October 6, ENS 29 rue d'Ulm, salle du DEC
 Bernhard Nickel (Harvard)
"Two approaches to generics: induction vs. inquiry"
I contrast two approaches to the interpretation of generics such as `ravens are black:' majority-based views, on which they are about what is the case most of the time, and inquiry-based views, on which they are about a feature we focus on in inquiry. I argue that majority-based views face far more systematic counterexamples than has previously been supposed. They cannot account for generics about kinds with multiple characteristic properties, such as `elephants live in Africa and Asia' or `roses are red, yellow, white, or black.' I then go on to sketch an inquiry-based view.
Talk is available here
November 6, ENS 29 rue d'Ulm, salle de réunion du Pavillon Jardin, 17h00-19h00
Yael Greenberg (Bar Ilan University)
“Contextual Dependency and Vagueness Systematically Interacting in the Interpretation of Two Types of Generic Sentences”
This talk deals with the way two types of generic sentences - with indefinite singular and with bare plural subjects (IS and BP generics, respectively) - tolerate exceptions, and with the connection of the exceptions-tolerance property of such sentences to some well-known observations about felicity differences between them (e.g. Lawler's 1973 Madrigals are popular vs. #A madrigal is popular). I show that whereas both IS and BP generics tolerate exceptional and contextually irrelevant entities (individuals and situations) in a strikingly similar way, which indicates the existence of a basically equivalent tolerance mechanism, there is also a difference between them, unnoticed so far, which has to do with the degree to which the properties of the legitimate exceptions can be characterized in advance. Following claims in Greenberg (2003) I argue that both this newly observed difference, as well as the traditional felicity differences result from an underlying contrast in the type of 'non-accidentalness' expressed by the two types of generic sentences, and more formally, in the accessibility relations that their generic quantifier (Gen) is compatible with. To capture the new difference in tolerance of exceptions I develop an improved version of the exceptions-tolerance mechanism for generic sentences suggested in Kadmon & Landman (1993) namely a restriction on the set of individuals and situations quantified by Gen, which is defined as partially vague to two different degrees using supervaluationist methods. The different degrees of vagueness in this restriction are shown to be systematically dependent on the two types of accessibility relations that IS and BP generics are compatible with.
Talk is available here

November 18, ENS 29 rue d'Ulm, salle de réunion du Pavillon Jardin, 17h00-19h00
Carmen Dobrovie-Sorin (Paris VII LLF)
 “Bare Nouns in Brazilian Portuguese”  
Talk is available here

 December 20, ENS 29 rue d'Ulm, salle de réunion du Pavillon Jardin, 17h00-19h00
Carmen Dobrovie-Sorin, Claire Beyssade, Alda Mari, David Nicolas  (LLF & IJN)
 “Genericity: Open issues”
Handouts are available here and here 
March 11, ENS 29 rue d'Ulm, salle de réunion du Pavillon Jardin, 12h00-13h30
Ivona Kucerova (UCL)
 “Habitual Readings of Imperfective Verbs and Maximize Presupposition
Abstract is available here and handout is available here

Alda Mari & Fabienne Martin (IJN & Université due Stuttgart)
"Generic (in)definites and types of judgments"
We propose an analysis of the French plural indefinite/definite (des/les)
distinction in generic sentences. We argue that in existential and generic
readings, (i) des is attached to a Maximality and a Novelty constraint,
whereas (ii) les is attached to a Maximality and Non-Novelty constraints.
This allows us not only to capture the rule of use of des vs. un, but also
to explain why the indefinite is preferred for expressing analytic
judgements, whereas the definite can express both analytic and synthetic
judgements. Finally, we also account for their different tolerance to
exceptions on a semantic basis.
A Paper is available here

June 11, ENS 29 rue d'Ulm, salle de réunion du Pavillon Jardin, 10h30-12h30
Richard Kayne  (NYU)
The definite article in Romance Languages. A comparative study.
L'article défini dans les langues romanes. Etude comparative.
Handout available here

June 26, ENS 29 rue d'Ulm, salle du DEC (on your right, right after the info desk), 13h30-15h30 
Bibhuti Bhusan Mahapatra  (Mumbai)
"Multiple Copulas and Genericity in Odia"

