The Difference Between Martin's 'Activity Sequence' And Halliday's 'Sequence'

  Martin, Matthiessen & Painter (2010: 283-4):

Both effective and middle clauses interact with the temporal relations discussed above to construct the activity sequences comprising a disaster and rescue operation of this kind. A transitive perspective on what happened is presented in Table 7.3 below. The experiential grammar builds the figures (participant, process and circumstance configurations) which constitute the predictable sequences the story recounts. Once the waves roll in, there are no surprises, which accounts for the absence of external concessive relations in FitzSimon's article. For more on ideational semantics in relation to field, see Working with Discourse Chapter 3.

Blogger Comments:

[1] To be clear, clauses don't "interact" with temporal relations; temporal is one type of enhancing relation, structural or cohesive, between clauses.

[2] To be clear, Martin's 'activity sequence' is theoretically distinct from Halliday's 'sequence'. In Halliday's semantics (Halliday & Matthiessen 1999: 49), a sequence (of figures) is the ideational phenomenon that is congruently realised grammatically by a clause complex (logical metafunction). 

In Martin's model, an activity sequence is variously construed as a unit of field, misconstrued as register, misconstrued as context (Martin 1992), evidence here, or as an experiential unit of discourse semantics (Martin & Rose 2007), distinct from the logical system of conjunction; evidence here.

[3] To be clear, 'figure' is not a discourse semantic unit (Martin 1992; Martin & Rose 2007), but a phenomenon in Halliday's ideational semantics that is congruently realised grammatically as a clause (Halliday & Matthiessen 1999: 49).

[4] This confuses supposed reader reactions (surprise) to events in the text with the author's (FitzSimons') use of logical relations (concessive). On the one hand, what happened after the waves rolled in was surprising to this reader. On the other hand, the author simply chose not to use any concessive logical relations, such as although everyone was surprised by the sudden horror of the unfolding events, the rescue operation began almost immediately. 

[5] For a close examination of the experiential discourse semantics in Working with Discourse Chapter 3, see the clarifying critiques here.

Misrepresenting The Ergative Model As The Transitive Model

 Martin, Matthiessen & Painter (2010: 283):

With all this activity going on, it is worthwhile exploring who is doing what to whom. An ergative analysis of transitivity reveals 14 effective clauses with an Agent impacting on a Medium. Agency is shared between the waves and the Bondi Boys, both of whom affect people. The main participants, in other words, don't act on one another; this is not a struggle between man and nature. Rather nature savages people, and then other people rescue them. In effect we have two protagonists, not a direct showdown with nature (antagonist) overcome by Bondi Boys (protagonist).

Blogger Comments:

[1] This confuses transitivity with ergativity. The 'impact' model is that of transitivity — Actor on Goal — not ergativity. Halliday & Matthiessen (2014: 334):

… the transitive model is based on the configuration of Actor + Process. The Actor is construed as bringing about the unfolding of the Process through time; and this unfolding is either confined in its outcome to the Actor or extended to another participant, the Goal. The Goal is construed as being impacted by the Actor’s performance of the Process.

In the ergative model, on the other hand, a Process is actualised through a Medium, and there may be an additional participant functioning as an external cause: an Agent; see Halliday & Matthiessen (2014: 342). That, is an Agent does not impact on a Medium; an Agent is the external cause of a Process that is actualised through a Medium.

[2] To be clear, if nature affects/savages people, then nature (Actor) is acting on people (Goal). What is true is the vacuous observation that people don't 'act on' three enormous ocean waves and overcome them.

Misrepresenting Ellipsed Subjects As Unmarked Themes

Martin, Matthiessen & Painter (2010: 282):

As noted above however, the text does feature a number of non-finite clauses, many with an implicit Subject/unmarked Theme. All except one (to be restored...) are imperfective and construct action that is underway and overlapping with the alpha clause on which they depend for temporal deixis (past) and to which they defer in terms of arguability (declarative). These clauses in effect help the text accelerate as a number of synchronous events are canvassed, thereby foregrounding the encompassing chaos of the waves' incursion and the cooperative intensity of the Bondi Boys' rescue.

Blogger Comments:

[1] This is misleading. Ellipsed Subjects are not unmarked Themes. The absence of elements through ellipsis iconically marks lack of textual prominence, and the analysis of these elements as Themes misrepresents them as textually prominent. Halliday & Matthiessen (2014: 635):

Ellipsis thus assigns differential prominence to the elements of a structure: if they are non-prominent (continuous), they are ellipsed; if they are prominent (contrastive), they are present. The absence of elements through ellipsis is an iconic realisation of lack of prominence.

