David Rose On Exchange, Marxian Theory And Halliday (1961)

David Rose responded to Kieran McGillicuddy on sys-func on 3 June 2022 at 11:08:

Perhaps [Halliday's] greatest contribution was to show us how ‘production’ and ‘exchange’ happen simultaneously, through grammatical metafunctions, that it is in fact a co-production between speakers. But this co-production is necessarily sequenced in time – one person then another. And each person evaluates the other as they speak or gesture. That’s what the ‘exchange’ metaphor attempts to capture... to technicalise and iconise.

I’m interested in the fields these metaphors are imported from. Put ‘exchange’ together with ‘powerhouse where meaning are made’, with what we know of the old man’s history and what motivated his career, and we get the field of Marxist theory of economic production and exchange, combined with Hjelmslev’s theory of semiotic strata.

In my view this helps to explain how he conceptualised the grammar/semantics relation in 1961, when his research focus was grammar... with grammar as the productive ‘base’ where meanings are made, that redounds with a semantic ‘superstructure’ of these meanings.

Blogger Comments:

[1] For some, Halliday's greatest contribution was his formulation of a scientifically rigorous theory of language.

[2] To be clear, Halliday did not show that language is  'a coproduction between speakers … sequenced in time'. This has been the experience of language users since time immemorial. This is one aspect of the data that is modelled in his theory.

[3] This is misleading, because it is untrue. The "exchange metaphor" does not attempt to "capture…to technicalise and iconise" any reciprocal evaluation by interlocutors. Exchange means giving and demanding (to be given) information or goods-&-services: the system of SPEECH FUNCTION. Whether or not interlocutors actually evaluate each other in their texts is another matter entirely.

[4] To be clear, this is nonsense. On the one hand, exchange is not Marxian theory; as a social phenomenon, it is data that can be interpreted in terms of Marxian, or any other, theory. On the other hand, the word 'powerhouse' was Matthiessen's addition to his editions of IFG (2004, 2014) and it was used to elaborate Halliday's original (1985, 1994) metaphor, drawn from the functioning of a computer, of grammar as the central processing unit of language, not to evoke "economic production and exchange". 

[5] To be clear, as levels of symbolic abstraction in linguistic theory, strata ultimately derive from Saussure's sign, where Signified is the higher level and Signifier is the lower level. Hjelmslev conceived of the higher level as content, and lower level as expression, in his characterisation of a denotative semiotic. Hjelmslev distinguished this from a connotative semiotic, which has a denotative system as its expression plane, and Halliday adopted this to model culture as the content plane and language as the expression plane of a connotative semiotic.

[6] This misrepresents the theoretical architecture of Scale & Category Grammar (Halliday 1961). Halliday (2002 [1961]: 39):

The words 'semantics' and 'semantic' were not used in Halliday (1961). The word 'context' in the above diagram was characterised by Halliday (ibid.) as follows:

The context is the relation of the form to non-linguistic features of the situations in which language operates, and to linguistic features other than those of the item under attention: these being together “extratextual” features. … Context is in fact (like phonology) an interlevel relating form to extratextual features.

And again, the compositional relation between a productive base and its superstructure is not analogous to the symbolic abstraction relation between grammar and semantics.

David Rose On Hjelmslev's Connotative And Denotative Semiotics

David Rose wrote to sys-func on 21 March 2022 at 1:48 pm:

The beauty of Hjelmslev’s interpretation is that ‘context’ can be described as a connotative semiotic (i.e. as systems of meanings), using the theoretical tools developed in SFL for describing language. This has enabled the description of field, tenor and mode as semiotic systems, whose options are configured (more abstractly) by selections in genre systems, and are realised by semiotic systems of language and other modalities. In Hjelmslev’s model the latter are denotative semiotics, as their expression planes are not separate semiotics (e.g. phonology in language). In contrast, connotative semiotics (e.g. genre and register) have other semiotics as their expression plane (e.g. language, image etc).

Blogger Comments:

[1] To be clear, what Hjelmslev (1969 [1943]: 114) actually says is:

In the last section, despite a considerable broadening of the perspective, we have still acted as if the unique object of linguistic theory were the denotative semiotic, by which we mean a semiotic none of whose planes is a semiotic. It still remains, through a final broadening of our horizon, to indicate that there are also semiotics whose expression plane is a semiotic — and semiotics whose content plane is a semiotic. The former we shall call connotative semiotics, the latter metasemiotics.

[2] This is misleading, because it is not true. As the quote above makes plain, it is not just context that constitutes a connotative semiotic, but both context (content plane) and language (expression plane).