Odia has a multi-copula system; it has four copulas, namely, aT/ø, ach, th(a:) and he. Binary distinctions such as ‘equative’ vs. ‘existential’ (cf. Laksmi Bai: 1986) or ‘equative’ vs. ‘pedicative’ (cf. Higginbotham: 1987, Heggie: 1988 and Rapoport: 1987) or individual level vs. stage level predicates (cf. Carlson 1977) are often adopted to distinguish copulas. Considering the fact that there are four of them in Odia, their distribution is not predictable by using simple binary distinctions. Across languages, copulas occur as auxiliaries of main verbs to carry aspect, tense and mood features. However, the aT copula never occurs as an auxiliary; it does not carry any aspect or tense or mood feature; it takes only the subject agreement features. Unlike aT copula, the ach and th(a:) copulas occur as auxiliaries. The ach copula goes only with the speech time which essentially marks the realis feature. The th(a:) copula carries the other tense and mood features which, in contrast with the speech time, can all be taken as irrealis features. Thus, the copulas ach and th(a:) jointly mark the [±realis] contrast. Like the aT copula, the th(a:) copula can drop the tense feature too. They, with their tenseless forms, behave as individual level copulas. However, they do not convey the same individual level meaning. Using the predicate distinctions of Krifka, Pelletier et al. (1995) it can be shown that aT goes with ‘lexical characterizing’ predicates which do not have ‘episodic’ counterparts, thus, it indicates an inherent property of the subject. In contrast, th(a:) goes with ‘habitual generic’ reading, which is an aspectual generalization over stages denoted by stage level (episodic) predicates. Since the copula ach goes with the speech time alone it just expresses a present episodic meaning which is available only with ‘episodic’ (stage level) predicates. The copula he (be/become/happen/come into being) is basically an unaccusative event functor. Of course, there are many more complicated grammatical facts related to ach and he which escape an explanation in terms of the ontological distinctions such as stage, object and kind either in the domain of entities or predicates. The facts are noted in Mahapatra (2002).

Abstract and references available here

September 17, ENS 29 rue d'Ulm,Pavillon Jardin 11h00
Ana Muller  (Sao Paulo)
The expression of genericity in Brazilian Portuguese
Handout is available here

November 12, ENS 29 rue d'Ulm, Salle C235; 14h00
Fabio del Prete  (Institut Jean Nicod)
L'interprétation générique des phrases au Présent et à l'Imparfait dans
l'italien (The generic interpretation of present and imperfect sentences in Italian)
Handout is available here

December 10, ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin
Alda Mari (Institut Jean Nicod)

There are two competing analyses of the epistemic modality in the present perfect in French: the syntactic view, which holds that the modal scopes  above the perfect, and the lexicalist view, which has provided new empirical evidence for claiming that the perfect scopes above the modal. We provide some new pieces of evidence to show that the major results of  these two positions can be reconciled if one considers the contribution of the past and the present component of the present perfect. We distinguish between two sources of modality and we propose an analysis that also explains why in Italian 'ha potuto' only has an implicative behavior.
Besides providing a formal compositional analysis, the paper offers some new results on the raising-control distinction in relation with different meanings of 'can' in FRench and Italian.
Handout is available here

January 20, ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin
David Nicolas (Institut Jean Nicod)

First- and higher-order logic contain singular quantifiers, like the existential and universal quantifiers (something, everything). But many natural languages have plurals and collective predicates, and as a result, plural sentences that cannot be reduced to ordinary singular sentences. How can their semantics be characterized? While most natural language semanticists are happy to use first- or higher-order logic together with sets or sums, various philosophers and logicians maintain that we should instead use logics enriched with plural quantifiers. I compare these approaches, presenting and discussing several arguments put forward by the partisans of plural logic.
Handout is available here

February 11, ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin
Marcelo Ferreira (University of San Paulo, Brazil)
In this talk, I explore the idea that the logical representation of certain
habitual sentences involve quantification/reference to plural events (see
Kratzer 2004). It has two parts:
In the fist part, I contrast the semantics and pragmatics of habitual sentences
with and without adverbs of quantification, as exemplified in (1):
(1) a. When Mary visits John, he always/usually cries.
b. When Mary visits John, he cries.
I claim that whereas a quantificational analysis is adequate for the ones with
Q-adverbs, with the adverbs being the counterpart of `every', `most' and so on,
the ones without them (which I call `Bare Habituals') should be analyzed as
involving plural definite descriptions of events.
In the second part, I discuss continuous and habitual readings of imperfective
sentences such as those in (2):
(2) a. Mary is dying her hair (right now).
b. Mary dyes her hair.
I argue that continuous and habitual readings share the same temporal and the
same modal ingredients. I assume the presence of an existential/indefinite
event determiner in both sentences and argue that the only difference between
the logical representations of (2a) and (2b) is the number (singular/plural)
of the event variables being quantified over. Continuous readings involve
quantification over singular events, whereas habitual readings involve
quantification over plural events. I also claim that cross-linguistic variation
within the domain of imperfectivity reduces to selectional number requirements
of an event determiner.
Slides available here

February 16, ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin
Marcelo Ferreira (University of San Paulo, Brazil)
In this talk, I provide evidence against the idea that
bare singulars in Brazilian Portuguese (BP) are morphologically
singular but semantically number neuter. I argue instead that they
are sometimes specified for number (singular) and sometimes
under-specified and that this has semantic consequences. I develop
an account according to which whether or not a bare noun can enter
a syntactic derivation under-specified for number depends on the
functional heads to which it is syntactically related. The
conclusion is that, as far as number is concerned,
morpho-syntactic specification should NOT be kept apart from
semantic interpretation.
Slides available here

