David Rose Misrepresenting Axis

David Rose replied to Ed McDonald on 1/8/22 at 20:42:

Axis pairs perceivable forms with conceivable meanings, which makes semiosis possible. Turning your question around, what the others all have is axis. It’s also what makes SFL systemic.

and on 2/8/22 at 12:41:

I liked my bon mot ‘Axis pairs perceivable forms with conceivable meanings, which makes semiosis possible.’

Blogger Comments:

[1] This is misleading, because it is untrue. In SFL Theory, axis is the local dimension whose orders are paradigmatic and syntagmatic, with system as the dimension of paradigmatic order and structure as the dimension of syntagmatic order (Halliday & Matthiessen 2014: 20, 32).

Importantly, axis does not "pair forms with meanings". Form and meaning are located on both axes in the lexicogrammar. (There is meaning, but not form, on the semantic stratum, and form, but not meaning, on the expression plane.) Lexicogrammatical form is modelled as a rank scale, and rank units are interpreted in terms of their function in realising meaning. Paradigmatically, each formal rank unit is the entry condition to systems of such functions, and syntagmatically, each element of function structure is realised by a formal unit of the rank below.

With regard to "perceivabilty", lexicogrammar is 'a purely abstract level of representation "in between" the two faces of the sign' (Halliday & Matthiessen 1999: 613).

[2] This is misleading. To be clear, the 'others' here are those identified by Ed McDonald:

I noted the reference to axis, and I'm a bit puzzled as to how and why this seems to have become the theoretical fons et origo. Do we really to trace everything back to that single distinction? And if so, why paradigmatic/syntagmatic (and not and / or system/text, content/form, synchronic/diachronic...)? Does it have a privileged relation to realisation and / or instantiation?

As McDonald implies, axis is but one dimension of language in the SFL model. Halliday & Matthiessen (2014: 32):

[3] This is misleading. What makes SFL systemic is the priority it gives to system over structure, just as what makes SFL functional is the priority it gives to function over form.

David Rose On The Absence Of Semantic Structures In Halliday's Model

David Rose wrote to sys-func on 1/8/22 at 17:09:

MAKH’s procedure for semantic networks was taken up by RH, who generalised their application, specifying the entry condition as messages, realised by clauses (Hasan, Cloran, Williams, & Lukin 2005)...

Naturally, the focus had to be not on a context specific semantic network, but on a language-exhaustive one, or contextually open...

What remained the same was classification of sub-types, still realised as features in grammar systems, rather than semantic structures. In RH’s 1983 system below, features classify sub-types of questions, by more general grammatical criteria than MAKH 1972.

Blogger Comments:

[1] To be clear, the procedure that Hasan (1983) took up from Halliday (1972) was simply the use of system networks to model semantics paradigmatically.

[2] This is misleading, because it is untrue. The networks in Halliday (1972) do not specify how semantic features are realised lexicogrammatically; see previous post. 

[3] This is misleading. As can be seen by examining the semantic network in Hasan (1983), the features are distinguished on the basis of meaning, not on grammatical criteria.

∞

It is important to understand the nature of Rose's argument in these posts. His claim is that Halliday's model of semantics does not include structures, and his method of arguing is to seek out old papers (1972, 1983) that were written before semantic structures were discussed in the modern framework. That is, Rose studiously ignores later evidence, such as Halliday & Matthiessen (1999), in which, for example, grammatical metaphor is modelled as a Token-Value relation between semantic structures. Halliday & Matthiessen (1999: 208-9):

It is the fact that metaphor multiplies meanings within the semantic system that opens up the possibility of metaphorical chains, with one congruent starting-point and another highly metaphorical end-point (A"' stands for A" stands for A' stands for A; e.g. 'engine failure' stands for 'the failing of an engine' stands for 'an engine failed'). The semantic system is being expanded along the dimension of the metaphorical token-value relation; but the expansion is still within the semantic system itself. 

 

As Halliday & Matthiessen (1999: 429) explain:

… in our model there are two system-structure cycles, one in the semantics and one in the lexicogrammar. Terms in semantic systems are realised in semantic structures; and semantic systems and structures are in turn realised in lexicogrammatical ones. As we saw in Chapter 6 in particular, grammatical metaphor is a central reason in our account for treating axis and stratification as independent dimensions, so that we have both semantic systems and structures and lexicogrammatical systems and structures.

∞

As Peter Medawar once wrote of another author:

Its author can be excused of dishonesty only on the grounds that before deceiving others he has taken great pains to deceive himself. 

David Rose Misrepresenting Halliday's Semantic Networks

David Rose wrote to sys-func on 1/8/22 at 16:25:

A related difference in the models is in the nature of the features. In the negotiation system, features are motivated from above by exchange type: knowledge/action, and within the same stratum, by sequencing of interactant roles. So the entry condition is [exchange], and the features are alternating exchange roles.

