Friday, June 1: Yenan Sun

Please join us this Friday as Yenan Sun from the Linguistics Department presents work on expressions of similarity and identity.

Date and time: Friday, June 1, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Two strategies of sameness: Chinese tong and xiangtong

This talk examines two ‘same’s in Chinese : the determiner-like tong (preceding Num-Cl) and the adjective xiangtong (taking the modification marker de). I propose they represent two ways of expressing sameness: identity vs. maximal similarity, which correlate with a structural difference — comparatives (DegP) vs. relational adjectives. I show why the identity/maximal similarity distinction cannot be reduced to the token/type-identity distinction (contra Liao&Wang 2014); and further show that the analysis captures two novel generalizations: [1] tong can (only) co-occur with numeral ‘one’ to license an internal reading while xiangtong cannot; [2] xiangtong can co-occur with plural numerals (e.g. two, three) to have a reciprocal reading while tong cannot.

This analysis brings together several lines of research on ‘same’: (i) how ‘same’ is modelled in semantics (the default option is coreference relation ‘λx.λy. x=y’ while Alrenga 2007 proposes the notion of ‘similarity scale’); (ii) whether different uses of ‘same’ (deictic use, internal use) can be unified under a compositional account (Barker 2007, Charnavel 2015, a.o.); and (iii) the connection between ‘same’ and comparatives (Heim 1985, Beck 2000, Alrenga 2007, Matushansky 2010, Oxford 2010, Hanink 2017, a.o.).

Friday, May 25: Peter Klecha

Please join us this Friday as Peet Klecha (Linguistics, Swarthmore) presents work on conditionals and speech acts.

Date and time: Friday, May 25, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Going performative again (with conditionals)


I’m developing a formal framework for modeling speech acts, including explicit performatives, which involves modeling things like assertions as special cases of explicit performatives; in other words, a revival of the Performative Hypothesis (Ross 1970, Lakoff 1968, McCawley 1968, Sadock 1968).

I’ll try this model out on an explicit formal analysis of conditionals (including normal conditionals like (1) and biscuit conditionals like (2)) which capitalizes on the intuition (DeRose & Grandy 1999) that in a biscuit conditional, what is conditionalized is the speech act itself.

1) If you need me, I’ll come find you.

2) If you need me, my name is Al.

Unlike prior approaches along these lines, my model makes clear precisely a conditional speech act is. I will also make and defend two far-from-ordinary claims:  i) speech act operators are present in the syntax, and ii) assertions are a species of exhortative or imperative. The analysis of both kinds of conditional then fall out from the standard restrictor approach to conditionals (Kratzer 1986).

Friday, May 18: Elisabeth Camp

Please join us this Friday as Elisabeth Camp (Philosophy, Rutgers) presents work on slurs and the divide between truth-conditional and expressivist semantics.

Date and time: Friday, May 18, 10:30 a.m. – 12:20 p.m.

Location: Stuart 209

Title: Expression, truth, and compositionality: slurs in discourse structure


Most theories of linguistic meaning are constructed with truth-conditional resources.  Some expressivists, like Gibbard, attempt to reformulate the entire truth-conditional structure to smoothly integrate the expression of non-truth-conditional contents or attitudes; others, like Potts, posit expressive meaning as a strictly segregated addendum to the truth-conditional machinery.  I argue that slurs demonstrate the need for an intermediate position: while compositional machinery tends to be focused on truth-conditional contents, the partition is not absolute, and incorporation within compositional meaning depends upon pragmatic as well as semantic factors.

Friday, May 4: Patrick Muñoz

Please join us this Friday as Patrick Muñoz presents work on metasemantics.

Date and time: Friday, May 4, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Experiential predicates: metasemantics and metaphysics


Previous research on experiential predicates (like frightening and tasty) has assumed that if such predicates are taken to have a ‘bare’ semantics, which does not somehow relativize their extensions to experiencers, then because there are in fact no monadic experiential properties (such as being frightening, or tasty, simpliciter), either discourse making use of such predicates must be radically defective, requiring a blindness or error theory (Stojanovic 2007; Hirvonen 2016), or the intensional parameters relative to which extensions are evaluated must be reinterpreted as something more metaphysically neutral than possible worlds that track ‘the way things are’ in some non-deflationary sense (Kölbel 2002; Coppock 2018). I demonstrate that a bare semantics for such predicates, taking English deverbal psych adjectives as exemplary, makes no such commitments, and argue that we instead should make sense of differences in truth judgments regarding such predicates as rooted in metasemantic variation among the speaker population as to which properties such predicates denote. This is normal, since in fact almost no predicates have consistent inter-speaker criteria of application; what differentiates experiential predicates is that the criteria for their application is constrained by the experiential reactions of speakers, and this fact can in turn be traced to their experiential semantics.

Friday, April 20: Manuel Križ

Please join us this Friday as Manuel Križ (École Normale Supérieure) presents work on bare plurals.

