>>Never mind the techniques for now, are there studies to indite
>>that the whole topic is valid in the first place?
>
> I once read a paper contending that the "Thomas More" codex was
> done by W.Shakespear, because the usage of dozen pairs of "skip
> words" (eg. a/an, all/any, better/best,...) gives no significant
> differences vs all the alledged authentic works of Shakespear.
The "skip word" distribution cited above may not have typical of
individuals, but of a certain period of the English language (Shakespear
and More lived in the same century. More was executed by the father of
Elisabeth the Great, Shakespear's patroness). The distribution of certain
parameters may be typical of period language. This, of course, must be
investigated by comparing several texts representative of different
periods. Such period language studies have been conducted by Douglas Biber.
It is easy to produce non-sensical results with the help of statistics. In
authorship studies, we ideally need a limited number of candidates to a
text with unknown author, let us say two or three, we further need material
known to be written by these candidates (and produced at about the same
time as the text in question), and we need to experiment with different
criteria. The idea is to find criteria that will identify texts know to be
written by a specific writer as written by that writer. When we have such
criteria, we can proceed to the study of the text with the unknown author.
It is important to remember that statistics do not deliver absolute
certainties, only probabilities. You cannot prove anything by means of
statistics, but you may be able to support an argument by such means.
Best regards,
Lars Martin Fosse
Lars Martin Fosse
Research Fellow
Department of East European
and Oriental Studies
P. O. Box 1030, Blindern
N-0315 OSLO Norway
Tel: +47 22 85 68 48
Fax: +47 22 85 41 40
E-mail: l.m.fosse@easteur-orient.uio.no