Misrepresenting Halliday & Hasan On Reference

Martin (1992: 145):

Except for structural cataphora (esphora), the chains are neutral with respect to grammatical structure; semantic dependency is noted between between items irrespective of clause or sentence boundaries.  The analysis of reference contrasts in this respect with that exemplified in Halliday and Hasan (1976: 340-55), which looks at cohesive ties between, not within, sentences. Here , as far as participant identification is concerned, the discourse structure of [3:85], [3:86] and [3:87] is treated as identical.

[3:85]

The boy was tired.
However, he kept looking.

[3:86]

Although the boy was tired,
he kept looking.

[3:87]

The boy kept looking in spite of his fatigue.

Blogger Comments:

[1] This is doubly misleading; see two earlier misleading statements in this regard (p19) here and here.

Firstly, in SFL theory, reference is, of course, "neutral with respect to grammatical structure" because, as a system of cohesion, it is not a structural relation.  Halliday & Hasan (1976: 8, 14):

Cohesive relations have in principle nothing to do with sentence boundaries. … Cohesion as we have said is not a structural relation; hence it is unrestricted by sentence boundaries …

Secondly, the reason why the text analyses in Halliday & Hasan (1976: 340-55) focus on referential relations between sentences was clarified by them (1976: 9), as already quoted by Martin (p19):

it is the intersentence cohesion that is significant, because that represents the variable aspect of cohesion, distinguishing one text from another.

The two sentences that immediately follow this excerpt in Halliday & Hasan (ibid.) — but which Martin omitted — are:

But this should not obscure the fact that cohesion is not, strictly speaking, a relation 'above the sentence'.  It is a relation to which the sentence, or any other form of grammatical structure, is simply irrelevant.

The reader can decide whether Martin's misrepresentation of Halliday & Hasan is merely accidental, or, rather, deliberate and strategic.

[2] To be clear, nowadays in SFL, the sentence is construed as a graphological unit rather than a lexicogrammatical one.

[3] To be clear, in the source of Martin's ideas, Halliday & Hasan (1976), all three texts display instances of anaphoric personal co-reference, contrary to the false implication of Martin.

[4] As previously demonstrated for another example, this analysis is inconsistent with Martin's model of participant identification as a relation between nominal groups.  On Martin's model, the relation obtains between the boy and his fatigue, which demonstrates the absurdity that results from mistaking nominal groups for reference items.