Confusing Metafunctions And Axes

Martin (1992: 462):

The challenge here is to interpret the meaning of Subject from a discourse perspective, by looking at the way in which discourse structures interact with Subject selection.

Blogger Comments:

[1] This confuses metafunctions.  Subject is an interpersonal function whereas discourse semantics has been demonstrated here to be largely a misunderstanding of Halliday and Hasan's model of lexicogrammatical cohesion, a non-structural resource of the textual metafunction.  The meaning of Subject is interpersonal, not textual.  Halliday & Matthiessen (1999: 528):

The “textual” metafunction is the name we give to the systematic resources a language must have for creating discourse: for ensuring that each instance of text makes contact with its environment. The “environment” includes both the context of situation and other instances of text.

[2] This confuses axes, and therefore levels of symbolic abstraction, since syntagmatic structures realise paradigmatic selections.

[3] As has been previously demonstrated here, the discourse "structures" reworked from Halliday and Hasan's model of lexicogrammatical cohesion are not structures, but cohesive relations.

Misconstruing Subject And Theme

Martin (1992: 461):

The concept of modal responsibility is less obvious in propositions, and the meaning of Subject is hard to isolate because of the fact that in English declaratives and interrogatives Subject conflates with unmarked topical Theme.  However, it is clear in contexts where marked Themes are used to scaffold a text's method of development that Subject selection is in principle independent of Theme selection.

Blogger Comments:

[1] This is not true.  The concept of modal responsibility is no less obvious in propositions.  In propositions, modal responsibility is responsibility for the validity of what is being stated or questioned (rather than offered or commanded, as in proposals).

[2] The meaning of Subject is modal responsibility.  The identity of Subject is "isolated" by Halliday & Matthiessen (2014: 149) as follows:

In fact the Subject in English has got a distinct identity, as we have pointed out; its identity can be established if we adopt a trinocular perspective, as suggested by the stratificational model of language. 

(i) From below, it is that nominal element (nominal group or nominalised phrase or clause) that is picked up by the pronoun in the mood tag. 

(ii) From round about, it is that which combines with the Finite (operator) to form the Mood element in the clause; it is also that which constitutes the unmarked Theme if the mood is declarative, and which switches place with the Finite if the mood is yes/no interrogative. 

(iii) From above, it is that which carries the modal responsibility: that is, responsibility for the validity of what is being predicated (stated, questioned, commanded or offered) in the clause.

[3] The conflation of Subject with Theme in the unmarked case in declarative clauses has no bearing on the meaning or identity of Subject.  Subject is an interpersonal function (modal responsibility of the clause as exchange), whereas Theme is a textual function (point of departure of the clause as message).

[4] Subject does not conflate with unmarked topical Theme in declarative clauses.  This wording betrays Martin's misunderstanding of un/marked Theme.  If Theme conflates with Subject in declarative clauses, then the Theme is said to be unmarked (whereas if it conflates with Adjunct or Complement, it is said to be marked).  Halliday & Matthiessen (2004: 73, 74):

In a declarative clause, the typical pattern is one in which Theme is conflated with Subject; … We shall refer to the mapping of Theme on to Subject as the unmarked Theme of a declarative clause. The Subject is the element that is chosen as Theme unless there is good reason for choosing something else. 

A Theme that is something other than the Subject, in a declarative clause, we shall refer to as a marked Theme. The most usual form of marked Theme is an adverbial group … or prepositional phrase … functioning as Adjunct in the clause. Least likely to be thematic is a Complement, which is a nominal group that is not functioning as Subject — something that could have been a Subject but is not … . Sometimes even the Complement from within a prepositional phrase functions as Theme … .

[5] Whatever the context, Subject selection is independent of Theme selection — regardless of whether the functions are conflated or not.  Subject selection is an interpersonal choice — one which identifies where the responsibility for the validity of the predication lies.  Theme selection is a textual choice — one which identifies the element that is highlighted as the point of departure for what follows in the message.

