File /Humanist.vol22.txt, message 50

Date: Sat, 31 May 2008 10:06:50 +0100
From: "Humanist Discussion Group \(by way of Willard McCarty              <>\)" <willard-AT-LISTS.VILLAGE.VIRGINIA.EDU>
Subject: 22.050 on firm and infirm ground
To: <humanist-AT-Princeton.EDU>

                Humanist Discussion Group, Vol. 22, No. 50.
       Centre for Computing in the Humanities, King's College London
                     Submit to:

         Date: Fri, 30 May 2008 18:56:45 +0100
         From: Willard McCarty <>
         Subject: on firm and infirm ground

On 2 April 1970 a rather full report on the 1970 symposium, "Uses of
the computer in literary research", organized by Roy Wisbey in
Cambridge, appeared in the Times Literary Supplement, p. 368, under
the title "The literary computer". The aim of this conference was to
report on the then current state of research, which not surprisingly
had much to do with concordance generation for print --
Wisbey's own focus, as some of you will know. "In the more suspect
world of stylostatistics", the reviewer comments, a number of
projects were presented, but if the report is accurate, did not
persuade the attendees that statistical analysis had much of a
future. Such analysis, which we now know to be very valuable indeed
as a set of experimental techniques, seems then to have been taken as
part of an attempt to construct a mathematical model (rather than to
conduct stylistic modelling with the aid of mathematics) for literary
text. The reviewer argues that "the real value of such applications
lies less in the tendencies they demonstrate (but can never
convincingly prove) than in the fact that they cause us to think more
deeply about what we mean by 'style'." It seems to me that today we'd
not see the reviewer's opposition as a choice we need to make. But
what I most want to put before you are the concluding two and a half
paragraphs of the review -- and ask you to consider how far over the
threshold sketched here we have been able to progress. The first
quoted sentence below refers, by the way, almost certainly to Joe
Raben, founding editor of CHum, whose literary research was as described.

 >However wide the vistas opened up by the computer as a means of
 >analysing the whole process of collocation, borrowing, adaptation
 >and affiliation underlying the literary tradition of a nation or
 >civilization, dwelt on at some length by an American scholar
 >comparing Shelley and Milton, the indispensable need for aesthetic
 >judgment to evaluate the mechanically produced evidence was fully
 >admitted by all exploring such possibilities.
 >In some respects these tasks are self-defeating, for the sheer
 >wealth of the tradition which the computer can help reveal is itself
 >proof that the parallels detected are far from supplying such
 >incontrovertible evidence for precise questions of literary
 >dependence and authorship as may at first sight appear. It is likely
 >that the disappearance of the common stock of traditional formulas
 >characteristic of much medieval literature, where they help little
 >in the convincing solution of such problems, is more apparent than
 >real; and that with the rise of individuality since the Renaissance
 >and the modern critical equation of originality with creative
 >achievement, the real facts of the continuing dependence of each
 >author upon his predecessors within a particular literary tradition
 >have sunk beneath the threshold.
 >If the computer helps to bring some of these facts to consciousness
 >by revealing 'parallels' everywhere, our knowledge of the processes
 >of literary creation will be enhanced; but the detailed and often
 >trivial questions of authorship and borrowing which occupy so many
 >scholars, both medieval and modern, will recede in significance.
 >This symposium clearly revealed that in the present state of
 >knowledge those concerned with the more modest problems of
 >lexicography and of information storage and retrieval stand on much
 >firmer ground than those who would harness the computer to explore
 >questions directly dependent on wider mental processes.

A thoughtful conclusion surely. But it seems to me that this reviewer
goes badly wrong. What would you say is his or her most fundamental
error? Or do you agree?


Willard McCarty | Professor of Humanities Computing | Centre for
Computing in the Humanities | King's College London | Et sic in infinitum (Fludd 1617, p. 26). 


Humanist Main Page


Display software: ArchTracker © Malgosia Askanas, 2000-2005