Three decades of relevance theory

Published in M. Padilla Cruz (ed.) (2016) Relevance Theory. Recent Developments, Current Challenges and Future Directions. Amsterdam: John Benjamins. Since this chapter is under copyright, the Publisher must be contacted for reproduction. I NTRODUCTION T HREE D ECADES OF R ELEVANCE T HEORY M ANUEL P ADILLA C RUZ Almost three decades have elapsed since the publication of Dan Sperber and Deirdre Wilson’s (1986) seminal and most influential book Relevance. Communication and Cognition, and exactly two since the publication of its second edition (Sperber and Wilson 1995). In them, the authors presented and revised a series of claims and principles about human verbal communication and the foundational postulates of what is now known as ‘relevance theory’. Over all these years, their two books and numerous subsequent publications have given rise to a strand of research in pragmatics with a psychological basis and cognitive orientation. Often alluded to as ‘relevance-theoretic pragmatics’, this strand seeks to unravel how the mind processes utterances –and, more widely, discourse– the contribution of diverse linguistic elements (e.g., discourse markers, particles, adverbials, intonation, etc.) to comprehension, why the mind arrives at a particular interpretation and the effects that may follow from understanding utterances in one way or another. Relevance theory has certainly awoken the interest of many pragmatists and linguists in general by posing many intriguing problems and thought-provoking questions. Relevance theorists’ continuous challenging of often-taken-for-granted assumptions, claims, generalisations, and even whole models, has also brought fresher air to those disciplines. Indeed, they have analysed in depth a wide variety of linguistic and communicative phenomena from a different perspective and with a new theoretical apparatus, which has shed much light onto underexplored or overlooked issues. This book celebrates these happiest anniversaries and, most importantly, the fact that relevance theory still continues to appeal to researchers, who find in it a very valuable model for understanding the intricacies of linguistic communication. This is a collection of papers, which very sincerely acknowledges the extensive work carried out not only by the authors of the theory themselves, but also by a large number of researchers who have elaborated on some of its postulates and distinctions, empirically tested some of its predictions or applied the theory to diverse domains or neighbouring fields, thus expanding its scope. Consequently, a tribute of deepest gratitude is paid to all of them for their brave efforts to answer questions related to an incredibly complex human activity. Ten papers are collected here. Five of them elaborate on issues that have traditionally concerned practitioners in relevance theory: the intentional nature of communication, how speakers guide hearers to recover intended meaning, how specific types of utterances or linguistic elements are interpreted, or the consequences of communication. Two papers apply the relevance-theoretic apparatus to account for features of some linguistic varieties or languages, thus helping to understand how they are structured and function. The remaining three papers present recent developments and further applications of the theory, as they consider epistemic vigilance mechanisms (Mascaro and Sperber 2009; Sperber et al. 2010). More precisely, those papers address the interaction of some linguistic elements with such mechanisms, explain their role in rhetoric and argumentation, or consider their importance in second language teaching and acquisition. Finally, the book concludes with a chapter that suggests possible future directions for research in relevance-theoretic pragmatics. These works offer new insights and seek to fuel what can be considered an authentic revolution in pragmatics, particularly in its cognitive branch, which had scarcely been convulsed since the various publications on meaning and communication by philosopher of language Herbert Paul Grice. Although most of the chapters summarise essential postulates and explain the key notions they rely on, in what follows Sperber and Wilson’s (1986/1995) views on communication, underlying assumptions, fundamental claims and basic concepts are presented. Since many readers will surely be familiarised with them, the following section is simply intended as a reminder. For those novel to the theory, this discussion supplies the necessary background that enables them to conveniently understand the works gathered here, as well as what relevance theory has meant. Next comes a review of research done in various areas from a relevance-theoretic perspective, which helps situate the chapters of this book in the wider panorama of relevance-theoretic pragmatics. Finally, this introductory chapter is closed with a description of the chapters. 1. The relevance-theoretic revolution In a field like linguistics, where most theories, models and frameworks have been constantly revised, refined, questioned and eventually abandoned in favour of others, referring to the contribution made by a particular one as a ‘revolution’ might sound presumptuous and even biased. Undeniably, proponents of new theories, models and frameworks have always been encouraged by a perennial and daring interest in unraveling how linguistic systems are organised and how they work at the service of communication, as well as in overcoming limitations in prior approaches. In a relatively young discipline like pragmatics, scholars have incessantly attempted to grasp the enormous complexity of human verbal communication, where linguistic systems are put to use to convey information and achieve goals that may crucially impact on human relationships, and the outputs of such systems have to be processed by an until fairly recently practically unknown mechanism like the human mind. Relevance theory originated as an exciting endeavour to find convincing answers to a series of only apparently simple and obvious questions for which many pragmatists could have thought there were final answers: - How do humans understand each other? - How do we arrive at a