Communication, comprehension, and interpretation

1` Communication, comprehension and interpretation Deirdre Wilson Published in H. Colston, T. Matlock & G. Steen (eds.) (2022), Dynamism in Metaphor and Beyond (Festschrift for Ray Gibbs), pp. 143–155. Amsterdam: John Benjamins 1. Introduction Ray Gibbs’s wide-ranging and influential studies of non-literal uses of language offer valuable insights into the way that tropes such as metaphor, irony, metonymy, hyperbole and sarcasm are understood (Gibbs, 1994, 1999, 2017; Gibbs & Colston, 2012). As one of the pioneers of the cognitive turn in pragmatics, Gibbs has emphasised the centrality of the overt expression and recognition of intentions to utterance interpretation in both everyday discourse and more literary modes, making it possible to envisage a unitary pragmatic theory, with a set of dedicated pragmatic principles or mechanisms which account for the recognition of speakers’ intentions across the full range of overtly intentional communicative acts. However, Gibbs has also expressed reservations about the possibility of developing such a theory. Here are some of his reasons: (a) Figurative utterances are heavily context dependent. Context dependence in general presents a challenge to pragmatic theory, and the context dependence of figurative interpretation casts doubt on an assumption common to several experimental approaches, that a given type of figurative utterance tends to achieve similar rhetorical effects and incur similar processing costs in all discourse contexts. As Gibbs and Colston (2012, p. 198) put it, People’s experience of figurative language is not easily reduced to lists of meaning or rhetorical effects. Individual figures of speech may evoke characteristic meanings, attitudes and emotions, yet it is not always clear that people always automatically experience a particular kind of trope in the same way in all circumstances. (b) Pragmatic performance varies not only across but within individuals. According to Gibbs (2017, p. 323), A closer look at the experimental studies also indicates significant variability in people’s pragmatic behaviors. Not only do some people fail to adhere to standard pragmatic norms, many people in the context of a single experiment vary considerably from trial to trial in their performance. While some of the resulting variations are dismissed as mere ‘noise’, or ‘outliers’ in reports of experimental studies, Gibbs suggests that they may arise from “self-organizing processes” at the level of the cognitive system as a whole, rather than from the working of a dedicated pragmatic system (ibid., p. 324). (c) The interpretations of figurative utterances are often hard to pin down, involving a mixture of conceptual, perceptual and sensorimotor information that cannot be adequately rendered in terms of a finite literal paraphrase, as is standardly proposed: 2` People’s experience of figurative meaning must include rich layers of conceptual and embodied knowledge, ranging from propositions to different sets of metaphorical and metonymic mappings. Acknowledging this fact presents a significant challenge to psychological theories aiming to characterize the processes and products of figurative language use (Gibbs, 2017, pp. 198–199). The points Gibbs raises in (a)–(c) underline the context-dependence, variability and indeterminacy of much utterance interpretation. At the least, they show that any adequate pragmatic theory must allow for the routine integration of linguistic information with various types of contextual information in the course of utterance interpretation. However, Gibbs draws the more radical conclusion that there is no need to appeal to specialised pragmatic principles or mechanisms in explaining how utterances are understood: people's choices of words, and how they are expressed, and listeners’ interpretations of linguistic utterances, unfold from dynamical processes that should not be reduced to some specialized set of norms, rules, or even principles which are presumably encoded in the minds of individuals. It may very well be true that people often act co-operatively in linguistic interactions, and appear to be interested in what others may know and are intending to do. Yet these regularities are emergent from the human system overall and not the output of some specialized ‘pragmatics’ part of the mind (Gibbs, 2017, p. 324). In other words, “there is no individuated set of pragmatic modules which drive our linguistic and real-world actions” (ibid., p. 323). Gibbs’s scepticism about the need for dedicated pragmatic principles or mechanisms is nicely illustrated by his analysis of allegory. In his paper ‘The allegorical impulse’, he surveys a range of empirical findings which suggest that the ability to discern allegory “is not a specialized