[2] To be clear, this is merely an impressionistic interpretation, unsupported by grammatical evidence. Synchronous events can be construed paratactically or hypotactically, and through finite or non-finite clauses. Moreover, foregrounding is a function of the textual metafunction, not of non-finite dependent clauses.

Confusing Two Distinct Texts

Martin, Matthiessen & Painter (2010: 282):

As far as mood is concerned, FitzSimons' text is overwhelmingly concerned with giving information, and uses declarative clauses to do so. It is only in the final two clauses that we find an interrogative (Do you...) and then an imperative (Write...) clause, signalling the shift from chronicling to solicitation:

Do you have a historical anecdote about a place in Sydney? |||

Write to Peter FitzSimons at pfitzsimons@smh.com.au |||

Working with Discourse, Chapter 7, introduces a framework for dealing with the function of mood in dialogue; see also Eggins & Slade 1997.

 

Blogger Comments:

[1] Clearly, these two clauses are not part of FitzSimons' text on the Bondi Beach tragedy.

[2] For a close examination of the account of 'the function of mood in dialogue' in Working with Discourse, Chapter 7, see the 25 clarifying critiques here.

Misrepresenting Internal Relations

 Martin, Matthiessen & Painter (2010: 282):

These internal relations, as we can see, tend to be realised between rather than within clause complexes and are oriented to phases of the generic structure of texts rather than what happened per se. Working with Discourse introduces conventions for analysing internal relations alongside external ones, rather than ignoring them (because they are between clause complexes) or subsuming some of them as delicate sub-types of expansion (inside clause complexes).

Blogger Comments:

[1] To be clear, as previous posts have demonstrated, the authors mistake instances of continuity for instances of internal conjunction.

[2] This is misleading, because internal relations also obtain structurally within clause complexes as well as cohesively between them. Halliday & Matthiessen (2014: 484):

… the enhancing relation may be internal rather than external (cf. Halliday & Hasan, 1976: Ch. 5; Martin, 1992: Ch. 4; Mann & Matthiessen, 1991); that is, the β-clause may relate to the enactment of the proposition or proposal realised by the α-clause rather than to the figure that it represents. For example, if it is not too personal an inquiry, what limits do you set ... means ‘if it is not ..., I ask you ...’; that is, the condition is on the act of questioning, not on the content of the question.

[3] To be clear, this is a bare assertion, unsupported by evidence. In the data analysed by the authors, it was the continuatives, not internal conjunctions, that marked new turns in the text — as is their function.

[4] For a summary of the many misunderstandings of internal conjunction in Working with Discourse, see Theoretical Problems With The System Of Internal Conjunction.

[5] This is very misleading indeed, because it falsely implies that, in Halliday's original model, cohesive internal relations between clause complexes are ignored, and that structural internal relations between clauses have less explanatory power by being subtypes of expansion. Martin's logical discourse semantic system lacks the general category of expansion, and its three subtypes: elaboration, extension and enhancement, because it derives from Halliday & Hasan (1976), at which time Halliday had not yet devised these categories. 

Importantly, contrary to the impression given above, the distinction between internal and external relations originates in Halliday & Hasan (1976), and others, such as Martin (1992) and Martin & Rose (2007), are merely trying to use their ideas.

Failing To Distinguish Continuity From Conjunction

 Martin, Matthiessen & Painter (2010: 282):

Similarly Ah, but marks the resigned counterexpectation of shifting from viewing the image in the present to reconstructing events from the past – the meaning is something like ‘although we've been looking at the picture for a bit, it's time now to begin our tale’:

Ah, but those 35,000 Sydneysiders [[who were lying in those very spots on the afternoon of February 6, 1938]], surely felt equally at peace. |||

Blogger Comments:

To be clear, from the perspective of SFL Theory, the Ah and but in this clause serve different functions, as recognised graphologically by the intervening comma that assigns them to different information units. The Ah is a continuative, and its function is to signal a new turn in the discourse. The previous clauses had described a contemporary photograph of Bondi Beach, and the continuative signals the beginning of the historical recount:

And so there they lie, happily sweltering in the summer sun on Australia's most famous beach, just as they have for so many generations past. It is such a wonderfully peaceful scene – of people and nature as a happy whole – that it is simply unimaginable that in a few seconds nature could ever rear up and savage the lot of them.

Ah, but those 35,000 Sydneysiders who were lying in those very spots on the afternoon of February 6, 1938, surely felt equally at peace.

The but, on the other hand, is a conjunction, and its function is to cohesively relate its clause to the previous text in terms of a logical relation of concessive condition ('even so').

