Our Evolving Work Using Cléirigh's Model

Martin & Zappavigna (2019: 27):

Our evolving work on these dependencies can be tracked through Martin et al. (2010), Hood (2011), Martin (2011), Martin, Zappavigna, Dwyer, and Cléirigh (2013) Martin and Zappavigna, 2018, Zappavigna and Martin (2018), and Hao and Hood (in press). From the perspective of SFL the most pertinent work on relations between modalities to compare with these studies is Painter et al. 2013 (on language and image in children’s picture books). Beyond these initiatives, multimodal discourse analysis research is best guided by Bateman et al. (2017).

Blogger Comments:

[1] This is misleading, since it misrepresents the authorship of this first publication featuring Cléirigh's model of body language.  The actual citation is:

• Zappavigna, M., C. Cléirigh, P. Dwyer & J. R. Martin. 2010. The coupling of gesture and phonology. In M. Bednarek, & J.R. Martin (eds.), New discourse on language: Functional perspectives on multimodality, identity and affiliation. London: Continuum. 219–236.

The function of misrepresenting Martin as the first author is to position Martin as the origination of "our evolving work". (The paper was primarily written by Zappavigna, using Cléirigh's model; Dwyer and Martin were the tenured academics who were granted funding for the project.)

[2] As the clarifying critiques on this blog demonstrate, "our evolving work" involves serious misunderstandings of Cléirigh's model of body language, with these misunderstandings now rebranded as Martin's model of paralanguage.

[3] To be clear, Bateman favourably reviewed Martin's first major publication, English Text (1992). However, in doing so, he neglected to check the provenance of "Martin's" ideas or to consider whether the work was consistent with SFL theory or even with itself; evidence here.  For some of Bateman's misunderstandings of SFL theory, see here.

Paralanguage As Language Expression

Martin & Zappavigna (2019: 26-7, 28):

In this model, the content form of face-to-face linguistic communication can be realised as phonology (of spoken language) or sign (including the sign languages of deaf communities and the ‘emblems’ of hearing ones), plus in both cases sonovergent and semovergent paralanguage; and for many languages we have a graphological system used for written communication. This leaves us with the terminological challenge of how best to name the sound quality and gestural resources we have been calling paralanguage in this paper (since they wouldn’t be para- anymore); we will not attempt to improve on our usage here. 

Blogger Comments:

[1] To be clear, content form is form, not function, and only refers to grammatical forms (clause, phrase/group, word, morpheme).  In terms of SFL theory, the authors here confuse 'form' with 'plane' (the level of symbolic abstraction that includes the strata of semantics and lexicogrammar).

[2] Here the authors reduce the expression systems of Sign languages to mere emblems, such as the thumbs-up sign and the middle finger salute.  To be clear, Sign languages are languages, and so, tri-stratal, and their gestural systems realise the lexicogrammar, just as phonological systems realise the lexicogrammar of spoken languages.  Emblems, on the other hand, are bi-stratal signs, and so their expressions only realise meaning directly, without the considerable semogenic power that a stratum of grammatical systems provides.  This is why emblems are types of epilinguistic body language in Cléirigh's model.

[3] To be clear, the expression systems of sonovergent paralanguage (Cléirigh's linguistic body language) only realise the linguistic content that prosodic phonology realises.  That is, it cannot realise the linguistic content that articulatory phonology realises, such as lexical items.  So the authors are seriously mistaken in presenting sonovergent paralanguage as an alternative to the entire phonological system.

[4] As previously explained, the claim that semovergent paralanguage (Cléirigh's epilinguistic body language) is an alternative to phonology and graphology is falsified by any clause that cannot be realised in semovergent paralanguage, such as the Leo Szilard quote:

A scientist's aim in a discussion with his colleagues is not to persuade, but to clarify.

To be clear, no grammatical structures can be realised in epilinguistic expressions (semovergent paralanguage) because epilinguistic semiosis is bi-stratal, not tri-stratal, and so expressions realise meaning directly, without the considerable semogenic power that a stratum of grammatical systems provides.  It is the absence of a grammar that the parlour game Charades exploits.

[5] To be clear, the authors began the paper by rebranding Cléirigh's model of body language as their model of paralanguage, and now, having satisfied themselves that their paralanguage is not paralanguage, the authors shy away from rebranding it with a more appropriate term, such as body language.