mode of interpretation, but a fundamental human impulse to draw diverse connections between concrete and more abstract experience” (Gibbs, 2011, p. 121). Gibbs sees the drive towards allegorical interpretation as so strong that “people sometimes can’t help but draw allegorical connections between what is said and broader themes”, regardless of the communicator’s intentions (ibid., p. 127). As a fictional example, he cites the character Chauncey Gardiner from Jerzy Kosinki’s novel Being There, “a plain, quite literal-minded gardener whose simple remarks are often over-interpreted as expressing allegorical meanings” (ibid., p. 127). For Gibbs, these over-interpretations arise not from the working of any specialised pragmatic principles or mechanisms but from a broader cognitive search for “connections between concrete and more abstract experience”. Among the pragmatic norms or principles whose status Gibbs questions are those proposed by Grice, Searle and Sperber and Wilson. Although neither Grice nor Searle sees these principles as linked to “some specialized ‘pragmatics’ part of the mind”, Sperber and Wilson explicitly reject the common view that pragmatic interpretation is simply a matter of applying general cognitive abilities to a particular (communicative) domain: Verbal comprehension presents special challenges, and exhibits certain regularities, not found in other domains. It therefore lends itself to the development of a dedicated 3` comprehension module with its own particular principles and mechanisms (Sperber & Wilson, 2002/2012, pp. 262–263). In this paper, I consider how relevance theory, an approach to pragmatics which incorporates such a dedicated comprehension mechanism or module, might deal with the contextdependence, variability and indeterminacy of utterance interpretation illustrated in (a)–(c) above. In section 2, I outline the main points of relevance theory, which sees human communication and cognition as intimately intertwined and treats general cognitive abilities including perception, memory and inference as playing a crucial role in comprehension. In section 3, using allegory as an example, I distinguish comprehension – the process of identifying a communicator’s intentions – from a broader interpretation process which includes drawing one’s own conclusions from an utterance regardless of the speaker’s intentions. In section 4, using metaphor as an example, I consider the issue of vagueness or indeterminacy in interpretation. Section 5 is a brief conclusion. 2. Relevance, cognition and communication Relevance theory’s central claim is that human cognition and communication are both governed by the search for relevance.1 Any stimulus (for instance, a sight, a sound, an utterance, a thought) that provides an input to cognitive processes is relevant to an individual when it combines with a context of available assumptions to yield positive cognitive effects (derivable from input and context together but from neither alone).2 According to the Cognitive Principle of Relevance, as a result of constant selection pressures, the human cognitive system as a whole has evolved a wide variety of mechanisms that conspire to allocate attention and processing resources to inputs with the greatest expected relevance (where relevance increases with cognitive effects and decreases with processing effort required) and to process these inputs in a context that maximises their relevance (Sperber & Wilson, 1986/1995, pp. 260–272). A single utterance may have a variety of cognitive effects for the addressee, not all of which will have been intended by the speaker. As I speak, you may be drawing conclusions about my social status and origins, my prejudices and preferences, the social relations between us, and so on, and these may help to make my utterance relevant to you, whether or not I intended you to draw them. It is therefore worth distinguishing between comprehension and interpretation, where comprehension is the process of recognising the intended import of an utterance – that is, the array of propositions that the speaker overtly intended to communicate – and interpretation, which includes a broader process of drawing one’s own conclusions as part of the overall cognitive search for relevance. Communication, then, takes place against a background where the addressee is permanently engaged in a cognitive search for relevance, and identifying the intended import of an utterance is only one of his goals (albeit an important one, since much of what we know 1 2 For brief outlines of relevance theory, see Wilson & Sperber (2004); Wilson (2017, 2019). Positive cognitive effects include warranted strengthenings or revisions of contextual assumptions, and contextual implications (derived from input and context together, but from neither alone). 