Misrepresenting IFG And Mistaking Continuity For Internal Conjunction

Martin, Matthiessen & Painter (2010: 281):

A further limitation of the grammatical approach outlined in IFG 3rd edition is its backgrounding of the important distinction between what Halliday & Hasan (1976) call internal and external conjunction. External conjunctions construct logical relations between events in the field of discourse; internal relations construct rhetorical relations among phases of unfolding text. FitzSimons for example begins his text with And so, even though he hasn't mentioned any events which could be construed as causing people to be lying on the beach. The causal relation here is a rhetorical one, related to us settling down with text and image and starting to read because we have the feature in front of us.

And so there they lie, ||

happily sweltering in the summer sun on Australia's most famous beach, ||

just as they have for so many generations past. |||

It is such a wonderfully peaceful scene – of people and nature as a happy whole – [[that it is simply unimaginable [[that in a few seconds nature could ever rear up|| and savage the lot of them]].]] |||

Blogger Comments:

[1] This is misleading, because it is untrue. Halliday & Matthiessen (2004: 542-3, 545-6):

Many temporal conjunctives have an ‘internal’ as well as an ‘external’ interpretation; that is, the time they refer to is the temporal unfolding of the discourse itself, not the temporal sequence of the processes referred to. In terms of the functional components of semantics, it is interpersonal not experiential time. Parallel to the ‘simple’ categories above we can recognise the simple internal ones set out in Table 9(6) above. These play an important role in argumentative passages in discourse.

Moreover, here the authors contradict their own former similarly false claim (p280n) that because this distinction is foregrounded, the distinction between simultaneous and successive relations is backgrounded:

Contrast IFG 3rd edition 542-3 which focuses grammatically on simple, complex and simple internal temporal relations, with several subdivisions within each of these; the opposition of successive to simultaneous events is not a major parameter.

Here the authors have ceased to provide a guide to IFG, and have switched to misrepresenting IFG in order to promote Martin's model (of Halliday & Hasan 1976) instead.

[2] This is misleading. Both external and internal conjunction are concerned with the logical relation of expansion. In the case of temporal relations, the distinction is between experiential time and the interpersonal time of the unfolding of the discourse.

[3] This is misleading, because the function served by And so is continuity, not conjunction, and as such, it does not construe a causal relation, internally or rhetorically. Its function is simply 'to signal that a new move is beginning' (Halliday & Matthiessen (2004: 145).

Misrepresenting Temporal Relations In A Text

 Martin, Matthiessen & Painter (2010: 279-80, 281):

As we might expect, successive (then, now, finally, after) and simultaneous (as, with) temporal conjunctions are used to time events; but there are many implicit temporal links with events connected 'neutrally' by simple juxtaposition, and, or the dependency of non-finite clauses on finite ones. The text, as do stories in general, thus takes advantage of field based understandings, in this case of what happens when big waves hit a beach and a rescue operation gets underway – a reasonable expectation in an article composed for beach-savvy Sydneysiders.

We can also note that many of these relations obtain between clauses inside clause complexes, and others between clause complexes – since successive and simultaneous temporal relations can be realised in either way. For example, the same phrase and then functions both within and between clause complexes during the waves' attack above, highlighting the borders of a purely grammatical approach to logical relations:

^ And then another wave hit, ||

^ and then another. |||

Blogger Comments:

[1] This is misleading, because here the authors misunderstand with as marking a temporal (enhancing) relation, instead of an additive (extending) relation, despite having previously correctly identified it as extension in Table 7.1.

[2] This is misleading. Firstly, simple juxtaposition does not express an implicit temporal relation, as demonstrated by all the juxtaposed clauses that are not related temporally. Secondly, the only temporal relations marked by and are those marked by and then, and these are all 'explicit temporal links'. Thirdly, the dependency of non-finite clauses, in itself, does not mark a temporal relation, and in this text, as Table 7.1 shows, none of the dependent non-finite clauses are temporally related to their dominant clause.

[3] To be clear, the (second-order) field of this particular situation, the subject matter, is construed by the text itself. The text explicitly describes 'what happens when big waves hit a beach and a rescue operation gets underway'.

[4] To be clear, temporal relations that obtain within clause complexes are structural relations and function logically, whereas temporal relations that obtain between clause complexes are cohesive relations and function textually. This distinction is lost in Martin's discourse semantic model.

Misrepresenting IFG On Temporal Relations

Martin, Matthiessen & Painter (2010: 280):

Since FitzSimons' text is a story of some kind, it is important to consider temporal relations in some detail (Martin & Rose 2007, Chapter 4). …  In part this is a question of delicacy, since temporal relations are just one kind of enhancement. Convention also plays a role since implicit temporal relations are not considered in IFG style clause complex analysis, so that relations marked by and are treated as extending rather than enhancing. In addition, relations between clause complexes are beyond the scope of the analysis. And from the perspective of genre, the fine distinctions between parataxis and hypotaxis are not immediately relevant to mapping the time line of events. Shifting our attention from grammar to discourse is one way to overcome these limitations.

