Martin (1992: 129):
As with SPEECH FUNCTION and MOOD in Chapter 2, IDENTIFICATION has been stratified with respect to nominal group structure. This raises a number of issues concerning the way in which descriptive responsibility is distributed across the two strata and how interaction between the strata should be interpreted.
 On the one hand, this is misleading. Martin here gives the false impression that the stratal distinction between the systems of SPEECH FUNCTION and MOOD is his, rather than Halliday's. On the other hand, this involves theoretical inconsistencies. To be clear, while the relation between SPEECH FUNCTION and MOOD is an interstratal relation between systems, the relation between IDENTIFICATION and nominal group structure is an interstratal relation between system and structure; the inconsistency is one of axis.
More importantly, the relation between IDENTIFICATION and nominal group structure is based on theoretical misunderstandings. These include
- the confusion of interpersonal deixis of the nominal group with textual reference,
- the confusion of referents ('participants') with reference items,
- the confusion of ideational denotation with textual reference,
- the omission of the adverbial group as a domain of reference items.
 The relation between strata is precisely defined by SFL theory as realisation. Realisation is an intensive (elaborating) identifying relation in which the Token (lower) and Value (higher) are levels of symbolic abstraction.
The 'interaction'* between strata is thus an identifying relation of elaboration.
The distribution of 'descriptive responsibility' across strata is defined by the stratal hierarchy itself:
- semantics is concerned with the meaning (Value) that is realised by wording (Token);
- lexicogrammar is concerned with the wording (Token) that realises meaning (Value).
*Later in this critique, it will be seen that use of the word 'interaction' betrays Martin's misunderstanding of the stratal and metafunctional dimensions of SFL theory as modules — e.g. here — a misunderstanding that is a source of serious theoretical errors.