My apologies, I missed this example, "A happy baby makes a happy mother"
This can only be identifying, as it is simple present (so not material) and reversible (so not attributive).
It seems odd because of the ‘cause-in-the-clause’ logical metaphor (‘if a baby is happy, then a mother is happy’). The identifying grammatical meaning is in tension with a consequential discourse semantic meaning.
A fuller account would ask the function of the metaphor in the discourse sequence. The consequential meaning is a link in a chain of reasoning. The identifying meaning is non-negotiable. It reads like a wise old proverb.
The sequence as a whole realises the ‘grounds’ for a proposal at register level...Grounds...Your child will thrive on Glaxowhen all else failsbecause it is a natural substitute for breast milk(now we all know) A happy baby makes a happy mother(so) Your happy Glaxo baby will make you a happy motherConclusion...(You should buy Glaxo)(Another point for your readers – realisation between strata)
Blogger Comments:
[1] This is misleading. To be clear, Rose didn't "miss" the example, he misanalysed it as an assigned attributive clause, as his own words make plain:
your second examples are caused attributive relations, in which Attributor is additional Agent from ergative perspective
[2] This misunderstands the grammatical metaphor. This instance is not a contrast between 'identifying grammatical meaning' and 'consequential semantic meaning', not least because both the identifying relation and the cause-conditional relation feature on both strata.
Grammatically the cause-conditional relation is construed metaphorically as a Process of an identifying clause, instead of congruently as a relator of clauses in a clause complex.
Semantically, the clause both congruently realises an identifying figure of being and incongruently realises a sequence of figures in a cause-conditional relation. See Halliday & Matthiessen (1999: 243, 244, 272, 278, 283) on ideational metaphor as a 'junctional' construct.
[3] To be clear, Rose has here invented an imaginary text. The original query (here) presented three separate instances, with no suggestion that they formed a continuous text:
Your child will thrive on Glaxo when all else fails because it is a natural substitute for breast milkA happy baby makes a happy motherYour happy Glaxo baby will make you a happy mother
[4] This is misleading, because it is not true. The presence of a Mood element in this declarative clause opens it up for negotiation:
A happy baby makes a happy mother— Yes, it does.— No, it doesn't.
[5] To be clear, as explained elsewhere, Rose follows Martin in misunderstanding register as a stratum of context rather than a sub-potential of language. In Martin's model, functional varieties of language are not considered language, which is analogous to functional varieties of dog — cattle dog, sheep dog — not being regarded as dogs.
