It’s also now 3 decades since JRM showed how grammar/semantics relations vary between systems that serve to organise discourse metafunctionally, and how congruent/incongruent contrasts display ‘stratal tension' between grammatical and discourse semantic functions. In this light, MAKH’s 1975 Hjelmslevian metaphor of 'splitting the content plane’ needs revising, since his evidence actually shows that grammatical metafunctions emerge with exchanges and figure sequences, i.e. discourse semantic systems.
Blogger Comments:
[1] This is misleading, because it is untrue on several counts. Firstly, Martin (1992) takes the relation between grammar and his discourse semantics to be invariably one of realisation: 'the realisation relationship between discourse semantics and lexicogrammar' (p57), and does not propose different relations, varying according to metafunction (but see the note on 'interaction patterns' below). Secondly, Martin does not understand the notion of realisation (or instantiation), as shown by the following quote (p5), where he misunderstands it as a relation between system and process (i.e. in SFL terms, between system and the process of instantiation):
As noted above, system is related to process through the concept of realisation — realisation formalises the instantiation of system in process.
Thirdly, despite using the term 'realisation' for the relation between strata, Martin does not understand strata as different levels of symbolic abstraction. Instead, Martin mistakes all strata as one level of symbolic abstraction, linguistic meaning, and misunderstands strata as interacting modules, and proposes different interaction patterns between grammar and discourse semantics: cohesive harmony, modal responsibility, method of development and point (p393). None of these are modelled as interaction patterns by the (misunderstood) intellectual sources of these ideas: Hasan (cohesive harmony), Halliday (modal responsibility) and Fries (method of development and point). Moreover, in later work, Martin & Rose (2007), cohesive harmony and modal responsibility are absent, and method of development and point are reconstrued as the textual discourse semantic system of periodicity (critiqued here), thereby dismissing the original proposals.
[2] This is misleading, because Martin's discourse semantic systems are his rebrandings of Halliday's interpersonal semantic system, speech function, and Halliday & Hasan's textual lexicogrammatical systems, cohesion, as previously explained. In Martin's model, the textual system of conjunction is rebranded as a logical system (conjunction/connexion), and the textual system of lexical cohesion is rebranded as an experiential system. That is, in terms of SFL Theory, Martin's discourse semantic systems are limited to two metafunctions, interpersonal and textual, and only one system is semantic.
[2] This is misleading, because Martin's discourse semantic systems are his rebrandings of Halliday's interpersonal semantic system, speech function, and Halliday & Hasan's textual lexicogrammatical systems, cohesion, as previously explained. In Martin's model, the textual system of conjunction is rebranded as a logical system (conjunction/connexion), and the textual system of lexical cohesion is rebranded as an experiential system. That is, in terms of SFL Theory, Martin's discourse semantic systems are limited to two metafunctions, interpersonal and textual, and only one system is semantic.
[3] To be clear, Martin's notion of 'stratal tension', which does not appear in Martin (1992), is simply a rebranding of Halliday's notion of an incongruent (metaphorical) relation between semantics and grammar. But more importantly, on Martin's model, there is stratal tension regardless of whether the grammatical realisation is metaphorical. For example, Martin's logical system, conjunction/connexion, is not organised according the three general types of expansion: elaboration, extension and enhancement, and the logico-semantic relation of projection is entirely absent. The reason for this is that Martin's source material, Cohesion In English (Halliday & Hasan 1976) was not organised on the three types of expansion, and being a model of cohesion, did not include projection.
[4] To be clear, this is 'metaphor' in the sense of theoretical model. Hjelmslev (1943) conceived of a content plane which distinguished content substance and content form. Halliday (1975), in describing the shift from protolanguage to language, relates meaning to content substance and wording to content form. Halliday (2004 [1975]: 55):
In Hjelmslevian terms, the functional basis of language has shifted from the “content substance” (in a system having no level of form) to the “content form”.
[5] This is misleading, because, to the extent that it is coherent, it is not true.
Firstly, even if it were true that Martin (1992) did provide evidence "that grammatical metafunctions emerge with exchanges and figure sequences, i.e. discourse semantic systems", it does not follow from this that the stratification of content into meaning and wording would need revising. The theoretical value of stratified content derives from the fact that it provides a systematic means of explaining grammatical metaphor.
Secondly, the notion that "grammatical metafunctions emerge" misunderstands SFL Theory. For Halliday, the metafunctions are highly generalised meanings that are used to interpret lexicogrammatical form. As Halliday (1985 & 1994: xvii) explains:
Firstly, even if it were true that Martin (1992) did provide evidence "that grammatical metafunctions emerge with exchanges and figure sequences, i.e. discourse semantic systems", it does not follow from this that the stratification of content into meaning and wording would need revising. The theoretical value of stratified content derives from the fact that it provides a systematic means of explaining grammatical metaphor.
Secondly, the notion that "grammatical metafunctions emerge" misunderstands SFL Theory. For Halliday, the metafunctions are highly generalised meanings that are used to interpret lexicogrammatical form. As Halliday (1985 & 1994: xvii) explains:
the form of the grammar relates naturally to the meanings that are being encoded. A functional grammar is designed to bring this out; it is a study of wording, but one that interprets the wording by reference to what it means.Thirdly, 'figure sequences' do not constitute evidence on this matter because they do not appear in Martin (1992). This is because 'figure' and 'sequence' feature in the ideational semantics of Halliday & Matthiessen (1999), which wasn't published until seven years after Martin's publication.
