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.

The Semovergent Paralanguage Of PERIODICITY

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

As noted in sections “Sonovergent paralanguage and Information flow (textual semovergent paralanguage)” above however the vlogger does end the episode with a contrasting high then lowered pitch (Fig. 35). The higher pitch penultimate tone group begins rhythmically speaking with a handclap foot and then a foot comprising the ‘filler’ / um /.

This is followed by the low pitch tone group; the vlogger is winding down. Following this there is a suspension of both language and paralanguage as her the vlogger’s eyes shut and her head slumps forward (Fig. 36).

The preceding episode to the one we are using to explore sonovergence here ends in a similar way (lowered pitch, with eyes shut, head down) (Fig. 37). So shutting down language and paralanguage and handing over to somasis is clearly a strategy for punctuating longer waves of discourse. It is at these points that the vlogger cuts from one filmic segment to the next (as she thinks of something more to say).

Blogger Comments:

[1] As previously explained here, in this instance, the authors mistook a (misanalysed) sequence of tones (language) as voice quality (paralanguage).  On Cléirigh's original model, any gestures consistent with the tone choices are instances of linguistic body language ("sonovergent" paralanguage), not epilinguistic body language ("semovergent" paralanguage), the concern of the present discussion.

[2] Here again, as above, the authors mistake a tone choice for paralanguage, and mistake the "sonovergent" gestures that are "in tune" with the pitch movement as "semovergent".

[3] The unarguable claim here is that ending discourse ('shutting down language and paralanguage and handing over to somasis') is one way ('strategy') of ending discourse ('punctuating longer waves of discourse').

∞

More to the point, the authors are here claiming to present paralinguistic examples of "longer waves of discourse".  In Martin & Rose (2007: 187-218), these are modelled in terms of:

• macro-Theme (introductory paragraph),
• hyper-Theme (topic sentence),
• hyper-New (paragraph summary), and
• macro-New (text summary).

To be clear, the authors have not identified any of the above in the text, and have not identified any semovergent paralanguage "in sync" with them.  Instead, the authors have merely mistaken intonation as paralanguage, and interpreted the unmarked tone for declaratives, tone 1 (and accompanying gestures) as "punctuating" a longer (unnamed) wave of discourse, while ignoring all the other instances of tone 1 (and accompanying gestures) at "non-punctuating" points in the discourse.

Again, the authors have tried to fit the data to their theory, instead of using theory to account for data.

The Need For A Metalanguage For Facial Expression

Martin & Zappavigna (2019: 18, 29):

Further work on this interpersonal aural dimension of paralanguage, drawing on van Leeuwen 1999, is beyond the scope of our current research.²¹ 

²¹ We also need to acknowledge that a metalanguage for facial expression, in some sense comparable in specificity to SFL work on attitude in the APPRAISAL framework, remains to be developed.

Blogger Comments:

To be clear, in Cléirigh's original model of body language, facial expressions can function:

1. protolinguistically (e.g. realising emotions), 
2. linguistically (e.g. realising features of KEY), or 
3. epilinguistically (e.g. realising 'uncertain' MODALITY). 

The authors (p29), however, have dismissed the notion of protolinguistic body language, on a misunderstanding, as previously demonstrated here, and reinterpreted it as either non-semiotic behaviour ("somasis") — which they nevertheless interpret as if semiotic — or as interpersonal epilinguistic body language ("semovergent" paralanguage).

Interpreting Averted Gaze As Non-Semiotic

Martin & Zappavigna (2019: 10-11):

The contribution of sonovergent paralanguage to the vlog is interrupted in tone group 15 of Appendix 1, suspended for tone groups 16–19, and resumes for tone group 20 – to allow for a somatic phase during which the vlogger uses her left hand to scratch her right arm. … 

The vlogger stops looking at her followers and begins scratching in the final foot of tone group 15 (Fig. 11). The scratching and absence of gaze continues for two tone groups (Figs. 12 and 13). Gaze resumes in the final foot of tone group 18 (Fig. 14). And the vlogger then resumes gesturing (Fig. 15). 