Misconstruing Modal Responsibility As Social Responsibility

Martin (1992: 461):

Halliday (1984/1988: 43-45; 1985: 76-78) stresses the fact that Subject is a meaning making function in the English clause, alongside the textual functions Theme and New or experiential functions such as Actor and Goal.  The meaning of Subject is glossed in terms of modal responsibility.  This meaning is most transparent in proposals, where regardless of voice, the Subject is the constituent responsible for seeing that goods are exchanged or a service performed.

Blogger Comments:

[1] Halliday & Matthiessen (2014: 148n):

… in our systemic-functional account of English, Subject is interpreted metafunctionally as an interpersonal grammatically [sic] category; but like other such categories, e.g. Theme and Actor, Subject is both grammatical and semantic at the same time – the notion of ‘modal responsibility’.

[2] This confuses semogenesis (making meaning) with stratification (wording realising meaning).

[3] New is a functional element of the information unit, not of the clause.

[4] Modal responsibility is the view of Subject 'from above' — from semantics. Halliday & Matthiessen (2014: 149):

From above, [the Subject] is that which carries the modal responsibility: that is, responsibility for the validity of what is being predicated (stated, questioned, commanded or offered) in the clause.

[5] The Subject is not "the constituent responsible for seeing that goods are exchanged or a service performed".  This social responsibility between interlocutors is not (semiotic) modal responsibility.  Modal responsibility is responsibility for the validity of the proposal or proposition.  The validity of a proposal is commanded or offered with respect to the Subject; the validity of a proposition is stated or questioned with respect to the Subject; see for example Halliday & Matthiessen (2014: 242).

Not Understanding Interstratal Realisation

Martin (1992: 460):

In Chapters 1 and 5 it was noted that one essential interpretation of "A realises B" in functional linguistics is that A is symbolically related to B — relationships between levels (beyond that between content and expression form) are natural.  The discussion of Theme and New in clauses, paragraphs and texts presented to this point in sections 6.3.2 and 6.3.3 illustrates clearly the symbolic way in which text can be said to be realised by clause.  Clause grammar is a metaphor for text structure; patterns of interaction between discourse semantics and lexicogrammar echo the textual organisation of the clause.  It is in this sense that a functional grammar is a critical resource for textlinguistics.

Blogger Comments:

This demonstrates most clearly that Martin (1992) does not understand the meaning of 'realisation' in the architecture of SFL theory.

[1] The clause A realises B construes an intensive identifying relation between A and B, such that A is the Token (the lower level of abstraction) and B is the Value (the higher level of abstraction).  Like verbs such as express, signify, spell, stand for, mean, the verb realise serves as a Process construing a symbolic identifying relation; see Halliday & Matthiessen (2004: 238; 2014: 269).

[2] The clause A realises B does not specify whether the relation of symbolic identity between levels is natural or otherwise.

A	realises	B
expression	realises	content
phonology	realises	lexicogrammar
lexicogrammar	realises	semantics
Token	Process: intensive: symbol	Value

[3] In terms of stratification, 'text' refers to the highest unit of the semantic stratum, and 'clause' refers to the highest unit of the lexicogrammatical stratum.  The semantic stratum (meaning) is realised by the lexicogrammatical stratum (wording).  The 'symbolic way' that 'text is realised by clause' is just this interstratal relation.

[4] The claim here is that clause rank lexicogrammar (textual metafunction) is a metaphor (incongruent realisation) of text structure.

[5] In terms of the architecture of SFL theory, 'patterns of interaction between discourse semantics and lexicogrammar' are patterns of realisation relations between the systems of each stratum.  Thus, the claim here is actually that patterns of interstratal realisation of discourse semantics in lexicogrammar reflect the Theme^Rheme structure of the clause.

[6] The reason why a functional grammar is a critical resource for textlinguistics is actually because of the relation between text and system: instantiation.  Halliday & Matthiessen (2004: 658):

A text is meaningful because it is an actualisation of the potential that constitutes the linguistic system; it is for this reason that the study of discourse (‘text linguistics’) cannot properly be separated from the study of the grammar that lies behind it.

Halliday (2008: 192):

The system and the text are not two different phenomena: what we call the “system” of a language is equivalent to its “text potential”. Analysing discourse means, first and foremost, relating the text to the potential that lies behind it.

