Sunday 19 April 2015

Misconstruing Register As Context [New]

 Martin (1992: 28):

Beyond this, as Halliday and Hasan note, the concept of cohesion is not in itself sufficient to define a text. Context is also critical and it is for this reason that Chapter 7 has been included in this volume to contextualise the work on discourse semantics developed in Chapters 2 through 6. Needless to say only a brief sketch of a model of the contexts in which English is used is provided at that stage. Since they have so often been misunderstood in this regard, we will close this introductory chapter with the following quotation from Halliday and Hasan (for further discussion of this point see Chapters 6 and 7 below).
The concept of COHESION can therefore be usefully supplemented by that of REGISTER, since the two together effectively define a TEXT. A text is a passage of discourse which is coherent in these two regards: it is coherent with respect to the context of situation, and therefore consistent in register; and it is coherent with respect to itself, and therefore cohesive. Neither of these two conditions is sufficient without the other, nor does the one by necessity entail the other. (Halliday and Hasan 1976: 23)


Blogger Comments:

Amusingly, here Martin accuses others of misunderstanding Halliday & Hasan's notion of context, while introducing a clarifying quote from them on the subject which he himself misunderstands.

To be clear, Martin mistakenly equates context and register, presumably on the basis of:

it is coherent with respect to the context of situation, and therefore consistent in register

despite the fact that on the previous page, Halliday & Hasan (1976: 22) make a clear distinction between context (situational features) and register (linguistic features):

The linguistic features which are typically associated with a configuration of situational features — with particular values of field, mode and tenor — constitute a register.

This simple misunderstanding invalidates Marin's model of context. For Martin, varieties of language — registers and genres — are context, not language. This is equivalent to claiming that varieties of climate — tropical, temperate etc. — are not climates, or that varieties of dog — cattle dog, sheep dog etc. — are not dogs. 

Martin On Cohesion, Coherence, Cohesive Harmony And Texture [New]

Martin (1992: 27):
Before going on to explore these systems it perhaps needs to be stressed that like Cohesion in English, English Text does not equate cohesion with coherence. Hasan's methodology for measuring coherence, cohesive harmony, will be introduced in Chapter 6, alongside other analyses which focus on patterns of interaction among discourse semantics systems and across strata. It is only by the end of Chapter 6 then that something approximating a comprehensive analysis of texture will have been achieved.


Blogger Comments:

[1] To be clear, here Martin is merely stating that he agrees with the sources of his work, Halliday & Hasan (1976) and Hasan (1989/1985) who, unlike Martin, originated and developed the notions of cohesion and coherence.

[2] It will be seen in the examination of Chapter 6 that Martin misrepresents Hasan's model of cohesive harmony in order to replace it with his own model. See the posts here.

[3] To be clear, Martin's notion of interactions among his metafunctional systems and across strata derives from his misunderstanding of metafunctions and strata as modules (p390).

[4] To be clear, in SFL Theory, 'texture' is the property of being a text (Halliday & Hasan 1976: 2) and it is created through the resources of the textual metafunction: the systems of COHESION, THEME and INFORMATION (Halliday & Matthiessen 2014: 650).

Martin, on the other hand, misunderstands texture as arising from the interaction of different metafunctions across different strata; see here, or here.

Martin's Experiential Discourse Semantic System Of IDEATION [New]

Martin (1992: 27):
Finally IDEATION attends to a variety of experiential relations among "lexical" items — hyponymy, antonymy, synonymy, meronymy and so on (the semantics of collocation if you will). The major influence on this chapter is once again Halliday and Hasan.


Blogger Comments:

[1] To be clear, Martin's experiential discourse semantic system of IDEATION is far more than just "influenced" by Halliday & Hasan (1976). Martin's IDEATION, which he characterises (p271) as 'the company words keep', is Halliday & Hasan's system of lexical cohesion, misunderstood, relocated from the textual metafunction to the experiential, and from lexicogrammar to Martin's discourse semantics stratum, and rebranded as Martin's work. The metafunctional confusion is maintained by Martin's use of a textual sub-unit 'message part' — realised by a lexical item — as his experiential unit (p325).

[2] To be clear, here Martin gives early warning that he does not understand his intellectual source by misconstruing the paradigmatic lexical relations of hyponymy, antonymy, synonymy and meronymy as types of the syntagmatic relation of collocation. Halliday & Matthiessen (2014: 644) provide the following useful classification of the relevant lexical relations:

Martin's Logical Discourse Semantic System Of CONJUNCTION [New]

Martin (1992: 27):
CONJUNCTION focuses on logical meaning — on relations of addition, time, cause and comparison between messages, as these are variously realised through paratactic, hypotactic and cohesive conjunctions (or metaphorically within a clause; see Chapter 4 for details). Once again, this analysis is inspired by Gleason (1968) and by Halliday and Hasan (1976).