In contrast, semantic networks are not concerned with sequencing of roles, but with classifying subtypes of what MAKH calls ‘verbal behaviour’. They are motivated from above by ‘behaviour potential’, and from below by grammatical realisations, but not axially by exchange structure. Their entry conditions are behaviour types, and their features specify behaviours realised by instantiating clauses, below...

Blogger Comments:

 [1] This is very misleading indeed. Rose is here misrepresenting a network from Halliday (1972) as a network of Halliday's current model, and comparing it with Martin's negotiation network (Martin & Rose 2007):

[2] This is misleading, because it is untrue. The features 'knowledge' and 'action' are located within the negotiation network, and so are not 'above' the network. Rose here confuses delicacy with stratification.

[3] This is misleading, because it is untrue. The 'sequencing of interactant roles' is how features are realised structurally, and so are 'below' the system.

[4] For the serious problems with the negotiation network, see

• Problems With The Exchange Structure NEGOTIATION System
• Problems With The Authors' Negotiation System Network

[5] To be clear, in this early paper, Halliday (1972) was exploring the possibility of a sociological semantics, where context was conceived as behavioural potential, with language as one means of realising behavioural potential. This is very different from the semantics of Halliday & Matthiessen (1999) where context is conceived metafunctionally in terms of field, tenor and mode, and semantic systems are the meanings construed by lexicogrammatical systems.

[6] This is very misleading. The reason why the networks in Halliday (1972) are not concerned with the sequencing of roles or exchange structure is that they predate Halliday's system of SPEECH FUNCTION, which Martin misunderstood and rebranded as his system of NEGOTIATION. Evidence here. It is Halliday's SPEECH FUNCTION network that would be the appropriate model for comparison.

[7] This is misleading. In Halliday (1972), the entry conditions to semantic systems are not behaviour types. Instead, behavioural potential is the higher-level context that semantic systems realise. Halliday (2003 [1972]: 341):

[8] This confuses the relation between strata (realisation) with the relation between potential and instance (instantiation).

David Rose Continuing To Misrepresent Halliday On Semantic Structure

David Rose wrote to sys-func on 1/8/22 at 13:40:

So how do the models compare at this point? Three significant cleavages are apparent. The first is axis. Both models posit realisations in grammar, but MAKH suggested semantic structures were unnecessary for the ‘limited purpose’ of semantic networks...

This [grammatical system/structure] is already fairly abstract, and it may be unnecessary therefore to interpose another layer of structure between the semantic systems and the grammatical systems, given the limited purpose of the semantic systems, which is to account for the meaning potential associated with defined social contexts and settings.

Nevertheless, describing semantic structure is possible...

It is important to emphasise here that structure is defined as the 'configuration of functions', since this is abstract enough to cover semantic structure if such a thing is to be formulated. The shape of a structure may vary; we may express it literally or hierarchically or simultaneously. But all such shapes have in common the property of being configurations of functions... All that the term 'structure' implies is that there will be some configurations of functions at that stratum, and that these will realise the meaning selections, the combinations of options in the meaning potential.

Accordingly, discourse semantic systems are motivated by structural variations, such as the negotiation system. Here from WWD, first with realisation statements, then with examples...

Blogger Comments: 

[1] This is misleading, because it is untrue. Halliday's semantics features both system and structure. Halliday & Matthiessen (1999: 429):

… in our model there are two system-structure cycles, one in the semantics and one in the lexicogrammar. Terms in semantic systems are realised in semantic structures; and semantic systems and structures are in turn realised in lexicogrammatical ones. As we saw in Chapter 6 in particular, grammatical metaphor is a central reason in our account for treating axis and stratification as independent dimensions, so that we have both semantic systems and structures and lexicogrammatical systems and structures.

Moreover, Rose not only cites Halliday (1972) to claim that the current model does not include semantic structures, but he misrepresents the content of the paper. In this early exploratory paper, Halliday is actually equivocal about the theoretical utility of semantic structures, as the text Rose omitted from his first quote makes plain. Halliday (2003 [1972]: 347):

This is already fairly abstract, and it may be unnecessary therefore to interpose another layer of structure between the semantic systems and the grammatical systems given the limited purpose of the semantic systems, 'which is to account for the meaning potential associated with defined social contexts and settings.

On the other hand, it is possible that one might be able to handle more complex areas of behaviour by means of a concept of semantic structure. It may be, for instance, that the study of institutional communication networks, such as the chain of command or the patterns of consultation and negotiation in an industrial concern, might be extended to a linguistic analysis if the semantic options were first represented in semantic structures — since the options themselves could then be made more abstract. Various complex decision-making strategies in groups of different sizes might become accessible to linguistic observation in the same way. But for the moment this remains a matter of speculation.

[2] This is very misleading indeed, because it falsely implies that Martin alone took up Halliday's idea of semantic structure.

[3] To be clear, the discourse semantic system of NEGOTIATION (Martin 1992) is a rebranding of misunderstandings of Halliday's semantic system of SPEECH FUNCTION. Evidence here.