4` could only have been acquired through communication). What pragmatic theory has to explain is not how the addressee draws his own conclusions from an utterance (this falls within the domain of a theory of cognition), but how he constructs a hypothesis about its intended import, by disambiguating ambiguous expressions, assigning reference to referential expressions, adjusting lexical meanings, assembling an appropriate set of contextual assumptions and deriving contextual implications and other cognitive effects, in a way the speaker could manifestly have foreseen and intended. Relevance theory’s account of communication centres on the notion of an ostensive act: an overtly intentional communicative act designed to attract the addressee’s attention and focus it on the intended import. Common cues to ostension include catching someone’s eye, touching them, pointing, showing them something, speaking and writing. According to the Communicative Principle of Relevance, ostensive acts create expectations of relevance not raised by other stimuli, and this regularity in the domain of ostensive communication provides the basis for an automatic heuristic that addressees can use (of course at a risk) to infer the intended import of ostensive acts addressed to them: Relevance-guided comprehension heuristic Follow a path of least effort in deriving cognitive effects. Test interpretive hypotheses (disambiguations, reference resolutions, lexical adjustments, contextual assumptions, contextual implications, etc.) in order of salience. Stop when you have enough cognitive effects to satisfy the expectation of relevance raised by the utterance.3 This heuristic yields reliable results only in the domain of ostensive communication, where the communicator wants her intentions to be recognised, is actively helping the addressee to recognise them, and would acknowledge them if asked. In other cognitive domains, the fact that a certain conclusion is highly salient has no epistemic advantage at all. The relevanceguided comprehension heuristic is therefore seen as part of a dedicated comprehension mechanism or module which exploits a regularity that exists only in this particular domain (Sperber & Wilson, 2002/2012). According to relevance theory, the Communicative Principle of Relevance is not (as Gibbs puts it) “encoded in the minds of individuals”. It is not a rule that individuals are expected to follow, or can violate at will: Communicator and audience need no more know the [communicative] principle of relevance to communicate than they need to know the principles of genetics to reproduce. Communicators could not ‘follow’ the [communicative] principle of relevance; and they could not violate it if they wanted to (Sperber & Wilson, 1986/1995, p. 162). 3 The literature on Questions under Discussion (QUDs) offers some insights into how linguistic and contextual factors may interact to create particular expectations of relevance (Beaver, Roberts, Simons, & Tonhauser, 2017). 5` The Communicative Principle of Relevance is a lawlike generalisation designed to capture a regularity in the communicative domain, which a dedicated comprehension mechanism or module has evolved to exploit. The comprehension module, by contrast, is seen as a “specialized ‘pragmatics’ part of the mind”, which increases the efficiency of intention reading in the communicative domain by providing an automatic heuristic for inferring the intentions behind ostensive acts.4 3. Comprehension and interpretation As noted in section 2, the addressee of an ostensive act is permanently engaged in a cognitive search for relevance which may yield conclusions that the communicator did not necessarily intend or endorse. Indeed, in satisfying the expectations of relevance raised by an utterance, the addressee necessarily has to go beyond the intended import and draw some conclusions of his own. As Sperber and Wilson put it, what our theory of relevance implies is that one of the speaker’s intentions (and a crucial one) is that the hearer, by recognising the speaker’s intentions, should be made capable of going beyond them and of establishing the relevance of the utterance for himself… A successful act of comprehension (which is what is aimed at by both speaker and hearer) is one which allows the hearer to go beyond comprehension proper (Sperber & Wilson, 1982, p. 78). Thus, comprehension – the process of constructing a hypothesis about the intended import of an utterance – is embedded within a broader interpretation process guided by the cognitive search for relevance and based on contextual information available to the addressee at the time. Here is a real-life example which bears out Gibbs’s comment that “people sometimes can’t help but draw allegorical connections between what is said and broader