Word Meaning and Concept Expressed: Philosophical and Linguistic Perspectives

Workshop Themes: 

The concept expressed by a word on a specific occasion often diverges from the concept encoded by the word. In recent years, this divergence has been the central focus for theorists working on lexical modulation, ad hoc concept construction and polysemy, especially in the contextualist tradition. This conference brings together philosophers and linguists to explore how occasion-specific meanings arise, how they relate to lexical semantics and pragmatics, and what their existence implies for broader issues in philosophy of language and mind (e.g. the nature of concepts, verbal disputes). Topics include the structure of lexical meaning, the theoretical distinction (or lack thereof) between polysemy and ad hoc concepts, and the implications of meaning flexibility for philosophical debates.

Schedule:

Talk 1:  9:30-10:50

Francois Recanati: Taking Polysemy Seriously: Two Approaches

Talk 2: 10:50-12:10

Tamara Dobler: Polysemy and the Semantics-Pragmatics Distinction

Lunch: 1 hour

Talk 3: 1:10-2:30

Robyn Carston: Word Meaning, Pragmatic Modulation and the Wrong/Right Format Issue

Talk 4: 2:30-3:50

Michelle Liu: Ad Hoc Concepts: Definition and Applications  

Tea break: 20 mins

Talk 5: 4:10-5:30

John Collins and Agustin Vicente: On What Constrains the Concept Expressed

----------------------------------------------------------------------------------------------------------

Abstracts

 Francois Recanati: Taking Polysemy Seriously: Two Approaches

Taking polysemy seriously is distinguishing it from ambiguity (in the sense of homonymy). An ambiguous word carries distinct lexical meanings, while a polysemous word carries a single (although possibly internally diverse) meaning. In that framework, how can we explain that a polysemous word can be used to talk about different things? Two theories (and their variants) will be discussed. One rejects the principle that a difference at the level of reference entails a difference at the level of concepts; it argues that the meaning of a polysemous expression is a unitary polysemous concept whose reference can shift from occurrence to occurrence. The other theory rejects the principle according to which the meaning of a linguistic expression is a concept, whether polysemous or not; it maintains that a polysemous expression may, in virtue of its unitary lexical meaning, express different concepts in different contexts.

 Tamara Dobler: Polysemy and the Semantics-Pragmatics Distinction

Polysemy has long been at the centre of debates about the interface between lexical semantics and pragmatics. A persistent question is whether the multiple aspects of meaning associated with words like book, school, or city are encoded in their lexical semantics, or whether they arise pragmatically in context (Recanati 2017; Vicente 2018; Carston 2020). Dual-nature approaches (Asher 2011; Gotham 2022) attempt to capture logical polysemy by positing type distinctions at the semantic level, while critics (Liebesman & Magidor 2023) argue that such approaches systematically mispredict certain data. This paper develops a new framework for understanding polysemy, grounded in a radical constructivist approach to the FLN-lexicon and a reconceptualization of dot-types. I argue that this model captures problematic data that challenge dual-nature accounts, such as the quantificational puzzle raised by Liebesman & Magidor (2023). If correct, this view reshapes the semantics–pragmatics distinction: semantics is not a matter of assigning determinate types to lexical items but of structuring conceptual files in ways constrained by core cognition. Pragmatics operates downstream, shaping how these structures are deployed in communication.

 Robyn Carston: Word Meaning, Pragmatic Modulation and the Wrong/Right Format Issue

Assuming a distinction between the (relatively) stable ‘meaning’ of a word and the various senses/concepts the word can be used to express on different occasions of use, there are multiple positions on what sort of thing that ‘meaning’ might be. On the thin side, it might be: (a) an abstract schematic core; (b) an instruction to fetch/construct a concept; (c) a pointer to a body of information. On the rich side, it might be: (a) a family/ network of interrelated concepts/senses; (b) a body of information (logical and/or encyclopaedic); (c) an array of memory traces of previously encountered senses/uses. Or it might be some thin/rich combination. Following Recanati (2004, 2017), these are all ‘wrong format’: they are semantically underspecified and cannot ‘go directly into the interpretation’ of an utterance; they all require some pragmatic process of ‘modulation’ (enrichment or selection). I will try to assess the adequacy of these views of word meaning while also arguing for a constraint on the nature of the pragmatic modulation process: that its input (as well as its output) must be a sense/concept (i.e. something that has the ‘right format’).

 Michelle Liu: Ad Hoc Concepts: Definition and Applications

A speaker often uses a word to communicate what linguists call an "ad hoc concept", an occasion-specific meaning that is different from the word’s stable, encoded meaning, and the hearer can usually construct the intended ad hoc concept through pragmatic inference (see e.g. Carston 2002; Wilson and Carston 2007). Ad hoc concepts feature prominently in lexical pragmatics. However, a precise definition is lacking in the literature. In this talk, I will put forward a definition of ad hoc concepts which distinguishes it from related phenomena such as polysemy. I will show how the notion of ad hoc concepts may be particularly useful for philosophers. I will illustrate this with respect to philosophical theorising about verbal disputes.

 John Collins and Agustin Vicente: On What Constrains the Concept Expressed

The work done by, inter alia, Carston (2012, 2019) and Recanati (2004, 2017), has shown that the semantic profile of words is radically misaligned with what is expected of concepts: words are semantically highly plastic, while concepts must be rigid to the extent that they resolve dimensions of word plasticity. How do word meanings constrain the concepts expressed by word-tokens? We examine three options with respect to the source of such constraints: (i) that they are imposed by a conceptual common core; (ii) that they relate to lexical non-conceptual features of words; (iii) that constraints emerge from prototypical meanings. Next, we evaluate the possibility that there are actually no constraints as such, but rather that what a word can express depends on what concepts and conceptual structures that word typically activates. After explaining why all these options fail, we consider the possibility that constraints on concepts expressed are architectural.