Regarding Figure 7.2 in Martin & Rose (2007), see

• Problems With The Exchange Structure NEGOTIATION System

Regarding Figure 7.6 in Martin & Rose (2007), see

• Problems With The Authors' Negotiation System Network

David Rose Misrepresenting Halliday On Meaning Potential

David Rose wrote to sys-func on 30/7/22 at 21:06:

This para bears reviewing as it seems pivotal to the history of the grammatico-semantic model...

The term «meaning» has traditionally been restricted to the input end of the language system: the «content plane», in Hjelmslev's terms, and more specifically to the relations of the semantic interface, Hjelmslev's «content substance». We will therefore use «meaning potential» just to refer to the semantic options (although we would regard it as an adequate designation for language as a whole).

Interestingly, linguistic tradition is presented as point of departure, and the reason for restricting the use of ‘meaning potential’ to the semantic stratum. 

Its conflict with MAKH’s theoretical position is added almost apologetically in brackets. Was it too radical to say out loud that language as a whole makes meaning? At any rate, associating meaning with just one stratum, for whatever reason, had far reaching consequences for the model’s development.

Blogger Comments:

[1] This is misleading, because it is untrue. The quote from Halliday (1972) merely specifies how the term 'meaning potential' will be used in the paper, relative to traditional usages of 'meaning' (interpreted in terms of Hjelmslev's model). It does not present linguistic tradition as the reason for the how the term 'meaning potential' will be used.

[2] This very misleading because it is very untrue. There is no conflict here. SFL models language as meaning potential, in terms of the dimension of instantiation, and stratifies the content plane of that potential into two levels of symbolic abstraction: meaning (semantics) and the wording (lexicogrammar) that realises meaning.

[3] Here Rose repeats Martin's confusion of semogenesis (making meaning) with stratification (wording realising meaning), encapsulated in Martin's characterisation of stratification as 'all strata make meaning'. As a consequence, Martin (1992) even misconstrues phonology as a stratum of meaning.

[4] This is potentially misleading. Even though semantics is the stratum of meaning, lexicogrammar is the stratum whose forms, modelled as a rank scale, are interpreted in terms of their functions in realising meaning. As Halliday & Matthiessen (2014: 49) say:

it is a semanticky kind of grammar. But the focus of attention is still on the grammar itself.

David Rose Misrepresenting Halliday's Development Of SFL Theory

David Rose wrote to sys-func on 29/7/22 at 13:21:

As quoted earlier, axis was the breakthrough that opened up metafunctions and rank scale system/structure cycles, which I’ve tried to symbolise with this diagram (hexagons borrowed from Giacomo Figueredo)...

But there was a tension with the earlier model predating axis, that assumed the abstraction relation lay between grammar as form and semantics as substance, and kept being repeated alongside axis...

The term «meaning» has traditionally been restricted to the input end of the language system: the «content plane», in Hjelmslev's terms, and more specifically to the relations of the semantic interface, Hjelmslev's «content substance». We will therefore use «meaning potential» just to refer to the semantic options (although we would regard it as an adequate designation for language as a whole).

This historically nuanced conditional formulation then gets reduced to a catchphrase...

Grammar is what the speaker CAN SAY, and is the realisation of what he means. Semantics is what he CAN MEAN

...which is later cemented as the commonsensical triad ‘meaning, wording, sounding’.

Blogger Comments:

[1] This is potentially misleading. Halliday (2002: 12) summarises the development of his theorising as follows:

The steps that have seemed to me perhaps most critical in this endeavour might be summarised as: the unity of lexicogrammar; the priority of the view ‘from above’, from meaning and function; the move into systemics (system networks), freeing the grammar from the restrictions imposed by structure; the metafunctional foundation, disentangling the strands of meaning that are woven together in the syntax; the construction of language by children, from protolanguage to mother tongue; the decoupling and recoupling of lexicogrammar and semantics – the phenomenon of grammatical metaphor; the conceptualising of the relation between system and text (instantiation) and the probabilistic nature of linguistic systems.

That is, according to Halliday, it was not axis, but giving priority to the view 'from' above (meaning and function over form), and to system over structure, that opened up the development of Systemic Functional Grammar.

[2] Some of the problems with this diagram were identified in the earlier post David Rose Misunderstanding Generalisation And Abstraction.

[3] To be clear, the notion of axis, system vs structure, has appeared in Halliday's work since his first paper Some Aspects of Systematic Description and Comparison in Grammatical Analysis (Halliday 1957), though the term 'axis' was not used until Class in Relation to the Axes of Chain and Choice in Language (1963). 

That is, there is no Hallidayan model that predates the notion of axis, and the model that predates the term 'axis' is not Systemic Functional Grammar, but a different theory: Scale & Category Grammar, expounded in Categories of the Theory of Grammar (Halliday 1961). Any supposed tension then, is between a discarded theory and its replacement.

