Giving Priority To Structure In Devising System

 Martin (2013: 93):

Blogger Comments:

[1] This is misleading because it is untrue. The system of MODALITY represents the choices at clause rank. If one of these choices is realised metaphorically as a projecting mental clause instead of an Adjunct, it can be specified by a realisation statement attached to the feature in the network. Martin's misunderstanding here again derives from giving priority to the view 'from below', structure.

Moreover, in order to account for grammatical metaphor, it is necessary to provide both the congruent and metaphorical analyses. Halliday & Matthiessen (2004: 613-4):

Here the cognitive mental clause I don’t believe is a metaphorical realisation of probability: the probability is realised by a mental clause as if it was a figure of sensing. Being metaphorical, the clause serves not only as the projecting part of a clause nexus of projection, but also as a mood Adjunct, just as probably does.

[2] To be clear, the entry condition to clause complexing is the feature clause, and in this metaphor, modality is realised within a clause complex instead of within a clause.

Misunderstanding 'From Above' And Rebranding Speech Function As Discourse Semantics

Martin (2013: 82-3):

Blogger Comments:

[1] Again, here Martin misunderstands looking 'from above' as looking 'above' — i.e. from below. Again, this is the direct opposite of SFL methodology. Instead of being concerned with how meaning is realised, Martin is concerned with what wording realises.

[2] Here Martin misleads by rebranding Halliday's semantic system of speech function as Martin's discourse semantic system, which Martin himself proposes.

Misrepresenting Interpersonal Metaphor Of Mood

Martin (2013: 80-1):

Blogger Comments:

[1] To be clear, the 'stratal tension between discourse function and MOOD' is the incongruent (metaphorical) realisation of SPEECH FUNCTION in MOOD. 

[2] To be clear, the metaphorical realisations do not "dress themselves up lexicogrammatically as commands". Lexicogrammatically, they are just like any other indicative clauses. The difference is semantic, not lexicogrammatical, since they realise proposals (demanding goods-&-services) rather than propositions (giving or demanding information) in terms of SPEECH FUNCTION.

Misunderstanding Congruent vs Metaphorical As Encoding vs Symbolising [1]

Martin (2013: 80):

Blogger Comments:

[1] Again, here Martin misleads by misrepresenting Halliday's system of SPEECH FUNCTION as Martin's discourse semantics.

[2] To be clear, grammatical structure realises (not "reflects") choices made in the MOOD system.

[3] To be clear, 'congruent' is Halliday's technical term for the agreement between semantics and grammar.

[4] To be clear, this seriously misunderstands realisation and grammatical metaphor. Realisation is the relation of symbolic abstraction, and 'symbolising' refers to the relation between semantics and lexicogrammar, irrespective of whether the realisation is congruent or metaphorical. In encoding, the lower level of abstraction (lexicogrammar) is used to identify the higher level of abstraction (semantics), and does not depend on whether the realisation is congruent or metaphorical.

Misrepresenting Halliday's Speech Function As Martin's Discourse Semantics

Martin (2013: 78-9):

Blogger Comments:

To be clear, speech function is Halliday's system of interpersonal semantics. Here Martin misleads the reader by misrepresenting speech function as a system of his model of discourse semantics.

Misrepresenting Comment Adjuncts

Martin (2013: 76-7):

Blogger Comments:

[1] To be clear, this network is poorly wired, since there is no need for two sets of wiring from the 'declarative' and 'interrogative' features if the two more delicate systems are represented as conjunctive systems (grouped by a curly bracket).

[2] To be clear, comment Adjuncts, according to Halliday & Matthiessen (2014: 190):

express the speaker’s attitude either to the proposition as a whole or to the particular speech function. In other words, the target of the comment may be either ideational (the content of the proposition) or interpersonal (the speech function).

That is, the distinction is not one of 'feeling' vs 'dialogism', since both types express an attitude, and both types — e.g. arguably vs broadly — can serve a dialogistic function.

Martin's Criticism Of Matthiessen's Network

Martin (2013: 73):

Blogger Comments:

[1] To be clear, here, yet again, Martin adopts the opposite perspective to SFL Theory by giving priority to structure over system.  A grammatical system is devised on the basis of the meaning being realised, not the structures that realise it.

[2] To be clear, as previously explained, Matthiessen's network properly restricts recursion to just those features that can be realised iteratively. Matthiessen (1995: 232):

Martin's clause network, in contrast, allows potentially unlimited insertions of Medium and Process, and potentially unlimited selections of both 'middle' and 'effective' if the feature 'agentive' is selected. So it is Martin's network that does not account for the structures he has presented.

[3] Matthiessen's network only presents recursion in effective clauses because that was the extent of its description. As Matthiessen (ibid.) made clear:

The system network in Figure 4-14 sets out a partial specification of options and their realisations.

