Fawcett's Argument That Rheme, Residue And Given "Do Not Contribute In Any Way To The Functional Structure Of The Clause"

Fawcett (2010: 135-6):

Let me summarise the stages of the argument that I have presented up to this point. We have established that the theoretical-generative model of a systemic functional grammar is always one of 'element conflation' and never one of 'structure conflation'. In other words, the 'mapping' between lines of analysis is performed on individual pairs and trios of 'functions' or 'elements', which from then on function as a single element of the clause. I would therefore say that those systemic functional grammars which suggest that the representation of a clause should include 'functions' such as 'Residue', 'Rheme' and 'Given' are in fact introducing what we might term 'pseudo-elements', i.e., chunks of text that do not contribute in any way to the functional structure of the clause. Their chief effect is to give the misleading impression that there is a 'structure' of roughly the length of a clause that corresponds to each of the strands of meaning. In other words, the only type of conflation that occurs in such a grammar — as it generates its single, 'flat-tree' structures — is the conflation of single, coterminous elements.

Blogger Comments:

[1] This is misleading. The fact that conflation involves only elements, and not structures, is not established by Fawcett's argument, because it is a principle of SFL Theory. The notion of 'structure conflation' is merely Fawcett's misunderstanding of SFL Theory, as demonstrated in previous posts.

[2] This is a serious misunderstanding of the conflation of functions. The conflation of functions, such as Theme, Subject and Actor does not entail that all three "function as a single element of the clause". The conflation of functions just means that all three are realised by the same nominal group, as happens in the case of such selections as [unmarked THEME, operative VOICE, declarative MOOD, material PROCESS TYPE]. Each of the three elements is a distinct function, each defined by its relations to other elements in its particular structure.

[3] To be clear, this is an invalid conclusion argued from a false premiss. The false premiss is that conflated elements "function as a single element of the clause"; see [2]. The argument is invalid, because, regardless of the truth of the premisses, the conclusion is not logically entailed by them. That is, the fact that one element is not conflated with another does not mean that such elements serve no function in their structural configuration. Function of an element in one metafunctional structure is not dependent on its being conflated with another element in a different metafunctional structure. As previously noted, a structure is the relation between elements, such as:

• the relation between Theme and Rheme in clause structure;
• the relation between Mood and Residue in clause structure;
• the relation between New and Given in information unit structure.

[4] This continues Fawcett's confusion between function structures and formal constituency. The 'flatness' here refers to formal constituency as modelled by the rank scale: ranked constituents (minimal bracketing), rather than immediate constituents (maximal bracketing). As Halliday (1995 [1993]: 273) makes clear:

'Rank' is constituency based on function, and hence 'flat,' with minimal layering;

It will be seen later that, while Fawcett argues for what he calls a 'flat tree structure', he simultaneously argues against the inclusion of a rank scale in SFL Theory.