Formal Semantics/Pragmatics

This monograph investigates the temporal interpretation of narrative discourse in two parts. The theme of the first part is narrative progression. It begins with a case study of the adverb ‘now’ and its interaction with the meaning of tense. The case study motivates an ontological distinction between events, states and times and proposes that ‘now’ seeks a prominent state that holds throughout the time described by the tense. Building on prior research, prominence is shown to be influenced by principles of discourse coherence and two coherence principles, NARRATION and RESULT, are given a formally explicit characterization. The key innovation is a new method for testing the definitional adequacy of NARRATION and RESULT, namely by an abductive argument. This contribution opens a new way of thinking about how eventive and stative descriptions contribute to the perceived narrative progression in a discourse.

The theme of the second part of the monograph is the semantics and pragmatics of tense. A key innovation is that the present and past tenses are treated as scalar alternatives, a view that is motivated by adopting a particular hypothesis concerning stative predication. The proposed analysis accounts for tense in both matrix clauses and in complements of propositional attitudes, where the notorious double access reading arises. This reading is explored as part of a corpus study that provides a glimpse of how tense semantics interacts with Gricean principles and at-issueness. Several cross-linguistic predictions of the analysis are considered, including their consequences for the Sequence of Tense phenomenon and the Upper Limit Constraint. Finally, a hypothesis is provided about how tense meanings compose with temporal adverbs and verb phrases. Two influential analysis of viewpoint aspect are then compared in light of the hypothesis.

The monograph is directed at graduate students and researchers in semantics, pragmatics and philosophy of language. The analysis of narrative discourse that is developed in the monograph synthesizes and builds on prior collaborative research with Corien Bary, Valentine Hacquard, Thomas Roberts, Roger Schwarzschild, Una Stojnić, Károly Varasdi and Aaron White.

It is our main claim that the time is ripe to link the dynamic turn launched by game-theoretical approaches to meaning with P. Martin-Löf's Constructive Type Theory (CTT). Furthermore, we also claim that the dialogical framework provides the appropriate means to develop such a link. We will restrict our study to the discussion of two paradigmatic cases of dependences triggered by quantifiers, namely the case of the Axiom of Choice and the study of anaphora, that are by the way two of the most cherished examples of Hintikka.

Cette étude porte sur les aspects morphosyntaxiques et sémantiques des proverbes bron, principalement le bron de Côte d’Ivoire. Elle a pour particularité de lever le voile sur les mécanismes de construction et d’interprétation de ces énoncés réservés aux initiés et aux amateurs de « l’art discursif ». Au niveau morphosyntaxique, l’analyse prend appui sur le cadre théorique de la Grammaire Générative. Les données empiriques nous ont permis de relever d’une part, divers modes de créations lexicales, en l’occurrence la dérivation par préfixation, par suffixation, et par relèvement tonale, la composition nominale et verbale (l’incorporation), les slogans, la lexicalisation, les noms mythiques et les onomatopées.
D’autre part, l’on a décelé les différentes têtes fonctionnelles rencontrées tant dans les structures simples que complexes, et les mouvements syntaxiques qu’elles suscitent. Aussi, avons-nous fait cas des marques aspectuelles régissant la description des procès avec les valeurs explicites ou implicites qu’elles portent. La deuxième partie nous a donné, par le biais des théories vériconditionnelle et pragmatique, de montrer que les locuteurs bron ont recours à des présuppositions sémantiques et pragmatiques, ou à des implicatures conversationnelles dans l’interprétation des proverbes. Cela nous a conduit à une réévaluation des aspects, partant de l’idée que le Télos contribue à cette interprétation. Nous terminons l’étude par une typologie des énoncés proverbiaux selon les quatre (4) critères qui se dégagent des particularités que révèlent l’ensemble de l’analyse.
Mots clés : morphosyntaxe, pragmatique, Programme Minimaliste, sémantique, Télos

There exist various sentence types in natural language that, under certain circumstances, are evaluated as neither true nor false. For instance, in a context in which the presupposition of a sentence is not satisfied, it is intuitively rather difficult to assess what the truth value of the sentence should be. A common theoretical approach is to characterize the status of such a sentence with a third value of one kind or another. In this paper, we consider children’s ac- quisition of four linguistic phenomena that can give rise to ‘gappy’ judgments that correspond neither to True nor False: scalar implicature, presupposition, homogeneity, and vagueness. We discuss how young children’s interpretations of such sentences can provide insight into how these phenomena should be treated within semantic theories.

Discourse relations link two different discourse units into a compound unit, and it is the presence of such relations that gives a discourse coherence. For example, the relations RESULT and NARRATION are responsible for the perceived narrative progression of the events described in the discourse. Building on Hobbs (1985) and Kehler (2002), we use the definitions of RESULT and NARRATION to derive constraints that pertain to the internal structure of discourse units, and we argue that the plausibility of these constraints lends a new type of support for the
definitions that we propose.

