Misrepresenting The Theoretical Architecture Of IFG2 And Halliday (1993)

Fawcett (2010: 102-3):

As we saw in Section 5.2 of Chapter 5, IFG is intended as an account of the outputs from the operation of the grammar, and not as an account of the grammar itself. It describes the instances, but only at the level of what is assumed to be the 'final' output at the level of form (in terms of the diagram in Figure 4 in Section 3.2 of Chapter 2). In contrast, the emphasis in "Systemic theory" is wholly on describing the potential of a language — and at the levels of both meaning (the system networks) and form (the realisation operations). This contrast in goals explains why the concepts of 'system', 'system network' and 'selection expression' play so little part in IFG. And, at the level of form, it explains why the focus in "Systemic theory" is on the generative apparatus of SFL and so on the 'realisation statements' that constitute the 'form potential', at the expense of the output structures that they generate. (Nonetheless, the lack of any mention of many of the categories and relationships that specify the outputs leaves a notable gap in what is intended as an account of the theory.)

Blogger Comments:

[1] This is misleading, because it presents IFG (Halliday 1994) in terms of Fawcett's model (Figure 4) which, as previously demonstrated, is inconsistent with the architecture of SFL theory — as well as being internally inconsistent, as exemplified by the modelling of syntagmatic structure as an instance of realisation rules:

In terms of his own model, Halliday (1994: xv) writes:

What is presented here, however, is not the systemic portion of a description of English, with the grammar represented as networks of choices, but the structural portion in which we show how the options are realised.

[2] This is misleading. The emphasis in "Systemic theory" (Halliday 1993) is on outlining the architecture of the theory in summarised form for an encyclopædia article.

[3] This is misleading, because, as Fawcett knows, Halliday locates realisation statements in system networks, not as Fawcett's 'form potential'.