Veranstaltungen

Bisher haben wir im Rahmen des Projekts die folgenden Vorträge organisiert:

Vortrag von Hana Filip und Peter Sutton (10. September 2019)

Informational Object Nouns and the mass/count distinction

This paper provides a novel contribution to the small but growing literature on the countability of abstract nouns (Grimm, 2014; Nicholas, 2010; Tovena, 2001; Zamparelli, 2018, amongst others).

Abstract nouns have so far been largely set aside in semantic theories of the mass/count distinction, which focus on the more tractable concrete Ns denoting material objects or stuff. Given the heterogeneity of abstract Ns (Grimm, 2014; Zamparelli, 2018), here we focus on one lexical subclass, which we call Informational Object Ns (IONs).

What needs to be explained is why some IONs are straightforwardly countable (e.g., three statements/beliefs/facts), they are count, but others are not (#three information(s)/knowledge(s)), they are mass. We draw on theories of the mass/count distinction developed for concrete nouns, especially on the semantics of Collective Artefact Nouns (CANs), e.g., furniture, equipment, and jewellery, which we argue share certain grammatical properties with IONs. Moreover, we use the type-generalised mereological approach in Schmitt (2013) and Schmitt (2017) which can accommodate, e.g., sums of propositions as well as the more familiar sums of entities. We also appeal to the semantic stative/episodic distinction which, perhaps surprisingly, turns out to be one factor in determining what interpretations are available for some IONs in counting constructions.

Vortrag von Swantje Tönnis (27. Juni, 2019)

German es-clefts in discourse – How clefts interact with the discourse question

In this talk, I will present a new perspective on German es-clefts that analyzes the effect of clefts on discourse coherence. In the context in (1), the cleft in (1-a) is better that the unclefted sentence in (1-b), at least when Peter is supposed to refer back to einem Typen (‘a guy’).

(1) Lena hat gestern auf der Party mit einem Typen1 gesprochen. Die beiden haben viel gelacht und sich direkt für den nächsten Abend verabredet. Dann ist Lena glücklich nach Hause gefahren. 'Lena talked to a guy1 yesterday at the party. The two laughed a lot and they immediately agreed to meet the next evening. Then Lena went home happily.’

a) Es war Peter1, mit dem sie gesprochen hat.

‘It was Peter1 she talked to.’

b) ?Sie hat mit Peter1 gesprochen.

‘She talked to Peter1.’

Previous literature extensively discussed exhaustivity of clefts, but also its relation to contrast and prosody marking. I show that none of the existing approaches can explain the difference between (1-a) and (1-b). Instead, I argue that the cleft is used to address a different question than the expected discourse question (as Simons et al. (2017) define it), but still needs an antecedent in the context. My analysis is based on corpus examples from novels, similar to (1), which nicely show the text coherence effect of German clefts.

Vortrag von Beáta Gyuris (25. Juni 2019)

Biased polar questions in Hungarian

The talk provides a survey of the basic form types that can be used to encode polar questions in Hungarian, which include

a) polar interrogatives marked with the help of the -e interrogative particle, and
b) polar interrogatives marked with the help of a final rise-fall (/\) intonation contour.

It is argued hat whereas both forms can be shown to represent the interrogative sentence type (e.g. due to compatibility with NPIs), and both are available to encode information questions in a neutral context, in other contexts their distribution is quite different. For example, -e-interrrogatives are excluded from contexts where evidence supporting the positive answer is present, and are not available for encoding indirect requests. In addition, whereas negative /\-interrogatives give rise to both inside and outside negation readings (cf. Ladd 1981), negative e-interrrogatives only give rise to the latter. The distributional facts are represented in a two-dimensional system of bias types, proposed by Sudo (2013). We look at how the distinctions could be accommodated into recent theoretical models of polar interrogatives (Farkas&Bruce 2010, Krifka 2015).