Peirce and Frege both distinguished between the propositional content of an assertion and the assertion of a propositional content, but with different notational means. We present a modification of Peirce's graphical method of logic that can be used to reason about assertions in a manner similar to Peirce's original method. We propose a new system of Assertive Graphs (AGs), which unlike the tradition that follows Frege involves no ad hoc sign of assertion. We show that axioms of intuitionistic logic can be derived from AGs, and argue that AGs analyse and represent assertions and illocutionary content in a way which is motivated both by its logical properties and its historical connection with the ideas that led to the development of the graphical method.

There is a longstanding debate in the literature about static versus dynamic approaches to meaning and conversation. A formal result due to van Benthem (1986, 1996) is often thought to be important for understanding what, conceptually speaking, is at issue in the debate. We introduce the concept of a conversation system, and we use it to clarify the import of van Benthem's result. We then distinguish two classes of conversation systems, corresponding to two concepts of staticness. The first class corresponds to a generalization of the class of systems that van Benthem's result concerns. These are the strongly static conversation systems. The second class is broader, and allows for a certain commonly recognized form of context sensitivity. These are the weakly static conversation systems. In the vein of van Benthem's result, we supply representation theorems which independently characterize these two varieties of conversation system. We observe that some canonically dynamic semantic systems correspond to weakly static conversation systems. We close by discussing some hazards that arise in trying to bring our formal results to bear on natural language phenomena, and on the debate about whether the compositional semantics for natural language should take a dynamic shape.

We investigate a novel use of the English temporal modifier 'now', in which it combines with a subordinate clause. We argue for a univocal treatment of the expression, on which the subordinating use is taken as basic and the non-subordinating uses are derived. We start by surveying central features of the latter uses which have been discussed in previous work, before introducing key observations regarding the subordinating use of 'now' and its relation to deictic and anaphoric uses. All of these data, it is argued, can be accounted for on our proposed analysis. We conclude by comparing 'now' to a range of other expressions which exhibit similar behavior.

What is the defining claim of Radicalism? That is, the one particular claim that sets it apart from the other views found throughout the literature, from Minimalism to Contextualism. That claim is the following: that semantics, while fully determined by the lexico-syntactic elements of a sentence, is not truth-conditional. Radicalism supports the idea of a sub-propositional semantics.
Both Contextualists, and Minimalists, are adamant that “semantics” – whatever else it is – is truth-conditional. That is, they both argue that a “semantic” theory deals in propositional contents.
For Contextualists, the desire to uphold a “truth-conditional semantics” leads them to dissolve the clear border between the “semantic” and the “pragmatic” and neglect the theoretically tractable, systematic level of context-free content. For them, there is pragmatic information – non-syntactically mandated information – involved in establishing truth-conditional “semantic” content. Thus, for them, “semantic” content is defined as the first intuitive literal proposition expressed by an utterance of a sentence that is consciously available to the hearer.
Minimalists, on the other hand, subscribe to the idea that the context-free linguistic content of sentence is properly truth-conditional. Thus, they maintain the traditional idea of “semantic” content as being context-free, compositional content and truth-conditional. That is, even in the face of underdetermination and incompleteness arguments (sentences containing expressions such as “ready”, tall’, “enough”, “continue” etc.), Minimalists argue for a full and complete truth-conditional level of context-free content. They do allow that this minimal content is often not what is meant or communicated by the utterance of a sentence, but nonetheless maintain its autonomy as a fully truth-conditional level of meaning.
Radicalism disagrees with both Contextualists and Minimalists on this issue. It maintains the traditional conception of a context-free “semantics”, but rejects the idea that this content is truth-conditional. Thus, according to Radicalism, the output of a semantic theory does not provide a set of truth-conditions for every declarative sentence in some language. This is not to say that there is no such thing as a theory of context-free linguistic meaning; it is only to say that such a theory is not a theory of truth-conditions.

(From proceedings of Amsterdam Colloquium 19)

A compositional analysis of a range of readings of comparison constructions, as well as the positive form is proposed, which, unlike previous accounts, is compatible with multidimensional adjectives and has the power to explain differences between them and nouns. To this end, adjectives are represented as properties of dimensional quantifiers, namely of sets of gradable properties; e.g., healthy  λGQ. n-many(λF.F is a health dimension, λF.GQ(F)), where many denotes a cardinality function and n sets up a standard. Comparison morphemes either set the standard of many or of the dimensions. Consequences are discussed for our understanding of the adjective-noun distinction and for the analysis of gradable morphology.

