Martin (1992: 419):
Hasan's definition of similarity and identity chains suggests a stratified approach to discourse semantics very like that proposed by English Text. Identity chains are based on co-referentiality, which is realised through pronominal cohesion, instantial equivalence, the definite article and demonstratives (or lexical repetition if the reference is generic); similarity chains are based on co-classification or co-extension, which are realised through substitution and ellipsis, lexical repetition and relations of synonymy, antonymy, hyponymy and meronymy. Discourse systems however, underlying these semantic chain structures, are not articulated.
Blogger Comments:
[1] This is manifestly untrue, on at least two counts. Firstly, Hasan does not take a 'stratified approach to discourse semantics'. Hasan's work is theoretically consistent with the SFL model in stratifying content into semantics (meaning) and lexicogrammar (wording). For example, as Hasan (1985: 71) makes clear:
The texture of a text is manifested by certain kinds of semantic relations between its individual messages. The nature of these relations and the lexico-grammatical patterns that realise them are discussed in the following sections.
Secondly, Hasan's work is also theoretically consistent with the SFL model in treating cohesive systems as resources of the textual metafunction. This is not the case with Martin's discourse semantic systems. The discourse semantic system of conjunction is presented as logical semantics, despite deriving from cohesive conjunction (textual and non-structural) mixed with clause complex relations (logical and structural). Similarly, the discourse semantic system of ideation is presented as experiential semantics, despite being derived from lexical cohesion (textual and non-structural) mixed up with logico-semantic relations within the clause and lexis as most delicate grammar.
[2] Any similarity between Hasan's approach and Martin's arises from Hasan's (1976, 1980, 1984, 1985) being the original work, and Martin's (1992) being merely a derived reinterpretation.
[3] Hasan does not provide discourse systems because her work is theoretically consistent with the SFL model in stratifying content into semantics (meaning) and lexicogrammar (wording), and in treating cohesive systems as resources of the textual metafunction. To be clear, in Hasan's model (1985: 73, 82, 84), the nature of cohesive ties is semantic, and the types of tie relation involves either co-reference, co-classification or co-extension. These semantic relations are realised by grammatical cohesive devices and lexical cohesive devices.
[4] Hasan's cohesive chains are not structures. Hasan's work is theoretically consistent with the SFL model in treating cohesive systems as non-structural resources of the textual metafunction. To be clear, Hasan (1985: 84):
[…] a chain is formed by a set of items each of which is related to the others by the semantic relation of co-reference, co-classification and/or co-extension. Taking the type of relation into account, we can sub-categorise them into two types: IDENTITY CHAINS and SIMILARITY CHAINS. […] The relation between members of an identity chain is co-reference […] the members of a similarity chain are related to each other either by co-classification or co-extension.