particular interpretation of what others say? - What makes us end up with a specific interpretation of an utterance? - Why do we select or reject an interpretation? In their search for answers to these questions, Sperber and Wilson challenged many tacitly accepted assumptions about how languages are used, how the human mind works and what might happen in it while processing input. Thus, they seriously questioned the code model of communication, which was deeply entrenched in the western linguistic tradition, and showed its many drawbacks, gaps and inconsistencies. Some progress had been made when the role of inference in communication was taken into consideration. Indeed, the also influential work by Grice (1957, 1975) – wherein relevance theory is rooted and to which it is greatly indebted– meant a huge leap. However, Sperber and Wilson (1986/1995) went well beyond the so-called Gricean pragmatics by proposing a model that draws from valuable insights from disciplines such as philosophy of mind, developmental psychology or cognitive anthropology, to name but some. Thus, relevance theory incorporated some of their notions and views to pragmatics in order to offer a profoundly cognitive, psychological perspective that has implications for the nature and role of semantics and pragmatics. 1.1. Intention, manifestness and cognitive environments Like Grice (1975), Sperber and Wilson (1986/1995) conceive of communication as an intentional activity, in which the speaker is prompted by two intentions: (i) her informative intention, which is her desire to make manifest to the audience a particular set of assumptions, and (ii) her communicative intention, which is her desire that the audience recognise that she actually has a particular informative intention. Recognising these two intentions requires theory of mind abilities –i.e., being able to attribute mental states to other individuals– and, therefore, metarepresenting the intentions and beliefs of others –i.e., creating representations of other public or private representations (Apperly 2012). These abilities make it possible for individuals to understand whether the communicator believes what she says is true –first-order mental states– or what the communicator knows the audience know, what the communicator intends the audience to believe, or what the communicator believes the audience believe –second-order mental states (Leekam 1991; Happé 1994; Sullivan et al. 1995; Sullivan et al. 2003; Wilson 2013). These layers of metarepresentation can be depicted as follows (Sperber 1994: 195): Speaker intends (attribution of communicative intention - 4) me to know (communicative intention - 3) that she intends (attribution of informative intention - 2) me to believe (informative intention - 1) that p The notion of manifestness is a remarkable innovation. A certain fact or assumption is manifest to an individual at a certain time if that individual is capable of creating a mental representation of it and accepting that representation as true or probably true (Sperber and Wilson 1995: 39). The set of facts or assumptions of which an individual actually has a mental representation at a certain moment, or of which he may potentially form a mental representation, make up his cognitive environment. If two (or more) individuals are capable of forming similar, though not completely identical, representations of the same physical or psychological objects, those individuals share a mutual cognitive environment. Through the notion of manifestness Sperber and Wilson (1995: 18) overcome the drawbacks of notions like common knowledge (Lewis 1969) or mutual knowledge (Schiffer 1972), which implied that, for communication to succeed, individuals should mutually share a certain amount of knowledge. It is virtually impossible to distinguish the amount (and type) of knowledge that two (or more) individuals merely share from the knowledge that is truly mutual. Perceptual systems and cognitive mechanisms act like filters and greatly determine what individuals perceive and how they represent it. What is more, checking that two (or more) individuals actually share some knowledge would require time-consuming and almost never-ending mental operations. However, the innovativeness of relevance theory does not end here. Though sharing the Gricean view of communication as an intentional activity, Sperber and Wilson also react against some of Grice’s (1957, 1975) own ideas by arguing that, instead of the Cooperative Principle, two other general principles govern human cognition and communication. 1.2. A theory based on two general principles Although based on Gricean pragmatics, relevance theory builds on it by rejecting some of its foundational postulates. Grice (1957, 1975) claimed that the Cooperative Principle and a series of maxims –those of quantity, quality, relevance and manner– govern communication. Communicators may abide by, covertly violate or blatantly flout those maxims in order to convey implicit contents and achieve certain effects. Also, Grice (1975) acknowledged the existence of other maxims of a social or aesthetic nature, such as that of politeness, which led to some adherents to his ideas to elaborate on those maxims and suggest different models to account for (im)politeness (e.g., Lakoff 1973; Leech 1983). Others, in turn, modified the number or raised the status of the original maxims to that of principles (e.g., Horn 1996; Levinson 2000), thus giving rise to neo-Gricean approaches. However, the Cooperative Principle and its maxims seemed to be based on intuitions about communication and observations of a series of regularities. Relevance theory, in contrast, is a post-Gricean model; it questions the existence of the Cooperative Principle and its maxims, since their origin is unclear, they do not seem to have universal validity and their operation seems to have different effects depending on circumstances (Wilson and Sperber 1991a: 381, 1991b: 586). If anything, the Gricean Principle and maxims would have the status of (cultural) norms, understood as “[…] internalised, unconscious patterns that the individual follows without even noticing that he is complying with an unwritten model” (Escandell Vidal 2004: 349). Relying