In place of Halliday's ‘verbal group’

Abstract

In Part 1 (Fawcett 2000b), I presented a number of explicitly syntactic reasons why a functional grammar of English functions better without Halliday's concept of the ‘verbal group’—reasons that follow logically from Halliday's own decision to promote the Finite to function as an element of the clause. Part 2 adds no fewer than three further sets of reasons for abolishing the ‘verbal group’. Two arise directly from the experience of building a large, computer-implemented, generative systemic functional grammar. The first of these comes from the challenge of generating the morphemes and ‘portmanteau realizations’ that constitute the elements of the supposed ‘verbal group’. The second is the evidence from nine different types of close interdependence between the meanings realized in (1) ‘verbal group’ elements and (2) established clause elements (such as Adjuncts). And the third type of evidence is the extraordinary frequency with which discontinuity would occur if the ‘verbal group’ were to be recognized as a unit. The paper concludes by considering the implications of these changes, both for linguistic theory and for the application of such descriptions in various fields.1