3.265 computers in the humanities, cont. (54)

Tue, 18 Jul 89 19:32:57 EDT

Humanist Discussion Group, Vol. 3, No. 265. Tuesday, 18 Jul 1989.

Date: Tue, 18 Jul 89 10:15:51 BST
From: Donald Spaeth (0532) 33 3573 <ECL6DAS@CMS1.UCS.LEEDS.AC.UK>

I am reading the discussion of the recent Chronicle of Higher Education
article with much interest. Unfortunately, I do not have access to
the Chronicle (no-one I've mentioned it to in Britain appears to have
heard of it!), so I can only comment second hand. But I am nervous about
the implication that the computer will make the humanities more
scientific, in that hard facts will render obsolete arguments which are
crafted to obscure lack of evidence. Jim McSwain's comments on this
point are well made, at least in the context of history. The documents
which survive are a sample, and unfortunately not a random one.
This means that 100 cases prove no more than 10 examples (the number
Keith Thomas tended to find, more or less, for each point). One common
problem is how to decide how significant a small number of documented cases
of a phenomenon is. They may be a small proportion of a much larger
number of episodes which occurred but were either not recorded or were
recorded in documents that were later lost. Or they may represent the
only occasions when such behavior occurred. There is no way of solving
this dilemma, although the historian may prefer to believe the first

With or without a computer, the historian must rely upon his/her overall
understanding of the documents, their provenance, their biases and
their flavor (an expression that may cause me trouble on my next
visit to the archives!) to judge how typical an example is. The
reason for giving a number of examples is less to swamp the reader
with the mass of evidence than to give a sense of how a trend being
described manifested itself in practice in various different ways and
to place the evidence in clear sight so that the reader may
evaluate it. In the case of Keith Thomas, for example, because he
was so liberal with his examples it is often possible for someone
who is familiar with an example he uses to dispute his interpretation.

If the computer will render obsolete the well-crafted bit of rhetoric,
based on weak foundations, I want no part of it. Such arguments
can be among the most enjoyable to read. (I'll cite no examples
lest I be had up for libel!) After all, that's what graduate school
is for: to teach us to see through such arguments.

Don Spaeth
Arts Computing Development Officer
University of Leeds
email: uk: d.a.spaeth at uk.ac.leeds
Janet/earn: d.a.spaeth at leeds.ac.uk