It would be interesting to know why he stuck with the same stratal model, of semantics without discourse, and context without systems, despite JRM’s 1992 massive empirical description of these strata’s systems.Particularly so, as he and RH made it abundantly clear in Cohesion in English in 1976, that the ‘cohesive’ resources they describe are discourse semantic systems, e.g...‘The concept of cohesion is set up to account for relations in discourse... what is in question is the set of meaning relations which function in this way: the semantic resources which are drawn on for the purpose of creating text.’They also made explicit the stratal relation between these systems and LG...‘The means of expressing these various types of cohesion are, as we have seen, drawn from a number of areas of the lexicogrammatical system, which have in common merely the fact that they contribute to the realisation of cohesion.’They had begun to flesh out the semantic stratum as discourse semantic systems, as he had done for LG and PH. But Cohesion in English was written as a basic introduction to the resources, without describing the systems. The imbalance in research and theory between LG and DS ‘cohesion’ systems is stark in Table 1 below. However CinE then became the point of departure and constant reference for JRM’s metafunctional description of the systems in English Text.
Blogger Comments:
[1] To be clear, the short answer as to why Halliday "stuck with his own stratal model", rather than adopt Martin's, is that Halliday understood his own model, whereas Martin never has (evidence here).
[2] These claims are misleading, because they are untrue. Firstly, Halliday's semantics is not "without discourse"; like all strata, it is a means of theorising and analysing discourse. For Halliday (2008: 78):
“discourse” is text that is being viewed in its sociocultural context, while
“text” is discourse that is being viewed as a process of language.
Secondly, Halliday's context is not "without systems"; like all strata, it is modelled as a cline from system to instance.
[3] This is very misleading. On the one hand, Martin's "massive empirical description" involved confusing context with functional varieties of language: register and genre (in the sense of 'text type' rather than 'mode'). And on the other hand, there are no genre systems in Martin (1992). See Martin's Reasons For Not Devising Genre Systems.
[4] This is very misleading. On the one hand, 'discourse' here refers to the data to be modelled, not to a discourse semantic stratum. On the other hand, the systems that realise cohesion are located on the lexicogrammatical stratum. Halliday & Hasan (1976: 6):
Cohesion is a semantic relation. But, like all components of the semantic system, it is realised through the lexicogrammatical system; and it is at this point that the distinction can be drawn. Some forms of cohesion are realised through the grammar, and others through the vocabulary. … We can refer therefore to grammatical cohesion and lexical cohesion.
This is also made plain in the second quote that Rose presents above, though he misunderstands the stratal location of the systems.
[5] This is very misleading indeed. The function of Table 1 is to locate cohesion in the overall model: within the textual metafunction on the lexicogrammatical stratum. The fact that Rose has removed the title of the table which makes this plain, suggests that this is a deliberate act of (self-)deception. Halliday & Hasan (1976: 29):