Vortrag von Hans-Martin Gärtner (18. Juni 2019)

"Varieties of Dependent V2 and Verbal Mood: A View from Icelandic"

I will discuss varieties of "dependent V2" with "broad" (bDV2) and "narrow" (nDV2) distribution − aka "generalized" and "limited embedded V2" − arising within Icelandic. This pattern is taken to correlate with construals of verbal mood as "dominant" in the former and "non-dominant" in the latter case, where dominance of verbal mood allows disregarding the illocutionary impact of V2. I further show that the variation fits into a model of historical stages with earlier variants "recruiting" verbal mood for clause combining and drift in later stages toward "autonomous" mood, i.e., toward a mood system with enhanced semantico-pragmatic transparency.

Vortrag von Jakub Dotlacil (7. Juni 2019)

"Dynamic inquisitive semantics"(joint work with Floris Roelofsen)

In this talk, I will discuss a new logical framework to analyze questions. The framework brings together two important strands in semantic theorizing, that of dynamic semantics on one hand (Kamp, 1981, Heim, 1982, Groenendijk and Stokhof, 1991, a.o.) and the analysis of questions using inquisitive semantics on the other hand (Ciardelli et al., 2018). The proposed framework preserves the advantages of earlier dynamic analyses of questions over static ones, as well as the advantages of earlier inquisitive theories over non-inquisitive ones.

The framework will be applied to several topics that played an important role in semantic accounts of questions. These topics include anaphora in wh-questions and polar questions, intervention effects, multiple wh-questions, and mention-some and mention-all questions. I will show that the proposed framework can provide an analysis of these and other phenomena and in some instances the analysis goes beyond the current state of the art.

Vortrag von Yasutada Sudo (20. Mai 2019)

"Plurality inferences as quantity implicatures"

Plural nouns typically give rise to 'plurality inferences', e.g. "Andrew wrote papers" implies that Andrew wrote multiple papers. But plurality inferences are not always present, e.g. "Andrew did not write papers". There are three types of approaches in the literature: (i) the scalar implicature approach (Spector 2007, Zweig 2009, Ivlieva 2013, Mayr 2015), (ii) the ambiguity approach (Farkas & de Swart 2010, Grimm 2013, Martí 2018), and (iii) the antipresupposition approach (Sauerland 2003, Sauerland et al. 2005). I propose a new scalar implicature account. The scalar implicature account assumes that the plural is semantically number-neutral, and the plurality inference arises as a scalar implicature in competition with the singular. However, the scalar implicature computation is not straightforward, given that pairs like "Andrew wrote papers" and "Andrew wrote a paper" would be truth-conditionally identical. Different versions of the scalar implicature account make use of different truth-conditional asymmetries, e.g. non-global levels of meaning, strengthened meaning, etc. I propose instead that the plurality inference can be derived as a global-level implicature based on non-truth-conditional aspect of the meaning. Specifically, the singular and plural sentences differ in anaphoric possibilities: the plural sentence introduces a discourse referent that ranges over singular or plural entities, while the singular sentence introduces a discourse referent that only ranges over singular entities. Based on this asymmetry, a quantity implicature is derived that the discourse referent is meant to only range over plural entities. I will formalize this idea in update semantics, and demonstrate that it makes correct predictions about negative sentences and quantified sentences.

Vortrag von Maik Thalmann (25. März 2019)

"Presuppositions At-Issue: Soft and Hard Triggers in Language Acquisition" (Joint work with Yuqiu Chen and Mailin Antomo)

Classically, the term presupposition was used to describe expressions that have rather homogeneous properties. More recently, however, a number of tests, among them felicity in epistemically deficient contexts, gave rise to a dichotomous subdivision into hard and soft presupposition triggers.

This sensitivity to explicit ignorance is closely related to Question Under Discussion (QUD) considerations and at-issueness, the latter of which has been implicated in being predictive of projection behavior (Xue & Onea 2011) and to affect presuppositions? felicity (Aravind & Hackl 2017) since presuppositions instantiate backgrounded utterance components. The present work seeks to investigate the following research questions in this regard: (i) is the above two-way split more of a continuum (Tonhauser, Beaver & Degen 2018), (ii) can we disentangle presuppositionality and at- issueness constraints, and (iii) are there reflexes in how children acquire or judge different triggers?