[4] This bare assertion bears no relation to the quote from Halliday (1972) that follows it as its support. The quote makes no mention of axis, and merely specifies how the term 'meaning potential' will be used, relative to traditional usages of 'meaning' (interpreted in terms of Hjelmslev's model).

[5] To be clear, the quote from Halliday (1972) that follows this misrepresentation of the previous quote as a 'historically nuanced conditional formulation' is not a reduced "catchphrase" of it. Where the previous quote is concerned with the term 'meaning potential', the quote that follows is concerned with the distinction between lexicogrammar and semantics, explaining that lexicogrammar is what can be verbally projected (say), semantics is what we can mentally project (mean), and the former realises the latter.

[6] To be clear, there is no "cementing" here, and these folk terms also cover phonology (sounding), which is not mentioned in the quotes that precede.

David Rose Misrepresenting Halliday (1972) On Semantic Systems

David Rose wrote to sys-func on 29/7/22 at 8:29:

MAKH explains here (1972) why it’s unnecessary to postulate semantic systems that restate grammatical systems...axis makes them already ‘semantic’...

The combination of system and structure with rank leads to a fairly abstract grammar (fairly 'deep', in the Chomskyan sense) and enables us to specify fairly accurately in theoretical terms - though not or course in rule-of-thumb terms - just how abstract it is. In principle, a grammatical system is as abstract (is as 'semantic') as possible given only that it can generate integrated structures; that is, that its output can be expressed in terms of functions which can be mapped directly on to other functions, the result being a single structural 'shape' (though one which is of course multiply labelled).

Blogger Comments:

[1] This is misleading, because it is untrue. On the one hand, this is an extract from Halliday's very early paper which considers arguments for and against the theoretical value of including semantic structure in the model. In the time since then, semantic structure has been part of SFL Theory. Halliday & Matthiessen (1999: 429):

… in our model there are two system-structure cycles, one in the semantics and one in the lexicogrammar. Terms in semantic systems are realised in semantic structures; and semantic systems and structures are in turn realised in lexicogrammatical ones. As we saw in Chapter 6 in particular, grammatical metaphor is a central reason in our account for treating axis and stratification as independent dimensions, so that we have both semantic systems and structures and lexicogrammatical systems and structures.

On the other hand, Halliday does not simply postulate 'semantic systems that restate grammatical systems'. Halliday & Matthiessen (1999: 26):

Thus when we move from the lexicogrammar into the semantics, as we are doing here, we are not simply relabelling everything in a new terminological guise. We shall stress the fundamental relationship between (say) clause complex in the grammar and sequence in the semantics, precisely because the two originate as one: a theory of logical relationships between processes. But, as we have shown, what makes such a theory (i.e. an ideation base as the construal of experience) possible is that it is a stratal construction that can also be deconstructed, every such occasion being a gateway to the creation of further meanings which reconstrue in new and divergent ways. Thus a sequence is not 'the same thing as' a clause complex; if it was, language would not be a dynamic open system of the kind that it is.

[2] To be clear, this misunderstands the extract from Halliday (1972). Halliday is here explaining that the system-structure relation (realisation) enables a systemic grammar to be more abstract — where 'more abstract' means 'more semantic' — since system is more abstract than (is realised by) structure.

Importantly, a functional grammar interprets grammatical form (e.g. verbal group) in terms of it function in realising meaning (e.g. Process). Halliday & Matthiessen (2014: 48-9):

We cannot expect to understand the grammar just by looking at it from its own level; we also look into it ‘from above’ and ‘from below’, taking a trinocular perspective. But since the view from these different angles is often conflicting, the description will inevitably be a form of compromise. … 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. Giving priority to the view ‘from above’ means that the organising principle adopted is that of system: the grammar is seen as a network of interrelated meaningful choices.

David Rose On Edelman's Theory of Neuronal Group Selection

David replied to Kieran McGillicuddy on sys-func on 28/7/22 at 17:27:

Re circularity, generalising seems to be a basic capacity of any organism: ‘food/not-food’, and later ‘food/mate’

Edelman explains how it happens neurologically in animals, as perception, and in larger brained animals as perception of perception.

and at 21:22:

To mansplain what [Edelman] means by perception of perception I’d have to drag up his whole theory of neuronal group selection.

Aside from his books there’s a bunch of papers that summarise it. You could start here which is free to download...

Edelman, G. M. (2003). Naturalizing consciousness: a theoretical framework. Proceedings of the National Academy of Sciences, 100(9), 5520-5524.

Blogger Comments:

[1] To be clear, ‘food/not-food’ and ‘food/mate’ are (linguistic) discriminations, not generalisations. 

[2] On Edelman's model, a perceptual category is established through instances being determined as alike in terms of both value-weighted perceptual features and adaptive motor response. This is achieved neurally through repeated value-guided firings of neuronal groups in global mappings for each instance. Each firing strengthens the synaptic connections within the group, making the group more likely — selecting it — to fire again under similar conditions in the future.

