Date: Tue, 22 Jul 1997 15:35:30 +0100
To: bhaskar-AT-jefferson.village.Virginia.EDU
From: ccw94-AT-aber.ac.uk (COLIN WIGHT)
Subject: Re: BHA: Non-experimental science (was "What must the ...")


Hi Marshall,

I take all of your points (if I have understood them correctly), and i'm
sure that RB deals with this critique somewhere, so I'll try and hunt out
his reply. 

>>
>>Yes but only in part. The important part about experiments is that they
>>allow us to identify mecahnisms that endure beyond the conditions of their
>>identification
>
>[in the field of science in which experiment takes place].

You see I think Bhaskar would argue that also in the non-experimental
sciences as well. i have often argued on this list that what marks out
science from other forms of knowledge is that it is explanatory, not that it
has a particular method, for example the use of experiments. As such I view
RB as deploying aguments derived from the necessity of experiments as
critiques of other philosophies of science. I'm sure he covers a lot of this
in the postscript to RTS, but I don't have it in the office.


>I agree and have never said otherwise.  The questions I've raised are: (1)
>the part in brackets above seems to be what really follows from RB's
>analysis,

We have to be wary of this though. Surely we can extrapolate form one domain
to the other, or from one instance to another or else science would be
impossible.

 (2) it adequately establishes his results in experimental
>sciences, 

That can and do apply beyond this instances of their identification hence
the things discovered must exist independent of their instances of discovery.

and (3) the bridge from experimental to non-experimental sciences
>remains problematic.  Physics (all of physics, not just experimental
>physics), and the world it studies, may be a certain way because otherwise
>physics' experiments would be unintelligible.  But this does not imply that
>paleontology is the same way. 

Well, Rb would not argue otherwise.  Paleontology constitutes a distinct
object domain that has its own methods, but it exists in this world not
another. Hence the general features of reality that RB gleans from his
analysis may persist in such a domain.

 In short, I'm arguing that ontology in one
>field does not transfer to another without further justification.  

The scientific ontology, of course, you are absolutley correct. But RB
claims to have discovered general features of reality. If I understand you
correct you are arguing that he has really only discovered features unique
(or possibly so) to the experimental sciences. But doesn't this supposition
rest on the assumption that the world of experimental sciences is not our world?


>Of course, I've never said otherwise.  I've questioned the legitimacy of
>way he extends his results to science as a whole, not the fact that he does
>so.

Yes this is a good point, but I think if you studied the nature of diffreing
sciences you would find that experiments figure in all. The issue might be
how close the can get to closure. Even in the social sciences we attempt
thought experiments.


>
>Perhaps; science as a unified practice is not sacred.  We should not let
>our wish for a generic notion of science blind us to important differences,
>even if this means we can come up with no generic notion.

Yes you are absolutley correct, that is why for Bhaskar the distinction is
not between natural and social science, but between many sciences, and for
that matter why I argue that what makes a certain form of knowledge
scientific is its explanatory content not its method of acquisition.

>But we still could come up with a generic notion that's different from
>RB's.  

Oh yes of course, the question then would be whether or not it would make
better sense than RB's.
>
>If we accept RB's account of reality as being "vertically" differentiated
>(i.e., layered), we ought not preclude the possibility of "horizontal"
>differientation (i.e., radically different ontological features in
>different realms).

Well he does insist on this. RB's ontology is stratified, relational and
differentiated. Part of his reason for the commitment to stratification and
depth realism BTW is his arguments for realism itself and many of these are
not derived from an analysis of experiments.


>Yes, at least as our starting point.  My question pertains to using a
>conception based on experimental sciences as our starting point.

This is the sticking point I fear. You read RB as dealing only with the
experimental sciences, whereas I see him as dealing with science and using
the necessity of experiments as one step on a wider project. 

Perhaps the contingent and
>variable nature of the objects of non-experimental science (dinosaurs don't
>become extinct every day) establishes for them what experiment establishes
>in the experimental ones.

But the palentologists I know do you experiments (carbon dating for
example). Also, how are we to construe digging for dinsosaur bones other
than as a commitment to a form of depth realism that doesn't reduce reality
to the experienced?


>I see your point and agree with what you say about other accounts.  The
>problem is that someone who accepts those accounts might not. 

Accepted, disagreement is a condition of being human, thankfully.

 For them,
>"the nature of science" is what their accounts describe.  We have to show
>that such accounts do not capture the nature of science, including
>non-experimental ones.

Yes good point. I think RB has moved onto this in his later work. Scientific
Realism and Human Emancipation shores up many of the weak points you
identify as does Plato etc. Still in terms of convincing most of the people
for most of the time the experimental sciences seem a goos place to start.

>>Again, for CR it is the object domain that will determine the particular
>>form a science will take. But experiments are not a condition of possibility
>>for science for Bhaskar.
>
>So why should the sociologist think Bhaskar's account of science applies to
>her field?

They don't have too. What RB is arguing here is not this straightforward. It
probaly goes something like this. 

1.      What does social science think it is?
2.      The answer to 1 depends in part upon what social science thinks
        science is.
3.      Social science thinks science = positivism.
4.      This belief leads to the application of positivist methods in SS or the
        outright rejection of them in favour of hermeneutics.
5.      Enter Bhaskar (and others) and says 3 is false.
6.      What does social science now think science might be.
7.      The best account of science is that of RB (supposedly)
8.      Socail science now thinks that TR = science.
9.      this should lead to a more scientific and nuanced view of SS, but
one         that takes seriously the difference in object domains.

RB does not think that you can simply plug TR into the social world. The
object will determine the particular form of a science.


>Good point.  I'm largely playing Devil's advocate and trying to find what
>makes RB's argument better at what seems to be its weakest point.

And doing a good job of it too.


>But couldn't someone who makes the denial "know" that science is something
>different from what RB says it is? 

Yes absolutley, but the attention paid to experiments provides a critical
tool for critical realist with which to say well your account does not
describe this activity too well. 


Anyway, once again thanks for your comments.


------------------------------------------------------------------------

Colin Wight
Department of International Politics
University of Wales, Aberystwyth
Aberystwyth
SY23 3DA

--------------------------------------------------------



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