themes”, regardless of the communicator’s intentions. It is taken from an interview with Philip Roth on his novel The Plot Against America (2004a), an alternative history in which Franklin D. Roosevelt loses the 1940 presidential election to the aviator Charles Lindbergh and the US becomes increasingly isolationist and anti-semitic: Philip Roth: That book was helped by a column in the New York Times by Frank Rich. (This was at the time Bush became President.) He said what I was doing was writing a kind of allegory about the Bush administration. And those years felt so powerless in the face of what they were up to that the book caught on. I had no intention at all of writing an allegory of the Bush administration. Interviewer: Can you understand, though, the fact that many readers and critics read that allegory into it – does that have any validity, or are they wrong? 4 There is some evidence that pre-verbal infants are sensitive to certain ostensive cues and treat them as indicating the presence of a communicative intention, leading to increased expectations of relevance (Csibra & Gergely, 2009; Csibra, 2010). 6` Philip Roth: No, they’re not wrong. It isn’t the way that I read the book, but each person makes use of the book in his or her own way (Front Row, BBC radio, June 2011). Roth makes clear that the allegorical interpretation was not part of the intended import of the novel, although he has no objection to readers constructing it on their own initiative. Indeed, in an article written before the book was published, he anticipates just such an interpretation and explicitly rejects it: Some readers are going to want to take this book as a roman à clef to the present moment in America. That would be a mistake (Roth, 2004b). While acknowledging this warning in his review, Frank Rich finds the allegorical interpretation hard to resist: As long as there's no explosive evidence to rain on that parade, Mr. Roth is entirely right to say that The Plot Against America cannot be squared with ‘the present moment in America.’ But what makes this book terrifying in its sly, even insidious way is that you can't read it without imagining how the combustible elements of our own home front might ignite if the present moment does not hold (Rich, 2004). The distinction between comprehension and interpretation helps in analysing this exchange. On the one hand, as Frank Rich points out, the allegorical interpretation added significantly to the relevance of the book for many readers at the time, for whom these parallels would be particularly salient.5 On the other hand, Philip Roth denies that the allegorical interpretation was part of the intended import, while acknowledging that “each reader makes use of the book in his own way”. This fits well with the idea that in identifying the intended import, the addressee must necessarily go beyond comprehension proper and draw some conclusions that the communicator need not necessarily either anticipate or endorse. How to draw the line between comprehension and interpretation: between conclusions that are plausibly treated as part of the intended import and those that are not? An addressee using the relevance-guided comprehension heuristic will treat as part of the intended import only those propositions that a speaker aiming to satisfy the expectations of relevance raised by the utterance could reasonably have intended him to derive. Consider the saying “A leopard can’t change its spots”, which has both a literal and an allegorical interpretation. In most circumstances, it is hard to see how a rational speaker could have expected this utterance to be relevant enough unless she intended to communicate the allegorical interpretation. By the same token, the simple literal remarks of Chauncey Gardiner in Being There are overinterpreted by sophisticated listeners with raised expectations of relevance, who cannot believe they could have been literally intended. The situation with The Plot Against America is quite different, since Roth spells out in a number of commentaries why, in his view, a literal interpretation of the novel is rich enough in implications to satisfy the expectations of relevance of many readers. In that case, the parallels to the Bush administration should not be 5 More recent readers tend to notice the parallels with Trump, which could clearly not have been foreseen or intended by the author (see Brody, 2017). 