The interim test results among German adult and child speakers show that at a descriptive level, backgroundedness and presuppositional status, while overlapping, need to be distinguished. Also, there is consistent variability between soft and hard triggers as well as different triggers within one category. Further, 4-to-6 year old children, although being sensitive to some of those contrasts, struggle with at-issue backgrounded content generally, while adults are very reluctant to accept hard triggers (and non restrictive relative clauses) but respond well to soft triggers when backgroundedness is violated.

Vortrag von Daniele Panizza (25. März 2019)

"The Nörten-Hardenberg school children conspiracy in the Neo-Gricean Paradise. On the processing and interpretation of indirect scalar implicatures in children and adults."

We report on a set of four experiments involving offline semantic judgment and online eye-tracking ambiguity resolution of sentences including indirect Scalar Implicature triggers (i.e. not all -> some) such as (1):

(1) Der Kapitän hat nicht mit allen Meerjungfrauen getanzt.
The captain has not with all mermaids danced.

A first set of experiments implementing unbiased prosody surprisingly showed that younger children (i.e. 6 year old) were faster in identifying the target of sentence (1) as compared to adults. A second set of experiments, in which the critical sentences were recorded with natural intonation, will shed light on the nature of the results of the first study, and tell whether it is the comprehension and processing of intonation, as compared to other factors, the potential culprit of the apparently unexplainable pattern of results of the first study.

Vortrag von Uli Sauerland (18. März 2019)

"Adverbs in collective conjunction" (joint work with Aron Hirsch)

Collins (1988) noted that the occurrence of sentential adverbs in NP-conjunctions (e.g. John and possibly Mary left) provides an argument that and in such cases can have a clausal type ⟨t,tt⟩ analysis. Sentential adverbs are possible even with collective predicates, (1). Schein (1993, 2017) proposed that and in collective conjunction is uniformly type ⟨t,tt⟩ with its sentential arguments created by covert event predicates, (2).

(1) John, Bill, and also / possibly Mary met.

(2) ∃e [John participates in e ∧ Bill participates in e ∧ {also, possibly} Mary participates in e ∧ meet(e)]

We show, however, that the distribution of sentential adverbs is more restricted than expected if all collective conjuncts were sentential and, in doing so, argue that a type ⟨e, ee⟩ analysis of conjunction is possible (Link 1983).

Vortrag von Kazuko Yatsushiro (18. März 2019)

"Children know the great scope inversion conspiracy"

Vortrag von Anja Müller (15. März 2019)

"Offline oder Online? Das (Nicht-)Verstehen von Sätzen mit Fokuspartikeln"

Die zielsprachliche Interpretation von Sätzen mit Fokuspartikeln wie (1) und (2) erfordert nicht nur Wissen um die Bedeutung und Funktion der Fokuspartikel, sondern ebenfalls Wissen über die Informationsstruktur des Satzes.

(1) Nur die Maus hat einen Ball. (Subjekt-Fokus)

(2) Die Maus hat nur einen Ball. (Objekt-Fokus)

Die Ergebnisse bisheriger Studien zeigen, dass Kinder bis ins Grundschulalter hinein Sätze mit Fokuspartikel nur nicht zielsprachlich interpretieren (u.a. Crain et al. 1994; Patterson et al. 2003). Bisherige Daten scheinen zu belegen, dass Kinder bei der Interpretation von Sätzen wie (1) und (2) keinen Unterschied machen und beide Sätze im Sinne Satz (2) interpretieren. Die nicht zielsprachlichen Leistungen der Kinder werden auf Grundlage eingeschränkter semantischer und/oder syntaktischer und /oder pragmatischer Fähigkeiten diskutiert. Die Ergebnisse lassen jedoch ebenfalls die Überlegung zu, dass die nicht zielsprachlichen Leistungen auf Schwierigkeiten im Bereich der Informationsstruktur zurückzuführen sind. Vor diesem Hintergrund werden Daten verschiedener Studien betrachtet, die das Verstehen von Sätzen mit der Fokuspartikel nur für monolingual deutsch aufwachsende Kinder, Kinder mit Deutsch als frühe Zweitsprache, Kinder mit Deutsch als späte Zweitsprache und Erwachsene untersucht haben. Den Daten der offline-Studien stehen Ergebnisse von online-Studien gegenüber, die die gefundene Subjekt-Objekt Asymmetrie nicht bestätigen. Die Daten von Paul et al. (2019) lassen sogar vermuten, dass Kinder Sätze wie (1) besser verarbeiten werden als Sätze wie (2). Eine ausführliche Diskussion der Ergebnisse erfolgt im Vortrag.