For Martin & Rose the basic distinction⁵ as far as temporal relations are concerned is between simultaneous and successive relations.

⁵ Contrast IFG 3rd edition 542-3 which focuses grammatically on simple, complex and simple internal temporal relations, with several subdivisions within each of these; the opposition of successive to simultaneous events is not a major parameter.

Blogger Comments:

[1] This is misleading, because enhancement is not a system in Martin & Rose (2007), since this work presents the model in Martin (1992), which derived from Halliday & Hasan (1976), which preceded Halliday's formulation of the most general logico-semantic relations. For some of the problems with Chapter 4 of Martin & Rose (2007), see the close examination here.

[2] This is misleading, because it is untrue. On the one hand, implicit temporal relations are covered by IFG — see Halliday & Matthiessen (2004: 420, 422-6) — and on the other hand, relations marked by and are those of parataxis: extending or enhancing. Halliday & Matthiessen (2004: 422):

[3] This is misleading, because there are no limitations (see above). Clearly, the misrepresentations above are a pretext for using Martin's derived model instead of Halliday's original model.

[4] This is misleading. For simple and simple internal temporal relations, the principal subtypes are following, simultaneous, preceding, conclusive:

This is an elaboration of Halliday (1994: 237):

The Problem With Macro-Theme As Linguistic Theory

Martin, Matthiessen & Painter (2010: 278):

Pushing this a step further, the Kicker of FitzSimon's text arguably functions as a macro-Theme for the text as a whole, with its long wave length reinforced graphologically through its superordinate position on the page and large bold font:

Blogger Comments:

[1] To be clear, this is a text about FitzSimon's text, and as such, not a section of FitzSimon's text.

[2] To be clear, macro-Theme is writing pedagogy ('introductory paragraph') masquerading as linguistic theory, and analysing a written text for macro-Themes is merely confirming the extent to which a writer has deployed this writing strategy. Moreover, in linguistic theory, Theme is one element of a structure along with Rheme. In the case of macro-Theme, no contrasting macro-Rheme is identified. A close examination of Martin (1992) on macro-Theme can be viewed here.

[3] To be clear, since Theme is a matter of higher textual status, the appropriate metaphor here would be amplitude, not wavelength.

The Problems With Hyper-Theme And Hyper-New As Linguistic Theory

Martin, Matthiessen & Painter (2010: 278):

Similarly, the waves' attack is introduced with a pair of clauses prefacing what is to come; in addition the results of the waves' incursion is summed once they've left the beach:

Technically speaking, forward looking topic sentences can be referred to as hyper-Themes (higher level Themes in other words), and retrospective summary comments as hyper-New. Both are a common feature of planned edited written discourse, and can be used to reinforce the partitioning of the text into phases signalled by marked Themes.

Blogger Comments:

[1] Again, hyper-Theme is writing pedagogy (topic sentence) masquerading as linguistic theory, and analysing a written text for hyper-Themes is merely confirming the extent to which a writer has deployed this writing strategy. Moreover, in linguistic theory, the function of Theme is to set the context of what follows, rather than to 'preface' it, and it is one element of a structure along with Rheme. In the case of hyper-Theme, no structure featuring a contrasting hyper-Rheme is identified.

[2] To be clear, hyper-New is also writing pedagogy (paragraph summary) masquerading as linguistic theory, and analysing a written text for hyper-News is merely confirming the extent to which a writer has deployed this writing strategy. Moreover, in linguistic theory, New information may either follow or precede Given information in a structural configuration. In the case of hyper-New, no structure featuring a contrasting hyper-Given is identified. A close examination of Martin (1992) on hyper-Theme can be viewed here.

[3] Again, Martin (1992) took the term 'hypertheme' from Daneš (1974), where it refers to the first of repeated themes (the theme that is 'over' or 'above' later repetitions).

[4] Again, this is writing pedagogy (proposals on how to write), not linguistic theory (propositions that model language).

The Problem With Hyper-Theme As Linguistic Theory

Martin, Matthiessen & Painter (2010: 278):

To this point, we've considered waves of information realised in ranking clauses. Martin & Rose (2007: Chapter 7) develop some of Halliday's ideas on textual meaning and periodicity by extending the analysis we've been exploring here to larger units of discourse. The Bondi Boys' rescue for example is introduced as their finest hour, in a clause which acts as a kind of topic sentence for their endeavours:

Blogger Comments:

[1] To be clear, the discussion of periodicity appears in Chapter 6 of Martin & Rose (2007). A close examination of this work can be viewed here.