Since Bott and Noveck (2004), there has been an ongoing discussion about whether scalar implicatures are delayed in online processing relative to literal meaning. Bott and Noveck (2004) provided Reaction Time evidence for such a delay, replicated in a number of later variations of their study (e.g., Bott et al., 2012). Breheny et al. (2006) found corresponding delays in self- paced reading. More recently, the issue has been investigated using the visual world paradigm, where results have been more mixed. Huang and Snedeker (2009, and subsequent work) have found delays for eye movements based on the ‘not all’ implicature of ‘some.’ But various others, (e.g., Grodner et al. (2010), Breheny et al. (2013), and Degen and Tanenhaus (2011)), report results which they argue show that implicatures are available immediately. Schwarz et al. (2015) added another angle to this picture, by using a sentence picture matching task using a Covered Picture (or Covered Box; henceforth CB; Huang et al., 2013), that allowed RT comparisons both within ac- ceptance (target) and rejection (CB) responses. While replicating the delay for implicature-based rejection responses, they find the reverse pattern for acceptance responses, with faster RTs for implicature-compatible conditions. They propose that delays associated with literal acceptances and implicature-based rejections result from a conflict between the two possible interpretations, rather than reflecting a cost of implicature-calculation. The present experiment extends this ap- proach beyond RTs by combining Visual World eye-tracking with the CB paradigm. The results a) are consistent with the notion that both literal and implicature interpretations are available in par- allel; b) show that literal acceptances are nonetheless only provided reluctantly, presumably due to a preference for implicature meanings, in line with Schwarz et al’s proposal; and c) suggest that for both literal acceptances and implicature-based rejections, there is a competition effect between the two interpretations. In addition, the RT data display an implicature-based block-priming effect, suggesting that the resolution of this conflict can be sped up through repeated task-exposure.

Donkey sentences have existential and universal readings, but they are not often perceived as ambiguous. We extend the pragmatic theory of nonmaximality in plural definites by Križ (2016) to explain how context disambiguates donkey sentences. We propose that the denotations of such sentences produce truth-value gaps — in certain scenarios the sentences are neither true nor false — and demonstrate that Križ’s pragmatic theory fills these gaps to generate the standard judgments of the literature. Building on Muskens’s (1996) Compositional Discourse Representation Theory and on ideas from supervaluation semantics, the semantic analysis defines a general schema for quantification that delivers the required truth-value gaps. Given the independently motivated pragmatic theory of Križ 2016, we argue that mixed readings of donkey sentences require neither plural information states, contra Brasoveanu 2008, 2010, nor error states, contra Champollion 2016, nor singular donkey pronouns with plural referents, contra Krifka 1996, Yoon 1996. We also show that the pragmatic account improves over alternatives like Kanazawa 1994 that attribute the readings of donkey sentences to the monotonicity properties of the embedding quantifier.

Arguably the biggest challenge in analyzing English tense is to account for the double access interpretation, which arises when a present tensed verb is embedded under a past attitude—e.g. "John said that Mary is pregnant". Present-under-past does not always result in a felicitous utterance, however—cf. "#John believed that Mary is pregnant". While such oddity has been noted, the contrast has never been explained. In fact, English grammars and manuals generally prohibit present-under-past. Work on double access, on the other hand, has either disregarded the oddity (e.g. Abusch 1997: 39) or treated it as a reflex of a particular dialect (e.g. Kratzer 1998: 14). The goal of the paper is to argue—based on a corpus study—that a present-under-past sentence is grammatical, but modulated by two, interacting pragmatic phenomena: cessation and parentheticality.

Abstract: In her book Fiction and Metaphysics [1999] Amie Thomasson, influenced by the work of Roman Ingarden, develops a phenomenological approach to fictional entities in order to explain how non-fictional entities can be referred to intrafictionally and transfictionally, for example in the context of literary interpretation. As our starting point we take Thomasson’s realist theory of literary fictional objects, according to which such objects actually exist, albeit as abstract and artifactual entities. Thomasson’s approach relies heavily on the notion of ontological dependence, but its precise semantics has not yet been developed. Moreover, the modal approach to the notion of ontological dependence underlying the Artifactual Theory has recently been contested by several scholars. The main aims of this paper are (i) to develop a semantic approach to the notion of ontological dependence in the context of the Artifactual Theory of fiction, and in so doing bridge a number of philosophical and logical gaps; (ii) to generalize Thomasson’s categorial theory of ontological dependence by reconstructing ontological categories of entities purely in terms of different structures of ontological dependence, rather than in terms of the basic kinds of entities the categorical entities depend on.

Multidimensionality in the grammar of gradability Galit Weidman Sassoon (20.7K words) In standard approaches to gradability, the interpretation of a gradable predicate is typically assumed to be generated by a single scalar dimension (Kennedy 1999). However, people seem to use multiple dimensions in a variety of cognitive tasks. This paper addresses the grammatical manifestations of this phenomenon. It presents data to the effect that dimension-sets and dimension-binding operations in the lexical representations of predicates affect the grammatical status, meaning composition, and truth conditions of utterances. The account is motivated by introspective data and a review of corpus patterns and experimental results. Multidimensional adjectives are quantificational; e,g,, to be healthy, one must be healthy in all respects (have no relevant disease). Given this meaning, questions arise as to what the interpretation of very healthy or healthier could be, and how it may be composed. The paper shows that degree-constructions with multidimensional adjectives involve counting of or quantification over dimensions. Data from the nominal domain further motivates this view. Based on psychological research, nouns are assumed to be associated with dimension-sets, like adjectives are. It is shown that when the semantic representations of nouns involve counting of or quantification over dimensions, their acceptability in degree-constructions improves. Moreover, the acceptability of nominal comparisons correlates robustly with that of constructions involving explicit quantification over dimensions. The paper presents a fully compositional account of multidimensional gradability using the standard semantic approach to gradability (Heim 2000; von Stechow 2009) plus a new respect-accessing operator.

