[Interpretationandmethods] How To Do Interpretive Social Science.

[Patrick Thaddeus Jackson]

Two quick comments on Bill's most recent post. First, Bill states:

> Actually, the message says nothing about answers, because its about  
> methods.


Of course it is. My point was that even the illustration doesn't  
answer a causal question; it answers an evaluative question. To me  
this says, quite simply, that Bill's proposed interpretive research  
method is a proposal for doing normative evaluations ("how rational is/ 
was this action?") instead of producing causal explanations ("why did  
this action occur?").

Second, Bill claims:

> One of Polanyi’s most significant contributions to the understanding  
> of social science methodology is his demonstration of the delusional  
> character of that distinction in both the natural and the social  
> sciences.  The notion that explanation is "value neutral" is an  
> instance of social scientists misunderstanding what they actually do  
> in social science.
>
> Personally, I think it’s a bad idea to perpetuate this=2 0self- 
> delusion by teaching it to grad students.  No social science  
> explanation is possible, much less complete, without numerous  
> evaluations at every step in the process.  Might as well do it openly.


I reiterate my claim that there is a great deal of difference between  
the observation that the concepts that we work with in the sciences  
have value-commitments embedded in them (of course they do, neither of  
us deny that), and the assertion that this presents some sort of  
problem for the fact/value distinction. I think this latter assertion  
is an instance of misunderstanding what "value neutral" means when it  
comes to explanations. Leaning on Weber as a place to start, I have  
teased this argument out at somewhat greater length elsewhere  
(actually, one web-accessible version of it is here: http://www.jpox.eu/component/streams/view,content/cid,192/) 
, but in a nutshell: the difference between a value-claim and a  
explanatory claim is that a value-claim simply states a moral position  
and then codes the world according to it, whereas an explanatory claim  
takes that position and tried to use it to generate some facts about  
the world -- facts that observers not sharing that value-commitment  
could still appreciate, because they would be able to see the logic of  
how the world looks when viewed through that value-commitment. This is  
not to say that explanatory claims are "falsifiable" or any of that  
other Popperian nonsense; rather, it is only to make the point that  
there is a logical difference between stating a value-commitment and  
using that value-commitment to produce facts out of empirical data and  
observation.

I'd also like to tease apart two senses of the term "evaluation":  
normative value-commitments, and practically competent discretion. I  
defy anyone to locate a social-scientific (or natural-scientific!)  
concept that isn't "evaluative" in the first sense; in that way I  
think that Bill is correct, even though I disagree with the  
conclusions he draws from this. But from my perspective, the most  
relevant "numerous evaluations at every step in the process" of doing  
social-scientific research are issues of technical competence; what we  
train students to do is to be able to utilize their professional  
sensibilities in ways that carry them past the necessary ambiguities  
of real-world research problems and puzzles and data. The world is not  
self-interpreting; we have to make sense of it. The fact that the  
dominant method of sense-making in the contemporary social sciences  
revolves around shoving observations and instances into little  
discrete boxes called "variables" and then devising more and more  
technically sophisticated ways of testing for systematic associations  
between those little boxes is testimony to the conceptual poverty of  
our line of work. Researchers always make practical evaluations when  
going from one step in a research project to the next -- they have to,  
since methodological rules and methodical procedures are always in  
some way insufficient to guide actual empirical work (and this is a  
general feature of formal rules and procedures, not a failing of  
social science methodologists; Wittgenstein figured that out decades  
ago, but news of this seems to be taking its own sweet time to  
percolate through the social sciences) -- and the fact that these days  
many if not most most social scientists are only taught to make  
statistical-comparative variable-based evaluations is a travesty.

End of sermon :-) and end of my comments on this listserv for a while  
-- I have some book chapters to write!

PTJ
===
Patrick Thaddeus Jackson
Director, General Education Program, American University
Editor-in-Chief, Journal of International Relations and Development
http://profptj.blogspot.com | http://www.kittenboo.com
calendar: http://ical.mac.com/onyxdr/Patrick

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserv.cddc.vt.edu/pipermail/interpretationandmethods/attachments/20080709/54219479/attachment-0001.html