Proceedings of Sinn und Bedeutung
Not a member yet
1139 research outputs found
Sort by
Extracting commitments: rising intonation and \u27ba\u27 in Mandarin
No full textThis paper investigates the particle ba and its interaction with the rising intonation in Mandarin. The paper presents novel data showing that ba can appear with rising tone only when the speaker is very certain about the prejacent, and that she invites the addressee to be committed to the prejacent as well by using the rising intonation. This is contrary to falling ba-declaratives, which typically express the speaker’s uncertainty. We formalize the effects of a rising ba-declarative by associating ba and the rising intonation to the following contextupdating conventions: one, ba adds the content of the prejacent to the speaker’s projected commitment set; two, the rising tone adds the content of the prejacent to the addressee’s projected commitment set. The account reconciles the seemingly contradictory effects of rising/falling ba-declaratives, and explains the ‘special’ effect of rising tone in ba-declaratives
A constraint on co-occurrence of partitive quanti?ers and gradable predicates
Get PDFQuanti?ers like most of are compatible with some predicates but not others, e.g., Most of the road is {narrow / #long}. In the context of collective predicates, this is known as the gather/numerous distinction. This paper focuses on gradable predicates—both singular and collective—and argues that only non-monotonic gradable predicates are compatible with quanti?ers like most of. I show that if a monotonic predicate co-occurs with such a quanti?er, the sentence does not have well-de?ned truth conditions, which makes it unacceptable
Event-related readings and degrees of difference
Get PDFI present an analysis of event-related readings that treats both them and object-related readings as resulting from quanti?cation over temporal stages of individuals, with the distinction between the two arising from a difference in the degree of maximality that quanti?ers require those stages to ful?l. The analysis accounts for the compositional-semantic effect of different in these examples, and the possibility of multiple event-related readings. I also examine the connection between this analysis and existing accounts of quanti?er domain restriction
Concealed questions without meta-meta-questions
Get PDFSentences like John knows the price that Fred knows are ambiguous between ‘Reading A’ and ‘Reading B’. When such sentences have a further embedding, existing analyses predict ‘Reading C’ that involves meta-meta-questions. This paper presents an analysis in which the escalating ambiguity disappears when the referent of concealed question DP has a proper name. I show that Reading C is reduced to a version of Reading B without meta-meta-questions. The proposal is based on the idea that questions are given different interpretations depending on the perspective taken to identify entities in the domain
Non-Manner ‘how’-Complements in English and in German
Get PDFThis paper compares the lexical and selectional semantics of non-interrogative, nonmanner how-complements in English and German. It argues that English has two types of such complements: how-complements that are licensed by perception, memory, and fiction verbs and allow paraphrase by a DP of the form ‘the event in which [TP ]’ (see Umbach et al., submitted), and how-complements that are licensed by presuppositional and ‘say’-verbs and allow paraphrase by a DP of the form ‘the fact that [TP ]’ (see Legate, 2010). The paper shows that German only has the first type of non-manner how-complement. It provides a characterization of these types of complements in terms of their observed entailment behavior
Situation partition and shifted evaluation of distributivity
Get PDFIn this paper, I discuss what I call group distributive readings. Japanese has a distributor “zutsu,” which can occur at the prenominal position or at the floating position. When it occurs at the prenominal position, distributivity is evaluated in a situation other than the situation in which the rest of the clause is evaluated. To account for this shifted evaluation ofdistributivity, I propose that “zutsu” partitions a situation and the shifted evaluation of distributivity comes from the presence of a situation pronoun. The syntactic distribution of group distributive readings follows from an independently motivated generalisation on situation pronoun binding. In addition, this situation-based account makes better predictions for the semantics of canonical distributive readings than previous analyses do
Why premise questions?
Get PDFThis paper is concerned with the interpretation of conditional questions introduced by the wh-word why (why-conditional questions, WCQs for short). In particular, we observe that WCQs – unlike the other conditional questions – cannot be uttered out of the blue and hence have a hypothetical reading; instead, the felicitous utterance of a WCQ requires some interlocutor’s commitment to the proposition in its antecedent (p). This makes WCQs necessarily premise conditionals in the sense of Iatridou (1991); Haegeman (2003). Along with Goebel (2017), we assume that premisehood is a speci?c use of a regular indicative conditional sentence. We explain the premise interpretation of conditionals as a combination of their thematic nature and the discourse constraints derived from the fact that p has its own discourse history. Additionally, we derive the necessary premise interpretation of WCQs from the impossibility of entertaining the question “What happens if ¬p?”, which is necessary for the hypothetical interpretation of indicative conditionals
Remarks on two approaches to NPI licensing
Get PDFIn relation to the notion of informativity, two types of approaches to DPs headed by any (any-DPs) have been distinguished. At a ?rst approximation, one approach takes an any-DP to be accompanied by a requirement that a clause containing it be more informative than all its relevant alternatives (cf. Kadmon & Landman 1993), while the other approach requires a clause containing it to be more informative than all its relevant alternatives that are true (cf. Chierchia 2013). The goal of this paper is to compare these approaches with respect to their predictions about the distribution of plural any-DPs in modal environments
The role of visual cues in detecting irony
Get PDFWe present four studies that aimed at investigating the contribution of purely visual cues for the detection of irony. In Study 1-3, we presented, without any preceding context, remarks (criticisms and compliments) uttered with sincere and with ironic intent, in three modalities: in the V modality participants could read the comment and see speakers’ facial expressions and bodily movements; in the A modality they could only hear the audio tracks of the uttered sentences; in the VA modality, both visual and auditory information were present. We found that purely visual cues were sufficient to discriminate the ironic intent of the speakers. In Study 4 we presented comments in the V modality, without showing the content of the remark: Accuracy in the detection of sarcasm dropped. We discuss that irony in Study 1-3 might have been recognized indirectly, by comparing the polarity of the remark with the polarity of the actors’ attitude, and we interpret Study 4 data as casting some doubts on the idea that there exist visual cues that specifically convey the speaker’s ironic intent
\u27Find\u27, \u27must\u27 and conflicting evidence
No full text\u27Find\u27-verbs — English find, German finden, French trouver and their counterparts in other languages — have figured prominently in the literature on subjective language, as they only allow complements that are about matters of opinion, rather than fact. This paper focuses on a lesser-studied property of \u27find\u27-verbs: the ban on must-modals in their complements and their interaction with epistemics and evidentials at large. The find-must ban has been attributed to a clash in subjectivity, with must-modals assumed to not be of the right type. We argue instead that the find-must ban is of evidential nature: \u27find\u27-verbs convey directness, \u27must\u27-modals convey indirectness, and their combination is a semantic contradiction. We couch our proposal in terms of von Fintel and Gillies’s (2010) kernels, modal bases responsible for direct knowledge. We show that \u27find\u27-verbs ban a variety of indirect markers across languages and further argue that \u27find\u27-verbs can embed epistemic modals, but only those that do not semantically encode indirectness, and thus draw a line between semantic vs. pragmatic evidential effects