Friday, November 13: Orest Xherija

November 10th, 2015

Please join us this Friday as Orest Xherija from the Linguistics Department presents work on the semantics of temporal connectives.

Title: Before without after: nonveridicality, disjunction and context dependence

Date and time: Friday, November 13, 10:30 a.m. – 12:20 p.m.

Location: Rosenwald 208 (Linguistics seminar room)


In this talk I will sketch a new, crosslinguistic semantics for the temporal connective before based on empirical observations from a number of languages. The semantics of before has been a longstanding problem in formal semantics and pragmatics due to the numerous properties that this connective seems to possess: (a) it allows for veridical, nonveridical and antiveridical readings of the before-headed clause (which reading we have depends (mostly) on context); (b) it licenses NPIs (in contrast to the alleged dual after); (c) it can license strong NPIs (Giannakidou 1998, for relevant examples from Modern Greek); (d) it seems to place an anti-PAST restriction on the verb of the before-headed clause (a fact that is observed in a number of diverse languages including Modern Greek, Albanian, Turkish and Japanese); (e) it appears to force a veridical reading when preceded by a measure phrase like three minutes. After showing that previous accounts (Ogihara 1995, Sánchez-Valencia et al. 1994, Condoravdi 2010, Krifka 2010 inter multa alia) have missed some of these empirical observations in attempting to formally characterize the semantics of before, I will sketch an account in which before is a function from contexts to meanings. This denotation aims to capture the numerous crosslinguistically robust properties of this connective with a uniform account that is language-independent. In sketching this account, I will provide evidence for the need of at least two lexical entries for before, to account for cases in which before is followed by a TP and cases in which it is followed by a DP, a fact that has not been addressed in any of the previous accounts.

Friday, October 30: Emily Hanink and Julian Grove

October 27th, 2015

Please join us this Friday as Emily Hanink and Julian Grove from the Linguistics Department present work on the German syntax-semantics interface. Please note that this will be our first meeting at our new default time of 10:30 a.m.

Title: Restrictive relatives, “same,” and the semantics of the German definite article

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

Location: Rosenwald 208 (Linguistics seminar room)


German definite articles are able to contract with prepositions under certain conditions. Schwarz (2009) argues that this contraction is blocked when a noun phrase is discourse anaphoric, otherwise it freely applies. In the current paper, we present a construction that counter-exemplifies this generalization: Restrictive relative clauses require the use of the non-contracted (strong) article form, despite their apparent lack of anaphoricity; both the determiner associated with the head noun and the relative pronoun (which is, in most cases, syncretic with the definite article) surface with the strong form. To account for this puzzle, we provide a uniform analysis of discourse anaphoric and relative clause uses that requires interpreting indices as features that may occupy their own projections in DP structure. In our analysis, the distinction between the strong and weak form is structural; the strong form contains an additional projection, which we call ‘idxP’, which hosts an index feature that may act either as a bindee, in the discourse anaphoric and relative-clause internal position, or as a binder, in the relative-clause external position. By building assignment functions into the semantic model (Sternefeld 1998, 2001; Kobele 2006, 2010; Kennedy 2014), we show that idxcan compositionally bind elements within its scope. We therefore unite anaphoric and relative clause uses by showing that both require the same additional structure, which is absent in the contracted (weak form), for binding purposes. We support this structure with morphological evidence for the presence of idx, which we argue can be realized overtly by the modifier same. An additional benefit of this proposal is that a single, Strawsonian denotation is preserved for all definite article uses, which is not possible in Schwarz’s system.

Friday, October 2: John MacFarlane

September 29th, 2015

Please join us this Friday as we host John MacFarlane, Professor of Philosophy from the University of California at Berkeley.

Date: Friday, October 2, 2015
Time: 12:30 – 3:20 p.m.
Location: MS 112 (Math-Stat Building)

Talk title: “Vagueness as Indecision”


“This paper motivates an explores an expressivist theory of vagueness, modeled on Allan Gibbard’s (2003) normative expressivism. It shows how Chris Kennedy’s (2007) semantics for gradable adjectives can be adjusted to fit into a theory on Gibbardian lines, where assertions constrain not just possible worlds but plans for action. Vagueness, on this account, is _literally_ indecision about where to draw lines. It is argued that the distinctive phenomena of vagueness, such as the intuition of tolerance, can be explained in terms of practical constraints on plans, and that the expressivist view captures what is right about several contending theories of vagueness.”

Friday June 5: Sophia Sklaviadis

June 3rd, 2015

Please join us for a talk by Sophia Sklaviadis this Friday (special time: 10:00 am):

Speaker: Sophia Sklaviadis (Philosophy, PhD Student)
Title: Dimensions and Degrees of Virtue
Date: Friday, June 5
Time: 10:00 am – 11:20 pm
Location: Rosenwald 208

Friday May 29: John MacKay

May 26th, 2015

Please join us this Friday for a talk by John MacKay, Assistant Professor of Philosophy at the University of Wisconsin.

