Sunday, June 06, 2010

Contrastive Explanations and Debates about the Strong Programme in the Sociology of Science

In a recent issue of Sociology of Science, Jeff Kochan examines the use of "contrastive explanations" in critiques of the Strong Programme. I really enjoyed this article because it explains the Strong Programme methodology in a clear, understandable manner. And, it makes some very interesting methodological arguments.

The article is a pseudo-response piece to critiques of the Strong Programme methodology that were advanced separately by Nick Tosh and Tim Lewens (referred to in the article as "TL", and the same herein).  Kochan starts with a description of contrastive explanation with the simple information request: "Why choose B, rather than any of the other alternatives in which A is true?" In this statement, "A" denotes the range of alternatives available from which "B" is chosen, which Kochan calls the "contrast space." The interesting thing about the idea of the contrast space is that it imposes structural conditions on an explanation. My take on it is that it is more than just "you cannot explain "B" without explaining "A"". Rather, the explanation of "B" can only be described within "A". That is, B, from the original information request, is a subset of A.

Methodologically speaking, this is fascinating because it very neatly explains how sociologists (especially my own teachers) examine scientific controversies and competing theories. My professor, Thomas Schlich, used to say that the most important step in any analysis is to show that you understand (or even assume) that your subject's perspective is internally consistent. Meaning, the scientist, physician, quack, holds to a countervailing opinion because, although they all see the same phenomenon, they "understand" it in the context of their own emotional-philosophical-social epistemic. If we return to the original information request, it means that different people will each have a different "A".  Thus, two people (1 and 2) can see the same thing, "X", but come to competing conclusions ("B1 and B2"), because their "A" is ultimately different (A1 and A2).

I did not know anything about the "Strong Programme" methodology before reading this article, but Kochan makes it into an accessible topic by offering brief explanations and by returning to the main text critiqued by TL, a book written by David Bloor.  Kochan's critique of TL also returns to Bloor's work, which used the electron-charge debate between Milliken and Ehrenhaft to explain the Strong Programme method.  Kochan's critique starts by identifying where he thinks TL misunderstand the Strong Programme as outlined by Bloor. In a bizarre twist (IMHO), TL's misunderstanding of Bloor's text is highlighted, in a Freudian slip, where Tosh omits a key set of scare quotes in his citation. For Kochan, this indicates that TL believe Bloor is a scientific realist, which is a clear contradiction of Strong Programme assumptions (relativism).

The section of the essay that I find most interesting comes after Kochan's critique of TL (unconscious) scientific realism, where he examines the contrastive approach itself. Kochan examines TL's approach to the contrastive explanation and discovers a methodological "short cut". Apparently, TL structure a contrastive argument that is really a conjunctive statement: Why did A choose P rather than Q? Because A believes P and not Q. In other words, his explanation actually avoids contrastive format by using a conjunction: meaning, he should have answered "A believes P rather than Q, because A rather than Q". This is particularly interesting because, as Kochan points out, you don't really need to use a contrastive explanation in order to achieve the same point. As I understand it, your argument can demonstrate the same goal by using a conjunctive structure instead of a contrastive one (ironically said using a contrast).

To me, this means that, if we return to our original information request (modified per above), "Why choose P rather than any other item in A (e.g. "Q")?", we can say "because P and not Q", instead of (quoting Kochan) "P simpliciter".  This returns to the earlier description of the contrastive space: we can explain why a scientist believes P rather than Q by explaining why P and why he does not believe Q, rather than explaining every single possibility in the contrastive space (A). Of course, this only works if your goal is to explain P and not Q. For instance, why did Milliken conclude that the negative charge he discovered applied to the electron and not the subelectron? This way, you do not have to entertain all of the other possible particles or explanations that could have won Milliken over.

Kochan's methodological argument is a fascinating one, and it works particularly well when examining the case made by different sides in a controversy.