From: "Tobin Nellhaus" <nellhaus-AT-gis.net>
Subject: Re: BHA: DPF C 2.10
Date: Thu, 12 Oct 2000 19:17:33 -0400


Hiya Ruth--

You wrote:

> Meanwhile, Tobin (Hi!) I wondered if you could spell the following out for
> me.  It seems important, but I don't know how to connect the dots on my
own!
> You wrote:
> >Most seriously of all,
> >doesn't your claim that reasons explain but don't generate actions imply
> >that retrodiction, which RB presents as crucial to science (science being
> >fundamentally explanatory, not predictive), fails to give a real account
of
> >causation?

Let me first back up a moment to quote the statement Howard made which
raised the question:

> Reasons, then, are a way we explain our actions, but this does not make
> them the powerful
> particulars that generate them.

Howard allows reasons as an explanation for actions, but not as a power that
generates them.  Such a claim involves distinguishing -- even divorcing --
factors that "explain" events from those that "cause" them.  (BTW, I'm
setting aside the issue of multiple causality.  It makes no difference to
this argument whether something is the sole cause of an event, or only one
of many.)  Granted, the two terms (explain & cause) are not synonyms, but in
my view the difference between them is one of perspective: the former is
retrospective, the latter prospective.  But in either case the "factor" --
again in my view -- is the *same*.  Not so for Howard: he says here that the
action or powers exerted by an entity can explain an event, without in any
manner having caused it.  So if reasons are merely "a way we explain our
actions" but do not in fact cause them, then reasons are -- well, I'm not
sure.  I'm tempted to say he means they are fictions, mythological beings,
excuses, or somesuch -- in any case, just words, not real causal powers --
but he seems to accept that they may explain correctly  Whatever he means,
such a division between explanatory elements and causal elements goes to the
heart of Bhaskar's view of scientificity, particularly that of the social
sciences, for which the criteria of adjuducation "*cannot be predictive* and
so must be *exclusively explanatory*" (PON2 p 21, and in many other places;
his emphases).  More generally, RB views retrodiction ("to possible
[antecedent] causes") as a key part of explanation -- the pivotal third "R"
of the "RRRE model of explanation" (PON2 p 129).  So for RB, causes and
explanation are closely linked.  And of course Bhaskar uses a causal
criterion of real existence: something is real if has causal powers.  But
Howard splits causes (which are real) from explanations (which apparently
aren't), which means that for him, retrodiction (i.e. the retrospective,
explanatory approach to scientific analysis) cannot provide an account of
real causation.

To be brief, I'm saying that if reasons do in fact explain anything, then
they must in fact cause them (or at least partly cause them).  To say that
something may correctly explain without supplying a cause means we have no
ability to identify a cause -- at least not in the social sciences, in which
knowledge can only be tested through explanations rather than predictions.

Hope that helps!  Best, T.

---
Tobin Nellhaus
nellhaus-AT-mail.com
"Faith requires us to be materialists without flinching": C.S. Peirce




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