Blogger Comments:

To be clear, Martin's logical discourse semantic system of CONJUNCTION is far more than just "inspired" by Halliday & Hasan (1976). It is a confusion of Halliday & Hasan's cohesive CONJUNCTION and Halliday's CLAUSE COMPLEXING, misunderstood and relocated from lexicogrammar to Martin's discourse semantics. That is, Martin confuses expansion features serving the textual metafunction (cohesive CONJUNCTION) with expansion features serving the logical metafunction (CLAUSE COMPLEXING). This confusion is maintained by Martin's use of Halliday's textual unit 'message' as his logical unit (p325).

The reason why Martin uses the relations of addition, time, cause and comparison instead of expansion and its most general subtypes elaboration, extension and enhancement is that these were the categories used in Martin's source, Halliday & Hasan (1976).

Moreover, because the logico-semantic relation of projection does not function cohesively, Martin's model of logical semantics omits the system of PROJECTION. That is, there is no semantic system to be realised lexicogrammatically by projection relations between units in unit complexes or metaphorically, for example, in circumstantial relational clauses such as the lecture covered topics in evolutionary biology.

Martin's Textual Discourse Semantic System Of IDENTIFICATION [New]

Martin (1992: 27):
IDENTIFICATION is a textual system concerned with tracking participants in discourse. At issue here is the way in which people, places and things are introduced in text and potentially referred to again once introduced (e.g. a robot...the android below). This work is based on Gleason's analysis of discourse structure within a stratificational framework (Gleason 1968) and Halliday and Hasan's (1976) description of referential cohesion.


Blogger Comments:

[1] To be clear, Martin's IDENTIFICATION, which he characterises (p93) as 'reference as semantic choice' is Halliday & Hasan's cohesive reference, misunderstood, relocated from Halliday's lexicogrammar to Martin's discourse semantics, and rebranded as Martin's work.

The fundamental misunderstanding that invalidates the model is Martin's confusion of textual reference with 'reference' in the the sense of ideational denotation. It is this confusion that leads Martin to misconstrue an ideational unit, the participant, as his textual unit.

This also leads Martin to mistake nominal groups — that realise participants — for reference items, which then leads Martin to confuse reference with nominal group DEIXIS.

[2] Consistent with the metafunctional confusion noted above, Martin here provides an example of his experiential system, IDEATION, instead of his textual system, IDENTIFICATION.

Misrepresenting Halliday On Mood And Misunderstanding Ineffability And Stratification [New]

Martin (1992: 31-2):
To begin, consider two central MOOD systems, which classify English clauses as three basic types of interact (Fig. 2.1):

 

… Feature by feature, the appropriate structural realisations are:

 

Traditionally, these systems and structures have come to be viewed as formal ones — the basic syntax of the English clause. Halliday (1984a, 1985a) on the other hand suggests that these systems and structures require a richer interpretation, beyond form, in spite of the fact that the content of the terms is less than transparent (largely because of their antiquity; see Halliday 1984b/1988 on the ineffability of linguistic categories). 
Rather, the systems engender two fundamental oppositions: information as opposed to goods and services (indicative vs imperative) and giving (declarative) as opposed to demanding (interrogative). He accordingly proposes a semantic perspective on the grammatical labels along the following lines, fleshing out the paradigm with an additional category which is not grammaticalised in MOOD at the very general level of delicacy outlined above (the Offer), as in Table 2.1.

 

Blogger Comments:

[1] This is misleading. To be clear, unacknowledged by Martin, this MOOD system and its structural realisations were originally devised by Halliday. For the full system, see Halliday & Matthiessen (2014: 162).

[2] This is misleading in two respects. On the one hand, because these systems and structures were devised by Halliday, they have not come to be viewed traditionally. On the other hand, Halliday's systems and structures have not come to be viewed as formal ones, since they model interpersonal functions of the English clause.

[3] This is misleading. Martin's juxtaposition of the 'adversative' on the other hand with Halliday reinforces the false impression that the MOOD system and its structural realisations were not originally devised by Halliday.

[4] On the one hand, this is misleading, since the MOOD system and its structural realisations are already "beyond form"; see [2]. On the other hand, the semantic interpretation of MOOD is not a "richer interpretation", but an interpretation at a higher level of symbolic abstraction. This is a foretaste of Martin's inability to understand stratification as levels of symbolic abstraction, as will demonstrated throughout this blog.

[5] To be clear, this is ineffable twaddle. By 'the ineffability of grammatical categories', Halliday (2002 [1984]: 303, 306) means that

The meaning of a typical grammatical category … has no counterpart in our conscious representation of things. … they do not correspond to any consciously accessible categorisation of our experience.

[6] To be clear, it is not so much that MOOD systems "engender" the SPEECH FUNCTION systems of COMMODITY (information v goods-&-services) and INITIATING ROLE (giving v demanding), but that selections of features in SPEECH FUNCTION systems on the stratum of semantics are realised selections of features in MOOD systems on the stratum of lexicogrammar.