7` seen as part of the point of the novel, but merely as one of several directions that the reader is encouraged to explore in the cognitive search for relevance.6 4. Indeterminacy in communication In most current approaches to pragmatics, what is communicated by an utterance is described as a speaker’s meaning and seen as having three distinctive properties. It can be • • • rendered without loss as a single proposition, or a small set of propositions; duplicated in the minds of speaker and hearer; and therefore added to the common ground and taken for granted in the rest of the conversation. However, most utterances also communicate something less determinate and more nuanced than a speaker’s meaning. Here is an example of indeterminacy, from a book by an award-winning writer on landscape and nature, which is discussed in a paper on figurative utterances and mental imagery by Robyn Carston (2018): (1) … a heron launched itself from low ground to our south, a foldaway construction of struts and canvas, snapping and locking itself into shape just in time to keep airborne… (Macfarlane, 2013, pp. 298–299). What (1) evokes in the reader might be best described as a loose impression, perhaps interspersed with images, of how the heron looked as it took off. According to Carston, we might not ourselves have seen exactly the kind of foldaway structure of struts and canvas an image of which the writer has in his mind; we can, however, use our memory of similar kinds of objects or their parts (e.g., a tent made of canvas, with poles that can snap into position), together with a degree of imaginative reconstruction, to supply a more or less complete image which informs our understanding of how the heron looked as it took flight (Carston, 2018, p. 201). What (1) communicates is often described as a non-propositional effect. Characteristic features of such effects include the fact that • • • • different audiences paraphrase them in rather different ways; no finite paraphrase captures all their nuances; they are often described as “open-ended”; and they typically involve the activation of perceptual, emotional or sensorimotor mechanisms. Non-propositional effects present a challenge to pragmatic theories, which typically abstract away from them on the ground that they are hard to formalise (see Wilson & Carston, 2019). 6 For interesting work on allegory from a relevance theory perspective, see Unger (2017, 2018, 2019). On relevance theory and literary interpretation, see Cave & Wilson (2018). 8` These are just the type of effects that Gibbs draws attention to in (a)–(c) above. In my view, they present a much more genuine challenge to the standard notion of a speaker’s meaning than they do to the existence of dedicated pragmatic principles or mechanisms. From the first, relevance theorists have seen it as a central goal of pragmatics to account for indeterminacy and open-endedness in communication (Sperber & Wilson, 1986/1995, pp. 54–60). They therefore assume that the intended import of an ostensive act is not necessarily a speaker’s meaning as standardly conceived, but an array of propositions which may vary indefinitely in size. At one extreme, this import may consist of a single proposition; at the other, it may be a vast array of propositions; and there is a continuum of cases in between. Moreover, some members of this array may be more salient and strongly evidenced than others, and are therefore more likely to be attended to by the addressee and accepted as true. In relevance-theoretic terms, a proposition is manifest to an individual to the extent that he is likely to entertain it and accept it as true. Then instead of saying that the communicator’s goal is invariably to induce a determinate set of beliefs in the addressee, we can say that her goal is to make a certain array of propositions more manifest to him (Sperber & Wilson, 2015, sections 4–6). The idea that the intended import of an utterance consists of an array of propositions that are more or less manifest to the addressee helps in analysing vagueness or indeterminacy in communication. Let us say that a proposition is strongly communicated when the communicator clearly intends that particular proposition to be strongly manifest to the addressee, and weakly communicated when there is an array of weakly manifest propositions with more or less similar import that the addressee is encouraged to explore. Weak communication typically involves a degree of indeterminacy. Consider the phrase “a foldaway structure of struts and canvas, snapping and locking itself into place just in time to get airborne” in (1). What the addressee experiences on reading this phrase is an increase in the manifestness of a wide array of propositions that all have a bearing on the issue of how the heron looked as it took flight. He may be aware of this increase in manifestness without individually enumerating and attending to all the propositions in the array – or indeed without attending to any of them apart from the conclusion that this is how the heron looked as it took off. We might therefore describe an impression as a noticeable change in the manifestness of an array of propositions that all have a bearing