Misrepresenting Halliday On Structural Realisation

Fawcett (2010: 129-30):

Surprisingly, perhaps, Halliday's own published fragments of generative grammars all take the same position as that to [be?] described now, as we consider the procedure for generating the element we in We would visit Mrs Skinner every Sunday (i.e., the clause analysed in the IFG-style in Figure 7). Drawing on Halliday (1969/81:142) and 1970/76b), we can say that Halliday would generate this composite element in the following manner: 

1. At some early stage, the 'function' of Theme' is inserted in the structure. 2. At a later stage of generation the 'function' of 'Subject' is inserted, and either then or by a later rule it is conflated with the 'Theme'. 3. Later still, a Participant Role is inserted, e.g., 'Actor,' and this is conflated with the 'Theme/Subject' to create a composite element 'Theme/Subject/Actor'. 

This is also essentially what happens in Matthiessen and Bateman's computer implementation of Halliday's Nigel grammar in Penman, according to their description of the process in Matthiessen & Bateman (1991:88-109). 

Blogger Comments:

[1] This is misleading, because it misrepresents a computer implementation of Halliday's theory, for the purposes of text generation, as his theory of human language.

[2] To be clear, technically, we is not an element of structure, but the data (language) to be accounted for by theory (SF grammatics). Elements of clause structure include Theme, Subject and Actor.

[3] This is misleading, because it misrepresents the realisation relation between paradigmatic system and syntagmatic structure as ordered in time. The relation of realisation is one of intensive identity, not circumstantial: temporal, identity.

[4] This is potentially misleading. The conflation is not a "composite element", but rather three distinct elements, each from a distinct structural configuration, that are integrated in the nominal group that realises all three.