March 18, ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin, 11h00
Francis Corblin
(Université Paris Sorbonne, IJN)

French quoi que ce soit , like English any, behaves, in some contexts as a
negative polarity item, and in others as a universal quantifier.
In this talk, I defend a unitary analysis of quoi que ce soit as a universal
quantifier set apart by two properties : it must take wide scope over any other
quantifier, and it has a context-proof maximal domain.
A compositional derivation of this semantics is provided by taking seriously the
fact that quoi que ce soit is not a lexical item, but has an internal structure,
and can be found in constructions with relative clauses (quoi que ce soit que tu
fasses, quoi que tu fasses) in contexts exhibiting a "concessive conditional"
reading. The talk assigns the origin of this concessive meaning to the
underlying disjonction, and support this claim by showing that only clauses
with an explicit disjonction like Que ce soit Pierre ou Marie qui vienne have
the same meaning.
This approach suggests an explanation for the "widening" effect (Kadmon &
Landman 1993), which is the central part of the recent analysis of quoi que ce
soit due to Vlachou (2007), by assigning it to two distinct features : 1)
concessive conditionals quantifies over all alternatives of their domain,
without exception; 2) in its absolute use (without a relative clause) there is
no restriction on the domain, hence its maximality.
The last part of the talk discusses the more intriguing feature of quoi que ce
soit, the fact that if analyzed as a universal quantifier, it must be analyzed
as a widest scope universal. The striking fact is that the extended
construction quoi que ce soit +relative, on the contrary, is open to any scope
hierarchy with quantifiers of its context.
This leads to postulate a correlation between maximality and widest scope, and
we give other arguments supporting this view, although no explanation for the
correlation is provided.
The talk tries to establish that maximality is the source of all the constraints
on the distribution of quoi que ce soit (unacceptable in episodic sentences in
increasing contexts, and in necessity modals).
Slides available here

Christian Retore (Université Bordeaux 1)
Sémantique et modélisation dans le cadre de la logique catégorique

Du point de vue de la logique, la théorie des catégories propose des modèles plus subtils que les modèles ensemblistes usuels,
ce qui a été parfois utilisée en linguistique et en sémantique formelle
(travaux de Petitot sur la prédication et l'espace, travaux récents de Pollard sur syntaxe et sémantique, travaux récents d'Asher sur la sémantique lexicale dans un cadre
Ces constructions peuvent s'adapter à des logiques plus raffinées comme la logique intuitionniste.
Elles peuvent aussi rendre compte non seulement des formules et de leur prouvabilité mais aussi des preuves elles mêmes (ou des lambda termes typés qui interviennent dans la sémantique de Montague).
Une des questions prises en charge par cette approche est celle de la quantification, aussi bien sur les individus que sur les propriétés, et aussi bien au niveau de l'interprétation des formules quantifiées que de celles des preuves de formules quantifiées.
Nous ferons un panorama de ces notions et proposerons un cadre logique pour la sémantique compositionnelle qui prenne mieux en compte la structure sémantique des expressions élémentaires, mots ou entrées lexicales.

Mikhail Kissine (Université Libre de Bruxelles)
Futurity and Necessity

Les marqueurs linguistiques des temps du futur, comme l’anglais will, se voient fréquemment analysés
comme des opérateurs de nécessité. Dans Kissine (2008), j’ai montré que de
telles analyses sont amenées à faire des prédictions erronées. En particulier,
si will est un opérateur de nécessité, soit (1) et (2) sont acceptables, soit (3) et (4) sont équivalents.

(1)  ? John will win, and it is possible that he will not win.
(2)  ? John will not win, and it is possible that he will win.
(3)  John will win.
(4)  It is possible that John will win.

Récemment, Portner (2009: 239-241) a objecté à cette analyse que
l’inacceptabilité de (1) et de (2) s’explique en termes dynamiques, qui évitent
l’équivalence entre (3) et (4). Dans cet exposé, je me propose de montrer que
mes arguments résistent à l’objection de Portner.