Vortrag von Edgar Onea (30. November 2018)

Gemeinsam mit dem Institut für Sprachwissenschaft organisiert.

"Mild rationality and communicative success. The case of focus exhaustivity"
In successful communication, the literal meaning of linguistic utterances is often enriched by pragmatic inferences. Part of the pragmatic reasoning underlying such inferences has been successfully modelled as Bayesian goal recognition in the recently developed Rational Speech Act (RSA) framework (Frank and Goodman 2012). In such a framework, the remarkable reliability of linguistic communication arguably boils down to the underlying rationality of speaker and addressee in the probabilistic inferencing process. This begs the question whether complexity of computational tasks will impact communicative success.
In this paper we model the interpretation of question-answer sequences with narrow focus in the answer in the RSA framework, thereby exploring the effect of computational complexity on interpretation. We present experimental data that suggest that the RSA reasoning appears to be applied imperfectly by interlocutors, but still leading to communicative success. We suggest a theoretical model in which communicative success uses rational reasoning as a starting point only, whereas actual success may rely on the predicted rather than the correct reasoning of interlocutors.

Vortrag von Guglielmo Cinque (21. November 2018)

"Some notes on 'meaningless movement'" Abstract [pdf]

Vorträge von Magdalena Kaufmann and Maša​ Močnik (13. Juli 2018)

Magdalena Kaufmann (University of Connecticut): "Stay strong and we'll still have a story"
The occurrence of imperatives in the first conjunct of conditional conjunctions, I(mperatives )a(nd )D(eclarative)s like “Take a step to the left and you’ll fall down the stairs”, constitutes a notorious puzzle for semantic theories of imperatives. In the previous literature, IaDs have been adduced as an argument for the ambiguity of the strings that occur in imperative clauses (Han 2000, Russell 2007), but also as evidence against a modal (aka “strong”) theory of imperative clauses (von Fintel & Iatridou 2017). In this talk, I will revisit some of the arguments and data, and I will propose a novel semantics for conditional conjunctions that builds crucially on the prosodic (Krifka 2004, Keshet 2013) properties of conditional conjunctions, on conditional antecedents as aboutness topics (Ebert, Ebert, and Hinterwimmer 2014), and on non-truth functional interpretative properties of “and” (Klinedinst and Rothschild 2015). I will argue that, in combination with general principles regarding the integration of linguistic material into the discourse, the resulting theory covers the full variety of conditional conjunctions, in particular IaDs with imperative first conjuncts interpreted along the lines of the strong theory I proposed in Kaufmann 2012, 2016.

Maša Močnik (MIT): "Where Force Matters: Life under Doxastic Attitudes"
There has been much recent interest in the analysis and distribution of embedded epistemic modals (Yalcin 2007, Anand and Hacquard 2013, a.o.). I present novel data using the embedding verb dopuščati (‘to allow for the possibility that’) from Slovenian, analysed as an existential doxastic attitude, and argue for a new analysis of epistemic modals that captures their restricted distribution under doxastic attitudes. In particular, I explain why embedding a universal epistemic modal is typically degraded under dopuščati (the Slovenian version of #I allow for the possibility that it must be raining) and negated doxastics. I build on Mandelkern’s (2018) semantics for epistemic modals and argue for a pragmatic competition story along the lines of Magri (2009, 2011), which is used to explain the oddness of sentences like #Some Italians come from a warm country.