It is interesting to note that the vlogger does not scratch in sync with the RHYTHM, TONICITY and TONALITY of the text; the scratching lasts for two and a half tone groups, and does not match the timing of salient and tonic syllables. But the paralanguage remains in sync, stopping precisely at the tonic syllable of tone group 15 (/ days ago //), resuming with a smile precisely at the tonic syllable of tone group 18 (/ Target //) and resuming with gesture precisely at the beginning of tone group 19. This indicates that synchronicity with prosodic phonology can function as a demarcating criteria for distinguishing somatic from semiotic behaviour.

Blogger Comments:

[1] To be clear, 'sonovergent paralanguage' is the authors' rebranding of Cléirigh's linguistic body language.

[2] Here the authors interpret the breaking of eye-contact with the addressee as non-semiotic behaviour (somasis). The problem with this is that breaking eye-contact with the addressee is just as meaningful (semiotic) as maintaining eye-contact.  In Cléirigh's original model, these are features of protolinguistic body language, the opposition being a human variant of the type of body language also recognisable in other species.

[3] This is hardly surprising, given that scratching an itch is not linguistic body language. In Peircean semiotics, a scratch might be interpreted as an indexical sign, indicating an itch or nervousness.

[4] To be clear, in Cléirigh's original model, a smile is another example of protolinguistic body language, interpreted as a threat in some social species. 

[5]  To be clear, the claim here is that the mere fact that gestures are speech-timed distinguishes them from non-semiotic behaviour.  The reason this is untrue is that, in Cléirigh's original model, only one of the three types of body language, linguistic body language, is speech-timed.  Consequently, 'synchronicity' merely distinguishes linguistic body language from everything else, whether semiotic (protolinguistic or epilinguistic body language) or non-semiotic ('somasis')

[6] Trivially, the singular form is 'criterion'.

Misunderstanding And Misrepresenting Protolinguistic Body Language

Martin & Zappavigna (2019: 7): 

We have drawn on this terminology to deal with two dimensions of the relation between language and paralanguage introduced by Cléirigh as ‘linguistic body language’ and ‘epilinguistic body language’.¹¹

Martin & Zappavigna (2019: 29):

¹¹ Cléirigh’s dimension of protolinguistic body language has been subsumed in our model as subtypes of somasis and interpersonal semovergent paralanguage. This avoids the problem of using the term protolinguistic for a paralinguistic system making meaning alongside language (protolanguage, as initial emergent semiosis, by definition cannot accompany language); and it makes room for paralinguistic systems enabled by the discourse semantic system of APPRAISAL, a system of meaning beyond the scope of Cléirigh's study.

Blogger Comments:

[1] The authors deliberately give low textual status to their argument for reinterpreting protolinguistic body language by placing it in an endnote rather than the body of the text.

[2] To be clear, the authors here reinterpret protolinguistic body language, the type of semiosis also found in other social species, as subtypes of non-semiotic behaviour — their 'somasis' — and epilinguistic body language — rebranded as 'semovergent paralanguage — the type of semiosis only possible for humans.

That is, they misinterpret the semiosis of rainbow lorikeets, for example, as either not semiosis at all, or as semiosis only humans can understand.

[3] This seriously misunderstands the source of the authors' ideas. On Cléirigh's model, protolanguage does not magically disappear with the ontogenetic move into language.  This is demonstrated by any facial expression that is interpretable to other members of a community, whether such protolanguage accompanies language — i.e as paralanguage — or not.  In terms of SFL expansion relations, the authors make the category error of misinterpreting temporal sequence (enhancement) as replacement (extension).

Moreover, interpreting the four microfunctions of protolanguage as one metafunction of epilanguage — interpersonal — creates further serious theoretical inconsistencies.

[4] To be clear, on the one hand, making the theoretical error identified above in [2] does not "make room" for these proposed paralinguistic systems, and on the other hand, there is no need to "make room" for them, because Cléirigh's model already accommodates them; see [6] below.

[5] To be clear, epilinguistic body language is body language made possible by language, not by theoretical models of it like APPRAISAL.

[6] To be clear, this is not beyond the scope of Cléirigh's theoretical model ("study"), merely beyond the authors' comprehension of it.