Date and time: Friday, April 20, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Bare plurals, multiplicity, and homogeneity


Existential bare plurals appear to have a plural meaning in affirmative contexts, but a singular meaning under negation.
  a. Mary saw zebras. ~> More than one.
  b. Mary didn’t see zebras. ~> Not even one.
I will present a novel theory of this phenomenon. Deviating from the conventional wisdom that competition with the singular indefinite is somehow involved, I argue that the facts can be conceived of in terms of the general trivalence of plural predication (“homogeneity effects”). In particular, both (1a) and (1b) have the third truth value when Mary saw exactly one zebra, just like (2a) and (2b) do when Mary read only some of the books.
  a. Mary read the books. ~> All of them.
  b. Mary didn’t read the books. ~> None of them.
This approach elegantly accounts for the behaviour of existential plurals using only independently motivated conceptual resources, and overcomes problems that have been raised for the competition-base accounts.

Friday, March 30: Mingya Liu

Please join us this Friday as Mingya Liu (Cognitive Science, Osnabrück University) presents work on negative polarity items.

Date and time: Friday, March 30, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Negative Polarity Items – semantics, pragmatics, and processing


In this talk, I will present my works on negative polarity items (NPIs) with three foci. First, I will review the literature and present my viewpoint on the long-standing puzzle as to why “only”, which is neither downward entailing nor non-veridical, licenses NPIs. With this case study, I will argue for a bidirectional examination of the licensor-licensee relation and for a hybrid semantic/pragmatic account of NPI licensing. Second, I will present two behavioral studies on NPIs and discuss their implications for the existing theories. Third, (if time permits), I will present an EEG study in German showing processing differences between NPIs and positive polarity items.

Friday, February 16: Carlos Cisneros

Please join us this Friday as Carlos Cisneros from the Linguistics Department presents more work on indiscrimnacy.

Date and time: Friday, February 16, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Dissecting indiscriminacy (and free chioce)


Indiscriminatives (depreciatives in Haspelmath 1997) are a crosslinguistic class of indefinite equivalent in meaning to English ‘just any’.  Haspelmath observed their strong relationship to free choice items crosslinguistically, calling them ‘semantic enrichments [of free choice meaning] by implicature’.  They are often derived from free choice items by means of special intonation and/or minor morpho-syntactic modification.  They are characteristically defined by their interaction with negation, which results in the denial that the indiscriminative’s restrictor serves as a sufficient condition for satisfaction of some predicate by an individual.

(1) It’s not that just any horror film should get nominated for an award.

In (1), the speaker denies that the predicate should get nominated for an award is satisfied by merely having the condition of being a horror film.  In this denial, the speaker also asserts that an individual must have some other quality besides being a horror film, thereby representing some proper subset of horror films yet to be suggested.  To my knowledge, there is no attempt in the literature to develop a semantic/pragmatic account of indiscriminative meaning.

The difficulty of approaching an semantic treatment of indiscriminative meaning is tied to the problem of free choice item semantics.  Outside of negation, both classes of items display the properties of anti-episodicity (aversion to episodic environments) and quantificational variability (a modal form of universal quantification which favors distributive over collective readings) (Giannakidou 2001).  There is a large literature on the semantics of free choice items, especially any, with much debate about their most proper, general treatment (Dayal 1998; Menendez-Benito 2010; Chierchia 2013; among many others).  But most of these approaches have neglected early insights that relate the meaning of free choice items to a greater systematic phenomenon in language.  Fauconnier (1975) noted the meaning similarity between any and the quantifying superlatives of English.  Konig (1991) later noted that quantifying superlatives are anti-episodic, in the same vein as free choice items.  Although it has not yet been discussed, quantifying superlatives also display quantificational variability and a strong resemblance to indiscrimatives while under negation.  Therefore, the meaning of free choice items, their link to indiscriminatives, and a larger set of phenomena including quantifying superlatives should result from a common set of semantic principles.

In this talk, I will draw on further data from my studies on minimal sufficiency just (Coppock & Beaver 2014) to argue that free choice items and quantifying superlatives both involve minimal sufficiency scales.  These are scalar inferences similar to downward entailment in which propositional strength is ranked according to the sufficiency of alternative individuals in their satisfaction of the same predicate.  Free choice items and quantifying superlatives both assert the truth of the strongest proposition, entailing the truth of all weaker propositions.  Free choice items additionally involve a mechanism for making subdomain alternatives accessible to minimal sufficiency scales.  The two primary options for this mechanism are reanalysis of NPIs as presupposing minimal sufficiency scales, or derivation of free choice items by means of ‘non-specific free relative clauses’ (Haspelmath 1997, 1995).  I will demonstrate how the two mechanisms work with minimal sufficiency scales to derive the meanings and distributions of both free choice any and Cuevas Mixtec free choice items.  The major payoff of this approach is how simply a semantics of indiscriminatives follows from it.   Indiscriminative meaning is free choice meaning as it is made available to negation (by focus), thereby denying the truth of the strongest proposition on a minimal sufficiency scale.

Friday, February 2: Yimei Xiang [pt. II]

Please join us this Friday as Yimei Xiang (Linguistics, Harvard) presents work on exhaustivity in attitudes.