Vortrag von Clemens Mayr (18. Juni 2018)

"Triggering presuppositions from L-analyticity"
Why are there presupposition triggers? For instance, why does Tony knows that it is raining presuppose that it is raining rather than assert so, as evidenced by the fact that the inference remains under negation? This question is at the heart of the so-called triggering problem. This puzzle has not received a fully satisfactory answer to this date. In this talk, I suggest that the presupposition of an expression α is a regular entailment of α. Its presuppositional status results from an obligatory restriction on the function denoted by α. In particular, this restriction is imposed by the deductive system – a system of interpretation only concerned with formal properties of language. Without such a restriction a sentence containing α would be ruled L-analytic by the deductive system, which would result in ungrammaticality. In order to avoid L-analyticity an operator is inserted that is interpreted as a restriction on the function in effect yielding the presupposition. In other words, presuppositions are not lexically stipulated. But certain regular entailments must be made presuppositional to avoid L-analyticity.

Workshop "Quantifikation und Pluralität", 13. Juni 2018

Programm

14:00-14:50
Kazuko Yatsushiro (Leibniz-Zentrum Allgemeine Sprachwissenschaft Berlin): "Plural is unmarked: evidence from Turkish, Hungarian, and German"

15:00-15:50
Uli Sauerland (Leibniz-Zentrum Allgemeine Sprachwissenschaft Berlin) "The Thought Uniqueness Hypothesis"

16:10-17:00
Nadine Bade (Universität Tübingen) "Exhaustivity in Questions - evidence from German particles so and alles"

17:10-18:00
Sigrid Beck (Universität Tübingen) "Constant Entailments and Compositional Change in Universal Quantification"

Vortrag von Marcin Wągiel (24. April 2018)

"Subatomic quantification in natural language"
In standard lattice-theoretic approaches to natural language (e.g., Link 1983, Landman 2000, Champollion 2017) singularities and pluralities are presumed to involve distinct mereologies and it is commonly supposed that quantificational expressions do not access subatomic part-whole relations. In this paper, I explore three hypotheses regarding natural language semantics: i) natural language is sensitive to topological relations holding between parts of singularities (cf. Grimm 2012), ii) there are general counting rules that presuppose such relations, and iii) quantification over parts is subject to identical restrictions as quantification over wholes. The evidence comes from the interaction between cardinal numerals and partitives involving irregular plurals in Italian as well as distinct classes of Polish `half' words and from the quantificational behavior of multipliers such as English `double'.

Vortrag von Jovana Gajić (12. März 2018)

"Negated conjunction and its implicatures"
Different intonational patterns correspond to different interpretations of the sentences with negated definite conjunctions in English and, consequently, to different possible continuations. The major parameter inside these prosodic patterns seems to be whether the stress is on the connective and or on the whole conjunction as a coordinated constituent. In the former case, the reading obtained is “not both” (¬[A∧B]), whereas in the latter, it is equivalent to a negated disjunction (¬[A∨B]), i.e. “neither”. This difference can be explained if different kinds of alternatives are activated and if exhaustification is applied a different number of times. Furthermore, contextual restrictions observed in the previous literature (Szabolcsi and Haddican 2004) fall out from this account.

Vortrag von Hilda Koopman (21. Juni 2017)

"On (some) Interface mismatches: when the syntax is not what it seems"
The problem: In (i), seem scopes above cannot, as shown in the paraphrase in (ii).

(i) Bill cannot seem to fix this.

(ii) It seems that Bill cannot fix this.

In the syntactic structure seem appears to be embedded under cannot, yielding what looks like a clear syntax-LF mismatch. I will show that the syntax of this construction must be more complex than it seems to be at first sight, and that there is in fact no syntax semantics mismatch. Arguments will be presented that (i) must be derived from the merge structure in (ii) seem > can > can. Insights into the syntactic derivation of (i) in English comes from the syntax of related Germanic OV languages. The analysis extends to other elements that raise in this construction (Tense and Adverbs, yielding an account for apparent adverb scrambling).
From English to German: I will suggest how the proposed analysis for English may in turn pave the way for a purely syntactic treatment of a well-known and curious syntax-PF mismatch with regards to zu-placement in German, which has called for a postsyntactic treatment, based on the DM mechanism known as "local dislocation" (Saltzman 2012). In sum, as the argument goes, once we understand the (independently motivated) syntactic derivations, these are instances of syntax phonology and LF interface matching, not instances of interface mismatches, neither of the syntax LF interface, nor of the syntax PF interface. This in turn can be taken as an argument in favor of the type of syntax that we should pursue.