The classic model of conversation based on the Common Ground (CG), introduced by Karttunen (1974), Lewis (1979) and Stalnaker (1978), was shown to be insufficient for accounting for various conversational phenomena (inter alia Portner 2004, 2007, Farkas & Bruce 2009, Murray 2014). This paper further strengthens this line by analyzing a type of non-truth conditional non-core dative termed the Discursive Dative (DD) as a discourse management device (Krifka 2008, Repp 2013). The DD signals that the asserted proposition p constitutes an exception to a normative generalization believed by the speaker to be shared by the speech event participants. In order to capture the notion of exception we propose to divide the CG into two sets of worlds, those consistent with previous assertions and their presuppositions (CGA) and those consistent with generalizations (CGG). The DD signals a non-inclusion relation between the asserted proposition and the CGG. This enables us to distinguish between different types of mirativity effects, by drawing a distinction between adding a proposition p that was not previously in the speaker’s expectation-set (inter alia DeLancey 1997, 2001, Rett 2009, Peterson 2013, Rett & Murray 2013) and the present case of the DD, where p can very well be in the speaker’s expectation-set, but objectively expected that ~p.

A popular view of the semantics of even takes it to presuppose that its prejacent, p, is less likely than all its contextually relevant focus alternatives, q. In this paper I point out
three novel problems for this `comparative likelihood' view, having to do with (a) cases where even p is felicitous though p cannot be considered less likely than q, (b) cases where
even p is infelicitous though p asymmetrically entails and is less likely than q, and (c) cases where even interacts with gradable predicates, indicating that merely requiring p to be higher on the scale than q is not enough to make even p felicitous. Instead, both p and q must also yield degrees which are at least as high as the standard of comparison.
In response to these problems I develop a revised scalar presupposition for even which resembles the semantics of comparative conditionals, and which requires that for a salient
x, retrieved from p, and a salient gradable property G, (i) x's degree on G is higher in all accessible p worlds than in all accessible q-and-not-p worlds and that (ii) in the latter
worlds this degree is at least as high as the standard on G. I show how this presupposition accounts for both traditional observations concerning even, as well as for the novel data and
propose that the common presence of `less likely' inferences with even can be indirectly derived from the common use of `distributional' standards of comparison with gradable
properties. A general contribution of the proposal, then, is in attempting to apply tools from research of gradability-based phenomena for a better understanding of scalarity-based
phenomena.

My thesis aims to jointly account for the non-bivalent truth value judgments, namely judgments which neither correspond to “True” nor to “False”, that are triggered by presuppositional sentences (like 1 in a context where Oscar is not French), vague sentences (like 2 in a context where Oscar is of an average age) or even hybrid sentences (like 3 in the same context).
1. Oscar has realized that you are French.
2. Oscar is old.
3. Oscar has realized that you are old.
Drawing on systems defining three logical values (true, false and other) and discussed both in the literature on vagueness and in the literature on presupposition, I develop a system with five totally ordered values, which thus defines three intermediate levels between true and false. After collecting experimental data conflicting with the predictions of this system, I propose a system with four partially ordered values, which is compatible with the experimental results and which defines vagueness and presupposition as entering in relation with truth along distinct specific dimensions.
In parallel, I conducted an experiment in collaboration with Paul Égré, which shows that speakers systematically reject contradictory descriptions of the form “old and young” but can accept contradictory descriptions of the form “neither old nor young”, “neither old nor not old” and “old and not old”. These results echo the idea that vague sentences like 1 can be judged “Neither true nor false” but also “True and false”, and allow us to discriminate between two competing theories of antonyms.

It is widely accepted that knowledge is factive, but two different understandings of "factivity" should be distinguished, namely, the implication version and the presupposition version. While the former only takes the truth of P as a necessary requirement for "S knows that P," the latter considers it also necessary for "S does not know that P." In this paper, I argue against presupposition and defend implication. More specifically, I argue against Wang and Tai's defense of the presupposition version as presented in a recent paper and propose a pragmatic response to the "persistence problem" of implication. In other words, my positive proposal is an account of implication plus pragmatic implicature. To conclude, I use my version to analyze Wang and Tai's distinction between inner skepticism and outer skepticism. My conclusion is that, after abandoning presupposition, we can identify two types of intermediate skepticism between Wang and Tai's inner and outer skepticism.

(From proceedings of Israel Association of Theoretical Linguistics (IATL) 26)

This paper uses Esperanto—a constructed language with transparent morphology but rich semantic-pragmatic components—to study antonymy and polarity. We investigate the distribution of the Esperanto antonymy morpheme ‘mal-’ (as in, for instance, ‘mal-alta’: antonym-tall, short) in a 4.3 million-word corpus, Tekstaro, and use it as an empirical basis to assess different theories of negative antonyms. Our methodology consists in in- vestigating the extent to which the antonymy morpheme ‘mal-’, which we take to denote negative polarity, bears the linguistic features predicted by traditional linguistic tests (such as incompatibility with measure and ratio phrases and low likelihood of nominalisation)