Misconstruing A Grammatical Reference Item As An Accumulation Of New Information

Martin (1992: 459):

Hyper- and macro-News, like hyper- and macro-Themes are more prominent in written than spoken English.  However, just as marked Themes were shown to functional [sic] in parallel ways to hyper-Themes in spoken mode, so extended reference can be shown to function in parallel ways to hyper-News.  Two excerpts from text [4:2] are re-presented below to illustrate this point.  This in [4:2ff & ii] and that in [4:2aaa & ccc] are used to accumulate the actions in the activity sequences that have gone before in a way which complements their scaffolding with marked Themes as discussed in 6.3.2.  Extended reference is in other words to hyper-New as marked Theme is to hyper-Theme:

Synoptic : Dynamic ::

(WRITTEN : SPOKEN ::)

hyper-Theme : marked Theme ::

hyper-New : extended reference

Blogger Comments:

[1] In text [4:2], or any other text, the grammatical items this and that do not "accumulate" (or summarise) previous New information.  As instances of anaphoric demonstrative reference, they refer back to earlier information, which may have been presented as Given and/or New.

[2] For Martin, activity sequences in the text are misconstrued as field, the ideational dimension of context, and this, in turn, is misconstrued as register.

[3] Glossing Martin's terms, these proportionalities become:

• an anaphoric demonstrative reference item in spoken mode is to
• a paragraph summary in written mode as
• a marked Theme of a clause in spoken mode is to
• a topic sentence in written mode.

When An Abstract Is A Summary And A Summary Is An Introduction

Martin (1992: 458):

The consideration of macro-Theme and macro-New in Matthiessen and Halliday's paper points to the difference between what are often referred to as an abstract and a summary. A[n] abstract is essentially a dislocated macro-New (typically "Theme marked" at a text's beginning or published elsewhere); it presents the essence of what can be learned by reading the text the abstract functions as macro-New for.  A summary on the other hand is more like a dislocated macro-Theme (often "Rhematic" at the end of a text or published elsewhere); it provides an outline of what can or has been read, re-scaffolding by way of review.

Blogger Comments:

[1] On this model, the difference between an abstract and a summary is that an abstract is a summary and a summary is an introductory paragraph; see the next two points, below.

[2] Glossing the term macro-New as 'text summary' (p454), this becomes:

A[n] abstract is essentially a dislocated text summary; it presents the essence of what can be learned by reading the text the abstract functions as text summary for.

[3] Glossing the term macro-Theme as 'introductory paragraph' (p437), this becomes:

A summary on the other hand is more like a dislocated introductory paragraph; it provides an outline of what can or has been read, re-scaffolding by way of review.

[4] An abstract or summary that is published elsewhere, rather than as part of the text in question, though related (intertextually), is not part of the logogenesis of the actual text it abstracts or summarises, and so does not function as its macro-New or macro-Theme.

Presenting Prescription As Theoretical Description

Martin (1992: 456):

Importantly, macro-News are not simply a replay of macro-Themes; they draw together new meanings which could not have been predicted by macro-Theme because they had not been made.

Blogger Comments:

[1] This translates as: 'text summaries are not simply a replay of introductory paragraphs'.  The field here, still, is writing pedagogy — prescribing how to structure written texts — rather than theorising the system of the English language by examining instances (texts).  Moreover, the situation is quite different when an actual text is closely examined.

In text [6:23], the introductory paragraph ('macro-Theme') is given by Martin as:

Wars are costly exercises. They cause death and destruction and put resources to non-productive uses but they also promote industrial and economic change. This benefit does not mean that war is a good thing, but that it sometimes brings useful developments.

and the text summary ('macro-New') is given by Martin as:

… it is clear that future generations not only enjoyed the security and peace won by their forefathers but also the benefits of war-time economic expansion.

It can be seen that, rather than accumulate the New meanings of the preceding co-text (Martin's claim), what actually happens in this particular text summary ('macro-New') is:

1. first, content not previously discussed in the text is introduced (it is clear that future generations not only enjoyed the security and peace won by their forefathers), and
2. second, one piece of New information from the introductory paragraph (promote industrial and economic change) is reworded (lexical cohesion) and again presented as New information (the benefits of war-time economic expansion).