Martin's Interpersonal Discourse Semantic System Of NEGOTIATION [New]

Martin (1992: 26-7):
NEGOTIATION is an interpersonal system concerned with discourse as dialogue. Given an exchange such as that presented below, an account will be developed which shows how a sequence of speech acts which we might gloss informally as question, nomination, answer and validation are syntagmatically related to each other and systemically related to other types of exchange. This work takes as its point of departure the work on classroom discourse developed by the Birmingham school (Sinclair and Coulthard 1975).


Blogger Comments:

To be clear, NEGOTIATION is Martin's only genuine semantic system, the other systems being rebrandings of Halliday & Hasan's lexicogrammatical systems of COHESION. However, NEGOTIATION is only "Martin's" system in the sense that it is his rebranding of Halliday's system of SPEECH FUNCTION, incorporating work carried out by Halliday's colleagues.

By the same token, NEGOTIATION is also Martin's only system that is realised by genuine structures, the others being, instead, cohesive relations misconstrued as Lemke's covariate structures, which Lemke has since (1989) conceded are not, in fact, a type of structure. However, despite being interpersonal structures, Martin models them in terms of constituency — e.g. exchanges consisting of moves — which is the favoured structure type of the experiential metafunction.

Negotiation As The Semantics Of Mood [New]

Martin (1992: 31):
This chapter is concerned with the semantics of MOOD in English (the grammatical description of MOOD proposed in Halliday 1976b:105-110 and Halliday 1985a will be assumed). Its basic concern will be to interpret MOOD from a discourse perspective as a resource for negotiating meaning in dialogue.


Blogger Comments:

To be clear, in citing only Halliday's system of MOOD, Martin gives the misleading impression that Halliday has not already provided 'the semantics of MOOD' in his system of SPEECH FUNCTION; see, e.g., Halliday (1985: 68-71). Cf. Halliday & Matthiessen (2014: 136):

"The Purpose Of These Descriptions Is To Enhance, Not Supplant, Those Offered In Cohesion In English" [New]

Martin (1992: 26):
The purpose of these descriptions is to enhance, not supplant, those offered in Cohesion in English; English Text functions for the most part to recontextualise Halliday and Hasan's conception of cohesion from the perspective of discourse semantics. Their very rich descriptions of texture in English will not in general be recapitulated, but rather presumed. This is particularly true with respect to SUBSTITUTION and ELISIPIS [sic], which will barely be touched on at all (see Chapter 6 below for further discussion).


Blogger Comments:

[1] To be clear, Halliday & Hasan (1976) analysed data and came up wth the theory of lexicogrammatical cohesion and its subtypes. Martin (1992), on the other hand, does not take data and come up with a theory of discourse semantics. Instead, he takes the hard won theorising of Halliday & Hasan and merely relabels (his misunderstanding of) it:

  • lexicogrammatical cohesion is rebranded as discourse semantics,
  • cohesive reference is rebranded as identification,
  • cohesive conjunction is rebranded as conjunction (later 'connexion'), and
  • lexical cohesion is rebranded as ideation.
By doing so, Martin gains institutional credit for the original intellectual work of Halliday & Hasan.

[2] To be clear, the reason why Martin was not able to reinvent Halliday & Hasan's ellipsis–&–substitution as discourse semantics is that it sets up a relationship that is lexicogrammatical, not directly semantic; see Halliday (1985: 296-7).

Confusing Stratification With Instantiation: Register And Genre [Revised]

Martin (1992: 26):
Finally, in Chapter 7, a model of context will be built up by sketching in the connotative semiotics of register, genre and ideology.

Blogger Comments:

To be clear, Martin (p493) claims to be following Hjelmslev in his use of the notion of a connotative semiotic. However, for Hjelmslev (1943), a connotative semiotic is a semiotic system whose expression plane is a denotative semiotic system. Martin's model of context misunderstands Hjelmslev in two crucial respects.

Firstly, as will be seen, Martin misinterprets Hjelmslev's connotative semiotic as merely its content plane, excluding its expression plane, the denotative semiotic.

Secondly, as will be seen, Martin misinterprets functional varieties of a denotative semiotic, registers and genres of language, as connotative semiotics.

These basic theoretical inconsistencies invalidate Martin's model of stratified context.

In SFL Theory, register and genre (text type), as functional varieties of language, are modelled as language, not as context. This means that, as varieties, they are located at the midpoint on the cline of instantiation of language. Register is text type (genre), viewed from the system pole of the cline, whereas text type (genre) is register, viewed from the instance pole.

However, as will be seen, Martin's model of genre adds further to the theoretical inconsistencies by also encompassing the SFL notions of MODE (e.g. recount, anecdote etc.) and semantic structure (e.g. Orientation^Record^Reorientation).