The aim of this book is to reassess the semantics–pragmatics interface by combining insights from Construction Grammar (Reference GoldbergGoldberg, 1995, Reference Goldberg2006; Reference Hoffmann and TrousdaleHoffmann and Trousdale, 2013a; Reference HilpertHilpert, 2019; Reference HoffmannHoffmann, 2022) and Relevance Theory (Reference Sperber and WilsonSperber and Wilson, 1995; Reference Wilson and SperberWilson and Sperber, 2012; Reference ClarkClark, 2013a).

The past seventy years have witnessed increasing attempts at describing linguistic knowledge and language use, from which various approaches gradually emerged. This growing interest can be traced back to Reference ChomskyChomsky’s (1965: 59) observation that appropriate descriptions of language use also necessarily require a good understanding of the underlying mechanisms, i.e. the cognitive abilities, that make communication possible. The extent to which performance and competence actually differ has caused a great deal of debate in the literature. Nevertheless, it is primarily this distinction that triggered a “cognitive turn in linguistics” (Reference Schmid, Middeke, Müller, Wald and ZapfSchmid, 2012: 380). Of course, with this new approach to language came a host of new research questions, the different answers to which resulted in the emergence of various frameworks. Construction Grammar and Relevance Theory developed out of this quest to provide cognitively plausible accounts of language use and, in their respective domains of application, gained enough importance to become respected landmarks in the linguistic scene. These two theories are the starting point of this book.

In spite of their common interest in cognition, the two frameworks generally focus on different aspects of language use. In Construction Grammar, the main goal is to provide an accurate description of what constitutes linguistic knowledge and to explain how this knowledge is actually exploited in practice. In contrast, Relevance Theory grew out of a general concern to understand the cognitive underpinnings that enable us to make sense of our world and in particular, when applied to linguistic communication, to how they contribute to understanding the speaker’s communicative and informative intentions. Although these two frameworks provide equally insightful understandings of verbal communication, there have been very few attempts to combine their perspectives. This is most probably due to the often-held assumption that grammar and pragmatics belong to different realms of cognition and deserve separate attention since knowledge about one can hardly provide a better understanding of the other. The aim of the book is precisely to show, however, that this combination is not only useful, but is indeed necessary in order to provide a richer description of the underlying mechanisms of both grammar and pragmatics and of their respective contributions to the interpretation of an utterance.

Because they were developed independently on the basis of different underlying assumptions with an eye to answering different research questions, the two frameworks sometimes provide opposite analyses of the same phenomena. For instance, consider the discussion between Bilbo and Gandalf:

Putting aside Gandalf’s wit for a moment, one might wonder why he needs to ask Bilbo what exactly he intends to communicate when using the phrase good morning. As a linguist, the answer to this question will vary depending on the theoretical background in which it is couched. Exaggerating somewhat, a constructionist might answer that Gandalf’s answer is indeed a bit odd given that good morning is a conventional construction of English which is a formula commonly used as a greeting when you meet a person for the first time early in the day. Given this convention, Gandalf should have known that Bilbo only meant to say hello and therefore answered good morning in return. In opposition, a relevance theorist might argue that Gandalf’s answer is quite appropriate since, in spite of the linguistic conventions, the meaning of a lexical item remains usually underspecific and needs to be systematically enriched in context via pragmatic inferential processes. As a result, if the intended interpretation was not clear, then Gandalf is indeed entitled to ask what it was that Bilbo actually meant. This of course is a very simplistic demonstration, and theorists in the different frameworks probably have more moderate views than the ones they are associated with here. Nevertheless, this example is meant to capture a general observation that will become clear throughout the following chapters, namely that Construction Grammar and Relevance Theory respectively tend to over-emphasize the role played by linguistic conventions and pragmatic inferencing, and typically so at the cost of the other. It could of course be argued that this tendency is an inevitable side effect of the respective aims of the two theories. Unfortunately, this therefore means that for a broad range of linguistic phenomena, it is unclear which of the two frameworks actually achieves descriptive accuracy (a goal they both set out to achieve) since their respective predictions sometimes come into conflict. For the sake of cognitive accuracy, it is therefore necessary to compare the two frameworks in a systematic manner so as to pin down more specifically the respective contributions of grammar and pragmatics during the interpretation of an utterance. It is my aim to do so.

In order to appreciate the respective contributions of Construction Grammar and Relevance Theory to the understanding of verbal communication, it is essential to provide detailed overviews of the two frameworks first. In Chapter 2, each theory will be introduced in turn. Their strengths will be highlighted and the weaker points needing particular attention (especially those that concern the semantics–pragmatics interface) will be identified. On the basis of this review, focus on the main points of contention will lead me to articulate the discussion around two facets of lexical semantics–pragmatics. In Chapter 3, the aim will be to define exactly how the notions of semantics and pragmatics apply to a lexical item. It will be shown that although the two frameworks describe the meaning of a lexeme in conceptual terms, their opposite views on the nature of concepts affects the way these concepts are argued to contribute to the understanding of the lexemes with which they are associated. I will assess the exact nature of conceptual content and the way this content is exploited in context on the basis of various arguments. I will generally argue that understanding a lexeme depends on rich semantic knowledge together with strong pragmatic principles, and the notion of lexically regulated saturation will be used to capture the interpretation process of a lexical item. In the following chapter, Chapter 4, the aim is to discuss the ways in which the direct linguistic environment of a lexeme contributes to this particular interpretation process. First, I will critically assess the nature of a mechanism known as coercion and argue that, although clearly semantically constrained, coercion is itself also essentially pragmatic. I will then show that the pragmatic roots of coercion are linked to the procedural nature of the semantic content encoded by the grammatical constructions in which lexemes occur. In doing so, both the formal nature of these constructions and the notion of procedural encoding will be (re)defined. It will be shown that constructions that have a coercive force are necessarily (semi-)schematic constructions and that procedural meaning might best be described in meta-conceptual terms. Second, it will be shown that the interpretation of a lexeme is also largely determined by more lexically fixed (i.e. idiomatic) sequences. Upon recognition of these patterns, the process of lexically regulated saturation may thus be suspended. I will argue that interpreting these larger patterns is a context-sensitive process and that the principle of relevance introduced in Relevance Theory can explain the underlying mechanism.

The concluding section of this book will show that integrating the perspectives of Construction Grammar and Relevance Theory proves to be particularly beneficial in the search for descriptive accuracy. In addition to increasing the respective explanatory power of the two frameworks, conjoining these two approaches provides additional insights into the underlying cognitive mechanisms which make verbal communication successful.