on a constant tendency that has propelled the evolution of the human species in general and the human mind in particular and greatly contributed to its efficiency –maximisation of gain in exchange of effort invested– Sperber and Wilson reject the Cooperative Principle and its maxims and propose two very general principles that govern communication and cognition 1 . Indeed, by ‘principle’ they understand a formalization of how a particular system works; in other words, a “[…] causal, mechanical explanation” (Escandell Vidal 2004: 349). These two principles are the cognitive and communicative principles of relevance. The former states that human cognition is geared towards the maximization of relevance and is argued to reflect how the human mind functions. The latter, in turn, claims that every act of intentional communication communicates a presumption of its own optimal relevance. This second principle operates in all cases of intentional communication and is responsible for the selection of an interpretation out of the many possible ones that an utterance may have: Communicators and audience need no more know the principle of relevance to communicate that they need to know the principles of genetics to reproduce. Communicators do not ‘follow’ the principle of relevance; and they could not violate it even if they wanted to. The principle of relevance applies without exception […] It is not the general principle, but the fact that a particular presumption of relevance has been communicated by and about a particular act of communication, that the audience uses in inferential comprehension. (Sperber and Wilson 1995: 162) These two new principles are based on the notion of ‘relevance’, which is the true cornerstone of the theory. Sperber and Wilson also characterise this notion in more precise terms than Grice (1957, 1975) actually did. 1.3. An underlying key notion Grice (1975) included the maxim of relation and worded it as “be relevant”, but unfortunately he did not define what he took ‘relevance’ to be, nor did he clearly explain its role in communication. In contrast, Sperber and Wilson (1986/1995) portray relevance as a property of the stimuli individuals produce and, therefore, of a very special sub-set thereof: utterances. They define it on the basis of two factors: (i) The positive cognitive effects that information yields, or the improvements to our mental representation of the world or set of beliefs about it. Those improvements may be strengthening of previous beliefs, contradiction and eventual rejection of those beliefs or the formation of new beliefs from the interaction of previous ones with new information. 1 In their 1986 work Sperber and Wilson only proposed a single principle. (ii) The processing or cognitive effort required by an item of information. This depends on the complexity of the linguistic form of the utterance that conveys that item of information (i.e., its syntactic structure, lexical items, etc.) or the effort of memory needed to retrieve or select a suitable context for processing it. These two factors are also essential for understanding what the presumption of optimal relevance involves: (a) utterances will normally be relevant enough –i.e., they will result in enough cognitive effects– for the hearer to decide to invest the cognitive effort necessary to process them, and (ii) the formulation of utterances will normally be the one the speaker thinks, given her abilities and preferences, will result in a satisfactory amount of cognitive effects (Sperber and Wilson 1995: 270). The first part of this presumption means that hearers normally expect some cognitive reward which they cannot obtain otherwise, i.e., positive cognitive effects or “[…] a worthwhile difference to [their] representation of the world […]” (Wilson and Sperber 2002: 251). In turn, the second part means that the speaker will normally be interested in producing utterances that are easily comprehensible and provide the hearer with enough evidences for the intended cognitive effects or additional cognitive effects rendering utterances optimally relevant (Wilson and Sperber 2002: 256-257). Nevertheless, the speaker’s performance will depend on her own cognitive skills and capabilities, which may be conditioned by absentmindedness, tiredness, boredom, etc., and her goals, among which is complying with norms dictating, for instance, register, amount or type of information to dispense, formality, etc. (Mazzarella 2013: 33-35). In addition to the two principles of relevance, Sperber and Wilson (1995: 166) initially proposed a criterion of consistency with the principle of relevance, according to which an interpretation is consistent with the principle of relevance if, and only if, the speaker expects it to be optimally relevant. This means that, if the hearer finds an interpretative hypothesis optimally relevant, he should not think that the speaker intended a particular utterance to be optimally relevant under another interpretation. The hearer will think that an interpretative hypothesis was intended upon checking that it yields enough cognitive effects that offset a reasonable amount of cognitive effort. By clearly defining relevance and arguing that the constant search for optimal relevance governs human cognition and communication, Sperber and Wilson go well beyond Gricean pragmatics. However, as pointed out, relevance theory also reacts against the well-established code model of communication. Indeed, Sperber and Wilson conceive of communication as a human activity requiring a great amount inference, which has significant implications for understanding the role of semantics and pragmatics. 