[3] To be clear, this is Martin's model of context, which mistakes varieties of a denotative semiotic (register and genre) for connotative semiotics (misunderstood as context only).

[4] This misunderstands Hjelmslev's notion of a denotative semiotic. As the quote above makes plain, a denotative semiotic is a semiotic where neither the content nor expression plane is itself a semiotic (i.e. stratified into content and expression).

[5] To be clear, this confuses variety with stratification. Genre and register are varieties of language, and as such, are at the same level of symbolic abstraction as language, not a higher level that is realised by language. Because both genre and register are at the same symbolic level as language, neither conforms to Hjelmslev's notion of the content plane of a connotative semiotic.

∞

Summary of Rose's misunderstandings of Hjelmslev:

David Rose Repeating Jim Martin's Misunderstanding Of Hjelmslev's Connotative Semiotic

David Rose wrote to sys-func on 2 Dec 2021, at 09:56:

If formal logic is derived from ‘natural' logic of language, then this kind of symbolism may be useful for a ’natural’ philosophy of language. It frees us to re-interpret choices we are given, beyond simply either/or. Are simultaneous choices possible? Are there more general or more delicate choices we can make?

For example, if we choose the metalinguistic feature [functional], then [systemic], must we then choose between lexicogrammatical or discourse semantic analysis, or are both possible? If we choose a [solidary] relation between language and social context, must we choose between a denotative or connotative perspective, or are both possible?

Selections from these metalinguistic systems are continually instantiated as linguistic fields unfold, and always couple features from both field and tenor systems. As with language systems, 'users share attitude and ideation couplings…to form bonds’ , to affiliate with one or another group (or both:)…

Blogger Comments:

[1] To be clear, in SFL Theory, the choice of stratum for analysis is a choice of level of symbolic abstraction: wording (lexicogrammar) or meaning (semantics). But as Halliday (1985: xvii) points out:

A text is a semantic unit, not a grammatical one. But meanings are realised through wordings; and without a theory of wordings — that is, a grammar — there is no way of making explicit one's interpretation of the meaning of a text.

Martin's discourse semantics, however, is another matter entirely. For evidence that the model of discourse semantics is, inter alia, rebrandings of Martin's misunderstandings of Halliday & Hasan's COHESION (textual lexicogrammar) and Halliday's SPEECH FUNCTION (interpersonal semantics) see the close examination of English Text (Martin 1992) here, and the close examination of Working With Discourse (Martin & Rose 2007) here.

Importantly, in SFL Theory, lexicogrammar is interpreted in terms of the meaning it realises — e.g. verbal group as Process — and in the absence of grammatical metaphor, lexicogrammar (wording) and semantics (meaning) are in agreement (congruent). Martin's discourse semantics, however, as a rebranding of COHESION, cannot provide a congruent relation between wording and meaning, because it actually models textual lexicogrammar rather than multifunctional semantics. This, in turn, undermines the systematic description of grammatical metaphor: the incongruent relation between wording and meaning.

[2] To be clear, 'solidary' is Martin's term — taken from Barthes (1977: 101) — for what he claims (1992: 20) is the 'natural' relation between discourse semantics and lexicogrammar. (The word 'solidary' actually means of a group or community: characterised by solidarity or coincidence of interests.)

However, in SFL Theory, the natural relation between semantics and lexicogrammar actually refers to the non-arbitrary relation between experiential meanings and grammatical form, such as that between process and verbal group, participant and nominal group, circumstance and adverbial group. It is this natural relation that makes a functional grammar possible. See Halliday (1985: xvii-xx).

Here Rose applies 'solidary' to the relation between language and social context, where social context is misunderstood, in Martin's model, as varieties of language (register and genre). The reader is invited to identify a linguistic theory that models the relation between language and context, whether understood as culture, or misunderstood as varieties of language, as 'unnatural'. 

[3] To be clear, here Rose is referring to Martin's stratified model, which identifies the context plane with a connotative semiotic system and the language plane with a denotative semiotic system. However, this is a serious misunderstanding of Hjelmslev's notion of a connotative semiotic, which Martin (1992: 493) claims to be using.

Importantly, for Hjelmslev, a connotative semiotic is a semiotic system that has a denotative semiotic system as its expression plane. Martin (1992), however, misunderstands the connotative semiotic to be merely its content plane, context, excluding its expression plane: language. 

Moreover, Martin compounds the error by locating register and genre, varieties of language, a denotative semiotic — the expression plane of the connotative semiotic — in the content plane of the connotative semiotic. That is, in Martin's stratification hierarchy, varieties of language are not modelled as language. This is analogous to claiming that varieties of dog — terriers, poodles etc. — are not dogs.

