On Martin's Controversial Move To Explicitly Consider Semantics

Bateman (1998: 23):

It is also certainly not the case that Martin’s approach is uncontroversial, even within his own functional tradition. The need for a move to explicitly consider semantics has sometimes been downplayed, aided in part by the rich functional semantic flavour that permeates grammatical descriptions such as that found in Halliday’s IFG; here “the grammar is infused with meaning, and a stratal distinction between grammar and semantics systematically blurred.’’[p33]. 

Blogger Comments:

[1] Here Bateman again frames the matter in terms of agreement among linguists instead of in terms of theoretical validity.  The distinction between agreement and validity is encapsulated in the following cartoon of B. Kliban:

[2] This is potentially misleading, since Martin does not "explicitly consider semantics", because his discourse semantics, as previously explained, is merely a combination of Halliday's textual lexicogrammar (rebranded cohesion systems) and Halliday's previously theorised semantic system (rebranded speech function).

[3] To be clear, as Halliday & Matthiessen (2014: 49) explain:

Being a ‘functional grammar’ means that priority is given to the view ‘from above’; that is, grammar is seen as a resource for making meaning — it is a ‘semanticky’ kind of grammar. But the focus of attention is still on the grammar itself.

[4] This quote from Martin demonstrates his misunderstanding of stratification, which is also encompassed in his mantra 'all strata make meaning', which confuses stratification ('all strata') with semogenesis ('make meaning'), and leads him to interpret every stratum as a stratum of meaning.  See, for example, Misrepresenting Stratification.

To be clear, SFL theory distinguishes between meaning (semantics) and wording (lexicogrammar), principally because the distinction is essential for the systematic understanding of grammatical metaphor.

On The Substantive Content Of Martin's English Text

Bateman (1998: 23):

Concerning the substantive content of the book, it is not possible that one would agree with all of the claims and analyses that Martin makes, nor with all aspects of the methodology, and some unclarities are introduced by the very position developed. For example, since now so much work is being done in the discourse semantics, there are probably ramifications for how a lexicogrammar should be organised: in Chapter 3, Martin notes his earlier critique of Halliday’s DEIXIS network in lexicogrammar, and in Chapter 5 the restriction of lexis as most delicate grammar to field-neutral oppositions is clearly a strong constraint. But there is little discussion of this in the book; it is generally stated that the account is compatible with Halliday and Matthiessen’s descriptions, which may be broadly the case. But particular differences in description may well be necessary. However, given that the focus of English Text is primarily on the discourse semantic stratum, this omission is probably quite justified.

Blogger Comments:

[1] To be clear, it is not a matter whether linguists agree about Martin's theorising, but whether Martin's theorising agrees with data and with the rest of the theory in which it is situated.

[2] To be clear, Martin's discourse semantics, apart from his rebranding of speech function, is rebranded lexicogrammar (textual cohesion):

• IDENTIFICATION is rebranded cohesive reference and ellipsis-&substitution, misunderstood, and confused with ideational denotation and with interpersonal deixis of the nominal group, inter alia (evidence here);
• IDEATION is rebranded lexical cohesion, misunderstood as experiential in metafunction, and confused with lexis as most delicate grammar and with misunderstood logical relations between ergative functions of the clause, inter alia (evidence here); and
• CONJUNCTION is rebranded cohesive conjunction, misunderstood as logical in metafunction, and confused with misunderstood logical relations between clauses in clause complexes (evidence here).

[3] To be clear, Martin confuses the interpersonal deictic function of determiners in the nominal group with the textual reference function of determiners in non-structural cohesion (evidence here).

[4] To be clear, Martin's proposal (pp289-90) is to locate "field-neutral" lexical taxonomies in lexicogrammar and "field-specific" lexical taxonomies outside language at the level of context (misunderstood as register) — with lexical relations at the level between them, discourse semantics.  

In terms of stratification, the proposal locates lexicogrammatical phenomena at three levels of symbolic abstraction, one of them outside language.  In terms of realisation, the proposals are that:

• field-specific lexical taxonomies are realised in lexical relations, and
• lexical relations are realised in field-neutral taxonomies

Or, taking metaredundancy into account, field-specific lexical taxonomies are realised in the realisation of lexical relations in field-neutral taxonomies.

[5] To be clear, Martin's account is not even broadly "compatible with Halliday & Matthiessen's descriptions"; it violates the theoretical architecture and misunderstands its basic concepts (evidence here).

[6] To be clear, the difference between modelling semantics and lexicogrammar is the level of symbolic abstraction, meaning or wording, that the theorising is concerned with.  A model of semantics specifies the systems of meaning that are realised in wording; a model of lexicogrammar specifies the systems of wording that realise meaning.

On Martin's System Of Nuclear Relations

Bateman (1998: 22-3):

Even I am not quite sure about Figure 5.23, which sets out the subtypes of nuclear lexical relations in IDEATION: It could be a summary overview of the material or a very interesting hypothesis about the relationship between the semantic region of IDEATION and its lexicogrammatical realisation. I suspect the latter, but this interpretation is certainly missed in, for example, Tucker’s (1998) review of approaches to lexis, Martin’s included. Tucker criticises Martin for not giving details of lexicogrammatical realisation, whereas the details of Figure 5.23 are very precise in the particular grammatical environments called for. Martin may well have intended these possible alternative interpretations; they certainly invite close consideration of the nature and form of a linguistic description!