1.4. A new conception of communication The code model metaphorically depicted communication as a process in which the code enables the speaker to package her thoughts (i.e., encoding) in a parcel (i.e., utterance) and send it over to the hearer (i.e., articulating and speaking). The hearer’s task is to open that parcel and match the encoded input to corresponding semantic representations (i.e., decoding) in order to decipher, so to say, what the speaker means. However, the code alone does not suffice to arrive at speaker meaning: not everything that the speaker means is encoded and there is no guarantee that the hearer interprets what she means correctly (Wilson and Sperber 1991b: 584-585; Sperber and Wilson 1995: 27). Hearers have to segment sounds and delimit words; parse and disambiguate constituents; narrow or broaden concepts; assign reference to some expressions; recover elided linguistic material; determine the attitude the speaker projects towards the proposition expressed, her degree of certainty about it or if she is performing a certain action, and supply any assumption that is necessary for arriving at implicit contents. All these tasks rely on inference, so Sperber and Wilson (1987: 698, 1995: 10) opt for describing communication as an ostensive-inferential activity. It is ‘ostensive’ because the speaker shows something to the hearer –an utterance, which is indirect evidence for her communicative intention– and uses it to attract and direct the hearer’s attention to something: her informative intention. Communication is ‘inferential’ because the hearer has to work out the speaker’s meaning and her underlying intention when drawing the hearer’s attention 2 . Grice (1957, 1975) emphasised the reliance of communication on inference, but he envisaged its role as limited to the determination of the spatio-temporal coordinates of an utterance, assignment of reference to some expressions and recovery of the implicit contents. Accordingly, hearers arrive at ‘what is said’ through decoding and very little inference, while arriving at ‘what is implied’ greatly depend on inference. This unveils a view of the role of semantics and pragmatics in communication in which the former contributes to the recovery of the proposition expressed, while the latter was crucial for arriving at implicit contents (Wilson and Sperber 1993: 3). On the contrary, by proving that inference intervenes in all the 2 As a cognitive-pragmatic model, relevance theory focuses on comprehension and aims to account for how hearers arrive at speaker’s meaning through a series of simultaneous inferential tasks. This should not mean that Sperber and Wilson (1986/1995) are not aware of the crucial role that inference plays in production. tasks listed above, Sperber and Wilson (1986/1995) present a more intricate picture of comprehension where pragmatic processes are indispensable for the recovery of explicit content. Showing that communication involves much more than an encoding-decoding process also leads Sperber and Wilson to contend that individuals may use utterances in order not to simply pack and send over their own thoughts. By means of utterances communicators may reproduce words or phrases, represent diverse states of affairs and even allude to the thoughts of other people. 1.5. Utterances as metarepresentations Sperber and Wilson (1986/1995) believe that individuals can employ any natural or artificial phenomenon in order to represent another phenomenon it somehow resembles in certain respects. By means of utterances, which are public representations –i.e., perceptible, audible– individuals represent their own private, mental representations. Moreover, utterances may represent existing or desirable states of affairs or the thoughts or representations that an individual thinks another individual, or group of individuals, entertains. This is the basis for Sperber and Wilson’s (1995: 228-232) distinction between descriptive and interpretive dimensions of language usage. Since utterances are representations of representations, they have metarepresentational uses (Noh 2000). One of these is metalinguistic, when utterances publicly reproduce other public representations like words, phrases or utterances, or their logical or conceptual content. Clear examples are direct speech and quotations. Another metarepresentational use is interpretive, when utterances publicly resemble other public or private representations, like other utterances or thoughts. A typical example is indirect speech. Furthermore, if utterances metarepresent the thoughts or utterances that (an)other individual(s) entertain(s) or say(s), they are used attributively or function as attributive metarepresentations, as the source of those thoughts or utterances can be identified more or less easily (Wilson 1999: 148). In contrast, if utterances only represent a word, phrase or sentence produced by an unidentifiable source, they work as non-attributive metarepresentations. Examples of non-attributive metarepresentations are (Wilson and Sperber 1988; Wilson 1999): a) Negative and disjunctive sentences, which metarepresent possible information or thoughts. b) Interrogative and exclamative sentences. These metarepresent desirable information or thoughts. Interrogatives are requests for information if the this volume is divided address topics that have traditionally intrigued them and areas where significant contributions and progresses are being made. Moreover, although the chapters in each part delve into specific issues or phenomena, they have different orientations and implications for fields as diverse as linguistic description, morphology, machine translation, rhetoric and argumentation, or interlanguage pragmatic development. 3.1. Issues on procedural meaning and procedural analyses The first part groups four papers that rely on the notion of procedural meaning. The first two papers look into the interaction of procedural elements with intonation, thus broadening the scope of procedural analyses, while the third paper presents what can be considered a traditional relevance-theoretic procedural analysis of some linguistic elements in an African language, thus contributing to its description. Finally, one paper explores the procedural meaning of verbal tenses and shows the usefulness of procedural analyses for a field like machine translation. The first chapter is “The speaker’s derivational intention”, by Thorstein Fretheim. Based on the notion of derivational intention, which alludes to the route the audience should follow in order to reach the content of the speaker’s informative intention, this chapter shows the usefulness of that notion and argues that it cannot be separated from that of informative intention. Fretheim claims that the speaker’s derivational