Daphna and Sarah talk for CycleLinguists

Daphna Heller (faculty) and colleague Sarah Brown-Schmidt (Vanderbilt University) are giving a talk for online occasional psycholinguistics group Cycle Linguists on Thursday, November 12, from 12 PM to 1:30 PM: "Common Ground is dead. Long live Common Ground!" Registration is available here.

Theories of the role of mental state representations during linguistic communication, inspired by early proposals by philosophers of language and formal linguists, posit representations of shared knowledge and beliefs, or common ground. Alternative accounts posit that successful communication does not require calculating representations of common ground, but instead, that more general cognitive mechanisms give rise to shared or coordinated representations. Despite their differences, what these views have in common is a focus on shared information. Here we argue that views that rely on what is shared as the basis for communication fail to capture many aspects of language use. We propose a novel account of the role of mental state representations for language, where the perspectives of the partners are compared. This proposal accounts for existing data, interfaces with findings from other cognitive domains, and makes novel, yet-to-be-tested empirical predictions.

Research Groups: Friday, October 30

10:00 AM - 11:30 AM: Language Variation and Change Research Group
Group discussion led by Marisa Brook (faculty) of a paper: Eckert, Penelope (2019). The individual in the semiotic landscape. Glossa, 4(1).

1:00 PM - 2:30 PM: Semantics Research Group
Dan Milway (Ph.D. 2019): "On the uncountability of possible worlds."

In possible worlds semantics the extension of a proposition in a given context is not a truth value but rather a function from possible worlds to truth values. Using Cantor’s diagonal argument - a method from mathematical logic - I show that this theory of semantics predicts that, although the set of possible propositions and the set of possible worlds are both infinite, the set of possible worlds is larger than the set of possible propositions. I further argue that this incommensurability between worlds and propositions renders possible world semantics semantically incomplete. I close by exploring alternative approaches to semantics, some of which (e.g., situation semantics) suffer from the same problems as possible worlds semantics, while others, which require us to limit the empirical domain of semantics, are still available to us.

Guest speaker: Ted Gibson (Massachusetts Institute of Technology)

The Cognitive Science Program at University College is hosting a talk by Ted Gibson, a faculty member from the Department of Brain and Cognitive Sciences at the Massachusetts Institute of Technology. He works on language processing, cognitive perspectives on the links between language and culture, and typological patterning from the standpoint of information structure. His talk, "Information processing and cross-linguistic universals", will be taking place on Thursday, March 14, at 4:30 PM, in UC 140.

Finding explanations for the observed variation in human languages is the primary goal of linguistics, and promises to shed light on the nature of human cognition. One particularly attractive set of explanations is functional in nature, holding that language universals are grounded in the known properties of human information processing. The idea is that lexicons and grammars of languages have evolved so that language users can communicate using words an sentences that are relatively easy to produce and comprehend. In this talk, I summarize results from explorations in two linguistic domains, from an information-processing point of view. First, I show that word lengths are optimized on average according to predictability in context, as would be expected under an information theoretic analysis. Second, I show that all the world's languages that we can currently analyze minimize syntactic dependency lengths to some degree, as would be expected under information processing considerations. Finally, we apply a simply information theory analysis to the language for color. The number of color terms varies drastically across languages. yet despite these differences, certain terms (e.g., red) are prevalent, which has been attributed to perceptual salience. Our work provides evidence for an alternative hypothesis: The use of color terms depends on communicative needs. Across languages, from the hunter-gatherer Tsimane' people of the Amazon to students in Boston, warm colors are communicated more efficiently than cool colors. This cross-linguistic pattern reflects the color statistics of the world: objects (what we talk about) are typically warm-colored, and background are cool-colored. Communicative needs also explain why the number of color terms varies across languages: cultures vary in how useful color is. Industrialization, which creates objects distinguishable solely based on color, increases color usefulness.

Workshop: New Perspectives on Mental State Attribution

The Department of Philosophy is hosting a workshop, New Perspectives on Mental State Attribution, early next week. This will be taking place at the Jackman Humanities Building (Room 100) on Monday, December 10, and Tuesday, December 11, from 9 AM to 5 PM both days. The program features a mix of guest speakers giving in-depth presentations. Of particular interest to the linguistics community are the two talks on the Tuesday afternoon, as follows:

2:00 PM to 3:30 PM:
Elizabeth Bogal-Allbritten (University of Gothenburg):
"Building attitudes in the grammar."