This paper provides a semantic-pragmatic answer to the question why some definite DPs are islands for wh-subextraction while others are not. While it was suggested as early as in Chomsky (1973) that the key to the problem are differences between determiners involved, there has been no analysis which would be based on independently attested properties of the determiners. This paper focuses on the contrast in wh-subextraction between DPs with two kinds of definite articles, the so called weak and strong ones, in Austro-Bavarian German, recorded in Brugger and Prinzhorn (1996). The analysis I offer makes use of the recent works showing that weak and strong definite articles can have different semantics. In particular, to account for the use and distribution of German strong articles, F. Schwarz (2009) assumes a semantics which routinely results in directly referential readings of the DPs headed by such articles. I show that, assuming a classic Hamblin/Karttunen semantics for questions, in cases of wh-subextraction out of directly referential DPs, it would have resulted in a trivial question which presupposes the asserted content of its possible answers. More broadly, this work aligns with a series of semantic-pragmatic analyses of constraints on island formation (Szabolcsi and Zwarts 1993, Fox and Hackl 2006, Oshima 2007, Abrusan 2008, Abrusan and Spector 2011, Schwarz and Shimoyama 2011, Mayr 2013).

In this paper I focus on a recently discussed phenomenon illustrated by sentences containing predicates of taste: the phenomenon of " perspectival plurality " , whereby sentences containing two or more predicates of taste have readings according to which each predicate pertains to a different perspective. This phenomenon has been shown to be problematic for (at least certain versions of) relativism. My main aim is to further the discussion by showing that the phenomenon extends to other perspectival expressions than predicates of taste and by proposing a general solution to the problem raised by it on behalf of the relativist. The core claim of the solution (" multiple indexing ") is that utterances of sentences containing perspectival expressions should be evaluated with respect to (possibly infinite) sequences of perspective parameters.

According to certain versions of predicativism, names denote metalinguistic predicates of a certain type, e.g. the name Perón as it occurs in the sentence Perón died in 1974 denotes a predicate more or less paraphrasable as "being called Perón" (Burge 1973, Matushansky 2008 and Fara 2011, among others). The metalinguistic theory of names is claimed to be superior in nontrivial ways to direct reference theories, according to which names contribute an individual without the mediation of descriptions (Kripke 1980). The alleged triumph of predicativism is that by assuming the "being called N" property as basic, both referential and non-referential uses of proper names can be given a uniform semantic analysis. By contrast, the referentialist needs to resort to homonymy or semantic ambiguity. In addition, there are some systematic connections between referential and predicative uses of names that at first glance also seem to favor predicativism. The goal of this paper is to present an alternative syntax, semantics and pragmatics of proper names. We assume that grammatical categories and the associated meanings that they are * For helpful comments and discussion on previous versions of this article, we thank Paolo Acquaviva, David Embick, Heidi Harley, Phoevos Panagiotidis, Stefano Predelli, one anonymous reviewer, and the people at The Philosophy of Language Group (at Sociedad Argentina de Análisis Filosófico), in particular, Ramiro Caso, Eduardo García-Ramírez and Eleonora Orlando. Our acknowledgment extends to the audience of the ICSO III workshop (Issues in Contemporary Semantics and Ontology workshop) held in Buenos Aires, 2016. Usual disclaimers apply. supposed to encode are not grammatical primitives, but epiphenomena that result from the particular way in which syntax combines functional material and lexical Roots. Before syntax, lexical Roots have no detectable meanings. On this account, there are no names before syntax, as there are no nouns, verbs or adjectives. Names are thus seen as the result of a particular syntactic configuration whose semantic realization is that of contributing an individual. Metalinguistic uses of names, and other derived uses, are involved in a different syntactic scheme, one that makes a name Root a predicate of a certain type. Besides their different syntactic basis, we argue that metalinguistic inferences in both referential and predicative uses of names have a pragmatic source. According to the theory we propose, then, the so-called Being Called Condition is neither a syntactic nor a semantic primitive. Under this conception of proper names the uniformity argument does not hold and the adduced linguistic evidence cannot lead to any (meta-)semantic consideration.

This paper is a reframing of the debate between Contextualism and Semantic Minimalism. By removing the focus from questions concerning what it takes to express truth-conditional content, I intend to show that Contextualism and Minimalism both account for important characteristics of natural language, but appear at odds due to the issue of truth-conditional content. The positive view that encapsulates this claim is called Radical Minimalism, or Radicalism, for short. It acknowledges the driving motivations behind each of the aforementioned theories, while ridding them of an unnecessary adherence to propositionalism. The concentration of the debate on whether or not context-independent linguistic content can express truth-conditions, or whether it needs optional-pragmatic processes, or hidden indexicals, or unarticulated constituents, has caused opposition and distance between theorists where it need not be.

  Radical Minimalism seeks out common ground and admits, along with both Contextualism and Semantic Minimalism, that in communication our sentences usually fall short of determining fully what we mean in using them; hence Contextualism is correct to point out that there is a gap between compositional, formal content, and communicated, speech act content. However, along with Semantic Minimalism, Radicalism, wants to account for the lexico-syntactically determined aspects of natural language (recursivity, systematicity, etc.). Essentially, Radicalism states that this minimal, formally constituted context-free, content can do all the work Minimalism wants it to do, without it being burdened with the claim that it also necessarily expresses truth-conditional content.
I will show that while each view does attempt to capture important linguistic phenomena, they fall short of a fully comprehensive picture of linguistic meaning and use.

