Imprecision, personae, and pragmatic reasoning

Beltrama, Andrea; Schwarz, Florian

doi:10.3765/salt.v31i0.5107

Cited by 5 publications

(11 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, and more centrally for our purposes, it's been shown that the social information available in the context affects the computation of the level of precision with which a numeral is interpreted. In particular, Beltrama & Schwarz (2021) provide evidence that comprehenders reason about the persona embodied by the speaker in inferring the range of facts compatible with a numeral description. Evidence supporting this claim is based on participants' indirect judgments about whether a certain fact is a likely candidate for being the one the numeral utterance was aiming to describe.…”

Section: Introductionmentioning

confidence: 92%

“…In a similar vein, a growing amount of work has been shedding light on the impact of social information on the processes involved when interlocutors resolve, and ultimately exchange, meaning in conversation. On an empirical level, it's been shown that pragmatic reasoning is impacted by a range of different social factors -e.g., politeness considerations (Bonnefon, Feeney & Villejoubert 2009;Yoon, Tessler, Goodman & Frank 2020;Mazzarella, Trouche, Mercier & Noveck 2018); affect (Kao, Wu, Bergen & Goodman 2014;Bergen 2016); speaker-specific properties such as linguistic nativeness (Fairchild & Papafragou 2018) and political orientation (Mahler 2020), and broader personae and identity categories (Beltrama & Schwarz 2021; see §2 for further details). Conversely, it has also been shown that comprehenders promptly infer identity and personality features of a speaker from the semantic and pragmatic properties of their utterances (Acton & Potts 2014;Beltrama & Staum Casasanto 2017Acton 2019;Glass 2015;Karawani & Waldon 2017;Jeong 2021;Thomas 2021;Hunt & Acton 2022; see Beltrama 2020 for an overview).…”

Section: Introductionmentioning

confidence: 99%

“…See in particular Huang, Spelke & Snedeker (2013). See Beltrama & Schwarz (2021) for further details on the norming study. In the experimental task, participants would see one of these two sets of characters engaging in a conversation, which always concluded with one character uttering a numeral description (e.g., "It's $200").…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Social identity, precision and charity: when less precise speakers are held to stricter standard

Beltrama¹,

Schwarz²

2022

SALT

View full text Add to dashboard Cite

Recent has begun to show systematic connections between social information and pragmatic reasoning. These findings raise the question of whether social information shapes comprehenders' assessments of the correctness of linguistic description in light of a single known and determined fact. We explore this question by testing the impact of speaker identity on T(ruth)-V(alue) J(udgment)s based on the interpretation of number words. We find that imprecise statements from speakers socially expected to be less precise – i.e. “Chill" ones – are rejected at a higher rate, and thus held to more stringent evaluation standards, than those from speakers socially expected to speak more precisely – i.e. “Nerdy" ones. We explain the new finding by appealing to the idea that, by virtue of generally being perceived to be more precise, Nerdy speakers are granted higher epistemic credibility than Chill ones. The emerging picture is one in which TVJ assessments are affected by social considerations in a different way from other experimental tasks, suggesting a nuanced interplay between social information and different interpretation tasks and processes

show abstract

Section: Introductionmentioning

confidence: 92%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Social identity, precision and charity: when less precise speakers are held to stricter standard

Beltrama¹,

Schwarz²

2022

SALT

View full text Add to dashboard Cite

show abstract

“…Specifically, our incorporation of persona in the prompt design is inspired by Beltrama and Schwarz (2021). They find that to compute the standard of precision required to interpret numeral expressions, human comprehenders reason about the speaker's social identity, particularly about the persona they embody.…”

Section: Introductionmentioning

confidence: 99%

“…Against this background, our goal is to examine the extent to which pre-trained LMs can "understand" implicit EVAL implicatures. We hypothesized that if pre-trained LMs are cognitively plausible, their performance should align with the human data in Brasoveanu and Rett (2018) and Beltrama and Schwarz (2021), namely: i) there should be no EVAL difference regarding adjective polarity, ii) LMs should predict POS constructions to be the most evaluative, iii) whether LMs consider the relative adjectives to be more or less evaluative than the absolute adjectives depends on construction type, iv) LMs should (at least) show a trend that the chill-persona prompt helps LMs' understanding of implicatures, relative to the nerdy-persona prompt.…”

Section: Introductionmentioning

confidence: 99%

Pre-trained Language Models’ Interpretation of Evaluativity Implicature: Evidence from Gradable Adjectives Usage in Context

Ye¹

2022

Proceedings of the Second Workshop on Understanding Implicit and Underspecified Language

View full text Add to dashboard Cite

By saying Maria is tall, a human speaker typically implies that Maria is evaluatively tall from the speaker's perspective. However, by using a different construction Maria is taller than Sophie, we cannot infer from Maria and Sophie's relative heights that Maria is evaluatively tall because it is possible for Maria to be taller than Sophie in a context in which they both count as short. Can pre-trained language models (LMs) "understand" evaulativity (EVAL) inference? To what extent can they discern the EVAL salience of different constructions in a conversation? Will it help LMs' implicitness performance if we give LMs a persona such as chill, social, and pragmatically skilled? Our study provides an approach to probing LMs' interpretation of EVAL inference by incorporating insights from experimental pragmatics and sociolinguistics. We find that with the appropriate prompt, LMs can succeed in some pragmatic level language understanding tasks. Our study suggests that socio-pragmatics methodology can shed light on the challenging questions in NLP.

show abstract

Certainly but not certain: The expression of subjective and objective probability

Yatsushiro

Trinh

Żygis

et al. 2022

Glossa: A Journal of General Linguistics

View full text Add to dashboard Cite

This paper investigates the interpretation of the epistemic modal adjectives possible, certain and their adverbial counterparts possibly, certainly, taking the perspective that the former express objective modality, whereas the latter express subjective modality. In support of this view, we report on two experiments that assess speakers’ acceptance of possible, certain, possibly, certainly in different contexts. Our results extend those of Lassiter (2016) in demonstrating a difference in interpretation between the adjectival and adverbial modals, with possibly less acceptable than possible but conversely certainly more acceptable than certain in the same situations. Furthermore, the results of our experiments suggest a new insight, namely that these differences depend, to some extent, on the probability of the eventuality in question and polarity of the prejacent sentence.

show abstract

Imprecision, personae, and pragmatic reasoning

Cited by 5 publications

References 43 publications

Social identity, precision and charity: when less precise speakers are held to stricter standard

Social identity, precision and charity: when less precise speakers are held to stricter standard

Pre-trained Language Models’ Interpretation of Evaluativity Implicature: Evidence from Gradable Adjectives Usage in Context

Certainly but not certain: The expression of subjective and objective probability

Contact Info

Product

Resources

About