4:00 PM to 5:00 PM
Rachel Dudley (École normale supérieure):
"Pragmatic effects in child's understanding of attitude verbs."

Invited talk for Philosophy: Craige Roberts (Ohio State University/New York University)

The Department of Philosophy and its group on Language, Epistemology, Metaphysics, and Mind are delighted to welcome Professor Craige Roberts. Now Professor Emerita at Ohio State University and a Visiting Professor at New York University, she is an esteemed scholar who works on formal semantics, pragmatics, and philosophy of language. Her talk, "The character of epistemic modals in natural language: Evidential indexicals", will be taking place on Thursday, November 8, from 3:00 PM to 5:00 PM, in room 418 of the Jackman Humanities Building.

I assume a central thesis about Modal Auxiliaries due to Kratzer, roughly as follows:

THE MODAL BASE PRESUPPOSITION: Natural language expressions that contain a modal       component in their meaning, including all English modal auxiliaries and Epistemic Modal Auxiliaries (EMA)s in particular, presuppose a modal base, a function that draws from context a relevant set of propositions which contribute to a premise-semantics for the modal.

Accepting this thesis for EMAs leaves open (at least) the following two questions about the meaning of English EMAs like must and might:

i. What constraints, if any, are there on the character of the premise set for an EMA?
ii. What is the nature of the relationship between premises and conclusion that is required for truth of the EMA statement?

I argue for at least a partial answer to (i), with two hypotheses about the proper constraints on the modal base for an EMA:

EVIDENTIALITY: The modal base for an EMA is evidential and doxastic, not truly epistemic (i.e., weak, not strong).

INDEXICALITY: EMAs, unlike some other types of modals, are indexical: They are anchored to an agent-at-a-time whose doxastic state is currently under discussion in the context of utterance.

These constraints are modeled as presuppositions triggered by the EMA, restrictions on the modal’s domain. The independently motivated indexical anchoring (a) correctly predicts the contextually limited range of candidates for the anchoring agent of such a modal, as attested in the literature, (b) thereby constrains what body of evidence is understood to be relevant (that of the anchor), and (c) in some cases plays a role in explaining the modal’s scope (not discussed here). The account sheds light on several puzzles, including (d) Yalcin’s (2007) version of Moore’s paradox for embedded epistemic modals, and (e) purported arguments for modal relativism (e.g., Egan, Hawthorne & Weatherson 2005).

Public lecture: Philippe Schlenker (New York University/École normale superieure)

The Jackman Humanities Institute welcomes Philippe Schlenker as a Distinguished Visiting Fellow next week. He is a Global Distinguished Professor at New York University and Director of Research at the Institute Jean-Nicod, Department of Cognitive Studies, École Normale Superieure. He holds two Ph.D.s: one in linguistics from MIT (1999) and one in philosophy from l'École des Hautes Études en Sciences Sociales (2002). His research interests encompass semantics and pragmatics, philosophy of language, morphosyntax, sign languages, and animal communication. On Tuesday, October 30 from 4 to 6 PM in room 100 of the Jackman Humanities Building, Professor Schlenker will be giving a public lecture: "Meaning in sign, in speech, and in gesture." ASL interpretation will be provided. There is no charge for attendance, and registration is not required.

Contemporary linguistics has established 3 results: 1. Sign languages, used by Deaf communities throughout the word, are full-fledged languages that share typological properties among themselves and also with spoken languages. 2. Sign languages have the same 'logical spine' as spoken languages, but sometimes they make the logical structure of sentences far more explicit than is the case in spoken language. A salient case concerns logical variables, which are covert in spoken language but are realized overtly in sign language by way of positions in signing space. 3. But in addition, sign languages have rich iconic possibilities, including at their logical core. For instance, logical variables can simultaneously function as simplified iconic representations of their denotations. By contrast, iconic possibilities exist but are limited in the spoken modality.

Should we conclude (from 2 and 3) that sign languages are more expressive than spoken languages, since they have the same logical spine but richer iconic possibilities? For the comparison to be complete, one must re-integrate into spoken language semantics the study of co-speech gestures, which have rich iconic capabilities. But we will argue that even when sign language is compared to speech-plus-gestures, sign languages have an entire class of expressive possibilities that spoken languages mostly lack.

