Must-Read/Thick Description/Pomo...

John L.McCreery (jlm@TANUKI.TWICS.CO.JP)
Thu, 16 Dec 1993 11:45:52 JST

First, to Rick Wilk, Dan Jorgenson, Dave Beer, James Carrier, thanks
a bundle for your additions to my "Must Read" list. I'll wait a few days
to see what else comes in, compile the list and post it. You will, I think,
find it interesting what other people have suggested.

Now, turning again to "Thick Description"

Richard Robbins writes,

"Just briefly, we can't lose sight of the fact that any criteria of
assessment is based on a social judgement." To which the answer is "Of
course." My point is that while there exist well-defined "social
constructions" that allow us to evaluate statistics and experimental
results in mutally agreed terms, we lack their equivalents for
evaluating interperative/critical/ hermeneutic/ humanistic/symbolic
(pick your own favorite word) research. The muddle of competing
opinions we see is typical of what Kuhn called pre-paradigmatic
science.

Dan Jorgenson writes,

"When we were grad students, we used to complain about Geertz's
"bean-bag effect" -- the sense that when you tried to grasp something
definite, he sagged. (We also used to joke about the title of Thick
Description -- we called it Thin Decryption.)" Again, I say, right on.
Then, on second thought, I realize that it's the sheer nastiness of the
words that I love. Personally, I get a lot more out of reading Geertz by
following his logic as far as it goes, then saying to myself, "Where can
we go with this?"

The way I formulate the problem goes back to the first Noam Chomsky
I read (_Syntactic Structures_ if my memory serves me right.) Here
Chomsky draws distinctions between three views of scientific method
conceived as three different types of black boxes.

The first (which I myself learned first in 8th grade General Science
class) depicts science as a "Discovery Procedure." Here the input to the
black box is data. The output is Truth (with a Tarskian capital "T"--pax
to Bob Graber). This view is a chimera, now totally discredited.

The second (of which Popper's "falsifiability" is a classic example) sees
science as a "Decision Procedure". The inputs to the black box are data
and a theory, the output a decision: right or wrong. This distinctly
"modern" view is rooted in syllogistic logic and conceives of scientific
results as indubitable. They may, however, be subsumed in later, more
comprehensive theories, as, for example, Newtonian mechanics are a
subset of Einstein's theory of relativity.

The third (which I have come to espouse myself) sees science as an
"Evaluation Procedure." Here the inputs to the black box are data and
at least two theories. The output is a ranking, which says that one
theory is better than another.

The question, of course, is what criteria get built into the black box.
Suggestions, please.

Which brings me at last to "PoMo" again. Here I can only applaud
Seeker 1's 16 theses, which seem to me to advance the discussion
considerably. It occurs to me that what I have written above is largely
addressed to No. 7.

John McCreery (JLM@Twics.co.jp)