Pragmatics and Non-Verbal Communication, by Tim Wharton: Cambridge: Cambridge University Press, 2009, 219 pp.: Reseña

PRAGMATICS AND NON-VERBAL COMMUNICATION BY TIM WHARTON Cambridge: Cambridge University Press, 2009, 219 pp. Reviewed by MANUEL PADILLA CRUZ University of Seville 1. Introduction Readers will certainly find in this recent volume a most timely work that stresses the importance that non-verbal phenomena and elements lying between the linguistic and non-linguistic, the coded and non-coded, have in communication. Thus, Wharton undoubtedly redresses the balance in their favour after decades in which they have been incomprehensively relegated to a second plane in linguistics and other neighbouring disciplines. By exploring what and how such phenomena and elements contribute to communication, he brings to the pragmatic arena a wide range of items that have often met controversial accounts or escaped systematic linguistic description but, particularly, missed pragmatic unitary explanations. The following sections summarise each of the chapters this volume comprises, after which follows a critical evaluation. 2. “Natural Pragmatics” The book opens with this introductory chapter (: 1-17), where Wharton starts by clearly explaining the reasons that encouraged him to write this most interesting work: controlled or unconscious vocal, facial and bodily gestures –which he refers to as ‘natural non-verbal behaviours– are omnipresent in human communication and largely contribute to or bias our understanding of discourse. Although they have been approached from different frameworks, such as functionalism, conversational and discourse analysis, sociology or anthropology, they have not been approached from a cognitive perspective that unveils how they might interact with linguistic properties of utterances. Such an approach must answer the following questions: a) What is the relation between natural non-verbal behaviours and intentional communication? b) How are non-verbal behaviours interpreted? c) What do they convey? d) What is the relation between natural non-verbal behaviours and those nonverbal behaviours that are not natural? (: 3-4) The answers to these questions depend on the definitions of notions such as natural, language, pragmatics and communication, so Wharton explains his conceptions thereof. Based on Grice (1957), he applies the term ‘natural’ to the way in which non-verbal communication means, and takes ‘natural meaning’ to be synonymous with ‘naturally indicates’, as opposed to ‘non-natural meaning’, often used to refer to arbitrary or conventional meaning. By ‘language’, he understands an Internal, Individual, Intensional object consisting of a mentally represented grammar governed by innately determined principles, so he adheres to the cognitive, Chomskyan view. Finally, as regards ‘pragmatics’ and ‘communication’, Wharton adopts the relevance-theoretic approach to language use, which centres on ostensive-inferential communication and the processes taking place in comprehension (Sperber and Wilson 1986, 1995). 3. “Natural and non-natural meaning” The second chapter (: 18-37) begins with a section dedicated to Gricean meaningNN. Wharton argues that, for Grice, cases of meaningNN contain a basic layer constituted by information pointed out, which cannot be derivable without a second layer of information that amounts to the intentional pointing of that first layer of information. He also comments on the tests devised by Grice to distinguish between cases of meaningN and meaningNN – paraphrasing and directly quoting (: 21-22). Next follows a section where Wharton challenges the Gricean description of meaningNN as requiring an intended response from the audience, the audience’s recognition of the intention to produce that response, the communicator’s intention that the audience recognises the intention to produce that response and the audience’s recognition of the communicator’s intention to produce a desired response. He also characterises intentional communication as “[…] deliberate and open […]” (: 29) in the sense that the communicator lets the audience know something and encourages them to think that she has done so for some reason. Thus, he distinguishes it from mere cases of showing, in which there is no real intention or reason on the part of the communicator to communicate anything, although the audience may draw their own conclusions. Wharton concludes this chapter by claiming that behaviours that can be regarded as cases of meaningN can be deliberately shown and “[…] recruited for use in overt intentional communication” (: 33). 