Vortrag von Mojmír Dočekal und Marcin Wągiel (19. Mai 2017)

"Derived collectives in Slavic: Decomposing groups, swarms, and aggregates"
In this talk we bring in novel evidence concerning semantic properties of derived bunch nouns in Czech and Polish. We discuss several classes of collectives derived by different suffixes. Some of them form standard group nouns sometimes referred to as committee nouns (Pearson 2011) whereas others seem to derive spatiotemporally groupings, i.e., swarms (Henderson 2017), or aggregates (Grimm 2012). Furthermore, we present a design for an experiment exploring anaphoric relations between group nouns and reciprocal expressions (Dotlačil 2012).

Vortrag von Uli Sauerland (15. März 2017)

"On Presupposition Projection from Quantifier Scope"
Quantifiers seem to differ in how they project presuppositions from their scope. Fox (2013) argues that a general mechanism that takes into account not only the quantifier, but also the environment it occurs in, is required. He develops such an account based on Strong Kleene Logic, but already discusses a problem of the account with indefinites. We adduce further problems modified numerals create for Strong Kleene Logic, and then suggest an alternative based on intermediate accommodation.

Vorträge von Fred Landman und Susan Rothstein (6. und 7. März 2017)

Gemeinsam mit dem Institut für Sprachwissenschaft organisiert.

Susan Rothstein (Bar-Ilan University): "Cardinality scales, measuring and approximation"
In this talk I explore and develop the concept of cardinality scales, originally proposed in Rothstein (2016), which allow us to compare the cardinal properties of pluralities without actually counting the atomic parts of the pluralities. These cardinality scales allow us to express the truth conditions of sentences such as (i) without making direct reference to the cardinalities of the sums of furniture that Mary and John possess.
(i) Mary has more furniture than John.
Cardinality scales allow us to give a single semantic representation for (i), while allowing it to be evaluated in terms of any contextually relevant parameter of comparison in contrast to (ii), which forces a comparison in terms of cardinality:
(ii) Mary has more pieces of furniture than John.
I suggest that cardinality scales are particularly useful in situations of approximation, where we are able to say whether plurality A or B is larger, without knowing what the exact cardinality of either A or B is, and I conclude by reviewing a number of different approximation techniques, comparing approximation using cardinality scales to imprecise and indirect counting.

Fred Landman (Tel Aviv University): "Why measures are mass and how mass counts"
Abstract [pdf]

Workshop "Semantik der Konjunktion", 19. Dezember 2016

Programm

12:00-13:00
Lucas Champollion (New York University): "The two senses of and in the coordination of nouns"

13:15-14:15
Viola Schmitt (Universität Wien): "Conjunction, plurals and non-local dependencies"

Kaffeepause

14:45-15:45
Yoad Winter (Universität Utrecht): "With without symmetry: the semantics of collective comitatives"

16:00-17:00
Moreno Mitrović (Universität des Saarlandes): "A historical typology of conjunction meanings"

Vortrag von Ryan Bochnak (24. Oktober 2016)

"Optional past tense in Washo and paradigmatic presuppositional alternatives"
Descriptively, Washo (language isolate; USA) is a language where past tense is optional, in the sense that a past tense morpheme is not obligatory in finite clauses with past temporal reference (Jacobsen 1964). I first argue that the past marker in Washo can be given an analysis as a vanilla past tense, contra Plungian and van der Auwera (2006). Second, I argue that tense features in Washo truly are optional, in the sense that morphologically tenseless clauses do not even contain a covert semantic tense (e.g., Matthewson 2006). To make this case, I provide a new argument based on Maximize Presupposition, which is made available to us precisely because Washo is an optional tense language. This analysis raises issues regarding the nature of optional tense marking in languages that instantiate the phenomenon, as well as tenselessness and the asymmetry between past and future reference. Along the way, I also highlight strategies for collecting the relevant data in a fieldwork situation.