intention is constrained by encoded conceptual semantics and encoded procedural semantics alike. However, a given piece of procedural information may be at odds with the speaker’s derivational intention in a given context, and hence with her informative intention. Accordingly, he contends that the procedural meaning encoded by one expression may override and set aside the procedural meaning encoded by a co-occurring expression, if there is an intuitively felt conflict between the constraints on interpretation that the two expressions encode. In such cases, the more powerful procedural constraint is consonant with the speaker’s derivational intention and the less powerful procedural constraint is not. By means of two sections on the pragmatic functions of certain intonational phenomena in Norwegian, Fretheim illustrates how native speakers differently ranked co-occurring procedural constraints on pragmatic interpretation. This caused their audience to adhere to the procedural information conveyed by the expression with the higher rank and to disregard the information coming from the more modestly ranked encoder of procedural meaning. The first section reports on results from a listening comprehension experiment. It showed that the more highly ranked procedural constraint prevailed. The second section discusses the extent to which Optimality Theory may be a suitable framework for handling conflicts between cooccurring procedural constraints. The second chapter is “Cracking the chestnut: How intonation interacts with procedural meaning in Colloquial Singapore English lah”, by Lee Junwen and Kim Chonghyuck. Also assuming that procedural meaning may interact with intonation, it looks into the different pragmatic functions of this particle, which are argued to result from the interaction between its unitary semantic meaning and the effect of pitch. This chapter questions previous analyses of this particle as a marker of solidarity, warmth or informality, attenuation or emphasis, assertion or accommodation. Due to the variety of pitch contours with which this particle can be pronounced, it had generally been regarded as either a set of homonymic variants, or as a unitary particle with a monolithic meaning despite tonal differences. For Junwen and Chonghyuck, ‘lah’ describes the preceding proposition as being of high strength. The falling tone characteristic of declaratives leads the hearer to interpret that proposition as referring to an actual situation, while the rising tone characteristic of interrogatives leads him to interpret it as alluding to a desirable thought. The various pragmatic functions ascribable to ‘lah’ then arise as a consequence of processing. Additionally, this chapter accounts for a particularly troublesome phenomenon: the ability of ‘lah’ to pragmatically strengthen declaratives but to weaken imperatives. Through the chapter, its authors show that relevance theory provides a useful framework for analysing the effect of intonation on the processing of this discourse particle, which opens up new ways of characterising other particles in Singapore English. In “Reference assignment in pronominal argument languages: A relevancetheoretic perspective”, Helga Schröder adheres to the relevance-theoretic view that reference assignment is part of the process of explicature construction and that pronominal expressions procedurally constrain it (Wilson and Sperber 1993; Sperber and Wilson 1995). By instructing hearers to pick out a reference, pronominals contribute to the computational side of comprehension. Based on data from Toposa, an Eastern Nilotic language spoken in South Sudan, and Kiswahili, spoken in Kenya, Shröeder demonstrates that two procedures are involved in reference assignment in African pronominal argument languages: an incorporated pronoun in the verb helps identify the referent, while that pronoun in the verb helps an attributive expression achieve referential status. Quite often, African languages do not mark their nominal expressions for definiteness or indefiniteness, so they enter the conversation as undetermined. Disambiguation of such expressions requires pragmatic enrichment. Those languages seem to follow a ‘double-strategy’ in terms of reference assignment. Pronominals can be marked in the verb or occur independently. Independent pronouns do not encode procedural information guiding the selection of the correct referent, whereas incorporated pronouns do. Processing independent pronouns, Schröeder concludes, requires extra effort but yields cognitive effects related to focus identification and contrastive focus. The last paper in this part has an empirical orientation and combines corpus work and linguistic experiments. In “Conceptual and procedural information for verb tense disambiguation: The English simple past”, Cristina Grisot, Bruno Cartoni and Jacques Moeschler seek to elucidate which features should be included in a model explaining and predicting cross-linguistic variation in the translation of tenses. Thus, the authors also try to explain how a source language verb tenses may be disambiguated in order to choose from among different translation possibilities and which features could contribute to improving the output of Statistical Machine Translation systems. Corpus analysis reveals a lack of correspondence between English and French tenses. One of the most frequent divergences is that of the English simple past. Its semantic and pragmatic domains may be rendered in French through three tenses: passé composé, passé simple and imparfait. The authors assume that the conceptual and procedural contents of the English simple past can be used as disambiguation criteria in order to search for a French equivalent. They test this hypothesis through annotation experiments, whose results partially confirm their hypothesis. 