Martin (1992: 433):
As a final caveat the reasons for developing a measure of textual coherence need to be cautiously examined. One might argue from the perspective of formal linguistics that just as the goal of a grammarian is to account for speakers [sic] intuitions about which sentences are grammatical, so the goal of the textlinguist is to account for intuitions about which texts are coherent. But just as a functional grammarian would reject this delimitation of the goals of grammatical theory, so a functional discourse analyst might view accounting for a speaker's intuitions about coherent text with some suspicion. Alternatively, one might pose questions about the effectiveness of a text in achieving it's [sic] interlocutors' social purpose. This rhetorical perspective is likely to prove more appropriate as far as functional linguistics is concerned. Or one might ask questions about a text's role in sustaining culture as a dynamic open system. These points will be taken up again in Chapter 7 below.
 The reason for raising formal linguistics here is to present Hasan's work as proceeding from Formalist (Cartesian, Chomskyan) principles and/or assumptions. Martin introduced the word intuition in the previous paragraph and uses it three times in this paragraph in order to establish the collocation with Formal linguistics. It is misleading in a number of ways. To begin with, here is what Hasan (1985: 72, 86, 88) actually wrote:By way of clarification, Halliday & Matthiessen (1999: 415-6) characterise the two linguistic traditions in the following way:
The property of texture is related to the listener's perception of coherence. Thus in common parlance, Example 5.1 would be described as possessing coherence while example 5.2 would be seen as lacking in coherence. …
Whenever I have presented these two texts to informants, they have unanimously agreed that text 5.2 is less coherent than 5.1… . An explanation of what this judgement correlates with in patterns of texture is difficult to find, so long as grammatical and lexical cohesion are examined separately. …
…can the listener's perception of varying degrees of coherence between Texts 5.1 and 5.2 be correlated with differences in texture…?A cautious examination of Hasan's work reveals that the purpose of the work on texture is to determine the linguistic correlates of the listeners' perception of coherence. In Hasan's work, the data are instances of language (texts). This is very distinct from the concerns of Formal semantics. In Formal semantics, it is intuitions about language that constitute the data. As Martin Stokhof (2011 Intuitions And Competence In Formal Semantics) explains:
In formal semantics intuition plays a key role, in two ways. Intuitions about semantic properties of expressions are the primary data, and intuitions of the semanticists are the main access to these data. The implication here is that Hasan's work is not framed within the rhetorical perspective. This is also misleading. Here is what Hasan (1985: 96) actually wrote:
The infra-structure of all assumptions about co-operative acts of doing and saying is, in the last resort, social. The assumption of coherence can be sustained so well because human language has the resource for indicating coherence, while the nature of language as a resource has developed in a particular way because it has had to serve the needs of the community. Our task is to understand the specific nature of these resources — not simply to hide behind the mind and the intention of particular speakers and listeners.
We can identify two main traditions in Western thinking about meaning (see Halliday 1977):
(i) one oriented towards logic and philosophy, with language seen as a system of rules;(ii) one oriented towards rhetoric and ethnography, with language seen as resource.
… Our own work here falls mainly within the second tradition — but we have taken account of the first tradition, and the general intellectual environment in which versions of our meaning base are being used also derives primarily from the first tradition. Indeed the two traditions can in many respects be seen as complementary, as contributing different aspects to the overall picture. Our own foundation, however, is functional.
 Incidentally, Martin provides no reasoned argument for or against the two linguistic perspectives, he merely expresses negative attitude (reject, suspicion) towards the Formalist tradition, and positive attitude (appropriate) towards the rhetorical tradition.
 In SFL theory, this is modelled by realisation, instantiation and semogenesis. A text realising a context of situation (an instance of culture) unfolds at the instance pole of cline of instantiation during logogenesis, each text nudging the probabilities of its register and the overall linguistic potential, and each context of situation nudging the probabilities of the situation type and the overall culture as potential. And, as Halliday & Matthiessen (1999: 18) put it, logogenesis provides material for ontogenesis, which provides material for phylogenesis (with phylogenesis providing the environment for ontogenesis, and this for logogenesis).
This extract from Martin (1992) is intended as the third and final part of a critique of Hasan's cohesive harmony, as a prelude to introducing his own version. As can be seen from the clarifications above, it is merely a misunderstanding and misrepresentation of Hasan (1985).