Modeling context and register: the long-term project of registerial cartography

Abstract

This paper presents the outlines of a long-term project concerned with the modelling of context and register along the lines originally drawn by M.A.K. Halliday and Ruqaiya Hasan in Systemic Functional Linguistics (SFL) — thus with “register” in its original sense of a functional variety of language, i.e. the meanings at risk in a given type of context. The  topic  of  this  project  can  be  characterized  as  register  cartography.  Section  1  presents the background to work on the modelling of context and register, noting different uses in SFL of the term “register” locating the notion of register as functional variation in terms of the two semiotic dimensions of the hierarchy of stratification and the cline of instantiation. Section 2 sets out alternative models of context and register, identifying the key semiotic dimensions involved in these models. It then compares and contrasts the models. Section 3 adds to this account of work on context and register by summarizing work that has been done in computational linguistics informed by SFL — important work that has sometimes been overlooked by researchers concerned mainly with manual discourse analysis. Based on the background sketched in Sections 1 through 3, Section 4 then reports on the long-term project of registerial cartography, giving a general account with examples of findings such as differentiation of registers in terms of relative text frequency of terms in certain major systems. Section  5  focusses  on  one  aspect  of  the  long-term  project  —  viz.  the  description of different fields of activity: eight primary fields of activity, each differentiated into more delicate subtypes — illustrated by means of variation in the deployment of pictorial resources.