3.2. Discourse issues The two chapters in the second part of this book are connected by the fact that they address discourse. While the first chapter deals with the clues that hearers can rely on in order to perceive discourse as ironical, the second one analyses how specific manifestations of discourse, namely, rhetorical and ironical questions, may be distinguished. Obviously, these two chapters are further connected by the fact that they address irony from a relevance-theoretic perspective. Francisco Yus Ramos had previously argued that irony comprehension involves the simultaneous or sequential activation of one or several contextual sources. A successful activation of these sources, especially the saturation of information that they may give rise to, leads to the so-called “criterion of optimal accessibility to irony”. Also, the simultaneous or sequential activation of these sources generates a number of prototypical cases in the comprehension of irony. This view of irony, the author thinks, is fully compatible with the extant relevance-theoretic approach based on the idea that irony is a case of echoic metarepresentational use of language, in which the speaker expresses a dissociative attitude towards certain words or thoughts attributable to some other (group of) individual(s). In “Relevance theory and contextual sources-centred analysis of irony: Current research and compatibility”, Francisco Yus Ramos has two aims. On the one hand, he checks whether those prototypical cases of irony are still valid and do cover all the possible ironic situations, together with the introduction of a more fine-grained notion of narrowed cognitive environment. On the other hand, he assesses to what extent his approach fits the latest relevance-theoretic research on irony such as the one involving different types of metarepresentation that are activated in successful irony comprehension. Next, “Distinguishing rhetorical from ironical questions: A relevance-theoretic account”, by Thierry Raeber, aims to offer criteria for differentiating rhetorical from ironical questions. Some of their features have been conflated on the grounds of the interpretative effects ironical questions seem to yield. Indeed, ironical questions have often been regarded as subtypes of rhetorical questions (Bonhomme 2005), or vice versa, rhetorical questions have been considered as subtypes of ironical questions (Gibbs 2000) owing to the obvious answer they call for, their persuasive power and the cognitive effects they result in. However, Raeber suggests that rhetorical and ironical questions are to be distinguished because the former implicitly assert a proposition bearing relevance – in most cases the answer itself or its consequences– while such implicit proposition, though manifest to the hearer, is not relevant at all by itself in ironical questions. The proposition implicitly rhetorical questions assert is recoverable thanks to a biased choice of answer. In contrast, ironical questions exhibit their in-context inaccuracy, thus giving rise to specific effects resulting from the contextual absurdity of pragmatic expectations, which are to be mobilized in order to motivate the question. Accordingly, Raeber proposes that failure at satisfying the hearer’s expectations of relevance triggers the attitudinal, non-propositional effects associated with ironical questions. 3.3. Interpretive processes The third part of the book includes two chapters that consider interpretation. While the first one centres on the interpretive processes by learners of a second language, the second one analyses how different linguistic elements are put to work in order to guide interpretation. The first chapter shows that relevance theory has significant implications for, and very helpful applications to, the field of interlanguage pragmatics. The second chapter, in turn, offers interesting insights into evidentials. Furthermore, both chapters are connected by the fact that, although in different domains, they take into account the role of epistemic vigilance mechanisms. These mechanisms have recently started to attract the interest of relevance theorists and are currently receiving due attention from them, as these works show. Assuming that little attention has been paid to the cognitive underpinnings of L2 learners’ pragmatic competence, Elly Ifantidou addresses the role of vigilance mechanisms in interlanguage pragmatic development in “Relevance theory, epistemic vigilance and pragmatic competence”. The author evaluates the effect of explicit pragmatic instruction in an EFL academic context and of language proficiency on learners’ metapragmatic awareness. This is an empirical study which considers different conditions of treatment, as determined by type of pragmatic input and level of language proficiency. It compares data from three groups of undergraduate students of English Language and Literature, which are used as control and experimental groups. Through newspaper editorials, Ifantidou examines how learners may exercise epistemic vigilance towards the source of information and thus avoid being accidentally or intentionally misinformed. In the metapragmatic awareness task she administered, learners relied on (i) relevance-theoretic mechanisms for comprehension of content and search for relevance, and (ii) epistemic vigilance for acceptance of both the content and source of communicated information. Since understanding is a precondition, even though not sufficient, for believing (Sperber et al. 2010: 368; Wilson 2010), learners’ pragmatic competence may benefit from sophisticated mindreading if they exercise epistemic vigilance towards the informant’s (i.e., newspaper journalist’s) epistemic states (e.g., acceptance, doubt, rejection) and intentions (e.g., to inform, to mislead). Therefore, the development of L2 learners’ pragmatic competence is contingent on the development of their epistemic vigilance mechanisms. The second paper analyses the interaction of epistemic vigilance with genre and evidentials –i.e., grammatical elements indicating the type of evidence the speaker has for making a statement. Authored by Christoph Unger, “Evidentials, genre and epistemic vigilance” shows that existing accounts have mainly explored the contribution of these elements to comprehension of speaker’s meaning. However, communicators may exploit evidentials in order to indicate that they are competent in distinguishing their information sources (Wilson 2011). Hence, evidentials help communicators portray themselves as trustworthy individuals and enable the audience to evaluate the evidential status of the