From: "Tobin Nellhaus" <nellhaus-AT-gwi.net>
To: <bhaskar-AT-jefferson.village.Virginia.EDU>
Subject: Re: BHA: Non-experimental science (was "What must the ...")
Date: Sat, 26 Jul 1997 19:03:54 -0400


Hi Ruth--

Uh-oh, I seem to be conducting heavy work avoidance again....

> > If social structures are (or have) generative mechanisms, then there is
no metaphor.  
> 
> I guess metaphor is the wrong term.  It's more like an analogy.

Same difference, in this context.  Either social structures are/involve
real generative mechanisms, in which case there's no metaphor, analogy,
similitude, etc; or they don't, but are/involve something *like* generative
mechanisms, in which case analogy (or whatever) does apply.  You seem to be
saying that, because social structures differ from natural ones, they
aren't properly called generative mechanisms; I'm saying they *are*
generative mechanisms, albeit of a different sort.  At the very least, you
appear to be taking natural mechanisms as paradigmatic, and I don't think
that's necessary or helpful.  That much is clear from the following:

> Then we get to social structures.  And we want to say that there are 
> generative mechanisms here, too.  Now the term does not carry with it 
> the same, or perhaps the full, sense that it had in the "original" (in 
> the sense of the logic of the early arguments for critical naturalism, 
> moving *from* RTS to PON) context.  

I really don't see why the sense is different.  All that is needed for
something to be a generative mechanism is for it to possess causal powers
and susceptibilities.  Social mechanisms have these in abundance.  It seems
to me that you think there are additional criteria, and that these somehow
disqualify social mechanisms from being generative mechanisms in the full
sense.  The leading contender appears to be in the following:

> Thus the contrast: in the social realm, unlike in the natural realm, 
> generative mechanisms are quite clearly *not* ontologically independent 
> of the intentions of human beings.

Well, of course, most of society consists of structures and conditions
that *nobody* intended (whence Engel's famous parallellogram of forces);
but I assume you're alluding to humans as intentional causal agents.  There
are a couple different issues involved here.  First there's the question of
whether this assertion is true, or rather, the sense in which it is true;
and I think Margaret Archer is right to argue that it must be taken
primarily with reference to the past--that is, what defines social
structures the most are the practices and concepts of the long dead, and we
the living must work with that inheritance and can only do so much.  (And
in that sense or to that degree, social mechanisms are in important
respects *independent* of living humans.)  But this leads me to my second
point: whether you accept that argument or not, it only concerns the *sort*
of mechanism that social structures might be, and the conditions for
studying them.  It does *not* concern the issue of whether social
mechanisms are full-blooded generative mechanisms.  Likewise, biological
entities are not ontologically independent of chemical entites, but they
are no less causal beings and generative mechanisms for all that.

What I'm trying emphasize, then, is that the issue of whether or not X is a
generative mechanism must be kept distinct from the questions of what sort
of mechanism X is, what the conditions of its existence are, or what
methods of studying it might be available.  The concept-dependence of
social structures affects the latter issues, but not the former.  *That*
hangs only on causal powers.

To return to the question of experimentation, I don't really follow your
argument.  You say:

>         If generative mechanisms in the natural realm are 
> ontologically dependent on the existence or interventions of human 
> beings, then what is the point of RTS?  So, for the purposes of RTS, 
> generative mechanisms *are*, if experimentation is to be both necessary 
> and intelligible, independent of humans.

[snip]

>          Now I don't have a problem with 
> this, particularly, but Marshall's point still seems to me to be kind of 
> obvious: the case for realism in the natural sciences (viz., that were 
> generative mechanisms *not* human-independent, experimentation would (a) 
> be unnecessary and (b) make no sense) --  has not a lot, logically, to 
> do with the case for realism in the social sciences.        

Are you saying that, in the case of human-dependent mechanisms,
experimentation is neither necessary nor rational?  Untangling your
double-negatives, that seems to be your assertion.  Farewell, experimental
psychology.  In fact, farewell social activism--we really must be crazy to
try to change things.  Sorry, I'm not trying to be mean, I'm just playing
out the problematic implications of what I'm looking at.  Actually my guess
is that, once put in the affirmative, you immediately saw the holes in
(what I took to be) your statement.

I wrote:

> > I still don't see any basis for saying that RB's analysis "depends" on
the "necessity" of experiment.  That account seems tantamount to 
> empiricism or positivism, which insists that everything must be 
> reducible to sense experiences.

to which you replied:

> But this is silly.  At a minimum, in my first post I noted that in 
> Dialectic he says he's done without the experiment argument.

Uh, Ruth, you keep talking about the "necessity and intelligibility of
experiment."  So is experiment necessary or is it not?  As for the rest, my
apologies, I was not replying exactly to what you had said, but to various
statements earlier in this discussion, to which you were responding.

Now I better go back into hiding, before I get a hiding.

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


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