POLYSEMY IN THE EVALUATIVE SPHERE
Seminar 4 | Invited talk
When this chef says pot: The importance of the speaker's identity in understanding ambiguous words
Marina Ortega-Andrés (University of the Basque Country)
21 November 2025 | 11h00-12h30 (WET) | Online
ONLINE | LINK ZOOM
In this talk I will explore how listeners interpret ambiguous words based on their previous experience with specific speakers. A widely accepted assumption in psychology and linguistics is that, over the course of life, speakers accumulate vast statistical information about language use—including not only the contexts in which certain words are typically used but also the relative frequency of their meanings. This information should guide (alongside other factors) our interpretations: in absence of more contextual information, we tend to assign words their most frequent or dominant meaning. However, recent studies have shown that these preferences are not fixed. A single exposure to a word in a disambiguated context can alter its later interpretation, at least temporarily. For example, after hearing bark referring to tree bark, listeners tend to associate the word with that meaning—even though it is less frequent—rather than its more dominant meaning ("dog’s bark"). This phenomenon, known as word meaning priming, can last for hours or even days, suggesting that previous experience can alter the listener’s probabilistic estimates about what a word probably means. In a series of experimental studies, we investigated whether this effect also depends on the identity of the speaker. Participants heard one speaker (e.g., a chef) consistently using ambiguous words (like pot) to refer to things related to cooking (cooking pot) in a thematically related context (cooking dinner for a birthday party). After that, participants heard a speaker (who could be the same or a different speaker) asking "which picture goes best with the word pot?" Participants had to pick the image that answered the question. We found that when the speaker was the same in both tasks, participants picked the image that was related to the primed sense (i.e. the cooking pot) more often than when there was no previous story. However, when the speaker was different in each task, the probability of selecting the primed meaning significantly decreased, even when both speakers were chefs. This result suggests that listeners sometimes retain previous uses of a word by one speaker for future encounters with the same speaker. However, if a different speaker—even another chef—uses the same word, listeners tend to access the more frequent meaning of the word (“plant pot”) again. We interpret this result as meaning that experience with specific individuals talking about a given topic shapes semantic expectations in future interactions with that same person. These expectations do not transfer to other speakers and even when the two speakers belong to the same group (such as being a chef) is not enough to generalize the interpretation from one speaker to another. Thus, access to meaning may be partly shaped by local, personal, and dynamic experiences with individual speakers. This invites us to consider that the mental lexicon is flexible and that semantic access may partially be speaker specific.
https://ifilosofia.up.pt/activities/seminar-4-polysemy-evaluative-sphere
Seminar| Polysemy in the Evaluative Sphere: https://ifilosofia.up.pt/activities/seminar-polysemy-evaluative-sphere
Organiser: Dan-Cristian Zeman (2023.05952.CEECIND)
CEECInd research project “Slurs and the Lexicon: A Rich-Lexicon Approach to Slurs and Other Evaluative Expressions (LEXISLUR)”
RG Mind, Language and Action Group (MLAG)
Instituto de Filosofia da Universidade do Porto - UID/00502/2025
Fundação para a Ciência e a Tecnologia (FCT)
____________________________________________
Instituto de Filosofia (UI&D 502)
Faculdade de Letras da Universidade do Porto
Via Panorâmica s/n
4150-564 Porto
Tel. 22 607 71 80
E-mail: ifilo...@letras.up.pt
http://ifilosofia.up.pt/
Financiamento, avaliação e auditoria do Instituto de Filosofia:

POLYSEMY IN THE EVALUATIVE SPHERE
Seminar 10 | Invited talk
What 'We' Can Mean
Katarzyna Kijania-Placek & Maciej Tarnowski (Jagiellonian University)
27 March 2026 | 11h00-12h30 (WET) | Online
ONLINE | LINK ZOOM
In this talk, we present a semantic account of the first-person plural pronoun -„we” in English - that comprises all systematic kinds of use of this expression. We argue that "we" exhibits five systematic types of meaning - directly referential, descriptive, deferred, anaphoric, and bound - each associated with a distinct Kaplan-style character. We show that these different meanings of „we" satisfy standard diagnostics for systematic polysemy, including non-zeugmatic co-predication across different senses, cross-linguistic robustness, and productivity across other plural pronouns and singular terms. Building on this, we introduce a two-dimensional model of polysemy in which lexical meaning consists of a set of rule-based characters capable of generating context-sensitive contents. This framework, which naturally extends to other singular terms, including demonstratives and proper names, preserves the Kaplanian treatment of indexicality while explaining descriptive and deferred uses of indexicals.
https://ifilosofia.up.pt/activities/seminar-10-polysemy-evaluative-sphere