4. “Pragmatics and the domain of pragmatic principles” The third chapter (: 38-69) opens with a section that summarises some of the basic tenets and fundamental claims of relevance theory (Sperber and Wilson 1986, 1995): the cognitive principle of relevance, the notions of relevance, informative intention, cognitive environment, strong and weak communication and strong and weak implicature. As a consequence of the different ways wherewith communicators make manifest their informative intention, Wharton proposes the existence of a continuum of cases between showing and meaningNN, at one extreme of which are cases of purely spontaneous showing, while at its other extreme are cases of authentic linguistic coding. In between lies a wide variety of cases in which more or less direct/‘natural’ and indirect/‘coded’ evidence mix to various degress (: 43-47). Then, Wharton addresses the problem of the semantic underdeterminacy of utterances and explains the relevance-theoretic notion of explicature and its implications for pragmatic theory. Since the conceptual structures obtained by decoding may be so imprecise not only at the sentence level, but at word level too, they must be inferentially developed, adjusted or ‘fine-tuned’. Openly shown natural behaviours, like shivers, intonation or gaze direction, Wharton argues, may affect the outcome of the processes of lexical adjustment taking place when explicatures are developed, thus contributing to explicit truth-conditional content and guiding hearers to certain conclusions (: 51). In other cases, such behaviours convey attitudinal information, which may also be conveyed in a more explicit way by recourse to linguistic elements, such as sentential adverbs, which involve encoding. For this reason, natural behaviours also contribute to higher-level explicatures, but in a less explicit way. Finally, Wharton introduces the distinction between translational and nontranslational activation of concepts, parallel to the relevance-theoretic distinction between conceptual and procedural meaning/expressions/encoding (e.g. Blakemore 1987, 2002), and reminiscent of the speech-act-theoretic between describing and indicating (Austin 1962; Searle 1969). Translational activation of concepts is based on the existence and usage of a code, and amounts to triggering off a concept when decoding takes place. However, whereas the notion of procedural meaning/expressions/encoding amounts to instructions constraining the comprehension process by reducing the search space for relevant interpretations, non-translational activation “[…] does contain a coded element that points the hearer in a direction they would not reliably take unless they knew the code” (: 61). This new distinction suggests a reinterpretation of procedural encoding in terms not just of instructions, but of “[…] the management of levels of activation (e.g. of conceptual representations, computations or expectations” (: 65), and of procedural expressions as involving different activations: inferential rules, conceptual representations (e.g. contextual assumptions or classes of candidate referent), or expectations of particular types of cognitive effects (: 65). 5. “Interjections and Language” Wharton applies some this new distinction and his showing-meaningNN continuum to the analysis of interjections in Chapter 4 (: 70-106). His major aim in this chapter is to answer these three questions: a) What do interjections communicate? b) How do interjections communicate? c) Are interjections part of language? The author starts by mentioning the controversy between the conceptualist and the non-conceptualist views of interjections, describing the major types of interjections –primary and secondary– and suggesting two general criteria to characterise them. Then, he discusses the problems he finds in the conceptualist approach: (i) difficulties to find satisfactory definitions for interjections, (ii) their vagueness, (iii) their context-dependence, (iv) their naturalness and spontaneity, (v) their lack of appropriate synonymous conceptual counterparts, and (vi) their nontruth-conditional nature. These problems lead him to claim that interjections are not conceptual elements. In the following sections, Wharton reviews anthropologist E. Goffman’s (1981) description of interjections as response cries, his classification of them and his proposal concerning a continuum between properly linguistic and non-linguistic response cries, or between ‘displaying’ and ‘saying’ –similar to Wharton’s showing-meaningNN continuum– although Goffman’s differs in that it seems “[…] to be based on the assumption that all communication involves at least some element of coding” (: 83). Even if Wharton assesses Goffman’s contribution positively, he criticises Goffman for not addressing how interjections communicate or present a clear alternative to the conceptualist approach (: 84). For this reason, he then explores the possibility that interjections are analysed as non-truthconditional indicators of “[…] higher-level explicatures containing the type of speech-act or propositional-attitude information the hearer is expected to infer” (: 85). Thus, he seeks to find an answer to the question about what interjections communicate. However, this analysis also poses some problems and seems quite restrictive, for interjections do not always appear in discourse with adjacent propositions that could yield the lower-level explicatures to be