Martin (1992: 156-7):
Finally, the major limitation on the account of participant identification in this chapter needs to be acknowledged — that is, it's lack of attention to the process of identifying participants. This is reflected in the concentration on textual relations at the expense of their interaction with experiential ones and the lack of a procedural orientation to determining what would be appropriate to identifying a participant at a particular place in the unfolding of a text and how exactly interlocutors might go about retrieving presumed information. The account thus falls far short of the specificity required for computer implementation and text generation.
 To be clear, the major limitation of this chapter is that it is not what it claims to be: reference as semantic choice. This is because it is founded on several serious theoretical misunderstandings. For example, Martin confuses:
- textual reference (identifiability) with ideational denotation (participant identification) — for the most part in the sense of Frege's reference to meanings transcendent of language (bedeutung);
- textual reference (cohesive) with interpersonal deixis of the nominal group (structural);
- reference items with nominal groups;
- reference with lexical cohesion;
- reference with ellipsis–&–substitution.
 Here Martin diverts the reader from personally identifying the actual problems with his theorising, and identifies the model's major limitation as issues of no importance — computer implementation and text generation — to his intended readership.
 This misunderstands SFL theory. The process of identifying participants (or indeed the process of referring to referents) is the process of instantiation: the selection of features in networks and the activation of realisation statements.
 This is misleading. Martin's model is a confusion of textual relations (reference) and experiential relations (lexical cohesion) and denotation (participant identification).
 The notion of "interaction" between textual and experiential relations betrays Martin's misunderstanding (1992: 390) of metafunctions as modules, instead of dimensions.
 On the one hand, the function of the grammatical metaphor and lexical density of this single nominal group — 11 lexical items — is to reduce its intelligibility to learners and outsiders (Halliday & Matthiessen 1999: 272). On the other hand, it is untrue, since Martin has demonstrated his procedure for identifying participants in text, in his own terms, throughout the chapter.
 This misunderstands reference. Reference presents items as identifiable; it does not inform the listener how to recover the identity. Halliday & Matthiessen (2014: 623):
The textual status at issue in the system of reference is that of identifiability: does the speaker judge that a given element can be recovered or identified by the listener at the relevant point in the discourse or not? If it is presented as identifiable, then the listener will have to recover the identity from somewhere else.