What You ( Fore ) see is What You Get : Thinking About Usage Paradigms for Computer Assisted Text Analysis

Abstract

Humanities Computing (HC) has been in existence since the beginning of computing, and was originally created as an attempt to transform aspects of humanities scholarship – a goal that it retains today. It would seem, however, that to date this central aim has not been achieved. In this paper we argue that HC might be more influential if it moved its operations closer to traditional scholarly methods. To illustrate the issue, we describe four usage paradigms for computing in humanities research. We argue that the conduit model is the one in the minds of most scholars who use comput- ing to support their research today – a model which supports established practice but fails in important ways to take advantage of much of what computing could offer. In contrast, the HC community is much more inter- ested in two other models (labelled here as the transformation and markup models) both of which, it is argued, do not connect well with the activities traditionally associated with the majority of humanities scholarship. We therefore propose a fourth model, the object manipulation model, which we believe is able to take better advantage of the possibilities computers provide to research than the conduit model and is, like it, more closely based on actual majority scholarly practice. We also mention that other models (such as those arising from the work of Druker and McGann) are also possible, and worthy of serious consideration within the HC com- munity.