[Humanist] 30.144 what is theory?

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Tue Jul 5 09:33:51 CEST 2016


                 Humanist Discussion Group, Vol. 30, No. 144.
            Department of Digital Humanities, King's College London
                       www.digitalhumanities.org/humanist
                Submit to: humanist at lists.digitalhumanities.org



        Date: Mon, 4 Jul 2016 10:23:32 -0400
        From: Ryan Deschamps <ryan.deschamps at gmail.com>
        Subject: Re:  30.143 what is theory?
        In-Reply-To: <20160704060038.6CBA1787D at digitalhumanities.org>


Willard,

As a social scientist (who flipped from the Humanities), what counts as
theory is something driven into our brains. A discussion in Paul Sabatier's
_Theories of the Policy Process_ helps me put the pieces together. In it,
Elinor Ostrom and others discuss "theory" at length (or worse "substantive
theory") and try to place the word in context with other
similar-but-different words like "framework," "model" and the word familiar
to the humanities, "metaphor." I could go into the definitions at length,
but perhaps it's just easier to say that a strong theory is backed by a
framework that states the important variables, and fronted by a number of
useful models that can help test whether a phenomenon fits or does not fit
into the theory.

Even now, it seems to me that computational methods of research in any
field are short on theory.

In the humanities, it is perhaps equally rare for the simple reason that
much of the humanities focuses on studying things that are intentionally
novel (including novels). Even when studying a particular period, like the
Victorian, what makes research exciting is understanding differences as
well as similarities in the field. In scientific terms, I always expect a
rejection of the null hypothesis in humanities research or i'm not
interested.  Of course Kierkegaard has a different view of Socrates than
Hegel - otherwise, why write it _On the Concept of Irony_? In science, the
approach would be to expect a stable result and report when something
untoward happens.

I see the theoretic process as different from the use of stochastic tools
in the humanities as well. If we hypothesize that sonnets are in iambic
pentameter, and Shakespeare goes with the line "And perspective it is best
painter's art," we do not interpret this as a unique phenomenon, but
Shakespeare using form to try and point something out to us. The surprising
result is not a surprise.

In network analysis, we have a similar problem. Networks form based on a
wide range of variables, both exogenous and endogenous to the network
itself. The differences between networks are usually based on common sense
and networks in my field (public policy) tend to change so frequently that
the only real theory we have is "things are complex." Thus, the network
"theory" is often accused of being merely "metaphor" (see Keith Dowding
"Model or metaphor? A critical review of the policy network approach" for
the whole story). In this context, "metaphor" means a way of describing a
phenomenon with no clear expectation of a particular result. I don't see
this as much of a problem for research, but in a field where you are
expected to make recommendations about things, it would be much nicer to
have something more predictive. (Except if people were predictable they
would be much easier to control and that would be a dangerous thing for
governments to know about, so a big qualifier on the word "nicer".)

It seems to me that computational research is more like poetry than
science. It uses science of course, but to me design and code is a complex
bricolage of metaphors that forms into a digital design.

Ryan. . .


-- 
Ryan Deschamps
PhD Candidate Johnson-Shoyama School of Public Policy
ryan.deschamps at gmail.com
Facebook: www.facebook.com/greebie  Twitter: www.twitter.com/ryandeschamps





More information about the Humanist mailing list