatory
power would not seem to be the exclusive determining factor, for example, in
adjudicating amongst moral theories. Now if all we are concerned with is
explanation, then this second limitation may not be germane. But when it
comes to not unimportant matters such as how we could better organise social
relations to foster human emancipation, we necessarily transcend mere
explanation (eg., of the causes of human misery) and enter the realm of
trying to project how new and untried ways of doing things could be better.
(This point tiptoes gingerly to the edge of the whole minefield of how fact
relates to value).

This leads me to think that there is no way around the fact that what we
judge to be better is in some sense tied to criteria that are necessarily
historically and culturally circumscribed. Even Alan's formulation
relativises the notion of better. Explanation A is better in relation to the
explanatory power of B, not in any absolute sense. Such a notion of better
is both contingent and necessary (another example, perhaps, of needing to
hold onto both aspects of a duality, that Alan refers to). Its substance is
contingent, while it is formally necessary to the possibility of comparing
theories.

I am inclined to think that this applies to judgemental rationality in
general. We know that we can make rational judgements, but we cannot define
in advance the content of the rationality that allows us to judge. We can
compare explanations to see which one is better, but we cannot define in
advance what being better means. Some notion of better is a transcendental
condition for the possibility of comparing theories, but it is a formal
condition. Critical realism can show us that there will always be one theory
that can be preferred to another, but it does not allow us to deduce from
philosphical first principles which theory that will be. I guess I'm not
sure that the formulation "explanation A is better than B when it explains
everything that B explains plus some significant aspect of the object under
investigation which B can't" captures this.

Howie Chodos



     --- from list bhaskar-AT-lists.village.virginia.edu ---

   

Driftline Main Page

 

Display software: ArchTracker © Malgosia Askanas, 2000-2005