[2] The claim here is that new meanings could not have been predicted by the introductory paragraph on account of the fact that these new meanings had not been made.  The logic here is as follows:

• Q could not have been predicted from P because Q had not yet happened.

The notion of prediction only applies to events that have not yet happened.

Confusing Writing Pedagogy With Linguistic Theory

Martin (1992: 456):

Macro-Theme, hyper-Themes and clause Themes project forward, scaffolding the text with respect to its rhetorical purpose (i.e. its genre); macro-New, hyper-News and clause News on the other hand look back, gathering up the meanings which have accumulated to elaborate a text's field.  The result is a textured sandwich in which texts project both forward and back as they unfold.  Texture of this kind, which is a tendency in writing (by no means a categorical rule) is outlined schematically in Fig. 6.12.

 

Fig. 6.12. Sandwich texture in abstract written discourse

Blogger Comments:

[1] The field here is writing pedagogy, not linguistic theory:

• macro-Theme is Martin's rebranding of introductory paragraph; 
• hyper-Theme is Martin's use of Daneš's (1974) term to rebrand topic sentence;
• macro-New is Martin's term for a text summary (and so 'New' is a misnomer);
• hyper-New is Martin's term for a paragraph summary (and so 'New' is again a misnomer).

[2] This confuses context (which is more abstract than language) with register (which is a functional variety of language).  The rôle of language in context is its rhetorical mode.  A genre, on the other hand, as a text type, is register viewed from the instance pole of the cline of instantiation.

[3] 'New' is a functional element of the information unit, not the clause. The domain of an information unit may be shorter or longer than a single clause.

[4] The meanings that 'elaborate' the field of a context of situation are all the meanings of a text, not just those that are presented as New to the addressee or re-presented in summaries.

[5] The relation between meaning (semantics) and field (context) is realisation, which is both an elaborating and an identifying relation.  It is the identifying component of the relation that marks the two as different levels of symbolic abstraction.  The use of the term 'elaboration' betrays the fact that Martin does not understand strata as different levels of symbolic abstraction, as has been demonstrated in numerous previous posts (here).

Confusing Textual Phase With Textual Status

Martin (1992: 456):

Public speakers (and even writers, at a time when composition was taught in schools) are commonly advised: "tell your audience what you're going to say, say it, and then remind them what you've said".  The generally cited Introduction ^ Body ^ Conclusion formula for expository texts makes the same point (as does Orientation ^ Complication/Resolution ^ Coda for narrative).  What this wise counsel represents is the tendency for whole texts in English to symbolise the Theme ^ Rheme and Given ^ New structure of the English clause.

Blogger Comments:

[1] As this demonstrates, the field here is writing pedagogy, not linguistic theory.

[2] What these three schemas actually have in common is that they are ways of naming the time-phases (inception, duration and termination) of a process — in this case, the unfolding of text (logogenesis).  This is quite distinct from Theme and New, which highlight the textual status of experiential and interpersonal content.

[3] Given ^ New is not a structure of the English clause.  They are functional elements of the information unit.

When New Information Isn't

Martin (1992: 455):

The pattern of interaction in [6:23] is outlined below, listing the text's News, hyper-New and macro-New.  The News of the introduction are listed (in plain face) but do not participate in this pattern; they participate instead in the complementary function of predicting the text's method of development reviewed above.

Blogger Comment:

To be clear, on this model, information that is presented by the writer as New to the reader in the introduction (Martin's macro-Theme) does not participate in the 'accumulation of New information'.

Misidentifying New Information

Martin (1992: 454-5): 

The next paragraph is not pulled together in this way; instead, [6:23r] functions as a macro-New, summarising the text as a whole. … macro-New [is …] underlined in the following representation of [6:23]; the minimal News of remaining clauses appear in bold-face.