A complicating factor here is that the meanings modelled as (graduated) ATTITUDE in the system of APPRAISAL are not confined to language.  For example, in the protolanguage of rainbow lorikeets, the personal expression of anger, realised, for example, vocally as a loud, harsh low-pitched squawk, can function interpersonally as a negative judgement of another bird, as when a lower ranked bird tries to feed before a higher ranked bird.  An example of the realisation of positive judgement is a brief entwining of necks, initiated by a female lorikeet or pigeon — a gesture that has only been observed after a male partner has defeated a male rival.

Turning to epilinguistic body language, a common realisation of positive judgement is the 'thumbs up' sign, and of negative judgement is the 'thumbs down' sign.  Because these are epilinguistic, made possible by language, such expressions can vary from linguistic culture to linguistic culture. 

That is, in short, attitudinal meaning can be instantiated protolinguistically, linguistically or epilinguistically.

Misunderstanding Semiosis

Martin & Zappavigna (2019: 6, 29):

But we have found it useful to try and compile a range of behaviours that border on semiosis and which can be interpreted by social semiotic animals as indexing purposeful activity. As Halliday and Painter have shown, early protolinguistic semiosis involves a reconstrual of some of these activities as the expression face of signs. And all of the behaviour outlined above has the potential to be used as signs – for example stamping one’s foot in frustration, coughing to remind a meeting of one’s presence, shivering to indicate one is cold, sniffing to object to an odour, kissing on the cheek as a greeting and so on. In these cases there is some degree of deliberation involved, as manifested in the fact that the behaviour will synchronise with the prosodic phonology and turn-taking structure of an interaction and will be responded to as meaningful by co-participants.ⁱ⁰

ⁱ⁰ To put this another way, we are arguing that the behaviours outlined in Figure 3 can be treated as paralinguistic or not depending on whether or not they are negotiated as meaningful in interaction. We also need to acknowledge that what we are calling somatic behaviour has the potential to be imbued with cultural norms (e.g. a style of walking, norms for coughing or spitting etc.); these need to be taken into account in future work on somatic behaviour.

Blogger Comments:

[1] To be clear, any behaviour that can be interpreted as "indexing" anything other than itself, purposeful or otherwise, is functioning semiotically.  The behaviour is the expression (signifier) and what it "indexes" is the content (signified) of Saussure's sign.

[2] This is all true.  However, what the authors are totally unaware of is that they have already interpreted most of these behaviours as the expression "face" of signs when presenting them as non-semiotic, as demonstrated in the previous post.  Note that none of the examples provided are exclusively paralinguistic — the model being developed in this paper — since they can occur in the absence of language.

[3] To be clear, deliberation — or indeed deliberateness — is not criterial in determining whether a behaviour is interpreted as semiotic (meaning something other than itself). The lack of intention of a meaning maker does not make a meaningful behaviour any the less meaningful to others.

[4] To be clear, this is a bare assertion, unsupported by evidence.  Consider what is being claimed.  The following behaviours are claimed to synchronise with the ongoing rhythm and intonation of the behaver's own speech:

• stamping one’s foot in frustration, 
• coughing to remind a meeting of one’s presence, 
• shivering to indicate one is cold, 
• sniffing to object to an odour, 
• kissing on the cheek as a greeting.

[5] To be clear, this is not a matter of interpersonal negotiation.  What matters is if a behaviour is construed as meaning something other than itself.  This does not depend on two interactants negotiating any agreement. 

[6] To be clear, assigning cultural values (content) to behavioural tokens (expression) is again treating the behaviours as semiotic, not as "somatic".

Interpreting Purportedly Non-Semiotic Behaviours As Semiotic

Martin & Zappavigna (2019: 6, 4, 29):

Physical activity covers material action involving some change in the relationship of one physical entity to another (walking, running, jumping, throwing, breaking, cutting, digging, pulling etc.). Biological behaviour can be divided into changes that restore comfort (sneezing, coughing, scratching, laughing, adjusting garments or hair etc.) and changes that index discomfort (nail biting, fiddling, fidgeting, wriggling, blushing, shivering, crying etc.). Social communion can be divided into mutual perception (sharing gaze, pitch, proximity, touch, smell etc.) and reciprocal attachment (tickling, cradling, holding hands, hugging, stroking, hugging, kissing, mating etc.). These proposals are outlined in Fig. 3.