[3] To be clear, Edelman (2003) makes no mention of "perception of perception", but in Edelman's model, the brain further categorises its own categorising performances. For example, in categorising perceptual categorising, the brain creates what Edelman terms 'concepts', but what SFL Theory would interpret as perceptual systems (perceptions organised into systems).

[4] For interpretations of Edelman's Theory of Neuronal Group Selection in terms of SFL Theory, see the 72 posts here.

David Rose Endorsing A Contradiction Of His Own Claim

David Rose wrote to sys-func on 28/7/22 at 15:20:

Note that generalisation increases down strata (not up), Martin & Matthiessen 1991...

The most powerful generalisation is, arguably, that there is a tendency towards generalisation as we move down either the stratal organisation or the rank scale. That is, to put it crudely, a given number of semantic systems is realised by a smaller number of corresponding lexicogrammatical ones; similarly, a given number of lexicogrammatical systems is realised by a smaller number of phonological ones. And the same holds for the rank scale; for example, a given number of clausal systems is realised by a smaller number of verbal group systems. The reason for this state of affairs is easy to see. A stratal descent or a rank descent always entails a generalisation across contexts and this generalisation is reflected in the relatively smaller number of realising systems.

Blogger Comments:

To be clear, here Rose unwittingly provides further evidence against his own previous claim (here) that 

[Halliday's] grammatico-semantic model assumes a bijective or one-to-one relation between LG features and semantic features…

David Rose Continuing To Misrepresent The Semantic System Of Modality

David Rose wrote to sys-func on 27/7/22 at 8:07:

Rather than a ‘semantic system of modality’, Martin & Matthiessen 1991 explain such motifs as topologies, or ‘fractal principles’ that recur across ranks and strata. They are generalisations of types of meanings that complement the typological generalisations of systems. Crucially they are not more abstract than the systems they generalise across. So does postulating a ‘semantic system of modality’ above the grammatical systems of modality, confuse abstraction with generalisation?

In any case these ‘semantic systems’ were not proposed in IFG1/2 but have been added to IFG3/4.

Blogger Comments:

[1] This is misleading, because it is untrue. Martin & Matthiessen (1991) are concerned with the semantic system of modality and the diverse ways it is realised in lexicogrammar. Martin & Matthiessen (1991: 376):

With MODALISATION, POLARITY in propositions is used to bound the semantic space under consideration, with degree of usuality and probabilily defining parameters between 'it is' (yes) and 'it isn't' (no). Similarly, obligation and inclination are used as parameters grading the semantic space in proposals between 'do' and 'don't'.

[2] This is misleading. Martin & Matthiessen (1991) don't explain 'such motifs' as topologies. The paper contrasts two perspectives in modelling: the typological and topological.

[3] This is misleading. The one instance of the word 'fractal' (p376) refers only to the system of EXPANSION.

[4] This is misleading, because it is untrue. With regard to expansion, Martin & Matthiessen (1991: 375) write:

All we can do at present is posit a generalised system that we 'abstract' from the specific grammatical environments in which it is manifested.

To be clear, logically, for there to be congruent and metaphorical grammatical realisations of MODALITY, MODALITY must be a more abstract system at the level of semantics that is realised in lexicogrammar, since congruent vs metaphorical relations obtain between strata.

[5] No. The confusion here is entirely Rose's. See [4].

[6] This is still misleading, because it is still untrue. Halliday (1994: 356):

The reason this area of the semantic system is so highly elaborated metaphorically is to be found in the nature of modality itself … Modality refers to the area of meaning that lies between yes and no – the intermediate ground between positive and negative polarity.

David Rose Misrepresenting IFG On Modality

David Rose wrote to sys-func on 25/7/22 at 17:44:

Thanks Chris, for another opportunity to compare the models.

Again, MODALITY is a described grammatical system, at clause rank in English, but the same term is used below for a semantic ‘domain’. And again, the account in IFG 1/2 differed from the account in IFG 3/4. In IFG1/2, MAKH described the implicit/explicit: subjective/objective orientations topologically, as interpersonal metaphors of modality. But in IFG 3/4 they are presented instead as features in the MODALITY system, within the MOOD system (Fig 4-13, Fig 4-23). The entry condition for this system is [clause], but the metaphorical realisations of subjective and objective orientations are projections and embedding (I think.../it is likely that...), blurring the line between grammar and semantics.

The described discourse semantic system is APPRAISAL: ENGAGEMENT: heterogloss, for which MODALITY is one resource to introduce additional voices into a discourse, alongside PROJECTION and CONCESSION.

Blogger Comments:

[1] This misleading. Halliday (± Matthiessen) explicitly describes MODALITY as a semantic system. Halliday (1994: 356) and Halliday & Matthiessen (2014: 691):

The reason this area of the semantic system is so highly elaborated metaphorically is to be found in the nature of modality itself … Modality refers to the area of meaning that lies between yes and no – the intermediate ground between positive and negative polarity.