[6:23]	(illustrating New … and macro-New)
	m	The war had also revealed inadequacies in Australia’s scientific and research capabilities.
	n	After the war strenuous efforts were made to improve these.
	o	The Australian National University was established with an emphasis on research.
	p	The government gave its support to the advancement of science in many areas, including agricultural production.
	q	Though it is difficult to disentangle the effects of war from other influences,
	r	it is clear that future generations not only enjoyed the security and peace won by their forefathers but also the benefits of war-time economic expansion.

Blogger Comments:

[1] This is the first mention of the security and peace won by their forefathers in the text, and so it does not satisfy Martin's definition of macro-New as 'summarising the text as a whole'.

[2] This does not satisfy Martin's definition of minimal New — the "highest ranking" clause constituent (p451) — because [6:23p] is a clause complex, rather than a single clause, with including agricultural production as a possessive attributive clause hypotactically elaborating the clause that precedes it.  Based on a spoken reading, there are likely to be information foci on many and agricultural.

Problems With The Complementarity Of Hyper-Theme & Hyper-New

Martin (1992: 454):

This complementarity of hyper-Theme predicting a text's method of development and hyper-New accumulating its point is outlined in Fig. 6.11.

 

Fig. 6.11. Complementarity of hyper-Theme & hyper-New

Blogger Comments:

[1] This is potentially misleading, to the extent that it represents Theme and New as always distinct.

[2] As previously explained, a hyper-Theme — Martin's rebranding of Topic Sentence, using a term coined for other purposes by Daneš (1974) — does not "predict" what experiential content will be highlighted in subsequent Themes.

Any relations between the experiential content of Topic Sentences and the experiential content of subsequent Themes are those of lexical cohesion and grammatical reference, two non-structural resources of the textual metafunction.

Misrepresenting New Information

Martin (1992: 453-4):

The pattern of hyper- and macro-Themes recognised in 6.3.2 above raises the question of whether a similar symbolic pattern arises based on New.  Reviewing text [6:23] it can be seen that hyper- and macro-New are also important aspects of texture, particularly in abstract writing.  The function of hyper-New is illustrated in [6:23l] below; this clause pulls together and summarises the information build up in the rest of the paragraph. … 

Hyper-New [is] single underlined … in the following representation of [6:23]; the minimal News of remaining clauses appear in bold-face.

[6:23]	(illustrating New, hyper-New …)
	a	Wars are costly exercises.
	b	They cause death and destruction
	c	and put resources to non-productive uses
	d	but they also promote industrial and economic change.
	e	This benefit does not mean that war is a good thing, but that it sometimes brings useful developments.
	f	The Second World War further encouraged the restructuring of the Australian economy towards a manufacturing basis.
	g	Between 1937 and 1945 the value of industrial production almost doubled.
	h	This increase was faster than otherwise would have occurred.
	i	The momentum was maintained in the post-war years
	j	and by 1954-5 the value of manufacturing output was three times that of 1944-5.
	k	The enlargement of Australia’s steel-making capacity, and of chemicals, rubber, metal goods and motor vehicles all owed something to the demands of war.
	l	The war had acted as something of a hothouse for technological progress and economic change.

Blogger Comments:

[1] The field here is again writing pedagogy, rather than linguistic theory.

[2] The paragraph is a unit of graphology, the expression plane of language realising written mode.  It is not a unit of linguistic content.

[3] New is not a functional component of the clause.  It is the obligatory element of the information unit, which is parallel with the clause.  The distribution of information units in a text is indicated by the phonological system of tonality, which specifies the distribution of tone groups, because an information unit is realised by a tone group.  The culmination of New information, the information focus, is indicated by the phonological system of tonicity, which specifies the location of tonic prominence, because the information focus is realised by tonic prominence.

In contrast, in the analysis of text [6:23], the New is arbitrarily assigned to the last functional element of the Rheme of the clause, regardless of intonation, graphological indications of tone group boundaries, lexical density, and so on.  The possibility of New information in the Theme, a very frequent feature of discourse, is ruled out by such a rigid, arbitrary approach.

Some of the likely foci of New information in text [6:23], based on a spoken reading, and overlooked by the arbitrary approach, are highlighted above in dark red.