Some special devices are designed to test the existence of the finiteness of a Chinese clause, though Chinese evidently fails to meet the general conditions on finiteness. These tests include the occurrence of aspectual markers and modals as AUX in finite but not in nonfinite clauses, the obligatory nullness of the subject of the nonfinite clause (Huang 1988(Huang [1982(Huang ],1 1984, the cooccurrence constraint, and the licensing of negative polarity items , as well as C.-C. criteria, which are not much discussed in the literature. In this paper we make a careful study of all these criteria specially devised for Chinese and find that none of them can make a valid distinction between finiteness and nonfiniteness. Since, as far as we know, there is no independent evidence to support such a distinction, we conclude that it in fact does not exist in Chinese.

Abstract of the presentation for the 4th meeting of the Réseau LACTO (Langage, Argumentation et Cognition dans les Traditions Orales)
Quatrième Rencontre du Réseau Lacto
CELHTO, bureau de l’Union Africaine, Niamey, Niger, du 22 au 25 septembre 2015
:

This very short paper, in spite of its vintage (1979), may still be of interest to anyone delving into the murky relationship between language and intention. What is presupposed by a speaker about the understanding of a listener goes to the very heart of how natural languages work, since when we get it wrong communication simply fails. For this reason the failure of presupposition is also the common bane of those working across languages, or even moving into another community where the same language seems to be spoken. When I began to look at linguistics seriously, a little before this paper was written, I was intrigued and dismayed to discover that the study of presupposition had apparently been hijacked by formalists who only seemed to find linguistic significance in symbolic regularities. There are indeed whole classes of expressions in English, related to certain verbs like “realize”, which are a delight to symbolic formalists who have written volumes on their tidy logic. Even at that early stage of research, I had dark suspicions that the ways in which real human beings used real language was nowhere near as tidy, and all the more interesting for that. The paper below was a first attempt to probe the fortress of symbolic formalism.

This paper investigates core semantic properties that distinguish between different types of gradable adjectives and the effect of context on their interpretation. We contend that all gradable adjectives are interpreted relative to a comparison class (van Rooij to appear), and that it is the nature of the comparison class that constitutes the main semantic difference between the different subclasses of such adjectives: some select a class comprised of counterparts of the individual of which the adjective is predicated, while others - an extensional-category of this individual. The role of the context is to determine the elements that comprise the comparison class. It is proposed, following Kennedy (2007), that the standard of membership is selected according to a principle of economy whereby an interpretation relative to a maximum or a minimum endpoint within a comparison class takes precedence over one relative to an arbitrary point. This proposal captures so-called “standard shift” effects, that is, the influence of context on the interpretation of gradable adjectives from all subclasses, in their positive form and when modified by degree adverbials. Additionally, this proposal captures cases of apparent lack of context sensitivity (e.g. intuitive inference patterns, unacceptability of for-phrases, etc.) Finally, we show that the type of comparison class is aligned with the well known distinction between stage-level and individual-level predicates.

Esta tese discute essencialmente o projeto lingüístico do lógico norte-americano Richard Montague (1930-1970), suas bases filosóficas e o formalismo lógico-matemático utilizado para implementá-lo. O objetivo de Montague é a construção de uma Semântica de Valor de Verdade para as línguas naturais através de uma semântica de modelo teórico e de mundos possíveis fazendo uso de uma linguagem lógica intensional.

(From the proceeding of Sinn und Bedeutung 13)

This paper presents a novel semantic analysis of unit names and gradable adjectives, inspired by measurement theory (Krantz et al 1971). Based on measurement theory's typology of measures, I claim that different predicates are associated with different types of measures whose special characteristics, together with features of the relations denoted by unit names, explain the puzzling limited distribution of measure phrases.

(From The proceedings of IATL 18)

In model theoretic semantics, we represent the core of predicate-sense by intension. Another notion, clusters of characteristic properties, serving as conceptual guidelines that help us identify the denotation in each context of use, has intrigued scholars from a variety of disciplines for years. Model theoretic theories often appeal to clusters of properties / features /criteria / propositions etc. However, there is no systematic account of the role of clusters. Stipulations are made in each case separately regarding the presence of clusters and/or their various effects. I propose a detailed formal model, which incorporates two kinds of clusters into our denotational representation of Predicate meaning. For example, the interpretation of the predicate chair includes in each context a set of individuals (chairs), necessary properties (like piece of furniture or solid) and stereotypical characteristics of chairs (like: has a back, four legs, is used to sit on it etc.) In this paper, I illustrate the use of this model in semantic analysis with one case study: contextual restrictions in universal generalizations with every, any and generic a.