Date and time: Friday, February 2, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Complete and true: attitudes held of questions


Attitudes (e.g., knowledge, memory, emotion, and so on) held of questions must be complete and true. For example, the sentence “Jenny knows who left” has two conditions: a Completeness condition, that Jenny knows an answer that completely addresses the question “who left”, and a false answer (FA-)sensitivity condition, that Jenny has no false belief relevant to “who left”. Completeness is standardly equivocated to exhaustiveness. For example, a complete answer of “who left” should exhaustively specify all the individuals who left. Adopting this assumption, recent works on question embedding treat the FA-sensitivity condition as a scalar implicature of Completeness and derive it via exhaustification (Klinedinst & Rothschild 2011, Uegaki 2015, Cremers 2016, Theiler et al 2016).
However, investigating into mention-some readings of questions with existential modals (e.g., “Who can chair the committee?”), I argue for a non-exhaustive definition of Completeness that unifies mention-some and mention-all readings of questions (Fox 2013). Further, drawing on observations validated experimentally, I argue against the exhaustification-based account of FA-sensitivity: the FA-sensitivity condition is much stronger than what it can be defined as in any exhaustification-based account. It is concerned with all the relevant false answers, including those that can never be complete. This generalization also suggests a non-trivial prediction against the reducibility view: to recover all the relevant false answers, questions under attitudes must be able to supply partitions; hence, attitudes held of a question cannot be reduced to attitudes held of one answer of this question.

Friday, January 26: Yimei Xiang [pt. I]

Please join us this Friday as Yimei Xiang (Linguistics, Harvard) presents work on questions.

Date and time: Friday, January 26, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: A hybrid categorial approach to question composition


Questions have been defined as lambda abstracts (as in categorial approaches), sets of propositions (as in Hamblin-Karttunen Semantics), or partitions of possible worlds (as in Partition Semantics). Recently, categorial approaches are less commonly used due to their technical difficulties: they cannot capture the existential semantics of wh-words, suffer type-mismatch in composing multi-wh questions, and cannot account for question coordinations. However, a few facts suggest that question denotations should be able to supply not only propositional meanings but also predicative and nominal meanings. For example, in the sentence Susan mostly knows who formed the committee, the adverb mostly quantifies over the set of committee members, which has to be recovered based on the nominal short answers of the embedded question. Moreover, wh-words have a strictly more limited distribution in free relatives (FRs) than in questions, which suggests that wh-FRs are  formed out of wh-questions: in languages with wh-FRs, any wh-word that can be used in FRs can also be used in questions, but not the other direction (Caponigro 2003). These facts leave lambda abstract the only possible denotations of questions.

To revive categorial approach, this talk proposes a novel hybrid categorial approach to compose questions. This approach carries forwards the merits of traditional categorial approaches in tackling short answers and predicative wh-constructions, and moreover overcomes their technical problems.

Friday, January 19: Julian Grove

Please join us this Friday as Julian Grove from the Linguistics Department presents more work on presuppositions.

Date and time: Friday, January 19, 11:00 a.m. – 12:50 p.m.

Location: Stuart 209 (Philosophy seminar room)

Title: Presuppositions as scope-takers


According to satisfaction-based accounts of presupposition, an expression’s presuppositions project out of a context surrounding the expression only if that context doesn’t satisfy (i.e., entail) the presuppositions. For example, the existence presupposition triggered by ‘his children’ in (1)

(1) If John has children, then his children are bald.

doesn’t project because the context provided by (1) itself is taken to include the information satisfying this presupposition contained in the antecedent.

One apparent difficulty for a standard satisfaction framework, like the one of Heim (1983; 1992), is that presuppositions are always assessed against their local contexts, which are determined through semantic compositional rules by the immediate syntactic contexts of their triggers. For example, apparently neither (2) nor (3) has presuppositions.

(2) Mary believes John read the book, and she believes he regrets reading it.

(3) John read the book, and Mary believes he regrets reading it.

But Heim (1992) predicts, based on the semantics alone, that the second conjuncts of (2) and (3) presuppose that Mary believes John read the book, not that he in fact read it; hence, while (2) should be presupposition-free, the same status for (3) requires extra explanation. Intuitively, we would like to say of the presuppositions in (3) that they are assessed, not in the local context provided by Mary’s beliefs, but against the common ground. It is as though they have QR’d:

[John read the book, and [“I’m the presupposition triggered by ‘regret’! Satisfy me here!”]_i [Mary believes he t_i regrets reading it]]

In this talk, I argue that the scope-taking behavior of presuppositions falls out naturally from a treatment of them in terms of graded monads. This treatment follows the work of Shan (2001; 2005) and Charlow (2014), who use (non-graded) monads in analyzing various other linguistic phenomena that appear to exhibit scopal side-effects. As a result, presuppositions can be handled through run-of-the-mill compositional principles (like functional application) enriched with a small set of operations for handling presupposition. I present analyses, within the monadic framework, of presupposition filtering in discourse-update and conditionals. Finally, I argue that allowing presuppositions to take scope allows for a natural response to Geurts’s (1996) “proviso problem”.