[2] To be clear, Martin's 'hyper-Theme' (1992) is simply his rebranding of 'topic sentence', and as such, is writing pedagogy masquerading as linguistic theory. Analysing a written text for hyper-Themes is merely confirming the extent to which a writer has deployed this writing strategy.

Martin (1992) takes the term 'hypertheme' from Daneš (1974), where it refers to the first of repeated themes (the theme that is 'over' or 'above' later repetitions). A close examination of Martin (1992) on hyper-Theme can be viewed here.

Misrepresenting The General Thematic Pattern Of A Text

Martin, Matthiessen & Painter (2010: 278):

The general pattern, as we can see, is for marked Themes to punctuate the text at intervals, shifting our gaze, and for unmarked Themes to sustain our orientation to the field in between.

 

Blogger Comments:

To be clear, this assessment is based on a misapplication of Theme to a single text. Compare the following SFL-consistent analysis:

To be clear, the marked Themes are:

• With a roar like a Bondi tram running amok
• In no more than 20 seconds
• In their long and glorious history
• As one
• On the shore

and the unmarked Themes are:

• it (an enormous wave)
• three huge waves
• the boiling surf
• yet more large waves
• some (Bondi boys)
• some (Bondi boys)
• the Bondi clubhouse
• ambulances from all over Sydney town

It can seen that the context for what follows can be switched by either marked or unmarked Themes, and that continuity is maintained by the ellipsis of Subjects that would otherwise have been Themes. Importantly, the authors ignore the role of textual Themes in organising the text.

Misrepresenting Rhematic And Ellipsed Subjects As Theme [4]

Martin, Matthiessen & Painter (2010: 277-8):

And then we shift focus again, via another marked Theme, to the rescue on shore:

Blogger Comments:

The authors' analysis is misleading. On the one hand, since the first clause has a marked Theme, the Subject many survivors is not thematic, but falls within the Rheme (Halliday & Matthiessen 2014: 105). On the other hand, since the final clause has an ellipsed Subject, it has no Theme. The absence of elements through ellipsis marks lack of textual prominence, and the analysis of ellipsed elements as Themes misrepresents them as textually prominent; see Halliday & Matthiessen (2014: 635).

Misrepresenting Rhematic And Ellipsed Subjects As Theme [3]

Martin, Matthiessen & Painter (2010: 277):

We then shift via a marked Theme to the Bondi Boys, who are unmarked Theme in every clause involving the rescue; in addition we find another marked Theme, as one, emphasising the intensity of their cooperation.

Blogger Comments:

The authors' analysis is misleading. On the one hand, since the first and last clauses have marked Themes, the Subjects this and they are not thematic, but fall within the Rheme (Halliday & Matthiessen 2014: 105). On the other hand, since the included dependent clause << ignoring their own possible peril >> has an ellipsed Subject, it has no Theme. The absence of elements through ellipsis marks lack of textual prominence, and the analysis of ellipsed elements as Themes misrepresents them as textually prominent; see Halliday & Matthiessen (2014: 635).

The analysis also falsely interprets the included dependent clause as a marked Theme within its clause complex. To be clear, such clauses are not marked Themes, because there is no agnate unmarked Theme for such instances. See Halliday & Matthiessen (2014: 549-53).

Misrepresenting Rhematic And Ellipsed Subjects As Theme [2]

Martin, Matthiessen & Painter (2010: 277):

Subsequently another marked theme shifts our attention to the waves' impact:

Blogger Comments:

[1] This is misleading, because it is untrue. The marked Theme In no more than 20 seconds highlights a temporal circumstance as the context for what follows as the body of the message.

[2] This is analysis misleading. On the one hand, since the first clause has a marked Theme, the Subject that peaceful scene is not thematic, but falls within the Rheme (Halliday & Matthiessen 2014: 105). 

On the other hand, the authors misconstrue an embedded clause serving as Qualifier, waving for help, as a ranking clause, and misrepresent its ellipsed Subject as Theme. To be clear, the absence of elements through ellipsis marks lack of textual prominence, and the analysis of ellipsed elements as Themes misrepresents them as textually prominent; see Halliday & Matthiessen (2014: 635).

Misrepresenting Rhematic And Ellipsed Subjects As Theme [1]

Martin, Matthiessen & Painter (2010: 277):

The incursion proper begins with a marked Theme highlighting the sound of the first wave, and an unmarked Theme introducing the wave itself. Subsequently, the waves are unmarked Theme in every ranking clause. Alongside explicit unmarked Themes we've added boxes for Themes that have been ellipsed in non-finite and paratactic extending clauses; these missing Subjects all refer to the waves as well. In this respect, English is operating here like many languages that use an unmarked Theme to establish an orientation to a field and then don't mention Theme again until this orientation needs to be reinforced or changed (Caffarel et al. 2004).