It is misleading of Martin to claim that Matthiessen restricts agency to effective clause, because the network is immediately preceded by instances of agency, all of which are middle clauses. Matthiessen (ibid.):

For instance:

The general kept the soldiers marching.
The general helped the soldiers march.
The general required the soldiers to march.
Well they got the army digging

∞

In summary, Martin here misleads the reader by deliberately misrepresenting Matthiessen's network, and promoting his own very poorly constructed network as successful in comparison.

Misunderstanding The Realisation Of Tense In English

 Martin (2013: 72):

Blogger Comments:

To be clear, in the English TENSE system, the elements of the logical structure are not the individual words. Halliday & Matthiessen (2014: 398, 399):

… 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.

Misrepresenting The Scope Of Recursion In The System Of Transitivity

Martin (2013: 71):

Blogger Comments:

To be clear, this system network is ill-conceived because, if the feature 'agentive' is selected, for the one clause, it allows potentially unlimited insertions of Medium and Process, and potentially unlimited selections of both 'middle' and 'effective'.

A system network that properly restricts the recursion in effective clauses is provided in Matthiessen (1995: 232):

Misunderstanding Recursive Systems And Misrepresenting 'Univariate'

Martin (2013: 70):

Blogger Comments:

[1] To be clear, this misunderstands a recursive system. The choice in a recursive system is either to go on or stop. Halliday & Matthiessen (2014: 438):

[2] To be clear, these structures are still referred to as univariate, and opposed to multivariate. Halliday & Matthiessen (2014: 390):

We refer to this kind of structure as a univariate structure, one which is generated as an iteration of the same functional relationship (cf. Halliday, 1965, 1979): α is modified by β, which is modified by γ, which is ... . By contrast, the type of structure exemplified by Deictic + Numerative + Epithet + Classifier + Thing we call a multivariate structure: a configuration of elements each having a distinct function with respect to the whole.

[3] To be clear, a complex is a type of univariate structure, one in which rank units are related to one another, as clauses in a clause complex. An example of a univariate structure that is not a unit complex is the logical structure of the verbal group, which is not a complex of rank units. Halliday & Matthiessen (2014: 398-9):

… the elements of the logical structure are not the individual words but certain rather more complex elements. … The logical structure of the verbal group realises the system of tense. … The primary tense is that functioning as Head, shown as α. This is the Deictic tense: past, present or future relative to the speech event. The modifying elements, at β and beyond, are secondary tenses; they express past, present or future relative to the time selected in the previous tense. Realisations are shown in Table 6-12.

Misrepresenting 'Rankshift' And 'Linker'

 Martin (2013: 65):

Blogger Comments:

[1] To be clear, this misunderstands 'embedded' as a later renaming of 'rankshifted', whereas both terms are current, and present different views of the same phenomenon. For example, a nominal group that is embedded in a nominal group has shifted from group rank to word rank — words being the constituents of groups.

[2] To be clear, the structural analysis misrepresents of as a linker, despite the fact that a linker is a marker of a paratactic relation, and no paratactic relation obtains between the embedded nominal group and what follows. As Halliday & Matthiessen (2004: 333) point out, the 'word of is the generalised marker of a structural relationship between nominals'. In such instances, of is a constituent of the prepositional phrase serving as Postmodifier. Halliday (1994: 196):

Note that in Matthiessen's editions of IFG, the term 'Facet' has been shifted from the embedded Thing to the embedded nominal group as a whole, and from a Pre-Deictic to a Numerative function. Halliday & Matthiessen (2004: 335):

Note also that 'Qualifier' here should read 'Postmodifier'.

Misrepresenting SFL Methodology And Nominal Group Potential

Martin (2013: 61):

Blogger Comments:

Firstly, this once again misrepresents SFL methodology by devising a system from below: on the basis of structure.

Secondly, not all nominal groups include a Thing function. For example, neither nominal group in the clause that was nice includes a Thing: the first has Deictic as Head, and the second has Epithet as Head. Martin does not account for this in any later discussion.

Thirdly, the choice of 'simple' vs 'complex' as a nominal group system is very poor pedagogically, not to say theoretically, since the notion of a simple nominal group and a complex nominal group invites confusion with a nominal group simplex and a nominal group complex. 

Misrepresenting System As Axis

Martin (2013: 59):

Blogger Comments:

This is misleading because it misrepresents system as axis. System is the dimension of the paradigmatic order of the dimension of axis, and it is system, not axis, that disclosed the notion of metafunction to Halliday. Halliday & Matthiessen (2014: 20, 32):

The misrepresentation is motivated, however, since it is presented as evidence for Martin's central claim in this monograph: that axis, misunderstood as 'axial relations', is foundational. 

