On Praising English Text

Bateman (1998: 25): 

Both differences reinforce the central role that discourse plays in the construction of a comprehensive semantics that can fully motivate functional differences found in the lexicogrammar. Indeed, any brand of functional linguistics considers it self-evident that the deployment of grammatical and lexical phenomena in texts (written or spoken) is to be explained by considering the communicative and/or social purposes involved. Less self-evident is the fact that it is not a priori clear how particular groupings of grammatical and lexical phenomena are to be isolated for purposes of explanation: Until the relevant ‘comparison sets’ are known, the functional alternations operating can scarcely be recognised, let alone analysed or explained. English Text therefore provides a detailed characterisation of those comparison sets that supports a richly integrating view of the discourse functions involved.

Blogger Comments:

[1] To be clear, in SFL theory, it is the grammar that plays the central rôle in "the construction of a comprehensive semantics".  As Halliday & Matthiessen (2014: 22) put it:

Grammar is the central processing unit of language, the powerhouse where meanings are created …

'Discourse', on the other hand, refers to 'the patterned forms of wording that constitute meaningful semiotic contexts' (Halliday & Matthiessen (1999: 512), and it is the textual metafunction that creates discourse.  Halliday & Matthiessen (1999: 528): 

The “textual” metafunction is the name we give to the systematic resources a language must have for creating discourse: for ensuring that each instance of text makes contact with its environment. The “environment” includes both the context of situation and other instances of text.

In reinterpreting Halliday & Hasan's (1976) textual systems of cohesion as semantic systems, Martin has confused a metafunction (textual) with a level of symbolic abstraction (semantics).

[2] In terms of SFL theory, this has the relation backwards.  It is not the semantics that motivates "functional differences found in lexicogrammar" but the lexicogrammar that creates the functional differences at the level of semantics.

Moreover, Halliday's (1984) paper On The Ineffability Of Grammatical Categories argues that semantics can only be validly theorised by encoding semantic values by reference to grammatical tokens, at the same time explaining why the grammar cannot be validly theorised by decoding grammatical tokens by reference to semantic values.

[3] To be clear, on the one hand, it is the architecture of SFL grammatics that provides an explanation of grammatical and lexical phenomena.  On the other hand, explanation in SFL does not involve isolating grammatical and lexical phenomena, but determining the relations between them.  SFL is a relational theory, not a modular one.

[4] To be clear, this obfuscatory reference to experimental procedure ('comparison sets') only reinforces the fact that Bateman does not understand the hierarchy of stratification in SFL theory, wherein semantics and grammar are different levels of symbolic abstraction, which only disagree in the case of grammatical metaphor.

[5] To be clear, the use of therefore here is misleading, since this conclusion is not entailed by the propositions that precede it.  Moreover, as the clarifying critiques on this blog and the blog reviewing English Text demonstrate, this is manifestly untrue.  Martin has merely taken Halliday & Hasan's (1976) grammar of textual cohesion, misunderstood it, mixed it up with other sources, also misunderstood, and rebranded the confusion as his model of discourse semantics.