Blogger Comments:

[1] Here the marked Theme is misanalysed for constituency; the embedded clause serving as Qualifier includes the conjunction like:

with a roar [[like a Bondi tram running amok]].

[2] This is misleading. Since this clause has a marked Theme, the Subject/Actor is not thematic, but falls within the Rheme (Halliday & Matthiessen 2014: 105). 

[3] This is misleading. Ellipsed Subjects are not unmarked Themes. The absence of elements through ellipsis marks lack of textual prominence, and the analysis of these elements as Themes misrepresents them as textually prominent. Halliday & Matthiessen (2014: 635):

Ellipsis thus assigns differential prominence to the elements of a structure: if they are non-prominent (continuous), they are ellipsed; if they are prominent (contrastive), they are present. The absence of elements through ellipsis is an iconic realisation of lack of prominence.

[4] To be clear, the ellipsis of the Subject signals continuity — see [3] — and its textual function is cohesive.

[5] To be clear, Martin's field, which he misconstrues as the ideational dimension of register, misconstrued as context, actually corresponds to ideational semantics in SFL Theory. Halliday & Matthiessen (1999: 407):

Thematic spaces in an ideational semantic network can be seen as a model of the systemic understanding of Theme and method of development articulated by Martin, where “field” corresponds to what has been discussed here in terms of ideational semantic networks in the ideation base.

Misrepresenting Theme

Martin, Matthiessen & Painter (2010: 276):

Let's begin with a theme analysis, focussing on ranking clauses. We'll box in Themes, shading marked Themes and treating as unmarked Theme any Subject Themes which follow them. Our reason for doing this is to highlight the complementary role played by marked and unmarked Themes – marked Themes to shift our orientation; unmarked Themes to sustain our perspective.

Blogger Comments:

To be clear, this seriously misrepresents Theme, and is even inconsistent with the authors' own account of Theme in Chapter 2. In SFL Theory, there is only one topical Theme to a clause: either unmarked or marked. In a declarative clause, a Subject that follows a marked Theme falls within the Rheme. Halliday & Matthiessen (2014: 105):

When some other element comes first, it constitutes a ‘marked’ choice of Theme; such marked Themes usually either express some kind of setting for the clause or carry a feature of contrast. Note that in such instances the element that would have been the unmarked choice as Theme is now part of the Rheme. …

The guiding principle of thematic structure is this: the Theme contains one, and only one, of these experiential elements. This means that the Theme of a clause ends with the first constituent that is either participant, circumstance or process. We refer to this constituent, in its textual function, as the topical Theme.

Misrepresenting Halliday On Grammar And Confusing Strata

Martin, Matthiessen & Painter (2010: 272):

Functional grammar analysis of the kind supported by this workbook enables a rich semantically oriented reading of clause, group and phrase structure. This is a critical step in text analysis since, as Halliday comments in his introduction to the 2nd edition, ‘A discourse analysis that is not based on grammar is not an analysis at all, but simply a running commentary on a text...’ By the same token, an analysis based solely on grammar, however rich, is not a complete analysis, since relationships between sentences (between clause complexes) still have to be accounted for. To interpret these relations beyond the clause, we need a model of text structure, and a model of social context. In this chapter we will take grammar as the point of departure and demonstrate how it can be extended in the direction of text structure¹ and context, drawing specifically on Martin & Rose's Working with Discourse (2nd edition, 2007) and Genre Relations (2008) to do so.

¹ For an alternative extension see Halliday & Hasan's 1976 Cohesion in English, which elaborates Chapter 9 of IFG (3rd edition).

Blogger Comments:

[1] This is very misleading indeed. The grammar includes the systems of cohesion, which are concerned with relations that obtain within and beyond clause complexes.

[2] This is misleading. Relations 'beyond the clause' are lexicogrammatical, and therefore require a lexicogrammatical model (cohesion). Text structure, on the other hand, is a semantic model, and 'social context', in SFL Theory, is a model of the culture as a semiotic system.

[3] For the serious theoretical inconsistencies in Working With Discourse (Martin & Rose 2007), see  the close examination of the publication here.

[4] For the theoretical confusions in Martin's model of genre, see here (Martin 1992) and here (Martin & Rose 2007).

[5] This is misleading. Halliday & Hasan's 1976 Cohesion in English is the intellectual source of both Chapter 9 of IFG (3rd edition) and the "extension" of it as discourse semantics (Martin 1992); see here.

Mental Clause Types With Act As Phenomenon

Martin, Matthiessen & Painter (2010: 244):

Embedded clauses can function as the Phenomenon of a mental clause under certain conditions. Embedded clauses representing acts can occur as the Phenomenon of perception in perceptive clauses.