Misunderstanding Delicacy And Prioritising Structure

Martin (2013: 56):

Blogger Comments:

To be clear, on the one hand, this misunderstands delicacy as "privileging". Delicacy is simply a hyponymic relation (Halliday & Matthiessen 1999: 145), and a hypernym (e.g. ape) is not more "privileged" than any of its hyponyms (e.g. human). Halliday & Matthiessen (1999: 14):

the paradigmatic network is ordered in delicacy (subsumption, classification, specialisation), from the least delicate (most general) to the most delicate (most specific types).

On the other hand, this, once again, misrepresents SFL methodology by taking the view 'from below', structural similarity, in determining how to organise a system.

Misrepresenting Polarity And Tag Potential

Martin (2013: 56-7):

Blogger Comments:

[1] To be clear, because of the unwarranted conditional marking, this network fails to account for either negative Mood Tags in negative clauses, such as So, he won't pay, won't he?, and negative exclamatives, such as How tidy you are…not!.

[2] To be clear, a realisation statement specifies the structural consequences of a systemic feature at the rank in question. If the feature 'tagged' is selected, the structural consequence is simply + Finite^Subject. Nominal and verbal group variation is accounted for at group rank, and is irrelevant to structure at clause rank.

Misrepresenting The Tagging Of Negative Clauses

Martin (2013: 55-6):

Blogger Comments:

This is misleading, because it is simply not true that negative clauses always reverse their polarity in the Tag, as demonstrated by instances like They won't pay, won't they?. Halliday & Matthiessen (2014: 173-4) provide the view 'from above':

Summary Of Exclamatives And Negative Imperatives

Martin (2013: 55):

Blogger Comments:

[1] To be clear, a proscription against negative exclamative clauses fails to account for instances like What a good boy you are…not!

[2] To be clear, the discussion of negative imperatives misrepresented Finite not as a Mood Adjunct.

See Halliday & Matthiessen (2014: 165-6) for the full paradigm of negative imperatives:

Misrepresenting The Speech Function Of Indicative Mood

Martin (2013: 52):

Blogger Comments:

To be clear, proposals are congruently realised by imperative mood. It is propositions that are congruently realised by indicative mood.

Misrepresenting Rank As Derived From System Networks

Martin (2013: 52):

Blogger Comments:

This is misleading, because it is untrue. Halliday's concept of rank was first published in Categories of a Theory of Grammar (Halliday 1961), five years before his first publication on system networks, Some Notes On 'Deep' Grammar (Halliday 1966), and nine years before the first publication on metafunctions, Language Structure and Language Function (Halliday 1970).

The rhetorical function of this misrepresentation is to support Martin's claim in this monograph that  "axial relations" are 'foundational'.

Misrepresenting Axis As the Origin Of Metafunction And Rank

Martin (2013: 52):

Blogger Comments:

[1] This gloss of the metafunctions is potentially misleading. The ideation metafunction is the construal of experience, not the construal of discourse, and the interpersonal metafunction is the enactment of interpersonal relations, not the enactment of discourse. Halliday & Matthiessen (1999: 7-8):

The ideational metafunction is concerned with construing experience — it is language as a theory of reality, as a resource for reflecting on the world. … The interpersonal metafunction is concerned with enacting interpersonal relations through language, with the adoption and assignment of speech roles, with the negotiation of attitudes, and so on — it is language in the praxis of intersubjectivity, as a resource for interacting with others. The textual metafunction is an enabling one; it is concerned with organising ideational and interpersonal meaning as discourse — as meaning that is contextualised and shared.

[2] The claim that 'SFL's distinctive approach to meaning and constituency in fact grew out of the development of axis (system/structure relations) from the early 1960s' is very misleading because it is untrue on several counts.

Firstly, Martin misunderstands axis as system/structure relations, as stipulated in realisation statements. In SFL Theory, axis is a local dimension whose two orders are the paradigmatic and syntagmatic (Halliday & Matthiessen 2014: 32), with system and structure as their respective dimensions. Realisation statements are located on the paradigmatic axis, and as the name suggests, the relation between the axes is simply realisation.

Secondly, the SFL approach to the metafunctions and constituency did not grow out of the development of either axis or system/structure relations from the early 1960's. 

On the one hand, the metafunctions are a distinct global dimension from the local dimension of axis (ibid.), and first appeared in Language Structure and Language Function (Halliday 1970). 

On the other hand, the SFL approach to constituency, a rank scale, is a distinct local dimension  from the local dimension of axis (Halliday & Matthiessen 2014: 32), and derives from the method of ranked constituency analysis, rather than immediate constituency analysis, as first set out in Categories of A Theory of Grammar (Halliday 1961), and explained more fully in Introduction to Functional Grammar (Halliday 1985).

The rhetorical significance of Martin's misleading bare assertion (the ipse dixit fallacy) is that it assumes the conclusion (the petitio principii fallacy) of the entire monograph. That is, it is one logical fallacy serving another in an invalid argument for a false conclusion.