Martin (1992: 536):
The semiotic system of field was introduced in Chapter 5 by way of contextualising the discussion of IDEATION. Field was introduced there in terms of sets of activity sequences oriented to some global institutional purpose (or more informally, field provides the semiotic interpretation of what counts as an answer to the question //1 What do you do// as put to strangers).
 As demonstrated in several previous posts, the discussion (p292, pp321-5) of activity sequences in Chapter 5 confuses the contextual category of field with the language that realises it.
 This is misleading. The answer to the question What do you do? is language, not field (context). As a dimension of context, field is not language. Language realises field; language and field are different levels of symbolic abstraction, with language as Token and field as Value. This stratal identity relation encodes field by reference to language; and decodes language by reference to field.
The confusion arises, most generally, from Martin mistaking semogenesis (all strata make meaning) for stratification (context/meaning/wording/sounding), and because of this, treating all strata as levels of (linguistic) meaning. This misunderstanding then leads, in the present chapter, to misconstruing context as types of language (register and genre). Martin's model of register and genre, as contextual strata, is thus based on sequenced theoretical misunderstandings, and is inconsistent with both stratification and the notion of text type (register/genre).