Blogger Comments:

To be clear, unknown to the authors — and IFG (Halliday & Matthiessen 2014: 252) — embedded clauses representing acts can also occur as the Phenomenon of cognition in cognitive clauses:

She envisaged [[ants crawling up her leg]]

as the Phenomenon of desideration in desiderative clauses:

She wouldn't want [[ants crawling up her leg]]

and as the Phenomenon of emotion in emotive clauses:

She hated [[ants crawling up her leg]].

Recommending Work Inconsistent With IFG

 Martin, Matthiessen & Painter (2010: 239, 271):

The ‘minimal length’ approach will allow for a consistency in the analysis, but has the disadvantage that some logico-semantic relations (particularly of elaboration) are likely to be missed. This may not be considered too serious a drawback if the text is also being analysed at the discourse-semantic level, in terms of CONJUNCTION. (See Chapter 7; also IFG2 9.4, IFG3 9.3 and Martin 1992: Chapter 4; Martin & Rose 2007: Chapter 4.) …

Further reading

Martin, J R 1992 English text: system and structure. Chapter 4. Amsterdam: Benjamins.

[The semantic resources of conjunctive relations are discussed here, with the clause complex as one type of realisation.]

Blogger Comments:

To be clear, Martin's discourse semantic system of CONJUNCTION is inconsistent with IFG in that it is a confusion of two grammatical systems: clause complexing (logical metafunction) and cohesive conjunction (textual metafunction). Evidence here (Martin 1992) and here (Martin & Rose 2007).

Univariate Structure

Martin, Matthiessen & Painter (2010: 230):

The kind of structure created in this way is a univariate one, a series of repetitions of the same grammatical unit (here, a clause), each related to another in various ways.

Blogger Comments:

A univariate structure is generated by recurrence of the same function. Halliday (1994: 193):

We refer to this kind of structure as a univariate structure, a structure generated by the recurrence of the same function: α is modified by β, which is modified by γ, which is ... .

To be clear, univariate structures do not necessarily involve repetitions of the same grammatical unit, as the authors later (p231) acknowledge. The units in question merely serve the same function, as when an adverbial group and a prepositional phrase in a complex serve as circumstance:

He threw it enthusiastically but without skill.

Problems With 'Further Reading'

Martin, Matthiessen & Painter (2010: 229):

Further reading

Matthiessen 1996 develops the reasoning around IFG’s approach to English tense in some detail; …

Matthiessen, Christian M I M 1996 Tense in English seen through systemic-functional theory. M Berry, C Butler & R Fawcett & G Huang (eds), Meaning and Form: systemic functional interpretations. (Volume 2 in Meaning and Choice in Language: studies for Michael Halliday) Norwood, NJ: Ablex. 431-498.

Blogger Comments:

To be clear, the misunderstandings of verbal group logical structure that have been identified in the preceding posts can also be found in the recommended reading Matthiessen (1996):

The logical structure of the verbal group realises the system of TENSE, so, because the Event does not constitute a choice in tense, it does not feature as an element in representations of logical structure; see Halliday & Matthiessen (2014: 411n).

The tense of the first of these verbal groups is 'present in future in past in future', and so makes 4 tense choices, not 5:

The tense of the second verbal group is 'present in past in future', and so makes 3 tense choices, not 4:

Misrepresenting The Mapping Of Verbal Group Structures

Martin, Matthiessen & Painter (2010: 227):

A verbal group realises a process in the clause structure, or, from an interpersonal perspective, a Finite + Predicator element. Verbal groups have a possible multivariate structure of Finite + Auxiliary/ies + Event, as in [Finite:] have [Aux:] been [Event:] avoiding, which maps onto a univariate structure in which the Finite is Head: [α:] have [β:] been [γ:] avoiding. Finites may realise tense or modality and Auxiliaries may realise secondary tense and/or passive.

 

Blogger Comments:

This is misleading because it misrepresents the mapping of the experiential ('multivariate') and logical structures of the verbal group, since the realisations of tense in the logical structure do not correspond to the words that make up the verbal group. Halliday (1994: 198, 199):

However, the structural labelling of the words that make up the verbal group is of limited value, not only because the meaning can be fully represented in terms of grammatical features (of tense, voice, polarity and modality), but also because it is the logical structure that embodies the single most important semantic feature of the English verb, its recursive tense system, and the elements of the logical structure are not the individual words but certain rather more complex elements.