The Argument That Paralanguage Is An Expression System Of Language

Martin & Zappavigna (2019: 26-7, 28):

But what about the stem -language which para- is prefixed to? The drift of consensus in gesture studies, as reviewed and promoted by Fricke (Fricke 2013) appears to be towards treating aspects of what we have been calling paralanguage here as part of language (in fact as part of grammar in Fricke’s work). From the perspective of SFL this argues for a re-interpretation of the taxonomy in Fig. 44 above as Table 6 below, with paralanguage positioned not alongside language but as part of its expression form.

Blogger Comments:

[1] To be clear, consensus is not argument, and to present it as argument is the logical fallacy known as Argumentum ad populum.

[2] To be clear, from the perspective of SFL, Fricke (2013) provides no argument with regard to theorising in SFL, because she does not proceed from the same theoretical assumptions as SFL.  That is, Fricke operates with a different conception of grammar, and a different conception what constitutes inclusion in a grammar, as the following quote (op. cit.: 734) makes clear:

[3] To be clear, the claim that paralanguage is an alternative to phonology, graphology and the expression systems of Sign languages is easily falsified by the fact that the following clause can be realised by genuinely linguistic expression systems, but not by Martin's semovergent paralanguage:

It is a miracle that curiosity survives formal education.

To be clear, in SFL, to be part of language, a semiotic system has to be systematically related to the grammar (Halliday 1985/9: 30).  This is why Cléirigh's linguistic body language ("sonovergent paralanguage") is called linguistic — because its gestures realise the same grammatical features as prosodic phonology, differing only in the parts of the body used to express them.

Epilinguistic body language ("semovergent paralanguage"), on the other hand, is not systematically related to the grammar, and like all semiotic systems other than language, is bi-stratal, not tri-stratal, which is why, unlike genuinely linguistic expression systems, it cannot realise the Einstein quote above.

∞

It is instructive to consider the overall argument of this paper:

• Firstly, Cléirigh's types of body language were rebranded as types of Martin's paralanguage, and then misunderstood and misapplied.
• Second, gestures realising numbers (claimed to be examples Kendon's emblems) were claimed to be expressions of language, not paralanguage.
• Finally, all paralanguage was claimed to be expressions of language (because people using other theories agree that it is).

To be clear, the reasons why the authors have interpreted their paralanguage as expressions of language are because

• linguistic body language ("sonovergent paralanguage") is language, and 
• the authors have confused intersemiotic relations ('convergence') with intrasemiotic stratification (realisation) and 
• have set out with the narrow intention to relate the expressions of epilinguistic body language ("semovergent paralanguage") to Martin's discourse semantic systems (his rebrandings of Halliday's speech function and cohesion).

What The Authors Have Done In This Paper

Martin & Zappavigna (2019: 25):

In this paper we have outlined a model distinguishing behaviour from meaning (somasis vs semiosis), and within semiosis, language from paralanguage. Paralanguage itself was then divided into sonovergent and semovergent systems according to their convergence with either the expression plane or content plane of language.

Blogger Comments:

[1] To be clear, the model the authors have outlined is (part of) Cléirigh's model, linguistic and epilinguistic body language, though misunderstood and rebranded as their own systems, sonovergent and semovergent paralanguage.

[2] As previously demonstrated here, having distinguished non-semiotic behaviour from semiosis, the authors then interpret non-semiotic behaviour as semiotic.

[3] As previously demonstrated, the paralanguage that the authors rebrand as 'sonovergent' is actually, in their own terms, 'semovergent', not 'sonovergent', because it instantiates the same meanings as language, but it diverges from language in the way it is expressed, gesturally rather than vocally.

On the other hand, the authors' approach to semovergent paralanguage has been merely an unsuccessful attempt to fit gestural data to Martin's discourse semantic systems, instead of using the gestural data to encode theory.  This will lead them to the erroneous conclusion (p26, 28) that paralanguage is an alternative expression form of language, alongside phonology, graphology and sign.

Taxonomy Of Sonovergent And Semovergent Paralanguage

Martin & Zappavigna (2019: 20, 24):

Multidimensionality (multiplying meaning)

The sonovergent and semovergent paralinguistic systems discussed thus far are outlined in Fig. 38 (including cross-references to Cléirigh’s original terminology).