subsequently embedded under higher-level explicatures (: 87-88). Based on Rey’s (1980) work on emotional states, feelings and sensations, he states that the question about what interjections communicate requires different answers: in some cases, they would convey information exploitable for higher-level explicatures; in other cases, emotional attitudes to propositions and not propositional-attitude or speech-act descriptions, and, finally, in other cases, feelings or sensations (: 88-89). Next, he turns to the question about how interjections communicate. Since interjections do not pass the tests about conceptuality, he suggests a procedural analysis, according to which interjections “[…] encode procedural information which ‘points’ in the general direction in which relevance should be sought” (: 90). The procedures interjections encode, Wharton says, activate “[…] various attitudinal concepts or classes of concepts, but not in the standard translational way” (: 90). Accordingly, wow might activate attitudinal descriptions having to do with delight, surprise or excitement; eh a variety of interrogative propositionalattitude descriptions; huh dissociative attitudes, etc. Prosodic information and paralinguistic information would determine the particular attitude involved and its intensity. With this proposal Wharton both resolves the problems the conceptualist account has and preserves the intuitions that interjections have a coded element and are more than natural displays (: 91). In order to answer the question whether interjections are part of language, the author takes into account their ‘paralinguistic’ nature, which places some of them close to gestures; their phonological atypicality, which prevents some of them from being reported by verbs of saying, and their syntactic independence and nonproductivity. He concludes that interjections constitute such a heterogeneous category, that a satisfactory answer cannot be given. Finally, Wharton closes this chapter by examining the naturalness of interjections. Quoting from Goffman, Darwin (1872), Sapir (1970), he shows that interjections occupy different positions along a continuum of naturalness, just as they occupy different positions along the showing-meaningNN continuum. Since some of them are instinctive and seem to be caused by certain states of mind, they may be viewed as developments of natural behaviours and, hence, as more natural (: 99). Other more stylised, iconic interjections, on the contrary, combine elements of coding and showing, which separate them from both proper cases of showing and saying, respectively (: 100101). After this, Wharton very accurately and clearly summarises his answers to the questions about interjections. 6. “Natural codes” After introducing what semiotics and the social sciences understand by code, in the fifth chapter Wharton (: 107-138) discusses two examples of natural codes – those used by honeybees and vervet monkeys– and compares them to some human natural behaviours –smiles, crying and shivering. He contends that some of these behaviours –smiles– carry ‘factive’ meaning, as they indicate something about their producers, and may convey messages without reference to their producers’ intentions. This does not exclude that in some cases their producers monitor them and may consciously produce, fake or exaggerate them, which is possible thanks to “[…] the adaptive functions of the behaviours themselves” (: 113). Wharton then explains the difference between signs and signals –the latter’s communicative function– and argues that some human natural behaviours –e.g. smiles– have evolved as signalling activities because they carry or indicate some meaning, whilst others –e.g. shivers– do not work in the same way and are just natural signs (: 114-115). Whereas human natural signs must be interpreted in inferential terms, human natural signals involve a certain element of coding, for they trigger off specific mental or emotional states corresponding to communicators’ mental or emotional states (: 115). However, human natural signals are special in that their interpretation is also supplemented by inferential processes. Next, Wharton reflects on the type of information natural codes convey. In order to do so, he comments on the distinction between digital and analogical coding, and illustrates that many human behaviours are interpreted analogically on the grounds of subtle discriminations of some of their features. Moreover, he states that analogue encoding lines up with the Peircean notion of index, i.e. a representation related to an object in a proportional or causal way. Nevertheless, he also acknowledges that, in addition to the notions of analogue encoding and index, something more is necessary to account for “[…] what the information conveyed by human natural codes looks like in cognitive terms” (: 122). His next step is to review the conceptualist approach to facial expressions, more specifically, Wierzbicka’s (2000) analysis in terns of a ‘Natural Semantic Metalanguage’. Although he