communicated information. Drawing from Wilson (2011), Unger argues that evidentials may be used as genre indicators. Indeed, Aikhenvald (2004) had shown that reported evidentials very often appear in traditional narratives. This suggests that reported evidentials and traditional narrative genres may be conventionally related. Accordingly, the author contends that, as opposed to general markers of metarepresentational use and other indicators of information source, true reported evidentials raise the activation of a whole array of cognitive mechanisms specialised in checking support and coherence of the communicated content with existing beliefs. Furthermore, the relevance of traditional narratives resides in the validity of cultural values or norms communicated through exemplification. For this reason, the processing of such narratives strongly engages the argumentation module, whose activation is raised by reported evidentials. 3.4. Rhetorical and perlocutionary effects of communication Before closing this book with additional directions for future research, two chapters discuss the effects of communication from a relevance-theoretic angle. The first chapter in this part also discusses the role of epistemic vigilance in argumentation, thus linking this part to the preceding one. “Rhetoric and cognition: Pragmatic constraints on argument processing” adopts a cognitive-pragmatic perspective on rhetorical effectiveness in order to hypothesise that the informationselection mechanisms in interpretation positively influence the outcome of subsequent argumentative evaluation. Moreover, author Steve Oswald moves for the inclusion of a cognitive-pragmatic component in a theory of argumentation, which has typically refrained from adopting cognitive standpoints (van Eemeren and Grootendorst 2004: 74). Indeed, individuals need to understand arguments before evaluating them, which makes comprehension, to some extent at least, responsible for the selection of premises that will constitute the input to the evaluative stage of argumentation. Accordingly, arguers who want to convince their audience can be said to seek to constrain their interpretative processes so that they only presumably select ‘rhetorically-friendly’ contextual premises. Rhetorical effectiveness is defined as the propensity for an argument to ensure that its conclusion appears to be optimally relevant with respect to its premises. In other words, successful arguments are those whose conclusions (i) are derived with little effort and (ii) generate significant cognitive effects given a series of premises. This definition encompasses both cases of sound argumentation, where potential counter-arguments are weighed and dismissed as epistemically weaker than the argument under consideration, and cases of fallacious argumentation, whose success rests on the audience’s inability to mobilise counter-evidence and discard the argument as fallacious (Maillat and Oswald 2009, 2011; Oswald 2010, 2011). Fallacious moves are therefore seen as attempts to manipulate those two conditions by making ‘rhetorically-friendly’ assumptions epistemically strong and accessible, and ‘rhetorically-unfriendly’ assumptions weak and less accessible. In turn, the second chapter focuses on perlocutionary effects, which should not be merely understood as consequences of the comprehension process, but as phenomena influencing the way communicated stimuli are actually processed. “Perlocutionary effects and relevance theory”, by Agnieszka Piskorska, narrows down the traditional Austinian definition of perlocution and confines it to intentionally evoked mental states, which may have a cognitive or affective nature. Although the notion of ‘perlocution’ was initially alien to relevance theory, there has recently been a growing interest in the speech-act theoretic concept (e.g., Jary 2010; see also Witczak-Plisiecka 2013 for a review). Sharing this interest, Piskorska briefly overviews relevance-theoretic insights into speech acts, such as its reductionism of speech act types or its account of illocutionary force in terms of procedural meaning. Then, she presents arguments supporting her claim. The main one is that on many occasions effects like amusement, warning/threatening or offending are the reasons why individuals interact. If relevance is assessed solely in terms of positive cognitive effects, something important is missing. Indeed, extant relevance-theoretic analyses of politeness phenomena (Escandell Vidal 1998, 2004) and metaphor (Pilkington 2010) question the view that meaning equals the sum of explicatures and implicatures. Emotion and cognition must be integrated in pragmatic analyses of meaning and its consequences, as argued in (neuro)psychology (e.g., Damasio 1994). Therefore, Piskorska makes two suggestions about the relationship between comprehension and affective states, which are consistent with the massive modularity model of the mind (Sperber 2005). Nevertheless, experiments on the role of perlocutions are needed, although their results could be limited due to the difficulties at modelling actual emotions in experimental settings. References Aikhenvald, Alexandra. Evidentiality. Oxford: Oxford University Press, 2004. Allot, Nicholas E. “Relevance and Rationality”. UCL Working Papers in Linguistics 14 (2002): 69-82. Apperly, Ian. Mindreaders. The Cognitive Basis of Theory of Mind. New York: Psychology Press, 2012. Berbeira Gardón, José L. “Posibilidad epistémica, posibilidad radical y pertinencia”. Pragmalingüística 1 (1993): 53-78. Berbeira Gardón, José L. “Relevance and modality”. Revista Alicantina de Estudios Ingleses 11 (1998): 3-22. Blakemore, Diane. Semantic Constraints on Relevance. Oxford: Basil Blackwell, 1987. Blakemore, Diane. “’So’ as a constraint on relevance”. In Mental Representations. The Interface between Language and Reality, edited by Ruth M. Kempson, 183- 195. Cambridge: Cambridge University Press, 1988. Blakemore, Diane. “Denial and contrast: A relevance theoretic analysis of but”. Linguistics and Philosophy 12 (1989): 15-37. Blakemore, Diane. Understanding Utterances. Oxford: Blackwell, 1992. Blakemore, Diane. “The relevance of reformulations”. Language and Literature 2 (1993): 