Blogger Comments:

This is potentially misleading.  To be clear, this taxonomy simply represents Cléirigh's original model with:

1. body language rebranded as paralanguage,
2. linguistic (body language) rebranded as sonovergent (paralanguage),
3. epilinguistic (body language) rebranded as semovergent (paralanguage), and
4. protolinguistic (body language) omitted altogether.

However, as demonstrated in previous posts, the authors seriously misunderstand and misapply the model they are rebranding here as their own.

The Claim That Units Of Speech Rhythm Realise Elements Of Writing Pedagogy

Martin & Zappavigna (2019: 19-20, 22-3):

As far as longer wave lengths of information flow are concerned,²³ our vlogger is seated and so whole body movement from one location to another is not a factor (as it would be for example for a lecturer roaming to and fro across a stage; cf. Hood 2011).

²³ van Leeuwen (1985, 1992) and Martinec (2002) argue that SFL’s phonological hierarchy can be pushed up several wave lengths beyond the tone group and their work suggests that higher level rhythm would converge with higher level periodicity in Martin's framework.

Blogger Comments:

To be clear, the claim here is that proposed higher level phonological units "converge" with Martin's discourse semantic functions of macro-Theme, hyper-Theme, hyper-New and macro-New.  There are several obvious theoretical inconsistencies here.

The over-arching inconsistency is that the authors are proposing that patterns of speech rhythm correspond to pedagogical suggestions on how to write.  This is because Martin's four discourse semantic functions are actually rebrandings of introductory paragraph, topic sentence, paragraph summary and text summary, as previously explained.

A second inconsistency is that speech rhythm can only identify potential New information, and bears no systematic relation to thematicity.

A third inconsistency is that the use of gesture to realise New information is linguistic body language ("sonovergent" paralanguage), not epilinguistic body language ("semovergent" paralanguage).

A fourth inconsistency is the matching of structural units (wavelengths beyond the tone group) with elements of structure (Themes and News).

A fifth inconsistency, in the authors' own terms, is the use of their term for a relation between the same stratum of different semiotic systems, converge, for an interstratal relation within language.

Textual Semovergent Paralanguage

Martin & Zappavigna (2019: 19, 30):

Information flow (textual semovergent paralanguage)

From a textual perspective²² we need to take into account how spoken language introduces entities and keeps track of them once there (IDENTIFICATION) and how it composes waves of information in tone groups, clauses and beyond (PERIODICITY).  Semovergent paralanguage potentially supports these resources with pointing gestures and whole body movement and position.

²² Martinec (1998) interprets textual meaning as realised through cohesion, following Halliday and Hasan (1976); here we follow Martin (1992) who reinterprets cohesion as discourse semantics, organised metafunctionally in Martin and Rose (2007) as ideational resources (IDEATION, CONNEXION), interpersonal resources (NEGOTIATION, APPRAISAL) and textual resources (IDENTIFICATION, PERIODICITY).

Blogger Comments:

[1] To be clear, despite this claim, it will be seen that the authors provide no instances of semovergent paralanguage in this paper that either introduce entities or keep track of them.

Moreover, IDENTIFICATION is Martin's rebranding of Halliday and Hasan's (1976) grammatical cohesive systems of REFERENCE and ELLIPSIS-&-SUBSTITUTION, misunderstood, confused with ideational denotation and the interpersonal DEIXIS of nominal group structure, and relocated to discourse semantics; evidence here.

[2] To be clear, on the one hand, this confuses content (information) with expression (tone group), following Martin (1992: 384).  On the other hand, on Cléirigh's original model, any aspect of body language that highlights the focus of New information, or delineates a unit of information, functions as linguistic body language ("sonovergent" paralanguage), not epilinguistic body language ("semovergent" paralanguage).

[3] To be clear, PERIODICITY is Martin and Rose's (2003, 2007) reinterpretation of what Martin (1992: 393) models as interstratal interaction patterns as a textual systems of Martin's discourse semantic stratum.  However, Martin's model misrepresents writing pedagogy as linguistic theory, such that:

• introductory paragraph is rebranded as macro-Theme,
• topic sentence is rebranded as hyper-Theme,
• paragraph summary is rebranded as hyper-New, and
• text summary is rebranded as macro-New.