on answering the same question, deciding the same issue, or understanding the same phenomenon (Sperber & Wilson, 2015). Then impressions fall squarely within the class of things that can be communicated by an ostensive act. Carston (2018) argues convincingly that the images evoked by figurative utterances can influence the impressions they convey. In interpreting the phrase “a foldaway construction …” in (1), for instance, retrieving the image of a canvas tent with poles that can snap into position might help to create the impression that the heron took off suddenly and haphazardly, in the manner of a tent blowing away, while retrieving a different image – say, of an early biplane with canvas wings supported by metal struts – might help to create the impression that the heron took off heavily and with difficulty, in the manner of an early biplane struggling to get off the ground. Although the author of (1) presumably had a 9` particular type of “foldaway construction” in mind, the phrase he has chosen does not provide enough evidence to pin his intentions down any further, and it is left to the addressee to come up with a hypothesis about what type of “foldaway construction” is involved. This is a case where the intended import is genuinely indeterminate and different addressees are likely to come up with different hypotheses, leading to different impressions of how the heron looked as it took off. Most current pragmatic theories assume that communication is a yes–no matter: a proposition is either communicated or it is not. However, the existence of weak communication brings that assumption into question. In strong communication, little initiative is left to the addressee, and the communicator is solely responsible for the truth of the propositions conveyed. Weak communication, by contrast, often calls for some creative input from the addressee, so that different addressees arrive at different interpretations for which they must take some of the responsibility themselves. As communication becomes weaker and more responsibility is left to the addressee, comprehension shades off into interpretation, and communication is no longer a yes–no matter but a matter of degree. 5. Concluding remarks Gibbs raises a number of issues which are often overlooked in pragmatics but which have to be dealt with in developing a unitary theory of communication. I have argued that it is possible to address these issues without accepting Gibbs’s conclusion that “there is no specialized ‘pragmatics’ part of the mind”. Certainly, communication and cognition are intimately intertwined: both are governed by the search for relevance, and the same perceptual, memory and inferential mechanisms play a central role in both. In particular, the context for utterance interpretation may be drawn from anywhere in the cognitive system, and crucially affects the outcome of the comprehension process.7 Since comprehension involves inferring the communicator’s intentions based on contextual information that happens to be available to the addressee at the time of utterance, it is no surprise that the outcome of the comprehension process may differ from individual to individual, or even for the same individual at different times, and that misunderstandings occur. As Sperber and Wilson (1986/1995, p. 45) put it, “failures in communication are to be expected: what is mysterious and requires explanation is not failure but success.” I have tried to show that what makes it possible for communication to succeed despite the diversity in the contents and organisation of individual cognitive systems is the existence of a “specialized ‘pragmatics’ part of the mind”: a dedicated comprehension module whose function is to detect the presence of ostensive acts and infer the intentions behind them. A key component of this module is a relevance-guided comprehension heuristic which increases the efficiency of intention reading in the communicative domain. I have also suggested that it is worth distinguishing between comprehension (identifying the intended import of an utterance) and interpretation (drawing one’s own conclusions from an utterance), and argued 7 For reflections on the role of context that fit well with the assumptions of relevance theory, see Degen & Tanenhaus (2019). 