SESSION 4: 10.00 – 12.30. Saturday, March 27
A Wikipedia page, considered as a dynamic textual object, would exhibit discursive features peculiar to its continuing augmentation and revision: which circumstance may well endow the page with a distinctive discursive history. The sheer volume of the archive that is Wikipedia necessitates, and its digital form abets, the automated assay of its contents, for evidence upon which to found such inference and interpretation as is proper to the eliciting of such a history: which should be particularly eventful when the topos or subject of a page admits incompatible founding premisses: and the automated routine that will be presented, to cluster the contributors or editors a page has had over some suitable length of time, is meant to uncover evidence of the conflicting agendas that different editors may now be pursuing. We shall dwell on the choices that were forced on us, in the design of the routine, by the exigencies of numerization and by considerations of efficiency: and we shall insist that the output of the routine is not, by itself, historiographically pertinent knowledge. Our clustering routine is a means, only, to the discovery of evidence: and as such would be useful only to those who bring considerable prior knowledge, and discursive agility, to the writing of a discursive history. The epistemological presumptions behind these claims, and the prejudice issuing in the caveat that follows, should be acceptable and natural to those who value qualitative understanding, in what are called the human sciences, over such objective consensus on hypotheses as may be obtained by their quantitative testing — particularly where the formulation of hypotheses is severely constrained by the demand for numerizable consequences — and this great methodological divide is the larger setting of what we shall venture to say.