It will be seen that, unsurprisingly, the authors provide no instances of semovergent paralanguage in this paper that 'compose waves of information', let alone gestural realisations of introductory paragraphs, topic sentences, paragraph summaries or text summaries.

[4] To be clear, here Martin and his former student follow Martin (1992) in rebranding misunderstandings Halliday & Hasan's (1976) non-structural textual systems of lexicogrammar as structural discourse semantic systems across three metafunctions.

[5] To be clear, IDEATION is Martin's rebranding of Halliday and Hasan's (1976) textual system of LEXICAL COHESION, misunderstood, confused with logical relations between experiential elements of nominal group structure, also misunderstood, and relocated to discourse semantics as an experiential system; evidence here.

[6] To be clear, CONNEXION does not feature in Martin and Rose (2007), or in Martin (1992). The term 'CONNEXION' is a rebranding of Martin's CONJUNCTION by Martin's former student, Hao. CONJUNCTION is Martin's misunderstanding of Halliday and Hasan's (1976) textual lexicogrammatical system of cohesive conjunction as a logical system at the level of discourse semantics.  Moreover, it confuses non-structural textual relations with structural logical relations, and misunderstands and misapplies the expansion relations involved; evidence here.

That is to say, CONJUNCTION was the only one of Halliday and Hasan's cohesive systems that Martin neglected to rebrand as his own system, and this oversight was finally addressed by his former student.

[7] To be clear, NEGOTIATION is Martin's (1992) rebranding of Halliday's SPEECH FUNCTION.

Interpersonal Semovergent Paralanguage

Martin & Zappavigna (2019: 15):

Evaluation (interpersonal semovergent paralanguage) 

From an interpersonal perspective we need to take into account how spoken language inscribes attitudes, grades qualities and positions voices other than the speaker’s own (APPRAISAL). We also need to account for how speakers exchange feelings, greetings, calls for attention, information and goods & services in dialogue (NEGOTIATION). Semovergent paralanguage potentially resonates with APPRAISAL resources through facial expression, bodily stance, muscle tension hand/arm position and motion (Hood 2011, Ngo n.d. in press) and voice quality. Whereas spoken language can make explicit attitudes of different kinds (emotional reactions, judgements of character and appreciation of things), paralanguage can only enact emotion. A further interpersonal restriction (as suggested by Cleírigh), setting aside emblems (discussed in Section “Emblems” below; Kendon 2004, McNeill 2012), is that semovergent paralanguage cannot be used to distinguish move types in dialogic exchanges (although sonovergent paralanguage can of course support TONE choice in relation to these moves).

Blogger Comments:

[1] To be clear, NEGOTIATION is not evaluation.

[2] To be clear, interpersonal semovergent paralanguage is the authors' rebranding of the interpersonal dimension of Cléirigh's epilinguistic body language.

[3] Here again the authors are looking to match semantic stratum systems instantiated in spoken language to expression plane instances of body language, instead of asking what linguistic meanings, of any mode, are being realised in body language expressions.  It will be seen that this procedural error leads the authors to wrongly conclude (Table 6, p28) that all paralanguage is an expression of language itself.

[4] Correcting for the error identified in [3], this claim becomes, in more coherent theoretical terms:

the instantiation of interpersonal meanings of semovergent paralanguage, realised in facial expression, bodily stance, muscle tension hand/arm position and motion and voice quality, "agrees with" the instantiation of interpersonal meanings of APPRAISAL systems.

However, since, in Cléirigh's model, the meanings of APPRAISAL can be instantiated both protolinguistically and epilinguistically, as well as linguistically, it will be seen that almost all of the instances to be discussed cannot be accurately described as epilinguistic ("semovergent").

[5] To be clear, with regard to the APPRAISAL system of AFFECT, emotional reactions only appraise if they enact an interpersonal assessment.  For example, the clause that surprised me construes an emotional reaction, but it does not in itself, even implicitly, enact a positive or negative assessment.

[6] To be clear, the APPRAISAL system of JUDGEMENT is not limited to assessing 'character'.  For example, the clause capitalism is immoral enacts a judgement, but not of 'character'.