The fact that it is construed as a semantic system is also demonstrated by it having both congruent and metaphorical grammatical realisations. 

[2] This is misleading, because it is untrue. Less importantly, it is only the explicit ORIENTATION that is metaphorical (Halliday 1994: 362). More importantly, the topological representation in IFG1/2 (and IFG3/4) does not describe MODALITY systems, but the relation of modality to polarity and mood. Halliday (1994: 367):

[3] This is misleading, because MODALITY is presented as a system network in IFG1/2 as well as IFG3/4. Halliday (1994: 358, 360):

Cf. Figure 4-23 in Halliday & Matthiessen (2014: 182):

[4] This is misleading. By including the system of MODALITY in his grammatical network of MOOD — Figure 4-15 in IFG3, Figure 4-13 in IFG4 — Matthiessen is interpreting grammatical form (modelled as a rank scale) in terms of its function in realising meaning, consistent with the modus operandi of SFL Theory. Halliday & Matthiessen (2014: 162):

Contrary to Rose's claim, the rank of clause is the appropriate entry condition for clause rank complexing (projection) and  clause rank elements realised by embedded clauses.

[5] For some of the misunderstandings in Working with Discourse (Martin & Rose 2007) in this regard, see:

• On The Heteroglossic Function Of Usuality
• On The Heteroglossic Function Of Usuality, Obligation, Inclination And Ability
• On The Heteroglossic Function Of Obligation
• On The Heteroglossic Function Of "Projections", "Polarity" And "Modality"
• On The Heteroglossic Function Of "Concessive But"
• On The Heteroglossic Function Of "Concession"
• On The Heteroglossic Function Of "Concessive Conjunctions"

Importantly, projection is not included as a logical system in discourse semantics (Martin 1992, Martin & Rose 2007). This is because the original intellectual source of Martin's CONJUNCTION is the system of cohesive conjunction in Halliday & Hasan (1976), and projection does not function cohesively. So, not only does discourse semantics fail to account for the semantics of projection, it also disables any attempt to account for grammatical metaphor in which projection is featured.

David Rose Misrepresenting Halliday On Semantic Structure

David Rose wrote to sys-func on 19/7/22 at 10:24:

Re semantics/grammar relations and axis...

Halliday (1972), cited in Hasan, Cloran, Williams, & Lukin (2005), is a key document in the history of SFL, as it is the foundational text for the semantic network research project, and excuses semantic networks from the axial realisation statements required for grammar and phonology. The key discussion on semantics/grammar relations are extracted here (p21-24). It concludes by suggesting approaches to describing semantic system and structure, but this was ‘a matter of speculation’. This conclusion could also be taken as foundational for the study of discourse semantics.

Halliday, M. A. K. (1972). Towards a sociological semantics. Urbino: Università di Urbino.

(cited as Halliday 1973b in...

Hasan, R., Cloran, C., Williams, G., & Lukin, A. (2005). Semantic networks: The description of linguistic meaning in SFL. In Continuing discourse on language: a functional perspective (pp. 697-738). Equinox Publishing.

Blogger Comments:

[1] This is misleading. In this very early paper from 1972, Halliday is considering arguments for and against the theoretical value of including semantic structure in the model. In the time since then, semantic structure has been part of SFL Theory. Halliday & Matthiessen (1999: 429):

… in our model there are two system-structure cycles, one in the semantics and one in the lexicogrammar. Terms in semantic systems are realised in semantic structures; and semantic systems and structures are in turn realised in lexicogrammatical ones. As we saw in Chapter 6 in particular, grammatical metaphor is a central reason in our account for treating axis and stratification as independent dimensions, so that we have both semantic systems and structures and lexicogrammatical systems and structures.

[2] This is misleading. On the one hand, in the model of discourse semantics (Martin 1992), the system networks do not include realisation statements that specify discourse semantic structures. On the other hand, in the model of discourse semantics (Martin 1992), the notion of structure is deployed inconsistently. 

For NEGOTIATION, the system derived from Halliday's semantic system of SPEECH FUNCTION, structure is the internal syntagmatic dimension of a unit, which is consistent with the notion of structure within and beyond SFL Theory. 

However, for IDENTIFICATION, IDEATION and CONJUNCTION, the systems derived from the lexicogrammatical systems of COHESIVE REFERENCE, LEXICAL COHESION and COHESIVE CONJUNCTION, units do not have internal structure, but instead link up to each other to form structures. This anomaly arises from the fact that cohesive relations are not structural relations.

David Rose Misrepresenting Axis As Key To Semantics-Grammar Relations

David Rose wrote to sys-func on 18/7/22 at 21:11:

Axis is key...