Speaker: John MacKay (U. of Madison, Philosophy)
Title: Explaining the Actuality Operator Away
Date: Friday, May 29
Time: 12:30 am – 2:20 pm
Location: Rosenwald 208


I argue that “actually” does not have a reading in English where it is synonymous with the actuality operator of modal logic, and offer an alternative account of the term. After reviewing the reasons why actuality operators were introduced into formal languages, I argue that the interaction of “actually” with modality and tense reveals that it is not interpreted as such an operator. I advance an alternative account of “actually” and “actual” according to which they are presupposition triggers, along the lines of “even” or “too”. The presupposition they trigger has to do with the discourse evolving in non-standard fashion, with multiple information states being live candidates to be the context set.

Subjectivity in Language and Thought Conference

April 21st, 2015

Here is the program for the Subjectivity in Language and Thought Conference:

Wednesday, April 22, Stuart 101

9:30 a.m. – 10:15 a.m.
Matt Mandelkern (MIT), “A Solution to Karttunen’s Problem.”

10:15 a.m. – 11:00 a.m.
Lilia Rissman (University of Chicago), “Intentionality and Modality in the Semantics of an Instrumental Verb.”

11:00 a.m. – 11:30 a.m. Coffee Break

11:30 a.m. – 12:30 p.m.
Cleo Condoravdi (Stanford University), “What Makes Detachment Fail?”

12:30 p.m. – 2:00 p.m. Lunch Break

2:00 p.m. – 2:45 p.m.
Igor Yanovich (Tübingen), “Multiple-Assertion Variable-Free Modal in Ukrainian.”

2:45 p.m – 3:30 p.m.
Jonathan Howell (Montclair State), “Decomposing Weak Necessity in English.”

3:30 p.m. – 4:00 p.m. Coffee Break

4:00 p.m. – 4:45 p.m.
Tamara Vardomskaya (University of Chicago), “Presentation and Assertion in Subjective Predicates.”

4:45 p.m. – 5:45 p.m.
Alda Mari (IJN, CNRS/ENS/EHESS – Chicago), “Epistemic Attitudes, Consensus and Truth.”

You can view the website here.

Friday April 17: Andrea Beltrama

April 16th, 2015

Please join us this Friday for a talk by Andrea Beltrama, PhD Candidate in Linguistics at the University of Chicago:

Speaker: Andrea Beltrama (Linguistics, PhD Candidate)
Title: Totally, unsettledness and questions under discussion: Exploring the pragmatic side of intensification
Date: Friday, April 17
Time: 12:30 am – 2:20 pm
Location: Rosenwald 208


Totally, unsettledness and questions under discussion. Exploring the pragmatic side of intensification. Intensification has received considerable attention in semantics. Yet, most authors mostly focused on cases of degree modification, where intensifiers operate over degree scales lexically encoded by their argument. Instead, little has been said about cases of purely pragmatic intensification, where intensifiers target scales recruited from non truth-conditional dimensions like commitment, certainty, expressivity, and as such operate outside the realm of regular at-issue semantic content. The current talk aims to take a step towards exploring this flavor of intensification by discussing the licensing and distribution of totally in American English in examples like the following.

(1) WTF Florida! Man in “I have drugs” shirt totally had drugs

(2) You should totally watch this movie.

(3) Dionne: There was a stop sign there!
Cher: I totally paused!

I argue that, in its usage as a pragmatic intensifier, totally operates as a device to resolve unsettledness surrounding the truth of a proposition. More specifically, by explicitly inviting the interlocutor to remove from the discourse model all worlds in which the proposition is false, totally intensifies the assertion itself, serving as device to close a question under discussion. Moreover, in contexts where no question under discussion is explicitly present, it can retroactively force the accommodation of one, while marking the proposition as conversationally relevant. I conclude by sketching out a possible way to bridge the connection between these cases and totally’s more canonic use as a degree modifier, suggesting that degree modification and pragmatic intensification are distinct flavors of the same underlying phenomenon.

Subjectivity in Language and Thought Conference

April 7th, 2015

Check out the website for our upcoming workshop on Subjectivity in Language and Thought, including logistical details and programme.

It takes place all day on Wednesday, April 22. Mark your calendars!

Friday April 3: Brian Weatherson

April 1st, 2015

Join us this Friday for a talk by Brian Weatherson (University of Michigan), who will be speaking about borderline cases:

SPEAKER: Brian Weatherson
TITLE: ‘Defending the Standard View of Borderline Cases’
DATE: Friday April 3
LOCATION: Rosenwald 208
TIME: 12:30 am – 2:20 pm


Many predicates, perhaps most of them, are vague. And most vague terms have, in some intuitive sense, borderline cases. There is a very common view about what it takes for something to be a borderline F: it is not determinately F, and it is not determinately not F. In her recent book, Diana Raffman has suggested a number of arguments against that standard view, and put forward an alternative. My main aim in this paper is to respond to those arguments, and defend the standard view.