(From proceedings of Israel Association of Theoretical Linguistics 21)

Parts 1-3 present and criticize Partee and Kamp’s 1995 well known analysis of the typicality effects. The main virtue of this analysis is in the use of supermodels, rather than fuzzy models, in order to represent vagueness in predicate meaning. The main problem is that typicality of an item in a predicate is represented by a value assigned by a measure function, indicating the proportion of supervaluations in which the item falls under the predicate. A number of issues cannot be correctly represented by the measure function, including the typicality effects in sharp predicates; the conjunction fallacy; the context dependency of the typicality effects etc. In Parts 4-5, it is argued that these classical problems are solved if the typicality ordering is taken to be the order in which entities are learnt to be denotation members (or non-members) through contexts and their extensions. A modified formal model is presented, which clarifies the connections between the typicality effects, predicate meaning, and its acquisition.

According to Ogihara (1995), the usage of the embedded present in a speech report such as John said that Mary is in the room is restricted by the cause of John's belief (the state that made John think that Mary is in the room): the present tense can be used only if this cause still holds at the time that John said that Mary is in the room is uttered. This paper presents experimental evidence demonstrating that this is only one of the factors that licenses a felicitous usage of the embedded present tense. In particular, we show that the cause of belief still holding is not a necessary condition, and identify two additional, sufficient (but not necessary) factors: in cases of false belief, who is aware of the falsity of the belief and duration of the reported state. While these factors are independent, they collectively support the idea that the present tense encodes 'current relevance', even in embedded contexts (e.g. Costa 1972; McGilvray 1974). This gives rise to the question of how we can derive 'current relevance' and, in particular, whether previous analyses of the embedded present tense are adequately equipped to do so.

It's widely accepted that knowledge is factive, but two different understandings of "factivity" should be distinguished, namely, the implication account and the presupposition account. While the former takes the truth of P as a necessary requirement for "S knows that P", the latter takes it further as necessary for "S does not know that P". In this paper, I argue against the presupposition account and defend the implication account. More specifically, I argue against Wang and Tai's defense of the presupposition account in [Wan10], and propose a pragmatic account to respond persistence problem of the implication account. In other words, my positive proposal is an account of implication plus pragmatic implicature.

This paper presents corpus-based evidence for a typology of multidimensional adjectives, like for example, healthy and sick. The interpretation of the latter is sensitive to multiple dimensions, such as blood pressure, pulse, sugar, cancer, etc. The study investigated the frequency of exception phrases, which operate on an implicit universal quantifier over adjectival dimensions, as in healthy, except for a slight cold, and not sick, except for high cholesterol. On the emerging typology, adjectives classify by the way their dimensions are glued together to create a single, uniform interpretation. The default interpretation of adjectives such as healthy involves implicit universal quantification over dimensions (dimension conjunction), while that of adjectives such as sick involves existential quantification (dimension disjunction). In adjectives like intelligent, the force of quantification over dimensions is context relative. Moreover, the paper presents support to the hypotheses that antonym polarity and modifier distribution guide our choice of quantifiers over dimensions in different adjectives. Thus, this research sheds new light on the nature of negative antonymy in multidimensional adjectives, and on the distribution of degree modifiers and exception phrases among multidimensional antonyms. Finally, it raises new questions pertaining to multidimensional comparisons.

This paper argues that combinations of gradable adjectives with for phrases are more interesting than they are normally considered to be. Important ingredients of the semantics of for phrases were largely neglected so far. These ingredients point against the popular analyses of for phrases. This paper defends an analysis of for phrases as modifiers of degree functions. The effect of this modification is shown to be significantly more fundamental than merely restricting function domains (cf. Kennedy 2007). Moreover, the new data and analysis have consequences to a variety of previous analyses that use a null morpheme, pos, to derive the interpretation of for phrases and the positive form, and/or to achieve a variety of other purposes. Based on new facts, this paper shows that the data does not support such a null morpheme; rather, postulating it creates a line of complications that need to be taken into account. The question, then, is – do we need it at all? We conclude that the answer should eventually be determined as part of a more general controversy regarding the level of representation usually called the logical form. This paper contributes to this controversy by providing two possible accounts of the facts, a degree-based analysis, albeit without a null-morpheme, and an alternative analysis without degrees at all. While these analyses are simpler, they appear to be at least as plausible.

This paper focuses on the meaning of degree modifiers such as slightly and completely, when they are either more prosodically prominent than the scalar adjective they modify or less so. Thus, one challenge is to explain the meaning, function and distribution of these modifiers. A second challenge is to explain the way accentuation (prosodic prominence vs. non- prominence) affects their meanings. The paper argues that the sensitivity of weak modifiers such as slightly to the type of membership norm of the modified adjective poses a challenge to semantic analyses of these modifiers in terms of quantification, scale-structure or norm- shifting (section 1.1), and suggests, instead, that these modifiers trigger granularity shifting (section 1.2). Two analyses of the role of accentuation in modifiers are then discussed (sections 1.3-1.4). Lastly, the paper presents an experiment that appears to support the granularity shifting account and a compatible treatment of prosodic prominence as generating local intensification of the meaning of the accented word (sections 2-3)

This paper describes a novel pedagogical software program that can be seen as an online companion to one of the standard textbooks of formal natural language semantics, Heim and Kratzer (1998). The Penn Lambda Calculator is a multifunctional application designed for use in standard graduate and undergraduate introductions to formal semantics: Teachers can use the application to demonstrate complex semantic derivations in the classroom and modify them interactively, and students can use it to work on problem sets provided by the teacher. The program supports demonstrations and exercises in two main areas: (1) performing beta reduction in the simply typed lambda calculus; (2) application of the bottom-up algorithm for computing the compositional semantics of natural language syntax trees. The program is able to represent the full range of phenomena covered in the Heim and Kratzer textbook by function application, predicate modification, and lambda abstraction. This includes phenomena such as intersective adjectives, relative clauses and quantifier raising. In the student use case, emphasis has been placed on providing "live" feedback for incorrect answers. Heuristics are used to detect the most frequent student errors and to return specific, interactive suggestions.