[7] To be clear, the APPRAISAL system of APPRECIATION is not limited to assessing 'things', either in the narrow sense of non-conscious material objects, or in the wider semantic sense in contrast to 'quality'.  For example:

• an ugly man — semantically: a conscious thing;
• a gorgeous blue — semantically: a quality;
• a breath-taking performance — semantically: a process;
• scoring that goal in extra-time was pure magic — semantically a figure;
• scoring one goal and setting up three more was sensational — semantically a sequence.

[8] On the one hand, this confuses the construal of experience as emotion (ideational metafunction) with the enactment of intersubjective relations through AFFECT (interpersonal metafunction), the latter being assessments made on the basis of emotion, such as She loves synchronised swimming.

On the other hand, this is misleading, because it falsely claims that paralanguage, in the authors' own terms, cannot realise the same meanings as tsk! tsk! (negative JUDGEMENT) or wow! (positive APPRECIATION).

[9] To the authors' credit, this is the only misspelling of Cléirigh in the entire article.

[10] The misunderstandings involved in the discussion of 'emblems' are very instructive, and will be examined in situ.

[11] If this is true, then, it is misleading to claim that NEGOTIATION "resonates" with semovergent paralanguage.

Interpersonal Paralanguage Resonating With Negotiation And Appraisal

Martin & Zappavigna (2019: 11):

interpersonal paralanguage is ‘expressive’, resonating with NEGOTIATION and APPRAISAL systems;

Blogger Comments:

[1] To be clear, this is the authors' rebranding of the interpersonal dimension of Cléirigh's epilinguistic body language, a type that can be deployed with or without spoken language:

	semantics	kinetic expression
interpersonal	eg modality and polarity	ø eg oscillating hand (modalisation),  nodding or shaking head (polarity)

[2] To be clear, the expression plane of any semiotic system is expressive — of content.

[3] To be clear, NEGOTIATION is a genuine semantic system, being Martin's rebranding of Halliday's SPEECH FUNCTION.  The authors provide no examples of gestures realising features from the system of NEGOTIATION, and despite the claim of "resonance" here, later dismiss the possibility (p15):

…semovergent paralanguage cannot be used to distinguish move types in dialogic exchanges…

[4] To be clear, APPRAISAL is also a genuine semantic system.  However,  the graduated attitudes of APPRAISAL are largely realised in protolinguistic body language, and every example of ATTITUDE provided by the authors is actually an instance of protolinguistic body language, not epilinguistic body language ("semovergent paralanguage").  (The examples claimed to be instances of GRADUATION and ENGAGEMENT will be examined later in situ.)

Ideational Semovergent Paralanguage Concurring With Ideation And Connexion

Martin & Zappavigna (2019: 11):

Ideational paralanguage is ‘mimetic’, concurring with IDEATION and CONNEXION systems;

Blogger Comments:

[1] To be clear, this is the authors' rebranding of the ideational dimension of Cléirigh's epilinguistic body language, a type that can be deployed with or without spoken language:

	semantics	kinetic expression
ideational	phenomena: elemental (& configurational?)	ø eg drawing shapes,  mimicking movements with hands

[2] To be clear, as Cléirigh originally observed:

When used in the absence of spoken language, this type of body language is called mime, and it is mimetic in this sense.

[3] The claim here is that the semogenesis of ideational paralanguage is similar to the semogenesis of Martin's discourse semantic systems of IDEATION and CONNEXION.  However, the problem here is that these systems are neither ideational or semantic.  These systems are Martin's rebrandings of Halliday and Hasan's (1976) lexicogrammatical systems of the textual metafunction: lexical cohesion (IDEATION) and cohesive conjunction (CONNEXION) — both of which, however, are  indeed confused with aspects of the ideational metafunction, as demonstrated here and here.  So the  proposed "concurrence" here is actually between textual grammar and ideational paralanguage.

In any case, as will be seen, the authors provide no examples of gestures realising CONNEXION features, although there is an instance of cohesive conjunction their data, which the authors misinterpret as an 'emblem'.

As will also be seen, the authors provide no examples of gestures realising features of three systems of IDEATION (Martin & Rose 2007: 76): taxonomic relations, nuclear relations or activity sequences.  Instead, the examples provided are all instances of gestures realising elemental or configurational phenomena (Halliday & Matthiessen 1999), in line with Cléirigh's original model.