finds points of agreement between this author’s work and his view, he finds the same problems mentioned in his review of interjections, which stem from Wierzbicka’s basing her analysis on the coding-decoding model and relegating inference to a secondary or minor role. He firstly admits that “[…] there may be a coded elements to some facial expressions” (: 124), but he contends that, for these expressions to communicate, they do not necessarily have to encode anything but to be exploited inferentially. Secondly, Wharton considers that the conceptual structures with which Wierzbicka characterises facial expressions are entirely digital and fall short of capturing what natural codes convey, their contextdependence and analogicity. Thirdly, he finds it hard to account for what facial expressions communicate on the grounds of encoded universal concepts, as, from a relevance-theoretic standpoint, not all concepts are lexicalised and, in the case of those lexicalised, they must always be narrowed or broadened. Finally, Wharton concludes this chapter arguing that, although natural signals such as facial expressions and affective tones are not part of a linguistic code, they are coded and may be best analysed in non-translational terms, as they do not contribute to the truth-conditional content of utterances, do not combine compositionally with other elements and are extremely context-dependent. Thus, as in the case of interjections, natural signals contribute to the construction of higher-level explicatures and convey attitudinal or emotional information (: 128131). Therefore, there would be different types of both linguistic and non-linguistic devices encoding non-translational information (: 133). Furthermore, he suggests that some natural behaviours may make more implicit or explicit contributions to communication, so they would also be placed along a continuum of explicitness/implicitness, and that they are interpreted by “[…] specialised, perhaps dedicated, neural machinery […]” (: 132). 7. “Prosody and gesture” The sixth chapter (: 139-154) is dedicated to two phenomena indispensable to understand what we say and our attitudes: prosody and gesture. As regards the former, Wharton says that prosodic inputs range from the natural to the linguistic and interact with information from different sources. Although their effects highly depend on context, prosodic inputs convey information about emotions or attitudes, create impressions or alter the salience of some interpretations. Accordingly, prosody interacts with lexical items so as to fine-tune their meaning (: 141-142), and unexpected stress patterns, costlier in terms of processing effort, divert hearers from expectable interpretations towards alternative ones (: 142). Discussing Gussenhoven and his colleagues’ ideas about increased articulatory precision, he argues that this is a natural sign exploitable in ostensive-inferential communication inasmuch as the saliency of the speaker’s effort may attract the hearer’s attention towards some assumptions and departures from expected pitch ranges, although increasing processing effort, may decrease effort to arrive at intended interpretations (: 143-144). As in the case of interjections or face expressions, Wharton also puts forward in this chapter that both natural and properly linguistic prosodic signals –lexical stress, lexical tone and grammaticalised aspects of sentence stress and intonation– encode procedural or non-translational information “[…] facilitating the retrieval of certain types of syntactic, semantic or conceptual representation” (: 146), and jointly interact with other linguistic signals, natural signals and natural signs. Nevertheless, he also concedes that all prosodic inputs may not be coded and that some of them may only stabilise in some languages or cultures, thus becoming emblems, which accounts for cross-cultural variations in their interpretations. Concerning gesture, Wharton finds clear correspondences with both the verbal and prosodic continua he discusses in previous chapters. On the basis of ‘Kendon’s continuum’, which he takes from McNeill (1992), he shows that gestures may range from more to less natural too. Thus, we have gesticulation, spontaneous movements accompanying speech; language-like gestures, which are integrated into speech and contribute to its interpretation; pantomimes, which resemble objects or actions; emblems, which are culture-dependent gestures conveying positive and negative meanings, and sign languages, which are rule-governed languages (: 149-151). Regretting that pragmatics has greatly ignored the role of gestures in communication, Wharton argues that the distinctions he traces in the book can be extended and applied to the study of gesture from a pragmatic viewpoint, as they can be used overtly. Accordingly, he concludes this chapter by suggesting that gesticulations are natural signs aimed to help the hearer and, therefore, are interpreted inferentially. They may be exploited in ostensiveinferential communication because they may convey information if the speaker uses and shows them intentionally (: 153). 