Lectures en linguistique expérimentale – 2020

Séminaire doctoral animé par Anne Abeillé et Barbara Hemforth. Habituellement le vendredi, de 14h à 16h, en salle 533.

en web seminar depuis le 27 mars

Programme

  • 17 janvier
    Yanwei Jin and Jean-Pierre Koenig, University at Buffalo, Expletive negation: From typology to on-line comprehension
    Abstract: Expletive negation (EN) has been discussed extensively within Romance linguistics. In this talk, we explore the hypothesis that EN is due to general properties of language production: the meaning of some lexical items leads to the concurrent activation of a proposition (p) and its dual (not p) and it is this concurrent activation that causes the erroneous production of not p rather than the intended p. We begin by presenting of a survey of the occurrence of EN across languages of the world (722 languages were surveyed overall); we then present a comprehensive list of EN-triggering contexts collected from French and Mandarin and compare that list with EN triggers in three other languages (4 language families in all). We then propose a language production and semantic account of the similarity of EN-triggering contexts found in these five languages. We propose that the meaning of EN triggers entails or strongly implies ¬p and that the activation of ¬p alongside p is what leads speakers to produce EN. Four semantic licensing conditions for EN triggers are identified and each EN-triggering context is semantically analyzed. We then show through a corpus study that speakers of languages said not to include EN (English) produce them with some frequency in the same contexts in which they appear in languages where EN is entrenched (French, Mandarin): EN interpretations of negators ranged from almost 0% to 100% in our corpus, depending on the trigger (mean = 28.64%). Finally, we report an on-line comprehension experiment that tests the hypothesis that native speakers of English interpret negators in complement clauses of EN triggers expletively. To test this hypothesis, we used the logic of the experiments on metaphorical class-inclusion statements reported in Glucksberg, Gildea, & Bookin (1982). We found that, as predicted, EN-trigger continuations elicited less logically accurate answers (p < .01) and longer response times (p < .01). This result suggests that English speakers do interpret negators expletively in the same contexts found to trigger grammaticalized EN in French or Mandarin. We also found that logical inaccuracy for EN continuations correlated with the frequency of EN interpretations in our corpus (r = .66, p < .01), suggesting that the more expletive interpretations of negators for a particular EN-trigger a speaker has encountered, the more likely she is to interpret expletively a new occurrence of that negator for that trigger. Overall, our research suggests that French EN is not just some odd historical artefact, but is rooted in general cognitive mechanisms. What is special about Romance (French ne, in particular) is the degree to which EN has become entrenched (and grammaticized).
  • 31 janvier (resp. Barbara Hemforth)
    Núria Esteve-Gibert, Amy J. Schafer, Barbara Hemforth, Cristel Portes, Céline Pozniak & Mariapaola D’Imperio - Empathy influences how listeners interpret intonation and meaning when words are ambiguous. Memory & Cognition, 2020. http://www.llf.cnrs.fr/fr/node/6475
  • 7 février:
    Abeillé, Hemforth, Winckel, Gibson:
    Subject island: extraction depends on discourse function, Cognition (2020)
  • 28 février (resp. Céline Pozniak)
    Walter Schaeken, Marie Van Haeren and Valentina Bambini. – The Understanding of Scalar Implicatures in Children With Autism Spectrum Disorder : Dichotomized Responses to Violations of Informativeness. – Front Psychol. 2018.
    https://www.ncbi.nlm.nih.gov/pubmed/30083124
  • 6 mars
    Suzanne Lesage
    , Barbara Hemforth, Olivier Bonami (LLF) : (Pseudo-)reflexive possessive binding : crosslinguistic experiments
  • 27 mars:
    Ira Noveck (LLF) and
    Camilo Rodriguez Ronderos (Berlin): How irony is revealing of intentions in utterance understandingreferences; ref: Nicola Spotorno, Ira Noveck  What’s behind an ironic utterance and its processing; 2018 Irony In N. Katsos & C. Cummins (Eds). Oxford Handbook of Experimental Semantics and Pragmatics. Oxford: OUP
  • 10 avril
    Gabriel Thiberge (LLF): Les interrogatives partielles en français : effets de contexte et stéréotypes sociolinguistiques
  • 15 mai
    Chiara Mazzocconi (LLF): Laughter development in mother-child interaction: from 12 to 36 months  References: Chiara Mazzocconi, Ye Tian, and Jonathan Ginzburg 2020 What’s your laughter doing there? A taxonomy of the pragmatic functions of laughter. IEEE TRANSACTIONS ON AFFECTIVE COMPUTING 1; Mazzocconi & Ginzburg 2020:Laughter growing up, Laughter workshop 2020
  • 22 mai Yingtong Liu (Harvard U): A Verb-frame Frequency Account of Constraints on Long-Distance Dependencies in English 
  • Abstract: Going back to Ross (1967) and Chomsky (1973), researchers have sought to understand what conditions permit long-distance dependencies in language, such as between the wh-word what and the object of the verb bought in the sentence ‘What did John think that Mary bought?’. In the present work, we attempt to understand why changing the main verb in wh-questions affects the acceptability of long-distance dependencies out of embedded clauses. In particular, it has been claimed that factive and manner-of-speaking verbs block such dependencies (e.g., ‘What did John know/whisper that Mary bought?’), whereas verbs like think and believe allow them. Here we provide 3 acceptability judgment experiments of filler-gap constructions across embedded clauses to evaluate four types of accounts based on (1) discourse; (2) syntax; (3) semantics; and (4) our proposal of verb-frame frequency. When analyzed using ordinal or logistic regressions, we see no evidence of interactions between verb-frame frequency and construction type (wh-question or cleft vs. declarative), and hence we see no evidence of an independent factor that predicts acceptability degradation.  The results are most simply explained by two factors: verb-frame frequency, such that verbs that rarely take embedded clauses are less acceptable; and construction type, such that wh-questions and clefts are less acceptable than declaratives. We conclude that the low acceptability of filler-gap constructions formed by certain SC verbs is due to low linguistic exposure. In addition, we observe that the wide application of linear models to Likert scale rating data in many previous island studies can sometimes lead to false positives - spurious syntactic “island” effects
  • 29 mai
    Laurel Perkins (LSCP, ENS & UCLA): How to grow a grammar: Syntactic development in 1-year-olds
    Perkins and Lidz 2019 : Filler-gap dependency comprehension at15 months: The role of vocabulary, Language Acquisition,Language Acquisition, http://ling.umd.edu/assets/publications/Perkins-Lidz-19-FillerGap-15mont...
  • 5 juin
    Verena Haser & Lars Konieczny (Universität Freiburg, Allemagne): Processing different types of irony
  • 12 juin