Michael Halliday is the only linguist who has given priority to the paradigmatic axis and organised the theory and description of language around this axis... but at the same time, he maintains the connection with the syntagmatic, by means of realisation statements specifying fragments of structure, and the account of structure he developed is a sophisticated unification of different metafunctional strands of organisation.

Matthiessen, C. M., Wang, B., Ma, Y., & Mwinlaaru, I. N. (2022). Systemic Functional Insights on Language and Linguistics. Springer

...two very basic theoretical concepts in SFL, metafunction and rank, are in fact based on the even more fundamental dimension of axis... stratification, alongside metafunction and rank, is based on way in which systems bundle together with one another - axis in other words is key.

Martin, J. R., Wang, P., & Zhu, Y. (2013). Systemic functional grammar: a next step into the theory–axial relations. Beijing: Higher Education.

In broad terms, a system is a set of relations of abstraction and generality. Its features are generalisations based on recurrent similarity of function at higher strata. From this instantial perspective, a system is an accumulation of instances over time scales of phylogenesis and ontogenesis.

Rose, D. in press. Designing pedagogic registers: Reading to Learn. In J.R. Martin, D. Caldwell & J. Knox (eds.) Developing Theory: A Handbook in Appliable Linguistics and Semiotics. London: Bloomsbury, 103-125

Blogger Comments:

[1] To be clear, this quote from Matthiessen et al. (2022) is a statement on how SFL gives priority to the paradigmatic axis, with metafunctional structures of the syntagmatic axis specified by realisation statements associated with features in systems.

[2] To be clear, this quote from Martin et al. (2013) makes the false claims that stratification is based on the way systems bundle together, and that axis is fundamental to stratification, metafunction and rank.

On the first claim, stratification is based on the principle of realisation. Different strata constitute different levels of symbolic abstraction.

On the second claim, axis, stratification, metafunction and rank are distinct dimensions, each with distinct ordering principles, independent of those of the other dimensions. Halliday & Matthiessen (2014: 20,32):

[3] Some of the problems with this diagram were identified in the earlier post David Rose Misunderstanding Generalisation And Abstraction.

[4] To be clear, this quote from Rose (in press) presents several misunderstandings. Regarding the first sentence, it is delicacy that is the ordering principle of system, whereas (symbolic) abstraction is the relation between systems of different strata.

The second sentence confuses instantiation ('recurrent') with stratification ('higher strata'). To be clear, the features of a system represent the potential choices of instantiation for a specified entry condition.

Regarding the third sentence, a system is built up at all three timeframes, including that of logogenesis, 'the creation of meaning in instantiation, maintained as a changing instantial system'. Halliday & Matthiessen (1999: 384-5):

If we look at logogenesis from the point of view of the system (rather than from the point of view of each instance), we can see that logogenesis builds up a version of the system that is particular to the text being generated: the speaker/ writer uses this changing system as a resource in creating the text; and the listener/ reader has to reconstruct something like that system in the process of interpreting the text — with the changing system as a resource for the process of interpretation. We can call this an instantial system. … the instantial system is built up successively by the generation process; but as it is developed, it in turn becomes a resource for further instantiation.

David Rose Misrepresenting The Semantic Systems Of Expansion And Figuration

David Rose wrote to sys-func on 18/7/22 at 16:05:

it’s always good to read IFG closely (and critically)...

It’s interesting that the terms FIGURATION and EXPANSION in Figure 10-2 below are printed in small caps, as though they are systems. It’s noteworthy that Halliday 1985/1994 (IFG 1/2) didn't describe expansion in general as a system, but as a ‘semantic motif’ that is dispersed across the grammar. A system in SFL is defined by axial relations between features and their realising structures. expansion is such a system in the grammar, at clause rank within clause complexing (IFG 7.4), and at group/phrase rank within verbal group complexing (IFG 8.6). But it is not a described semantic system.

‘Figuration’ on the other hand corresponds to a described semantic system, of lexical relations within figures, entitled nuclear relations in ET 5.3, WWD 3.3. Figures are typically realised as clauses (IFG 5.1.1)...

The canonical description of nuclear relations, and their stratal relations to grammar and field, is

Hao, J. (2020). Analysing scientific discourse from a systemic functional linguistic perspective: A framework for exploring knowledge building in biology. Routledge.

Jing also carefully relates the discourse semantic ideation systems to the ideational semantics in Construing Experience.

Rose soon qualified this at 17:24:

Correction...

Hao, J. (2020), re-analyses nuclear relations as the systems of FIGURE and ENTITY, to clearly distinguish systems of LG, DS and field.

 

Blogger Comments:

[1] To be clear, even a casual glance at IFG reveals that the caption of Figure 10-2 identifies FIGURATION and EXPANSION as semantic systems.

[2] This is misleading. Halliday (1985, 1994) describes expansion as a type of logico-semantic relation, along with projection, that has three subtypes: elaboration, extension and enhancement. The term EXPANSION appears in small capitals (e.g. 1994: 219), the convention for representing system names. This constitutes a verbal description of a system rather than a diagrammatic representation of a system as a network. Halliday (1985, 1994) did not feature any system networks.