Research Groups: Week of May 16-20

Thursday, May 19 - 11:00 AM to 1:00 PM in SS 1086
Syntax/Semantics Group
Two practise talks for CLA-ACL: Neil Banerjee (BA) and Nicholas Welch (postdoc).

Friday, May 20 - 2:00 PM to 3:00 PM in SS 560A
Semantics/Pragmatics Group
Rory Harder (Department of Philosophy): "Deontic modals: Inference patterns and understanding."

Here are two desiderata that a theory of meaning for deontic modals should meet: (1) given that P entails Q , it does not follow that Ought(P) entails Ought(Q); and (2) if, in a given context c, a speaker who is semantically competent with the term Ought can make a sincere judgment that Ought(P), a theory of meaning for Ought should not rule out the truth-hood of Ought(P) in c. The first is called non-upward monotonicity, and the second semantic neutrality. It is surprising to me that no theory of meaning for deontic modals has been presented in the literature that meets both of them. In this paper, I first summarize some results that show that the semantics from Kratzer (1981) and Lassiter (2011) fail to meet at least one of the desiderata. I then present some novel results in this vein. A couple recent theories, from Cariani (2013) and Fusco (2015), designed specifically to meet constraints like (1), fail to meet (2). And a couple recent neutral theories, from Carr (2015) and Charlow (forthcoming), designed specifically to meet (2), fail to meet (1), and also, thereby, fail to fully meet (2) as well. In conclusion, I make some general suggestions for how an adequate semantics for deontic modals could be constructed.

Guest speaker: Zsófia Zvolenszky (Slovak Academy of Sciences/Eötvös Loránd University)

The Department of Philosophy is hosting a talk on the philosophy of language being given by Zsófia Zvolenszky of the Slovak Academy of Sciences and Eötvös Loránd University in Hungary: "A common problem for possible-worlds analyses of deontic and fictional discourse." This will be taking place on Monday, April 11, from 3:15 PM to 5:00 PM in room 418 of the Jackman Humanities Building.

In my talk, I will bring together considerations from formal semantics and philosophy of language about two seemingly very different types of discourse: deontic modal discourse (about what is necessary and possible according to a certain set of rules, for example, rules of etiquette or U.S. traffic laws), and discourse about fiction (claims like 'Anna Karenina is Russian', true in the work of fiction but not true simpliciter). My aim is to draw out commonalities of these two types of discourse, and lessons they hold for one another. Specifically, I focus on (i) a common problem that affects influential possible-worlds-based accounts of the two types of discourse; and (ii) the advantages of an alternative, operator-based approach to the two types of discourse. I'll say a bit more about these in reverse order. 
A fiction-operator ("in/according to the relevant work of fiction") is widely used in analyses of fictional discourse; I argue that an analogous corpus-operator ("according to the relevant corpus of rules/laws") has advantages when accounting for deontic modal discourse. Crucially, neither operator should be cashed out in terms of truth with respect to possible worlds—for short, neither should be pw-based. The reason: a thorny problem (about intuitively false deontic modal claims that nonetheless come out as logical truths: for example, many claims of the form ‘if p then it ought to be that p). I (and others) have been discussing this problem over the past decade and a half in the context of deontic modal discourse, showing that it afflicts a benchmark pw-based analysis of modality due to Angelika Kratzer (she discusses the problem in her 2012 book Modals and Conditionals). I will argue that the thorny problem is even more general than previously thought (indeed applicable to deontic claims that aren’t conditional) and calls for giving up a basic tenet of pw-based analyses of deontic modality (like Kratzer’s): that the universal truth of p across a selected set of possible worlds is sufficient for ‘it ought to be that p’ to be true. A corpus-operator—if it steers clear of possible worlds—promises to avoid the various versions of the thorny problem at hand. Yet if we look to fashion the corpus-operator based on its far more widely discussed cousin, the fiction-operator, we find that one prominent approach to it, David Lewis’s (in his classic 1978 "Truth in Fiction"), gives a pw-based account of the fiction-operator. I argue that it is well to steer away from a Lewisian approach in the fictional discourse realm due to an analog of the thorny problem already familiar in the deontic modal discourse realm. 
My conclusion is that there are reasons—in many ways parallel reasons—for adopting a non-pw-based operator approach to analyzing fictional discourse as well as deontic modal discourse.