       Yanis da Cunha (LLF) L'alternance actif/ passif en français

  • 19 juin
    Efrat Levant (Tel Aviv University): Default sarcastic interpretations of attenuated and intensified similes

    What causes an utterance to be interpreted as sarcastic? Does it have to do with familiarity, context, or structure? Are some constructions always sarcastic, regardless of other factors? Many theories have attempted to find out what makes an utterance sarcastic. The Defaultness Hypothesis suggests that attenuating highly positive concepts (of novel, ambiguous similes, involving no internal incongruity) will generate sarcastic interpretations by default (i.e. automatically, regardless of context or salience). This theory was previously proven on two constructions – negations (X is not the mostY)and rhetorical questions (Do you reallybelieve he is smart?).   

    In this work, we test the effect of attenuators (about/almost/at least) and intensifiers (absolutely /really/so) on the sarcasm levels of similes within the framework of the Defaultness Hypothesis. Five offline experiments were performed. Our results support the prediction of the Defaultness Hypothesis, showing that indeed strongly attenuating highly positive concepts generate sarcastic interpretations by default.

    Additionally, our experimental results diverge from previous corpus-based findings. This prior work found that attenuated similes involving internal incongruity are significantly more sarcastic than their unattenuated counterparts. In contrast to these findings, we show that no significant difference in sarcasm rating was detected between attenuated/unattenuated similes when containing internal incongruity. However, when comparing attenuated/unattenuated similes without internal incongruity, a significant difference in sarcasm rating was found.

    We conclude that, as predicted by the Defaultness Hypothesis, it is attenuating highly positive concepts that affect sarcasm by default, i.e., even without relying on supportive context or internal incongruity.