Page 1 of 16 Towards a Flexible Functional Proposition for Dynamic Discourse Meaning Chi-Hé Elder and Kasia M. Jaszczolt University of East Anglia, U.K. University of Cambridge, U.K. Forthcoming in: Tanaka, Hiroaki, Kaori Hata, Etsuko Yoshida and Masataka Yamaguchi (eds). 2024. Towards a Dynamic Pragmatics, volume 4. Tokyo: Kaitakusha (2023 pre-print) 1. Introduction: Desiderata for propositions The focus and scope of theory of meaning are still widely discussed topics. As is well known, Grice’s concept of ‘non-natural’ meaning combined both ‘what is said’ and ‘what is implicated’. Since then, Grice’s non-natural meaning has long been considered to delimit what a theory of meaning ought to represent, or, in other words, to delineate “an outer boundary on the communicational effects that a theory of communication is responsible for” (Levinson 2000: 12-13, as quoted and discussed in Wilson and Carston 2019). But while Levinson suggested that Grice’s non-natural meaning did an adequate job of carving out, and carving up, what is communicated, the concept of what is said – that pertains to truth-conditional, semantic, propositional content – has been subject to a lot of scrutiny in terms of the units that most appropriately reflect how people communicate in natural language, and various post-Gricean accounts have redrawn the said/implicated boundary, including more and more content in what is said/explicit. In addition, Wilson and Carston (e.g. 2019) have challenged the current perception of the entire scope of pragmatic theory by focusing on creative metaphors and the nuanced contributions to meaning they exert: contributions that, albeit standardly deemed non- propositional, in fact display an array of forms from what can be captured in terms of propositional content after all to more elusive imagery. The purpose of this paper is to further engage in this metapragmatic inquiry into the scope of a theory of communication and how it ought to be divided. In what follows, we will be focusing on the question of what counts as propositional content, asking about (i) its scope in relation to the standard said/implicated divide, as well as (ii) its relation to the speaker’s intended and addressee’s recovered meaning, as well as the meaning that can emerge somewhere in between, so to speak, as jointly constructed. We propose a concept of a flexible functional proposition that allows us to more accurately represent the content of the main message conveyed by an utterance, reflecting what really matters to both the speaker and the addressee, allowing for the input of each side to meaning construction, and respecting the multimodal nature of communication—something that past post- Gricean accounts that were constrained by intention recovery on the one hand, and the fixation on the logical form of the sentence (even if moulded to fit with communicative intentions) on the other, failed to do. 1 In post-Gricean contextualism, in the tradition of Atlas-Kempson (Atlas 1977, 1979, 1989; Kempson 1975, 1979, 1986), Relevance Theory (Sperber and Wilson 1986), Recanati (2010), and others, we have seen various configurations of the semantics/pragmatics boundary, where the focus of attention has been on the delimitation of truth-conditional content, and as such, propositional content. The key idea is that context is needed to provide information such that the logical form of the uttered sentence may be further elaborated on – ‘saturated’, ‘modulated’, ‘enriched’, or ‘developed’ (to use some of the popular, albeit not synonymous, labels) – in order to produce a unit that reflects how natural language users themselves understand utterance meanings. We begin by taking a radical stance on the role of the said/implicated boundary in delimiting propositional content. Jaszczolt’s (2005, 2010) theory of Default Semantics pushes the concept of 1 For a discussion of what counts as ‘being post-Gricean’ see Jaszczolt 2019.