10` that in identifying the intended import of an utterance, the addressee necessarily has to go beyond the speaker’s intentions and draw some conclusions of his own. Finally, I have tried to show that communication may be stronger or weaker. In stronger forms of communication, the communicator makes clear that she intends the addressee to treat a certain proposition as part of the intended import, and thus takes full responsibility for its truth. In weaker forms of communication, the addressee is encouraged to explore an array of propositions with roughly similar import – the wider the array, the weaker the communication – and in choosing to accept any proposition from that range, he takes on some of the responsibility for its truth. In coded communication (the strongest form of communication), no initiative is left to the addressee as to the outcome of the comprehension process. In inferential communication, and particularly in weaker forms of inferential communication, much more is asked for from the addressee. The common practice in pragmatics of abstracting away from the challenges that arise in accounting for the full range of forms of commumication brings certain advantages from the point of view of formalisation and experimental testing, but there is much to be gained from facing these challenges head on. Acknowledgements I would like to thank Ray Gibbs for his originality as a thinker, his brilliance as a communicator and his generosity in reaching out across the boundaries of theories and disciplines. I would also like to thank Robyn Carston for extremely helpful comments and suggestions on an earlier version of this chapter. References Beaver, D., Roberts, C., Simons, M., & Tonhauser, J. (2017). Questions under discussion: Where information structure meets projective content. Annual Review of Linguistics, 265–284. Brody, R. (2017). The frightening lessons of Philip Roth’s The plot against America. New Yorker, February 1, 2017. Carston, R. (2018). Figurative language, mental imagery and pragmatics. Metaphor and Symbol, 33: 198–217. Cave, T., & Wilson, D. (Eds.). (2018). Reading beyond the code: Literature and relevance theory. Cambridge: Cambridge University Press. Csibra, G. (2010). Recognizing communicative intentions in infancy. Mind & Language, 25, 141–168. Csibra, G., & Gergely, G. (2009). Natural pedagogy. Trends in Cognitive Sciences, 13, 148– 153. Degen, J., & Tanenhaus, M. (2019). Constraint-based pragmatic processing. In C. Cummins & N. Katsos (Eds.), Handbook of experimental semantics and pragmatics (pp. 21–38). Oxford: Oxford University Press. Gibbs, R. (1994). The poetics of mind: Figurative thought, language, and understanding. Cambridge: Cambridge University Press. 11` Gibbs, R. (1999). Intentions in the experience of meaning. Cambridge: Cambridge University Press. Gibbs, R. (2011). The allegorical impulse. Metaphor & Symbol, 26, 121–130. Gibbs, R. (2017). Experimental pragmatics. In Y. Huang (Ed.), The Oxford handbook of pragmatics (pp. 310–325). Oxford: Oxford University Press. Gibbs, R., & Colston, H. (2012). Interpreting figurative meaning. New York: Cambridge University Press. Macfarlane, R. (2013). The old ways: A journey on foot. London: Penguin Books. Roth, P. (2004a). The plot against America. New York: Houghton-Mifflin. Roth, P. (2004b). The story behind The plot against America. New York Times, September 19, 2004. Rich, F. (2004). President Lindbergh in 2004. New York Times, September 23, 2004. Sperber, D., & Wilson, D. (1982). Mutual knowledge and relevance in theories of comprehension. In Neil Smith (Ed.), Mutual knowledge (pp. 61–85). London: Academic Press. Sperber, D., & Wilson, D. (1986). Relevance: Communication and cognition. (2nd edition, 1995.) Oxford: Blackwell. Sperber, D., & Wilson, D. (2002). Pragmatics, modularity and mindreading. Mind & Language, 17, 3–23. Reprinted in Wilson, D., & Sperber, D. (2012). Meaning and relevance (pp. 261–278). Cambridge: Cambridge University Press. Sperber, D., & Wilson, D. (2015). Beyond speaker’s meaning. Croatian Journal of Philosophy, 15, 117–149. Unger, C. (2017). Towards a relevance theory account of allegory. In A. Piskorska & E. Walaszewska (Eds.), From discourse to morphemes: Applications of relevance theory (pp.152–174). Newcastle: Cambridge Scholars Publishing. Unger, C. (2018). Cognitive pragmatics and multi-layered communication: Allegory in Christian religious discourse. In P. Chilton & M. Kopytowska (Eds.), Religion, language and the human mind (p. 333–352). Oxford: Oxford University Press. Unger, C. (2019). Allegory in relation to metaphor and irony. In K. Scott, B. Clark & R. Carston (Eds.), Relevance, pragmatics and interpretation (pp. 240–252). Cambridge: Cambridge University Press. Wilson, D. (2017). Relevance theory. In Y. Huang (Ed.), The Oxford handbook of pragmatics (pp. 79–100). Oxford: Oxford University Press. Wilson, D. (2019). Relevance theory. In Oxford research encyclopedia of linguistics. Oxford: Oxford University Press. doi:http://dx.doi.org/10.1093/acrefore/9780199384655.013.201 Wilson, D., & Carston, R. (2019). Pragmatics and the challenge of ‘non-propositional’ effects. Journal of Pragmatics, 145, 31–38. Wilson, D., & Sperber, D. (2004). Relevance theory. In L. Horn & G. Ward (Eds.), The handbook of pragmatics (pp. 607–632). Oxford: Blackwell.