We discuss minimal sufficiency readings of exclusives like "just", as in "Just the thought of him sends shivers down my spine", which does not mean the same thing as "Only the thought of him sends shivers down my spine". We provide a set of diagnostics for identifying minimal sufficiency readings in conditionals and in simple clauses, and identify a generalization as to where the latter type appear: only in arguments that have a 'causer' thematic role. For this reason, we see minimal sufficiency readings in conditionals as basic, and provide an analysis of them building on Kratzer's notion of a modal base.

I examine  two Hebrew particles which intuitively correspond to the English particle 'while', namely ' be-'  (literally 'in', as in (1)) and 'beodo' (literally 'while-he', as in (2)). In both cases  the non-tensed adjunct clauses get their tense marking from the matrix:

(1) be-[xacoto et ha-kviS]adjunct [pag’a bo mexonit]matrix
in-cross-he acc. the road hit him car
(2)  beod[o xoce et ha-kviS]adjunct [pag’a bo mexonit]matrix
while-he  cross acc. the road hit him car
Both: “While he was crossing the road a car hit him

Yitzhaki 2003 proposes an equivalent semantics for be- and beodo where in both the temporal location of the matrix event (e.g. crossing  the road) interrupts / is located within the interval where the adjunct event (e.g. being hit by a car) holds.
In this paper  I argue that despite the apparent similarity between the two constructions, exemplified in (1) and (2) their semantics is different. I start by pointing out three differences between be- and ¬beodo. In then  give a brief semantics for be- in terms of temporal coincidence, and claim that the semantics of beodo is composed of that of be- (temporal coincidence) plus the semantics / pragmatics of odo – an inflected form of the Hebrew word for still (od / adayin). I develop an analysis of odo / still which is based on both traditional claims about the assertion and presuppositions of sentences with 'still', as well as on some novel claims. Specifically I propose that (a) The reference time of sentences with still / odo must be salient / anaphoric (The 'reference time anaphoricity requirement'), and (b) that this 'anaphoricity requirement' is a conversationally triggered presupposition.

This paper discusses the semantic role of demonstrations in deictic uses of demonstrative descriptions. The main question is: are they essential or non-essential to the semantics of demonstratives? In some theories (such as Kaplan 1989a; Roberts 2002 and King 2001), demonstrations are treated as " essential " in the sense that they correspond to a particular logical element in their formalization. In Wolter (2006) " s approach, demonstrations are not essential by themselves, but they do determine, pragmatically, an element essential for the interpretation of demonstratives – the index they must be interpreted relative to, which Wolter calls a " non-default situation ". Here we compare these two views; we conclude that Wolter " s approach is on the right track. Our main argument is that demonstrations can be replaced by other elements (such as salience or additional descriptive content) depending on the context. Thus, demonstrations are not essential to demonstratives, but they perform a pragmatic function that is essential – the identification of a " non-default situation " .

This paper will examine the relationship between minimal semantics and the modularity theory of mind. The modularity of mind thesis claims that the mind is composed, at least in part, of discrete, relatively autonomous modules each with their own specialized bodies of information and rules of operation. That is, there are specific areas of the brain that deal specifically with certain types of information and do so in a relatively automatic or reflexive way. Linguistic competence is a potential candidate for such a modular explanation.
The reasons for this examination are twofold. Firstly, I wish to argue, alongside Emma Borg, that our linguistic abilities do seem to be explainable on a modular account of mind. Further, by showing aspects of our linguistic abilities to be underpinned by a specific modular process, it helps secure this content as an autonomous level of content (contra Contextualism) that is distinct from the general intelligence that is used when interpreting pragmatic information.
Secondly, however, I wish to depart from Borg’s analysis by arguing that such a language module does not exclusively deal only in truth-conditional entities. That is, minimal content, that is not necessarily truth-conditional, can do all the things that Minimalists want it to do, including being governed by a distinct linguistic module, without the need to maintain that it expresses truth-conditional or propositional contents.  Thus Radicalism is, in fact, better suited to compliment the modularity of mind thesis, than is Borg’s Minimalism.

Diese Studie zur Struktur, Bedeutung und kommunikativen Funktion von Frageausdrücken und anderen zum Vollzug von Fragehandlungen geeigneten Formulierungen des Deutschen wurde im August 1980 der Philosophischen Fakultät für Sprach-und Literaturwissenschaft II der Universität München als Dissertation vorgelegt und für den Druck nur geringfügig überarbeitet. Vor allem wurden zur Erhöhung der Lesbarkeit dem formalen zweiten Teil Erläuterungen hinzugefügt.