Thanks John But it's not just Gordon’s phrasing… it’s normal SF talk, as in your own “the Medium isn't implicit: it's the empire;” which, forgive me, seems to imply the role and lexical item are identical. Is it possible our shorthand masks some unresolved slippage? Hence my question about the relation. ‘Filled’ is too syntax and vocab isn’t it? And how can a grammatical function structure be ‘realized’ by a lexical item? Where in the theory’s architecture is that possible? The problem is two distinct construals of field that come together in an instance like 'the hurricane struck the coastal area of Haiti’… lexical and grammatical
Blogger Comments: To say the Medium is the empire does not imply that 'the role and lexical item are identical' — not least because the empire is not a lexical item, but a nominal group. In construing an element of clause structure (the Medium) and a nominal group (the empire) as higher (Value) and lower (Token) levels of abstraction, the clause is consistent with SFL theory:
But this misses the point. The clause construes an identifying relation between linguistic theory and language data, such that a theoretical value (the Medium) is encoded by reference to a token of language (the empire).
 The notion of a grammatical structure being realised by a lexical item is nonsensical. A lexical item is the synthetic realisation of the most delicate features of the lexicogrammatical system, whereas a grammatical structure is constituted by the relations between its functional elements. Halliday & Matthiessen (2014: 83-4):
The significance of any functional label lies in its relationship to the other functions with which it is structurally associated. It is the structure as a whole, the total configuration of functions, that construes, or realises, the meaning. … It is the relation among all these [functions] that constitutes the structure.