Proceedings of Sinn und Bedeutung
Not a member yet
1139 research outputs found
Sort by
Limited variable-force modals and the interactions of scal{ar,eless} implicatures
Get PDFIn this paper I present novel data from Kinande (Bantu J, DRC) attesting a previously undiscussed type of variable-force modal: limited variable-force modals. The Kinande modal prefix anga can be interpreted as a possibility modal or as a weak necessity modal, but never as a strong necessity modal. I show the prior analyses, whether they focus on domain restriction or exhaustification, are alone insufficient to account for this kind of modal force pat- tern, and develop a unified account of modal force that explains the Kinande facts as well as the previously-attested modal systems, generating a typology of modal force
Anaphoric reference in descriptions of surprising events
Get PDFThe choice of anaphoric reference is a complex process regulated by a combination of linguistic and cognitive constraints. This paper experimentally addresses the impact of world knowledge on the types of references speakers produce, focusing on the predictability of event progressions. In order to avoid confounding event predictability and the predictability of words in their descriptions, we created an artificial virtual world and trained participants to recognize typical event progressions within. Speakers then described novel scenes, which either conformed to their expectations or violated them, in a free production experiment. The data reveal that surprising event progressions lead to a more frequent production of definite noun phrases, in contrast to reduced linguistic expressions (pronouns and zero anaphors). We further introduce a Bayesian inference model, which offers an explanation of why a definite noun phrase and not a pronoun is used for the description of surprising events. We conclude that speakers choose more informative descriptions to talk about surprising events because this strategy allows them to avoids misunderstanding with higher likelihood
Comparative and superlative differentials: Experimental evidence from Czech
Get PDFThis article brings new experimental evidence for the treatment of Czech ne více než ‘no more than’ comparative differential modifier as the class B numerals modifier. At a more general level, our experimental data bring support for such theories of modified numerals which distinguish among them based on their semantics (like Kennedy 2015)
Mandarin bare indefinites
Get PDFThe recent literature on bare nouns in articleless languages suggests that they are subject to cross-linguistic variation. In this paper, we add Mandarin bare nouns to the debate. We identify a hitherto unnoticed restriction on their indefinite readings in object position and argue that it is related to the notion of stereotypicality. We develop a formal analysis of stereotypical verb-noun combinations, and show how it allows us – in combination with the independently motivated assumption that Mandarin yī ‘one’ is developing into an indefinite article – to derive the relevant restriction. We conclude that Mandarin bare nouns cannot freely take on indefinite readings and that Mandarin can no longer be considered a run-of-the-mill articleless language
(Non)Attitude verbs and control shift: Evidence from German
Get PDFRecent work on obligatory control (OC) phenomena (following, in particular, Landau 2015) holds that certain non-canonical OC construals such as partial control, implicit control or control shift are generally possible with matrix attitude verbs but not with nonattitude verbs. The crosslinguistic validity of this empirical generalization however is subject to on- going research. With regard to German, it has been disputed for the case of implicit control (Pitteroff and Schäfer, 2019), but supported for partial control (Pitteroff et al., 2017). The study reported in this paper contributes experimental evidence for an attitude/nonattitude contrast in the availability of control shift in German. The results also indicate that the relevant difference concerns the interpretation rather than the acceptability of the triggering construction, thus adding some nuance to the empirical picture on control shift
Polyadic cover quantification in heterofunctional coordination
Get PDFIn Heterofunctional Coordination conjuncts bear different grammatical functions, as in the attested \u27What and when to eat before workout\u27. While such English constructions are analysable in terms of ellipsis (cf. \u27What to eat [before workout] and when to eat before workout\u27), in Slavic they are often argued to involve direct coordination. The aim of this paper is to provide a semantic analysis of this phenomenon in Slavic, one that is based on a generalization of the cumulative polyadic lift, namely, on the cover lift
When a quantifier is not a quantifier: Non-conservative percentage expressions in German
Get PDFIn this paper we discuss percentage expressions of the type ‘fifty percent’ in German and two readings they give rise to, a conservative and a non-conservative one. Based on a questionnaire and a corpus study, we show that the non-conservative reading does not just need a bare, non-genitive-marked nominal to arise, but is also conditioned by word order (it has to appear low) and by the types of predicates that allow for it (in essence, predicates that can be used as existential or HAVE-predicates). We propose an account, under which the percentage expression is a type of scalar modifier that has an apparent effect on the predicate due to semantic incorporation ((in)transitives) and to an existential structure (intransitives)
On the gradient acceptability of quantifiers scoping over questions
Get PDFQuestions with quantifiers such as Which book did every student read? can allow for pair-list answers. However, whether or to what extent such pair-list answers are acceptable varies for different types of quantifiers. Recently, van Gessel and Cremers (2021) experimentally tested the acceptability of pair-list answers for questions with different types of quantifiers, and discovered a full gradient in their acceptability judgments. In this paper, we start with the intuitive idea that pair-list answers are expected if we let quantifiers take scope above questions, and show that we can extend inquisitive semantics in a principled way using independently motivated ingredients to derive such wide-scope readings that expect pair-list answers for various quantifiers. We also identify factors that contribute to their gradient acceptability
Present counterfactuals and the indicative-subjunctive divide
Get PDFThis paper aims to add to the longstanding debate on the relation between subjunctive and indicative mood in conditional sentences. In doing so, the theoretical range is narrowed down to three options, each of which takes at least one of the two moods to be presuppositional. Based on an experiment reported in (Wimmer, 2020), we take a few steps towards a competition-based analysis on which the subjunctive is vacuous (Schlenker, 2005; Leahy, 2011, 2018), contrary to its morphological markedness
The left-CONS2 constraint
Get PDFWe propose a new generalization over determiner denotations. We claim that, for any determiner Det, Det(P)(Q) entails Det(P⋂ Q)(Q) (the left-CONS2 Constraint, as we elaborate in the paper). We discuss potential counterexamples to the validity of this constraint including restricted universals (e.g. every ... but John, approximately/almost all/every...) and proportional expressions with an upper bound (e.g. fewPROP, at most one-third..., fewer than seventy percent...). Following earlier proposals, we provide evidence that problematic inferences associated with determiners arise not from their denotations but either from the operator Exh or from degree operators