In the authors' example, the elements of the logical structure are -s, have…-en and be…-ing:

Rebranding Pre-Deictic And Pre-Numerative As Focus

Martin, Matthiessen & Painter (2010: 226):

In addition to the functions shown in the table (Deictic, Post-Deictic, Numerative, Epithet, Thing, Classifier, Qualifier), the meaning potential of the group also includes the possibility of a Focus function (e.g. the tip of the iceberg, two cups of black coffee, a dream of a holiday job).

Blogger Comments:

As previously explained, the authors' 'Focus' is an unnecessary, internally-inconsistent, and poorly named rebranding of Halliday's Pre-Deictic and Pre-Numerative.

Misrepresenting Identifiability As A Textual Function Of Nominal Group Structure

Martin, Matthiessen & Painter (2010: 225):

For example, the textual metafunction deals with thematicity at clause rank (in terms of thematic structure) but with identifiability at group rank (in terms of deictic items, as in this hall vs. a hall).

Blogger Comments:

This is misleading because it misrepresents identifiability as the group rank analogue of thematicity at clause rank.

Firstly, in the nominal group, the textual metafunction is realised throughout the entire structure. Halliday (1994: 191):

Textual meaning is embodied throughout the entire structure, since it determines the order in which the elements are arranged, as well as patterns of information structure just as in the clause … .

Secondly, the initial position of both Theme and Deictic reflects the same textual principle. Halliday (1994: 187):

So the principle that puts the Theme first in the clause is the same as that which puts the Deictic first in the nominal group: start by locating the Thing in relation to the here-&-now – in the space-time context of the ongoing speech event.

But, importantly, this the structural deictic function — the system of DETERMINATION — not the cohesive reference function ("identifiabilty"). That is, the authors have once again confused determination with reference. Halliday & Matthiessen (2014: 365):

The Deictic element indicates whether or not some specific subset of the Thing is intended; and if so, which. The nature of the Deictic is determined by the system of DETERMINATION. The primary distinction is between (i) specific or (ii) non-specific.

Thirdly, the irrelevant notion of 'identifiability' is itself a misunderstanding of cohesive reference, because it misconstrues the recoverable identity of a reference item as the identifiability of the nominal group featuring the reference item.

Confusing Formal Constituency With Function Structure

Martin, Matthiessen & Painter (2010: 225):

Groups and phrases are units of the rank below that of the clause. That is, they are the smaller units that realise the functional parts – the constituents – of the clause.

Blogger Comments:

To be clear, here the authors confuse formal constituency — the rank scale — with function structure. It is the groups and phrases that are the constituents of the clause. The "functional parts" are the elements of function structure that the constituents realise.

Problems With Exercise 4: Analysing (Projecting) Verbal Group Complexes

Martin, Matthiessen & Painter (2010: 221-2):

Exercise 4 Analysing (projecting) verbal group complexes

Analyse the verbal group complexes in the following clauses, and then the verbal groups making up the complexes: 

Blogger Comments:

This demonstration analysis is misleading because it misrepresents the logical structure of a verbal group as including an Event, and misrepresents the non-finite verbal group by assigning it a logical structure, despite there being no tense to be realised. According to SFL Theory, the analyses are as follows:

Problems With Exercise 3: Analysing (Expanding) Verbal Group Complexes

Martin, Matthiessen & Painter (2010: 221):

Exercise 3 Analysing (expanding) verbal group complexes

Analyse the verbal group complexes in the following clauses, and then the verbal groups making up the complexes:

Blogger Comments:

The demonstration analysis is misleading because it misrepresents the structures of both the verbal group complex and its verbal groups. The two verbal groups in the complex are could make and dream, and only the finite verbal group has tense, and thus, a logical structure:

Problems With Exercise 2: Analysing Elliptical Groups

Martin, Matthiessen & Painter (2010: 221):

Exercise 2 Analsying [sic] elliptical groups

Analyse the elliptical verbal groups in the responses in the following examples.

Blogger Comments:

This example analysis is misleading, because it misrepresents the logical structure of this elliptical verbal group in two ways. Firstly, the structure realises future tense, not modality, because it is not an expression of inclination ('want to'), obligation ('has to') probability ('maybe') or usuality ('sometimes'). Secondly, as an expression of the simple future, there is only one tense choice, and only one structural element:

A Problem With Exercise 1: Analysing Groups With More Complex Tenses

Martin, Matthiessen & Painter (2010: 220):

Exercise 1 Analysing groups with more complex tenses

Analyse the verbal groups in the following clauses:

Blogger Comments:

Here again the authors misrepresent the logical structure of the verbal group by including the Event as an element, and thereby mislead the reader to do likewise. Halliday & Matthiessen (2014: 411n):

A major point of difference between the verbal group and the nominal group is that the Event (unlike the Thing) is not the point of departure for the recursive modifying relationship. Hence it does not figure as an element in the notation.