Ignoring Content And Getting The Phonology Wrong

Martin & Zappavigna (2019: 8, 7):

The phonological system of TONE is realised through pitch movement. In example (1) the vlogger’s eyebrows move up in tune with the rising tone (tone 2) on the syllable prev (Fig. 5).

 

Blogger Comments:

To be clear, this presents Cléirigh's linguistic body language as the authors' sonovergent paralanguage, though in doing so, the authors only present the expression plane of the system, ignoring the content being realised.

Since this is linguistic body language, the interpersonal meaning that the pitch and eyebrow movement both realise depends on the grammar, the choice of MOOD, which in this instance is declarative:

//4 but /^ I could / not / find the /hair dye that I //2 bought / previously when I //3 dyed my / hair which I //3 loved I //3 loved the/ first time

On Halliday's model, the combination of tone 2 with declarative mood realises a protesting or contradicting statement (Halliday 1994: 305), as in

//2 that / can't be / true // ('so don't try and tell me!')
//2 ^ it / didn't / hurt you ('so don't make a fuss') 

In the authors' data, however, the speaker is not making a protesting or contradicting statement, and so this casts doubt on the phonological analysis. Listening to the data reveals that the pitch movement on the tonic is fairly level, tone 3.

The basic meaning of ('low-rising') tone 3, on the other hand, is that the information being realised is dependent on something else (op. cit.: 303).  In this monologic instance, it could be take to mean 'hold on, there's more to come'.  (In dialogue, tone 3 can function as a as a turn-keeping device: 'I'm not finished yet, so don't interrupt!'.) On this basis, the instance in question is better analysed as an emphatic variant of tone 3, just like the three tones that follow (and also the  preceding "tone 4").

Two trivial errors can be noted:

• the tonic syllable is pre, not prev, and
• a 'rise-fall' eyebrow movement corresponds to tone 5, not tone 2.

Misrepresenting Cléirigh's Work As The Authors' Work

Martin & Zappavigna (2019: 8, 6):

Sonovergent paralanguage converges with the prosodic phonology of spoken language (Halliday 1967, 1970; Halliday and Greaves 2008; Smith and Greaves 2015). From an interpersonal perspective, it resonates with tone and involves a body part (e.g. eyebrows or arms) moving up and down in tune with pitch movement in a tone group (TONE and marked salience). From a textual perspective it involves a body part (e.g. hands, head) beating in sync with the periodicity of speech – which might involve beats aligned with a salient syllable of a foot, the tonic syllable of a tone group, or a gesture co-extensive with a tone group (i.e. in sync with TONALITY, TONICITY or RHYTHM). An outline of this sonovergent paralanguage is presented in Table 4.

 

Blogger Comments:

[1] To be clear, as previously explained, 'sonovergent' paralanguage is the authors' rebranding of Cléirigh's 'linguistic' body language, justified on the basis that the invented word 'sonovergent' is more transparent.  As previously explained, linguistic body language, as the name implies, is "convergent" with language itself and differs from language only in its mode of expression — the opposite of the authors' claim.  As Cléirigh originally elaborated:

	lexicogrammar	prosodic expression
		phonology	kinetic
textual	LEXICAL SALIENCE	RHYTHM	gesture (hand, head) in sync with the speech rhythm
	FOCUS OF NEW INFORMATION	TONICITY	gesture (hand, head) in sync with the tonic placement
	INFORMATION DISTRIBUTION	TONALITY	gesture (hand, head) co-extensive with tone group
interpersonal	KEY	TONE	gesture (eyebrow*, hand) in tune with the tone choice

* also: rolling of the eyes for tone 5.

[2] This is misleading.  Here the authors elaborate the details of Cléirigh's model of linguistic body language — see [1] — as if it is their own development of it as sonovergent paralanguage.  Proof that the omission of attribution is not accidental is provided by the misleading claim (p3) identified earlier which primes the reader for the interpretation of this work as the author's innovation:

We will in fact suggest that SFL’s tone group, analysed for rhythm and tone, provides an essential unit of analysis for work on paralanguage as far as questions of synchronicity across modalities are concerned.