This paper is part of a larger project that I’ll gesture at towards the end, and will happily talk about. The idea is to use a familiar kind of algebraic semantics to develop a theory that has the best features of some existing approaches, without their weaknesses. It’s relevant to the present talk because the theory does something Raffman says can’t be done: namely keep the standard view of borderline cases without qualifying classical logic.

Friday February 20: Patrick Munoz

February 17th, 2015

Please join us this Friday for Patrick Munoz (PhD Student, Linguistics)!

SPEAKER: Patrick Munoz
TITLE: ‘Names in de re belief reports: problems with conceptual covers and a new analysis of identity’
DATE: Friday 2/20
LOCATION: Rosenwald 208
TIME: 11:30 am – 1:20 pm


Ambiguities between de dicto and de re interpretations of many belief reports are elegantly captured by analyzing the distinction as one of scope with respect to a doxastic modal operator. But many well-known paradoxes – among them the famous ‘double vision’ puzzles, such as those presented in Quine (1956) and Kripke (1979), in which the belief-holder is familiar with some individual under two distinct guises, and fails to recognize their identity – present problems for applying such an analysis to de re belief reports about individuals denoted by proper names. If proper names are taken to denote individuals without the mediation of any conceptual or non-trivial intensional content, then their being evaluated within, or beyond, the scope of a modal operator that shifts the world of evaluation is irrelevant to their semantic contribution. This has two unwelcome results: (1) de re belief attributions containing proper names collapse into their de dicto counterparts, making capturing the ambiguity impossible; and (2) it is impossible to capture the intuition that a belief-holder may simultaneously have, and not have, a certain de re belief about an individual in different respects without being ‘incoherent,’ i.e. without entertaining logically impossible doxastic alternatives.

This has led researchers to suggest that identifying individuals under quantification is impossible without reference to some conceptually mediating guise or other: the choice of guise then determines whether or not the de re belief holds. Aloni (2001) identifies problematic results in previous theories attempting to cash out this approach formally, and resolves them by proposing her theory of conceptual covers, according to which quantification occurs not over bare individuals, but over ‘ways of identifying’ those individuals. This is done by introducing the notion of a conceptual cover, which is a set of individual concepts such that relative to each possible world, exactly one individual falls beneath each concept, and each individual falls beneath exactly one concept; there is thus a one-to-one mapping between concepts and individuals, such that a conceptual cover can be seen as a certain way of uniquely and exhaustively partitioning the domain. Some conceptual cover or other is determined by the discourse context, and variables are in turn assigned to individual concepts that are members of this conceptual cover. While this explains the data with which Aloni is concerned, I argue that the way the theory is presented makes the prediction that relative to the use of any single conceptual cover, there are no de re belief attributions whose truth requires that two such individual concepts denote the same individual in the same world (otherwise, the uniqueness condition would be violated). But such belief attributions do exist, viz. those reporting de re beliefs about identity, which I argue can be interpreted ‘doubly de re,’ such that the expressions denoting both individuals being equated are evaluated outside the scope of the doxastic operator. Such attributions can be true, although on Aloni’s account this would require that the same individual be denoted by two individual concepts within a conceptual cover relative to the belief-holder’s doxastic alternatives, which is impossible.

The fact that the primary appeal for conceptual covers (the uniqueness condition) leads to this problematic result, in addition to the problems that previous attempts to treat quantification over individuals as mediated by conceptual content have engendered, leads me to adopt a new approach to analyzing de re belief reports containing proper names, according to which names are once again stripped of all conceptual content. I argue that the problem with such belief reports is not in our analysis of names at all, but rather in our analysis of identity: I show that such reports can be captured using the traditional scopal ambiguity analysis if the identity relation used to symbolize their logical forms is treated not as a matter of logical identity between members of a domain, but rather as a descriptive identity amounting to an indiscernability of identicals and an identity of indiscernables. This allows us not only to capture all of the data with which Aloni and her predecessors were concerned, but also belief reports about identity statements, and further allows us to drastically simplify the semantic and pragmatic mechanisms accompanying proper names.

The result of adopting this new approach to natural language identity has several important consequences, first among which is that the domain is now better conceptualized as a set of discourse referents rather than individuals. I argue that this is a positive result, and has promising applications for broader issues in the semantics of names, including a straightforward solution to Frege’s Puzzle and the dissolution of worries involving negative existential statements containing names. It thus points the way to a promising defense of a simple, Millian semantics of proper names.