Thursday, August 9, 2012

2-D Framework for Explanations in Why

Version 0.2

An Interesting 2-D Framework for Explanations in Why? What Happens When People Give Reasons... and Why by Charles Tilly

Not doing justice to a rather interesting book of applied epistemology, but wanted to record the key analytic notion because it struck me as unusually interesting. The book is much longer of course, with a long chapter on each of the four cells, but it's also a rather heavy read and I don't even feel I could do justice to it on the basis of one normal reading. This is the kind of book that you could easily build a college course around, probably in psychology, but perhaps in another department such as philosophy or sociology or even within a law school or medical school (since many of the examples are specific to those two professions).

It's actually somewhat difficult to clarify his two dimensions. He labels the horizontal dimension popular versus specialized, where "popular" is in the sense of general use by the entire population while "specialized" is for use by specialists. The other dimension is for formulas versus cause-effect accounts, but I think of the "formulas" more in terms of purely descriptive records of observation.

As he applies the typology, popular formulas defines the cell for "conventions", which are reasons that are basically just standard expressions and somewhat mindless. In the morning, everyone says "good morning", and that's a convention that doesn't involve a lot of thought. The popular cause-effect accounts usually take the form of "stories", where the key is a reduction from the complexities of life and the storyteller's insertion of reasons and rationales for the events, often leading to a moral or illustrating some teaching point. The specialized formulas are the "codes", explicitly in the case of codes of law, but implicitly in such situations as standards of good medical practice that physicians are expected to follow. In applying the code, the intention is to fit the real-world situations into the closest matches of the code and then follow the specified procedures. The judge attempts to match prior cases and follow the precedents of earlier decisions, or the physician attempts to match prior diagnoses and follow the standard treatments for those illnesses or injuries. The specialized cause-effect accounts are called "technical accounts", where the experts are talking to each other about their deep justifications of what was observed in the real world where the codes were applied.

Not sure if writing that will help me internalize the deep concepts here, but it's definitely a thought provoking book and worth the reading, if you are interested in having your thoughts provoked. The part that is harder for me to follow is how our selections among these kinds of reasons determine or are determined by the relationships among the people giving and receiving them. The most concrete example is how physicians relate to other physicians as peers when they exchange technical accounts, whereas they relate to patients from a superior perspective when they present the code of standard medical practice.

In the chapter on technical accounts, he includes some general works targeting non-specialists, and based on my reading of Blood, Germs, and Steel by Jared Diamond, I regret that none of my accessible Japanese libraries seem to have the books he mentions by Charles Pasternak and Luigi Luca Cavalli-Sforza--at least not in English.