If we look at his examples of Nigel’s protolanguage, each microfunctional utterance is a simple sign, initiating or responding, as David K quotes below. The mother-tongue examples are then multi-move texts. The Phase II breakthrough of recombining tones with ‘words’ enables multi-move exchanges of knowledge as well as action. It is a simultaneous breakthrough into the metafunctions of grammar and discourse semantics. This is not foregrounded in MAKH’s discussion, as his systems focus on cataloguing nascent grammatical features. But it is apparent in the examples. The disassociation and recombination of content and expression planes enables the bifurcation of the content plane into grammar and discourse. It also enables the bifurcation of the context plane into register and genre.
Blogger Comments:
[1] To be clear, here Rose is presenting his argument that the semantics that emerges in the ontogenetic move into tri-stratal language is the discourse semantics of Martin, not the semantics of Halliday. The argument rests on the unstated assumption that initiating and responding moves in multi-move exchanges is consistent with Martin's discourse semantics, but not with Halliday's semantics. This, of course, is untrue. Halliday models this through the semantic system of SPEECH FUNCTION, and Martin's discourse semantic system of NEGOTIATION is merely Halliday's system misunderstood and rebranded as Martin's model. The evidence for this assessment is available here (English Text) and here (Working With Discourse).
[2] This is misleading. Kellogg's quote from Halliday (1974) says nothing whatsoever about initiating and responding with "microfunctional signs" in protolanguage. It concerned with the transition stage (II) between protolanguage (I) and language (III):
"There is, with Nigel, a discontinuity in the expression, as well as, of course, the discontinuity that arises from the introduction of a third level of coding into the system. But there is no discontinuity in the content. The social functions that have determined the protolanguage--satisfying immediate needs, controlling people's behaviour, being 'together', expressing the uniqueness of the self, exploring the world of the non-self and creating a world in the imagination — all these evolve gradually and naturally into the social contexts and situation types that we characterise as semiotic structures; and the semantic system, the meaning potential that derives from these functions, evolve likewise. The progressive approximation of the child's meanings to those of the adult through interaction with and reinforcement by older speakers, begins before these meanings are (necessarily) realised through the words and structures of the adult language, and continues without interruption. Without this continuity, the semantic system could not function effectively in the transmission of the social system from the adult to the child."("A sociosemiotic perspective on language development", in Collected Works of MAKH Vol. 4, p. 109)
[3] This is misleading. It is not true that Halliday failed to detect discourse semantics because he was focused on the grammar. As his publications on ontogenesis confirm, he was concerned with both lexicogrammatical form and its function of realising meaning — and necessarily so, since he claims that 'the two originate as one'. Headings in the article cited by Kellogg, A Sociosemiotic Perspective On Language Development (Halliday 1974), include:
2 A Functional Semantics3 Meaning And The Environment5 Lexicogrammatical And Semantic Structures
Moreover, with the exception of NEGOTIATION (rebranded SPEECH FUNCTION), the discourse semantic systems in Martin (1992) are not semantic, but rebrandings of the lexicogrammatical systems of REFERENCE (confused with DEIXIS), LEXICAL COHESION (confused with clause nuclearity) and COHESIVE CONJUNCTION (confused with CLAUSE COMPLEXING). The evidence for this assessment is available here.
[4] To be clear, on the one hand, this a bare assertion, unsupported by argument, and on the other hand, it misrepresents Halliday's research. Halliday (2004 [1993]: 336-7) identifies the dissociation of associated variables as one of three semogenic strategies that become available after the content plane has become stratified into semantics and lexicogrammar. And it is semantics, not discourse semantics that emerges in the "bifurcation of content", for reasons that include those provided in [1] and [3] above.
[5] This is a bare assertion, unsupported by argument, a mere statement of belief. Moreover, Martin's model of context is invalidated by multiple self-contradictions. For example, Martin's model of stratification asserts that functional varieties of language, register and genre, are context not language. This is analogous to claiming that functional varieties of cattle, beef vs dairy, say, are not cattle.
And, as previously explained, Martin's model mistakes the content plane of a connotative semiotic for the entire connotative semiotic, and fills this connotative with varieties of a denotative semiotic. That's three interlocking inconsistencies. Or, most abstractly, Martin's model mistakes a token–type relation (instantiation) for a value–token relation (realisation).
Adding to these multiple levels of confusion and inconsistency, Martin's model also misconstrues the instantiation relation, at the level of context, between culture and situation as a realisation relation between genre and register. Martin (1992: 495):
The tension between these two perspectives will be resolved in this chapter by including in the interpretation of context two communication planes, genre (context of culture) and register (context of situation), with register functioning as the expression form of genre, at the same time as language functions as the expression form of register.