Anamaria Falaus (Unversité de Nantes)
9 juin 2010 - salle du  DEC
Alternatives as sources of semantic dependency: the case of epistemic indefinites

This talk focuses on the properties of the Romanian dependent indefinite vreun,
whose constrained distribution is not easily amenable to known patterns of
semantic dependency. Examining a wide range of contexts of occurrence (in
particular modal and attitude verbs), I argue that the distribution of vreun is
sensitive to epistemic alternatives: vreun can only occur below operators which
cannot be used in situations where their complement proposition is established
to be true. Next, I seek to situate this distributional pattern with respect to
arguably similar items described in the literature, in particular existential
free-choice items (Kratzer & Shimoyama 2002), and the Romance determiners
quelque (studied in detail in work by Jayez & Tovena) and algún (Alonso-Ovalle
& Menéndez-Benito). The interest of vreun with respect to this emerging
typology is twofold: (i) it provides clear empirical support for the existence
of a class of dependent items sensitive to epistemic modality only (as opposed
to modality in general) and (ii) unlike quelque or algún, it also exhibits a
negative polarity behavior, a property which shows that the cross-linguistic
split we find between free-choice items that double as negative polarity items
and free-choice items that don’t (any vs qualsiasi, cualquiera) is also present
in the realm of existential items. In order to capture the many existing
connections between the (still emerging) patterns of ‘epistemic’ indefinites
and to derive the ‘epistemic constraint’ governing the distribution of vreun, I
defend a unified alternative-
based view of polarity sensitivity, mainly building
on Chierchia (2006 et seq). This account seeks to reduce empirical variation to
a small set of semantic primitives, by pursuing the hypothesis that the
different patterns of sensitivity can be derived from the interaction between
the alternatives these items bring about and the inferences they give rise to.

James Hampton (City University London)
9 juin 2010 - salle du  DEC
How modifiers affect the judged likelihood of generic

Connolly et al. (Cognition, 2007) reported an effect that
generic sentences (e.g. ravens are black) are judged less likely
when the subject noun is modified (e.g. jungle ravens are black). A
series of experiments investigated this phenomenon. 1) When people
justify the judgment, three different types of explanation are
offered based on pragmatics, knowledge and ignorance. 2) It is shown
that the effect holds just as strongly when universal quantifiers
are added to the sentence, thus yielding a fallacy. 3) We
investigated whether the effect would be diminished for more central
properties (ravens have wings) compared with surface properties
(ravens are black), and whether it would disappear all together for
categorical sentences (ravens are birds).
slides available here

Luciana Storto (Université de San Paulo)
January 24th -
ENS 29 rue d'Ulm, Salle RdC, Pavillon Jardin

The absence of determiners, quantifiers, classifiers and plurality in Karitiana and the clausal nature of demonstrative and universally quantified phrases

    Muller, Storto & Coutinho-Silva (2006) have shown that noun phrases in Karitiana (Arikém family, Tupi stock) are not specified as singular, plural, definite or indefinite (Muller 2010), and that the language lacks determiners, quantifiers, number and classifiers altogether. Nouns can be massive or unspecified for number in that only the latter may be counted without making reference to a unit of measure or containement. Numerals are construed with bare nouns through the use of postpositions (Muller, Storto & Coutinho-Silva 2008). If there is a notion of plurality in the language, it applies to events, rather than nouns (Sanchez & Muller 2007, Storto to appear-a). Demonstrative and universally quantified phrases involve non-finite clauses headed by a copula (Muller, Storto & Coutinho-Silva 2006).
    This presentation discusses possible meanings and structural analyses to account for what is known at the present moment about the properties of demonstrative and universally quantified clauses in Karitiana. Since such clauses are used in the language in place of what in other languages are phrases headed by determiners and quantifiers, the investigation of such phenomena may have important consequences for theories of linguistic typology and universals.
    Quantification in the language has been shown to be mostly adverbial in that quantifiers often distribute as sentence adjuncts (Muller, Storto & Coutinho Silva 20006, Sanchez 2008), adjoining to any maximal projection (Storto 1999). Universal quantification, however, is obtained through the use of a special kind of subordinate clause headed by the subordinator tyym which takes as a complement a verb phrase headed by a copula (Muller, Storto & Coutinho-Silva 2006, Coutinho-Silva 2008). Such clauses have been analyzed by Coutinho-Silva (2008) as maximizing head-internal relative clauses, displaying movement of the head NP to a left-periphery position inside the clause followed by movement to a position outside the clause (as in Grosu & Landman 1998). We consider the alternative possibility that they are adverbial subordinate clauses (Storto to appear-b), for two reasons: (1) relative clauses are never introduced by subordinating heads in Karitiana, whereas universally quantified clauses must invariably occur with the subordinator tyym, also used in adverbial embedded clauses and translated roughly as ‘when’, (2) the universally quantified clause may occur as an adjunct embedded clause whose subject is anaphoric with the subject or object of the matrix.
Demonstrative clauses are formed by a deictic element, a noun phrase and a copula verb. Coutinho-Silva (2008) analyzes them as head-internal relative clauses in which the deictic element is an adjunct. Although we take that analysis to be basically correct, we point out that these deictic heads do not seem to be lexical – as opposed to functional - in nature, because some of them also occur in the language as heads of auxiliaries used to express imperfective aspect.

Octobre 2008 - Webmaster : Sophie Bilardello
Mises à jours: Alda Mari