These superficial difficulties can all, by and large, be unravelled or contextualised by close reading, but for the student who is perhaps still struggling with the overall map of the system, the difficulty is an unnecessary overhead.

Blogger Comments:

[1] As this blog demonstrates, Bateman's "modest" use of even I (thematised counter-expectancy: exceeding) is entirely unwarranted.

[2] To be clear, Martin's Figure 5.23, Nuclear Relations In English, is the system network for nuclear relations, one of the more delicate systems of Lexical Relations In English (Figure 5.10), and so, in that sense, it is a "summary overview of the material".

The reason why Figure 5.23 is not "a very interesting hypothesis about the relationship between the semantic region of IDEATION and its lexicogrammatical realisation" is that it does not specify how any of its features are realised grammatically.

One reason why Bateman was fooled, in this respect, is that Martin falsely presents examples — chase + cat, her + cat, etc — in the form of realisation statements.

Another reason why Bateman was fooled, in this respect, is that the network presents a view of what is purported to be a semantic system from the perspective of the view from below, lexicogrammar, as demonstrated by the "discourse semantic" features [clausal], [verbal], [nominal].  The view from below is the opposite perspective of that taken in theorising in SFL.

In this regard, Bateman has mistaken a network of grammatical features for the grammatical realisations of a discourse semantic system.

The irony, of course, is that, in modelling lexical relations, Martin's system is a proposal for a lexicogrammatical system.  It's just that neither Martin nor Bateman realises it.

[3] As the argument in [2] demonstrates, Tucker was entirely justified in criticising "Martin for not giving details of lexicogrammatical realisation".

[4] Strictly speaking, these are not alternative interpretations, since the system in Figure 5.23 is presented by Martin as both, "a summary overview of the material" and as a discourse semantic system realised in lexicogrammar.

[5] This is true.

[6] To be clear, "superficial difficulties" such as those identified in [2] are more accurately described as serious theoretical inconsistencies, which even Bateman's close reading failed to identify.

On Martin's Use Of System Networks

Bateman (1998: 22):

An example of one of the more superficial presentational problems is the ready use made of system networks for a number of different purposes as well as for linguistic descriptions at all of the strata treated in the book — i.e., lexicogrammar, discourse semantic, register, and genre. It is as a consequence not always immediately clear at what level of description a particular system network is being presented—thus Figure 2.2 is for MOOD (i.e., from the lexicogrammar) while the figure following, Figure 2.3, is speech function (i.e., from the discourse semantics); similarly, Figure 5.10 is a discourse semantic network, whereas Figures 5.11 and 5.12 are drawn from the field component of register and Figures 5.3–5.9 from the lexicogrammar. Sometimes, it is not even clear if a network is part of the intended linguistic description at all; some networks are simply presenting overviews of the material introduced in the book: e.g., Figure 1.16’s overview of types of linguistic structure, or Figure 3.10’s summary of types of phoricity (although these latter can then always, of course, be considered as field-specific taxonomies defining the technical terms involved as set out in the chapter on IDEATION: they are not earmarked as such in the text however).

Blogger Comments:

[1] Bateman's confusion in this regard stems from the fact that Martin uses the system network schema both for genuine system networks and for simple classifying taxonomies (e.g. Figures 1.16, 3.10, 5.11, 5.12).  

[2] To be clear, Halliday's speech function is a genuine semantic system — one that can be realised congruently or metaphorically in a grammatical system of the same metafunction (mood) — rather than Martin's rebranding of textual grammatical systems (cohesion) as discourse semantic systems of various metafunctions.

[3] To be clear, Martin's Figure 5.10 presents relations between lexical items (lexicogrammar) as a discourse semantic system.

[4] To be clear, the fact that these latter taxonomies are not identified in the text by Martin as taxonomies specific to the field of discourse semantics demonstrates that Martin did not realise that they could be interpreted as such in terms of his own model.

On Subsequent Works Relating Genre To Grammar

Bateman (1998: 22):

Fortunately, some of the discussions of English Text and the work out of which English Text has grown have already started making their way into introductory text books in systemic functional linguistics, of which there have recently been a number of important publications. Eggins (1994), for example, assumes the English Text model as the basic starting point for her introduction, while notions of genre and its relation to grammar are explicitly introduced and summarised in both Gerot and Wignall (1994) and, in more detail, Butt, Fahey Spinks and Yallop (1995). This orientation gives students a clear additional reason and motivation for learning the intricacies of grammatical description, for only then is one in a good position to search out genre. They will also help substantially when beginning to comes to terms with English Text itself.

Blogger Comments:

This is potentially misleading.  Unlike Martin's students Eggins and Wignell, Butt et al. (1995) do not build on Martin's model of genre, but on that of Martin's source, Hasan, distinguishing, for example, between context (4, 7) on the one hand, and genre as text type on the other (8-14).