19.390 collegiality

               Humanist Discussion Group, Vol. 19, No. 390.
       Centre for Computing in the Humanities, King's College London
                   www.kcl.ac.uk/humanities/cch/humanist/
                        www.princeton.edu/humanist/
                     Submit to: humanist_at_princeton.edu

         Date: Wed, 02 Nov 2005 07:06:32 +0000
         From: Ryan Deschamps <Ryan.Deschamps_at_Dal.Ca>
         Subject: Re: 19.385 collegiality

Williard,

Having gone through the gauntlet of economic paradigms and models, I
feel I can speak a little as to how economics can get war-like, even
among academics.

Neophyte economists go through a process of learning economic models
-- things like supply and demand, aggregate demand curves, income and
consumption-side GDP forumlae and so on. These become the basis for
empirical inquiry -- you may even say that this becomes the language
of inquiry. THe problem is that I don't really believe economists
really know what to do when their empirical data behave differently
from the models that they learned. The analogy I make to this is a
colonial one -- as if the English speaking economist encounters
non-English data, then they frame it in terms of the model, suggest
the incorporation of new dimensions, yet another greek symbol to
represent a variable which confuses the heck out of anyone with less
than 3 years of study in the discipline and frankly, are just
impossible to teach.

So what you get are students who need to learn "fundamental" models
that are simple, but sketchy -- and if they want to get out of the
"sketchy department" they have to learn, well, Greek.

Ideologies form around the "sketchy" models, and they have serious
implications for groups of people. For instance, whether you
believe ala Keynes that "wages are sticky on the downward side," "ie
that workers and organizations are slow in agreeing on prices that
are lower than the status quo," you might argue that a "tight"
monetary policy to control inflation would create excessive and
prolongued hardship on the workforce. This view, in ideological
terms would be highly favored by unionist ideologues who in turn,
fund certain political parties and not others. The opposite view, a
view purported by Monetarists, would imply controlling inflation is
the primary goal for monetary policy, something that would favor
business profits which would have its own ideologues.

In a interdisciplinary sphere, like Public Administration or
International Development, you learn just enough of this to be
dangerous. And there is rarely an opportunity to explore the "true"
experience of monetary policy in a society, and if you do look at
some data and it doesn't quite suit model, the assumption is that you
did something wrong, or that the model is imperfect, but works
sufficiently to cover most situations.

Another challenge for the economist who thinks his/her data has
"special" characteristics is that saying the data supports x position
(which will be what policy makers will be looking for) will make
certain friends in certain circles and certain enemies in
others. And perhaps the bigger problem is that economists make
really bad anthropologists.

Jane Jacob's book _Dark Age Ahead_ is an interesting study in the
cross between science and policy and has definite relevance to this
list, although the example here is tied to the models assumed by
"traffic engineers."

She cites the example from the Febrary 16 1998 issue of _Chemical &
Engineering News_ of the assumption among traffic engineers who
assumed via a computer model, that "closing a road causes traffice
using it to move elsewhere." The empirical tests actually showed
that when the road is closed 20-60% of the traffic simply
vanishes. (Jacobs, Jane (2004). _Dark Age Ahead_ Random House Canada_. p. 75.

So, I think there are a number of questions coming out of this.
1. What is the relationship between/among scientific models &
ideologies? 2. Does IT play a role in catalysing these
ideologies? 3. What is/are the academic's role(s) in mediating
between models, data and the political sphere? 4. Can computing
humanists play a role here? 5. How do you train the student who
knows just enough to be dangerous to be critical of the standard models?

Ryan. . .

Ryan Deschamps
Received on Wed Nov 02 2005 - 02:20:52 EST