Confusing Structure With Instantiation

Martin & Rose (2007: 199):

As analysts, we tend to treat texts as objects, and reify the structure that in fact unfolds as spoken or written discourse is produced. So it is important to keep in mind that the periodicity we are discussing here is an unfolding process, not a rigid structure linking parts to wholes.

Blogger Comments:

[1] To be clear, this follows Martin (1992) in confusing structure with instantiation; evidence here. In SFL Theory, structure is the relation between elements on the syntagmatic axis, whereas the unfolding of discourse is the logogenesis of text through the instantiation of potential. But see [2] below.

[2] To be clear, as far as (lowest level) Theme and New is concerned, this is accidentally true. This is because, unknown to Martin & Rose, the process of selecting Themes and News forms not structures, but patterns of instantiation in the logogenesis of text.

[3] To be clear, Martin & Rose have not presented any structures; they have merely presented single elements — (macro- & micro-) Theme and New — as if they were structures. There has been no account of the complementary elements — (macro- & micro-) Rheme and Given — with which they would form structures.

[4] To be clear, this confuses constituency (part-whole relations) with structure (part-part relations). In SFL Theory, constituency is modelled as a rank scale: a clause consists of groups ± phrases, which consist of words, which consist of morphemes, whereas structure is modelled as the relations between functional elements at each of these ranks: Theme to Rheme, Senser to Process to Phenomenon, etc.