8. “Mindreaders” The seventh and penultimate chapter (: 155-170) underlines the importance that the attribution of mental states to other individuals has in both cognition and communication, and reviews the extensive literature evidencing mind-reading. Wharton devotes some pages to summarise contributions on the consequences that impairments in mind-reading abilities have on, e.g. autistics (: 156-158). He also underlines Dan Sperber and Deirdre Wilson’s proposal that verbal comprehension might be carried out by a mechanism or module forming part of our mind-reading ability and specifically dedicated to the interpretation of ostensive stimuli (: 159160). After this, he addresses some criticism against relevance-theoretic claims about the role of mind-reading by Breheny (2006) and Recanati (2002), and provides evidence supporting that in both very basic acts of ostensive-inferential communication, in which communicators give direct evidence of their intention to inform to their audience, and other acts of ostensive communication in which the evidence provided is indirect, attribution of mental states is essential to recognise what has been shown and why, as well as what has been said and why, respectively. Since mind-reading plays such a crucial role in communication, Wharton concludes that people having problems reading other individuals’ minds will also have problems understanding gestures and other non-verbal behaviours intentionally employed in interaction (: 163-164). For this reason, he next reviews some experiments that show that autism and right hemisphere damage result in problems to understand emotional, attitudinal, inarticulate and intrinsic prosody and contrastive stress (: 165-167), and suggests two test cases aimed at investigating the prosodic difficulties arising in autism, Asperger’s syndrome and right hemisphere damage (: 168). 9. “The showing-meaningNN continuum and beyond” Finally, Wharton rounds up his work the last chapter (: 171-194). He starts by remarking that the continua proposed by Goffman, Gussenhoven and Kendon are based on the code model, whilst the continuum he proposes in this book is based on the role played by the inferential attribution of intentions. For this reason, he calls the former types of continua ‘Code-continuum’ (C-continuum) and his ‘Ostensive behaviour-continuum’ (O-continuum) (: 171-172). Both continua represent the evidences used in communication, which range from cases of display to those of linguistic coding. However, the C-continuum cannot explain how communicative behaviours are used and the varied ways in which different behaviours can be exploited to convey information. On the contrary, the advantage of his O-continuum is its applicability to the elements included in the Ccontinuum, as it can account for the ostensive uses of language to display and of display to meanNN (: 173). Besides, the O-continuum captures diachronic evolution of some phenomena, as “[…] it can represent the fluidity and constant change that results in expressions coming to form part of language” (: 174). Nevertheless, as Wharton acknowledges, more research is needed so as to elucidate if it “[…] has an evolutionary-diachronic as well as a historical-diachronic dimension” (: 175). The author goes on to deal with the debate between those who contend that communication began as a coding-decoding activity and those who argue that it required metarepresentational abilities, and gives sound reasons about why metarepresentational abilities might have developed before, independent of communication (: 176-179). After this, he also addresses the problem about why communication might have stabilised, following Dan Sperber, who argues that this might have happened in a panorama in which factors such as the development of the human ability to present arguments for conclusions the audience is intended to draw or the ability to evaluate the argument of others concurred and laid the foundations for the development of complex metarepresentational abilities and a logical vocabulary (: 180-183). Owing to the manifold uncertainties about the emergence of language, he states that we can only account for it in terms of myths, so he then reviews one suggested by Grice himself. This myth portrays the evolution of language and communication as a sequence of stages in which human beings were able to attribute and recognise intentions behind certain behaviours in which they used progressively less direct evidences of their intentions until they reached a point at which communication did not need to depend on natural connections between ostensive stimuli and intended meanings (: 184-190). Finally, Wharton closes this chapter summarising how he has answered the initial questions that motivated this book and suggesting that his ideas may be extended and applied to other disciplines, such as cognitive science, psychology, discourse analysis, sociolinguistics, aesthetics or music.