Moreover, a close reading reveals that the term 'semantic motif' does not appear until Halliday & Matthiessen (1999: 72, 223, 562, 612), where it explicitly refers to systems. Halliday & Matthiessen (1999: 223):

[3] This is misleading because a system is not 'defined by axial relations'. in SFL Theory, system constitutes the dimension of paradigmatic order (Halliday & Matthiessen 2014: 20). This a distinct dimension from structure, the dimension of syntagmatic order (ibid.), and axis is the dimension whose orders are the paradigmatic and syntagmatic (op. cit.: 32).

[4] To be clear, Halliday's Introduction to Functional Grammar (1985, 1994) is focused on grammar, not semantics, which means modelling grammatical form in terms of its function in realising meaning. A focus on semantics, mainly ideational, is provided by Halliday & Matthiessen (1999).

[5] This is very misleading indeed. As Figure 10-2, above, makes clear, FIGURATION is the semantic system that is congruently realised by the grammatical system of TRANSITIVITY. Martin's discourse semantic system of IDEATION, which is a rebranding of lexical cohesion, but within which 'nuclear relations' are modelled, does not feature the figure as either a system or structural unit. 

For some of the problems with 'nuclear relations' in Martin (1992), see the close examination here. For some of the problems with 'nuclear relations' in Martin & Rose (1992), see the close examination here.

[6] Clearly, any "canonical" description of 'nuclear relations' lies not in the work of Martin's student, Hao, but in her source: Halliday & Mathiessen (1999). For the source of Hao's system of FIGURE, see Chapter 4 Figures (Halliday & Mathiessen 1999: 128-76), and for the source of Hao's system of ENTITY, see Chapter 5 Elements (Halliday & Mathiessen 1999: 177-226).

David Rose On Martin Crediting Hasan

David Rose replied to Annabelle Lukin on sys-func on 14/7/22 at 20:58:

My only interest is to understand relations between the models, which I suggested below may be complementary.

The MESSAGE system was quoted from Hasan 1983, in your paper with RH, Geoff and Carmel, that I promoted as ‘a useful history and synopsis’. I thought it was interesting to see how the MESSAGE and MOOD features presented as bijective differ in their systemic valeur, and how the MESSAGE features presage later developments.

The NEGOTIATION system wasn’t published until 1992, extensively credited to Margaret Berry. The APPRAISAL system was published a decade later. So it was impossible to know in 1983 that the features described as MESSAGE options were also features in NEGOTIATION and APPRAISAL systems. So certainly not a re-run.

On the other hand, the whole of discourse semantics is extensively credited to RH, including 197 mentions in ET. … 

 

Blogger Comments:

[1] This is misleading. What Rose actually claimed (here) was that Hasan's semantic system attempted to synthesise the complementarity of two discourse semantic systems:

In sum, the MESSAGE system attempts to synthesise the complementarity of two discourse semantic systems, NEGOTIATION and APPRAISAL.

[2] This is misleading, because it is untrue. As previously demonstrated here, Hasan's network allows for multiple grammatical realisations of a semantic feature.

[3] To be clear, the source of Berry's model, as well as Martin's rebranding of it as his discourse semantic system of NEGOTIATION, is Halliday's semantic system of SPEECH FUNCTION.

[4] To be clear, if there are any "reruns", they are necessarily in the later discourse semantic systems.

[5] This is misleading. What is true is that Cohesion in English (Halliday & Hasan 1976) is the data that Martin (1992) uses to theorise his discourse semantics, rebranding their REFERENCE as his IDENTIFICATION, their LEXICAL COHESION as his IDEATION, and their cohesive CONJUNCTION as his CONJUNCTION (now CONNEXION); evidence here. Given this, it is hardly surprising that Hasan should be referenced in Martin (1992).

However, the term 'credited' here is potentially misleading. See, for example:

• Misrepresenting Hasan's Work On Speech Function
• Misrepresenting Halliday & Hasan On Reference
• Misidentifying The Main Differences Between Martin And Hasan
• Misrepresenting Hasan And Confusing Strata And Metafunctions
• Confusing Text Type With Text Structure And Misrepresenting Hasan's Work On Cohesive Harmony
• Misrepresenting Hasan's Work On Cohesion
• "One Apparently Unresolved Problem With Hasan's Technique"
• Misrepresenting Hasan's Cohesive Harmony
• Presenting Misunderstandings Of Hasan's Cohesive Harmony As Deficiencies In The Model
• Misrepresenting Hasan's Work On Coherence As Formalist
• Misrepresenting Hasan On Text Structure
• Not Acknowledging Hasan As Intellectual Source
• Under-Acknowledging Hasan As Theoretical Source
• Misinterpreting Hasan And Proposing Theoretical Inconsistencies
• Strategically Misrepresenting Hasan
• Misrepresenting Hasan