Trained as we are as linguists and semioticians we are not ourselves in a strong position to further develop this model.⁹ 

⁹ For exemplary forays into this realm of inquiry see Martinec (1998, 2000a, 2000b, 2001, 2004), who models somasis as if it was semiosis, drawing on SFL theory to do so.

Blogger Comments:

[1] To be clear, these are construals of experience as material processes.

[2] To be clear, contrary to the authors' stated intention, interpreting these behavioural and material processes as meaning something other than themselves, as

• changes restoring comfort,
• indexing discomfort,
• social communion,
• reciprocal attachment

is modelling them as semiotic, not "somatic".

[3] This is misleading. To be clear, the authors mistakenly model semiosis as if it were 'somasis', whereas Martinec models semiosis as semiosis.

Confusing Levels Of Complexity With Types Of Behaviour

Martin & Zappavigna (2019: 5-6):

As far as somasis is concerned we have found it useful to draw on Halliday’s proposals for an evolutionary typology of systems (Halliday 1996: 388, Halliday 2005: 67–68). He recognises four orders of complexity, with semiotic systems evolving out of social systems, social systems out of biological ones and biological ones out of physical ones. We have adapted this framework in our classification of somatic behaviour, distinguishing physical activity, biological behaviour and social communion.

Blogger Comments:

This is a very serious misunderstanding. Here the authors map levels of emergent complexity onto types of behaviour:

		systems	types of non-semiotic behaviour
levels of  emergent  complexity	physical	sub-atomic particles, atoms, molecules etc.	activity (walking, running etc.)
	biological	anatomy and physiology of organisms	behaviour (sneezing, coughing etc.)
	social	eusocial insect colonies	communion (hugging, mating etc)

On the authors' model, in the history of the universe, activities like walking and running etc. eventually complexified to the point where a new level of organisation emerged, behaviours like sneezing and coughing etc., which eventually complexified to a point where a further new level of organisation emerged, communion through hugging and mating etc.

To be clear, the authors have simply taken the names of the three levels of complexity that are not semiotic systems and used them to label their types of non-semiotic behaviour, without regard to the meanings of the source model or the implications when applied to behaviour.

In SFL theory, types of behaviour are construed by the ideational metafunction, and transitivity analyses of clauses, identifying the dimensions that differentiate subtypes of behavioural processes, would constitute evidence in support of any proposed categories.

Misunderstanding The Difference Between Semiosis And Non-Semiosis

Martin & Zappavigna (2019: 5):

Non-semiotic behaviour (somasis)

One basic challenge that has to be faced when working on paralanguage is how to distinguish it from behaviour – separating semiosis from non-semiosis in other words. … For Halliday and Painter the key criteria are that 

i. the act in question is interpretable as one of a systematic system of content/ expression pairs (i.e. signs with valeur), and 

ii. the act in question is used on a number of different occasions (i.e. not simply iterated in a single interaction). 

Halliday (1984/2003: 240) for example notes three signs oriented to action in his son’s protolanguage at 8 months of age (Fig. 2). The signs are constituted as the following content/expression pairs: 

‘I want it/grasp firmly’,
‘I don’t want it/touch lightly’ and
‘do that with it/touch firmly’. 

The contrast between semiosis and non-semiosis is evident here, perhaps most clearly in the contrast between touching something lightly (semiosis) and pushing it away (non-semiosis). In Peircian terms we might say that the semiosis symbolises the intention of the speaker while the non-semiosis indexes it. 

From this point on we will use the term somasis for non-semiotic behaviour, and semiosis for systems of signs.

Blogger Comments:

[1] To be clear, for behaviour to be semiotic, it must mean something other than itself, as when the act of pushing something away (also) means 'I don't want it'.

[2] Here the authors merely take a gesture that is not included in Halliday's model of his son's protolanguage, pushing something away, and conclude that it therefore cannot be semiotic.  To be clear, if a child is old enough to push something away, rather than just touch it firmly, and it is taken to mean 'I (emphatically) don't want it', then the gesture is semiotic.  Moreover, it is an instance of graduated attitude being realised protolinguistically, rather than linguistically.

[3] This is a serious misunderstanding. Here the authors misinterpret Peirce's semiotic distinction, symbolising meaning vs indexing meaning, as a distinction between semiotic and non-semiotic behaviour, respectively.

[4] To be clear, this is 'intention' in the sense of 'meaning'.