17.545 CS and humanities computing

From: Humanist Discussion Group (by way of Willard McCarty willard.mccarty@kcl.ac.uk)
Date: Tue Jan 20 2004 - 03:01:52 EST

               Humanist Discussion Group, Vol. 17, No. 545.
       Centre for Computing in the Humanities, King's College London
                     Submit to: humanist@princeton.edu

         Date: Tue, 20 Jan 2004 07:50:01 +0000
         From: Willard McCarty <willard.mccarty@kcl.ac.uk>
         Subject: CS and humanities computing

Many here may know that during the last few decades much ink and toner and
many pixels have been deployed within computer science in an ongoing
identity crisis. Although in some ways intellectually barren -- one can
readily sympathize with the let's-just-get-on-with-it reaction -- it is
useful to understand the tensions and tendencies within CS, for they are
much like our own. It is also quite salutory and head-clearing to resolve
what may seem a monolith into a complex variety of interrelated activities.
As William A. Wulf has said, "One of the wonderful things about this
discipline we call computer science -- and I call it computer science only
because everyone else does -- is that it spans an enormous intellectual
distance, going all the way from very theoretical and abstract mathematics
to very crafty application programming" (Ubiquity 1.28, 2000,
http://www.acm.org/ubiquity/interviews/w_wulf_1.html). As you may know,
those who have thought about how to construe these activities as part of a
single field have in the past tended to put them under the rubrics of the
(physical) sciences, mathematics and engineering. Sometimes the grouping
has been simpler (science and engineering only), sometimes more elaborate
(adding in information theory as a close cousin to mathematics). More
recently some in CS have broadened the field considerably, e.g. by
including "interaction design", which requires (if I understand these
matters) that a large social science component be added to the mix. The
picture is more complicated than this, of course, but my point is the
evidently cohesible *plurality*.

Given this plurality, it would seem to me that a term such as "humanities
computer science" is rather problematic. There are at least two problems I
can see for those of us outside CS.

First it is easy for us silently to assume that we know the meaning of the
term "computer science". We are apt to take the term at its word, as
denoting another one of the "sciences" (and so among the sources of "real
knowledge"), and so for it to serve the same mind-numbing honourific
function that (at least in English) the term "science" so often serves,
e.g. in "management science". In thus honouring what is happening in the
humanities, its application to them is likely to deprive them of their own
honour by implying that the real thing takes place elsewhere and is then
merely *applied* to the humanities, where it has thus and such an "impact".
As I suspect all engineers know only too well, the appellation "applied"
can be profoundly misleading, since the central problems in such a field
may be only tangentially related to the imported knowledge, and the field
itself has many other sources of knowledge unique to itself. See Wulf's
interview for more on this topic.

The second problem is harder and more subtle. Given his or her inclination
to construe CS as having a singular essence, an outsider is especially apt
to construe "computer science" as if it were reducible to one of its
subfields. In historical terms, that is, he or she construes a particular
essence, then constructs a history to match, suppressing the other
histories in the process. For example, the theoretician, wanting "the"
computer to be an engine of logic, writes it into the history of
logic-engines, dating back at least to Raymond Lull, perhaps (as John Sowa
does). This then becomes demonstrably what the machine has *really* been
about all along. To take quite a different example, a more socially
inclined person may argue that "the" computer is no longer a device that
calculates but an "interaction machine" which forms part of a social
environment; a prominent ancestor to this computer is the telephone. Terry
Winograd argues in this direction. Another may argue similarly that "the"
computer is now prominently and is increasingly destined to be an
appliance, invisible and ubiquitous; the very word "appliance" tells you
what its history will look like. And so forth and so on. The universality
of the fundamental Turing Machine design means, I take it, that there may
be no strict limit to the number of "the" computers possible.

The error here I take to be the exclusivity in the essentialist argument.
There is no "the" computer. So what possible could "humanities computer
science" be? What in balance is the benefit from taking on this potentially
confusing label? Epistemologically what becomes of knowing by means of a
computer when it is used to help us understand the data of the humanities?
What computer do we have then?

As a mightily refreshing antidote and guide to "our young and slightly
paranoid discipline", as he calls his own, allow me to recommend something
else by Wulf, a very brief opinion-piece entitled, "Are We Scientists or
Engineers?", ACM Computing Surveys 27.1 (March 1995): 55-7, which he has
put online at http://www.cs.virginia.edu/~wulf/documents/sci.or.eng. "It
might be more comfortable to define ourselves by analogy to the
establishment," he observes in closing -- and he means the physical
sciences cum mathematics, viewed as outsiders do, as a unitary method for
research -- "but what we do is too important to be limited by that. Let's
kick the inferiority complex and find our own way, define our own
paradigms. Let's adopt the applicable best.... Let's be inclusive rather
than exclusive. Let's build even more bridges to other disciplines and to
industry; let's amplify the special strengths of each."



Dr Willard McCarty | Senior Lecturer | Centre for Computing in the
Humanities | King's College London | Strand | London WC2R 2LS || +44 (0)20
7848-2784 fax: -2980 || willard.mccarty@kcl.ac.uk

This archive was generated by hypermail 2b30 : Fri Mar 26 2004 - 11:19:38 EST