[4] To be clear, what Rose refers to as 'metalinguistic systems' are systems that classify linguistic theories. In SFL Theory, these are systems whose entry condition is the FIELD feature 'linguistics', where 'field' means the ideational dimension of context; that is: the culture as semiotic system. Rose, on the other hand, follows Martin in misunderstanding FIELD as a system of register. In SFL Theory, register is a subpotential of language, not a system of context.

[5] To be clear, from the perspective of SFL Theory, the claim here is that selections from linguistic FIELD systems ("metalinguistic systems") are continually instantiated as linguistic fields unfold. Firstly, this misunderstands instantiation. Selections from systems are not instantiated; the selection of features is the instantiation.

Secondly, there is no "continual" instantiation of FIELD system features. The instantiation of the context of culture as a context of situation is the selection of a specific configuration of FIELD, TENOR and MODE system features. If a different FIELD system feature is selected, a different context is instantiated.

Rose's misunderstanding here derives from the misunderstanding of field in Martin's model. Although, Martin misunderstands field as a dimension of register, he further compounds the problem by confusing field with ideational semantics; see the evidence here. So, from the perspective of SFL Theory, what Rose is actually concerned with here is the instantiation of ideational semantic features in the logogenesis of a text.

[6] To be clear, from the perspective of SFL Theory, the claim here is that selections from linguistic FIELD systems ("metalinguistic systems") always couple features from both field and tenor systems. Clearly, this is incoherent when expressed this way, since it is not the selection of FIELD features that causes the "coupling" (co-selection of features across FIELD and TENOR systems). More coherently, any situation involves the instantiation of a configuration of contextual features (from the systems of FIELD, TENOR and MODE).

[7] To be clear, this bears no relation to any of the text that precedes it. The terms 'coupling', 'bonding' and 'affiliation' are from Martin and his students, and the unremarkable claim here is simply that people form social alliances on the basis of shared values. The same might be said of social insects; see also:

David Rose Negatively Appreciating The Deployment Of Logic

David Rose Positively Judging Jim Martin's Empire Building.

David Rose Crediting Jim Martin With Michæl Halliday's Ideas

David Rose replied to Ed McDonald on SYSFLING on 27 Aug 2020 at 10:01:

It’s also 6 decades since MAKH proposed ’solidary’ relations between metafunctions and field/tenor/mode as contextual dimensions (following Firth and Malinowski), and 3 decades since JRM proposed field/tenor/mode systems as connotative semiotics realised metafunctionally by language and other modalities (following Hjelmslev). The last decade has seen rapid progress in describing these ‘register’ systems and modalities, by rising stars like Jing Hao, Yaegan Doran, Erika Matruglio, Michelle Zappavigna ...

Blogger Comments:

[1] This is misleading, because it is untrue on several counts. Firstly, Halliday did not project the metafunctions onto context 6 decades ago, because metafunctions did not feature in Scale & Category Grammar (Halliday 1961).

Secondly, Halliday did not project the metafunctions onto context 6 decades ago, because context, in the SFL sense, did not feature in Scale & Category Grammar (Halliday 1961). Scale & Category Grammar was organised as follows:

Thirdly, Halliday did not follow Firth in projecting the metafunctions onto context 6 decades ago, because, although Halliday's Scale & Category did "follow" Firth in many respects, Firth died long before Halliday had theorised the metafunctions or context as the culture as semiotic system.

Fourthly, Halliday did not follow Malinowski in projecting the metafunctions onto context 6 decades ago, because Malinowski died long before Halliday had theorised the metafunctions. What is true is that Halliday later took Malinowski's notions of context of culture and context of situation and built them into his model of stratification, using Hjelmslev's notion of a connotative semiotic.

[2] This is very misleading indeed, because it falsely credits Martin with Halliday's work. Firstly, it was Halliday, not Martin (1992), who first proposed that the culture as semiotic system could be modelled, using Hjelmslev's ideas, as the content plane of a connotative semiotic, with language, a denotative semiotic, as its expression plane.

To be clear, for Hjelmslev, a connotative semiotic is a semiotic system that has a denotative semiotic system as its expression plane. Martin (1992), while claiming to be following Hjelmslev (p493), reduces the connotative semiotic to only its content plane, context, and models it as varieties of a denotative semiotic, register and genre, which on Hjelmslev's model, are located on the expression plane of the connotative semiotic.

[3] To be clear, since Martin misunderstands context systems (field, tenor, mode) as register (sub-potentials of language that realise context), any descriptions by these former students of Martin cannot be regarded as progress in the development of a coherent theory, however rapid.

∞

Postscript: Martin has not publicly corrected any of the false attributions credited to him by Rose.