101-120. Blakemore, Diane. “Relevance, poetic effects and social goals: A reply to Culpeper”. Language and Literature 3 (1994): 49-59. Blakemore, Diane. Relevance and Linguistic Meaning. The Semantics and Pragmatics of Discourse Markers. Cambridge: Cambridge University Press, 2002. Blakemore, Diane. “On the descriptive ineffability of expressive meaning”. Journal of Pragmatics 43/14 (2011): 3537-3550. Blass, Regina. “Pragmatic effects of co-ordination: The case of ‘and’ in Sissala”. UCL Working Papers in Linguistics 1 (1989): 32-51. Blass, Regina. Relevance Relations in Discourse: A Study with Special Reference to Sissala. Cambridge: Cambridge University Press, 1990. Bonhomme, M. 2005. “Flou et polyvalence de la question rhétorique: L’exemple des fables de La Fontaine”. In Les états de la question, edited by C. Rossari et al., 191-209. Québec: Nota Bene. Carston, Robyn. Thoughts and Utterances. The Pragmatics of Explicit Communication. Oxford: Blackwell, 2002. Carston, Robyn and Seiji Uchida (eds.). Relevance Theory. Applications and Implications. Amsterdam: John Benjamins Publishing Company, 1998. Clark, Billy. “Let and let’s: Procedural encoding and explicature”. Lingua 90 (1993): 173-200. Clark, Billy. Relevance Theory. Cambridge: Cambridge University Press, 2013. Curcó, Carmen. “Some observations on the pragmatics of humorous interpretations: A relevance theoretic approach”. UCL Working Papers in Linguistics 7 (1995): 27-47. Curcó, Carmen. “The implicit expression of attitudes, mutual manifestness, and verbal humour”. UCL Working Papers in Linguistics 8 (1996): 89-99. Curcó, Carmen. The Pragmatics of Humorous Interpretations: A Relevance- Theoretic Approach. PhD diss., University of London, 1997. Damasio, Antonio R. Descartes’ Error. New York: Putnam, 1994. Eemeren, Frans van and Rob Grootendorst. A Systematic Theory of Argumentation. The Pragma-dialectical Approach. Cambridge: Cambridge University Press, 2004. Escandell Vidal, M. Victoria. “Towards a cognitive approach to politeness”. Language Sciences 18 (1996): 629-650. Escandell Vidal, M. Victoria. “Politeness: A relevant issue for relevance theory”. Revista Alicantina de Estudios Ingleses 11 (1998): 45-57. Escandell Vidal, M. Victoria. “Norms and principles. Putting social and cognitive pragmatics together”. In Current Trends in the Pragmatics of Spanish, edited by Rosina Márquez-Reiter and M. Elena Placencia, 347-371. Amsterdam: John Benjamins, 2004. Escandell Vidal, M. Victoria, Manuel Leonetti and Aoife Ahern (eds.). Procedural Meaning. Problems and Perspectives. Bingley: Emerald/Brill, 2011. Fodor, Jerry. The Modularity of Mind. Cambridge, MA: MIT Press, 1983. Gibbs, R. W. 2000. “Irony in talk among friends”. Metaphor and symbol 15: 5-27. Grice, Herbert P. “Meaning”. Philosophical Review 66 (1957): 377-388. Grice, Herbert P. “Logic and conversation”. In Syntax and Semantics vol. 3: Speech Acts, edited by Peter Cole and Jerry L. Morgan, 41-59. New York: Academic Press, 1975. Groefsema, Marjolein. “Can, may, must and should: A relevance theoretic account”. Journal of Linguistics 31 (1995): 53-79. Gutt, Ernst-August. “Translation and relevance”. UCL Working Papers in Linguistics 1 (1989): 75-94. Gutt, Ernst-August. Translation and Relevance. Cognition and Context. Oxford: Basil Blackwell, 1991. Hall, A. 2007. Do discourse connectives encode concepts or procedures? Lingua 117(1): 149-174. Happé, Francesca G. E. “An advanced test of theory of mind: Understanding of story characters’ thoughts and feelings by able autistic, mentally handicapped, and normal children and adults”. Journal of Autism and Developmental Disorders 24/2 (1994): 129-154. Horn, Larry. “Presupposition and implicature”. In The Handbook of Contemporary Semantic Theory, edited by Shalom Lappin, 299-320. Oxford: Blackwell, 1996. Ifantidou, Elly. “Sentential adverbs and relevance”. UCL Working Papers in Linguistics 4 (1992): 193-214. Ifantidou, Elly. “Parentheticals and relevance”. UCL Working Papers in Linguistics 5 (1993): 193-210. Ifantidou, Elly. Evidentials and Relevance. Amsterdam: John Benjamins, 2001. Ifantidou, Elly. Pragmatic Competence and Relevance. Amsterdam: John Benjamins, 2014. Imai, Kunihito. “Intonation and relevance”. In Relevance Theory. Applications and Implications, edited by Robyn Carston and Seiji Uchida, 69-86. Amsterdam: John Benjamins Publishing Company, 1998. Itani, Reiko. “A relevance-based analysis of hearsay particles: Japanese utterancefinal tte”. UCL Working Papers in Linguistics 6 (1994): 379-400. Iten, Corinne. Linguistic Meaning, Truth Conditions and Relevance. The Case of Concessives. Basingstoke: Palgrave Macmillan, 2002. Jary, Mark. “Is relevance theory asocial?” Revista Alicantina de Estudios Ingleses 11 (1998a): 157-169. Jary, Mark. “Relevance theory and the communication of politeness”. Journal of Pragmatics 30 (1998b): 1-19. Jary, Mark. 2010. Assertion. Palgrave Macmillan. Jary, Mark. “Two types of implicatures: Material and behavioural”. Mind & Language 28/5 (2013): 638-660. Jodłowiec, Maria. The Role of Relevance in the Interpretation of Verbal Jokes: A Pragmatic Analysis. PhD diss., Jagiellonian Uiversity, 1991. Jodłowiec, Maria. “What’s in the punchline?” In Relevant Worlds: Current Perspectives on Language, Translation and Relevance Theory, edited by Ewa Wałaszewska, Marta Kisielewska-Krysiuk, Aniela Korzeniowska and Małgorzata Grzegorzewska, 67-86. Newcastle upon Tyne: Cambridge Scholars Publishing, 2008. Jucker, Andreas H. “The discourse marker well: A relevance-theoretical account”. Journal of Pragmatics 19 (1993): 435-452. Korzeniowska, Aniela and Małgorzata Grzegorzewska (eds.). Relevance Studies in Poland. Vol. 2. Warsaw: University of Warsaw, 2005. Lakoff, Robin T. “The Logic of Politeness; or, Minding Your P’s and q’s”. In Papers from the Ninth Regional Meeting, edited by Claudia W. Corum, Thomas C. Smith-Stark and Ann Weiser, 292-305. Chicago: Chicago Linguistic Society, 1973. Leech, Geoffrey. Principles of Pragmatics. London: Longman, 1983. Leekam, Susan R. “Jokes and lies: Children’s understanding of intentional falsehood”. In Natural theories of mind. Evolution, development and simulation of everyday mindreading, edited by Andrew Whiten, 159-174. Oxford: Basil Blackwell, 1991.