Languagre. doi: (to appear). With Ryan Bennett and Meg Harvey
Uspanteko is an endangered Mayan language spoken in Guatemala. Unstressed vowels in Uspanteko often delete, though deletion is variable within and across speakers. Deletion appears to be phonological: it is sensitive to foot structure, morphology, and certain phonotactics; and occurs in slow, careful speech. But deletion also has characteristics more typical of a phonetic process: it is intertwined with a pattern of gradient vowel reduction, and is insensitive to most phonotactics. Electroglottography data shows that even 'deleted' vowels may contribute voicing to [C(V)C] intervals when flanked by voiceless consonants. This suggests that 'deleted' vowels are represented in the input to speech production, even when they are acoustically masked by articulatory overlap with adjacent segments. We conclude that vowel deletion is grammatically controlled gestural overlap, consistent with the claim that phonological representations encode information about the relative timing and coordination of articulatory gestures (e.g. Browman & Goldstein 1986, Gafos 2002). At a minimum, language-specific phonetic processes must have access to more fine-grained, abstract grammatical information than is usually assumed.
With Ryan Bennett and Meg Harvey
In Prosody and Prosodic Interfaces, edited by Haruo Kubozono, Junko Ito, and Armin Mester.
Uspanteko (Mayan) is unique among the languages of Guatemala in having a full-fledged system of lexical tone. Tone is thoroughly integrated into the morphology of the language, and triggers several predictable phonological alternations. The Uspanteko tone system is also relatively simple, involving at most two contrastive categories. However, tone in Uspanteko shows a high degree of phonetic variability, and there is no consensus as to what tones are phonetically or phonologically active in the language. This paper argues that at least some of this unclarity owes to the effect of intonation, which can obscure lexical tone on vowels. A phonetic study attempting to control for intonation suggests that lexical tone in Uspanteko involves a privative [H]~0 contrast on short vowels, and either a [H]~0 or [H]~[L] contrast on long vowels. The relationship between the phonetics of tone and its low functional load in Uspanteko is also discussed.
With Jérémy Pasquereau and John W.W. Powell
Presented at the Workshop on Mesoamerican Languages, WCCFL 29, and Sinn und Bedeutung 26
Piipaash (Yuman) has what, at first pass, look like standard dependent definites (e.g., Balusu 2006;Farkas 1997; Henderson 2014). Looking more broadly we see that the dependent indefinite marker -xper- has a wider distribution than markers of dependent indefinites in other languages discussed in the literature. Moreover, this distribution introduces twopuzzles that we will solve in this talk by proposing a unified account of -xper- in terms of a novel kind of pluractionality that we dub dependent pluractionality. The core proposal is that in most previously discussed languages the relevant morphology marks an individual variable as dependent (i.e., the variable quantified over by a numeral or indefinite). In Piipaash (aka Maricopa), -xper- marks an an event variable as dependent. What is special about Piipaash is that a wide variety ofexpressions are verbal, including numerals, and have an event argument.
With Ryan Bennett, Megan Harvey, and Tomás Alberto Méndez López
accepted at Tlalocan.
Uspanteko is one of the smallest and most threatened Mayan languages spoken in Guatemala. While Uspanteko belongs to the K'ichean branch of the Mayan family, it bears a striking number of linguistic features not seen in its sister languages, across major aspects of the grammar, including phonetics/phonology (i.e., lexical tone) and morphosyntax (i.e., null verbal inflection). After introducing an Uspanteko text about the sacred hill Xoqoneb' and placing it in its cultural context, including surveying related Uspanteko texts, we use Xoqoneb' to illustrate those aspects of the Uspanteko language which stand out relative to other K'ichean languages. We also comment on the possible sources of these divergent features, which include both language contact and innovation.
With Ryan Bennett, Megan Harvey, and Tomás Alberto Méndez López
accepted at Language and Linguistic Compass.
Uspanteko is an endangered Mayan language spoken by up to 6000 people in the Guatemalan highlands. We provide an overview of the phonetics and phonology of Uspanteko, focusing on phenomena which are common in Mayan languages and/or typologically interesting. These include glottalized consonants (ejectives, implosives, and glottal stop), uvular consonants, vowel length contrasts, syllable structure, stress, and lexical tone. Tone is unusual among Mayan languages, especially in Guatemala, and the phonetic description here complements the small handful of existing descriptions of tone in Uspanteko and within the Mayan family.
With Elin McCready
Proceedings of Probability and Meaning Conference, Gothenburg 2020.
Judgements about communicative agents evolve over the course of interactions both in how individuals are judged for testimonial reliability and for (ideological) trustworthiness. This paper combines a theory of social meaning and persona with a theory of reliability within a game-theoretic view of communication, giving a formal model involving interactional histories, repeated game models and ways of evaluating social meaning and trustworthiness.
With Elin McCready
Proceedings of Probability and Meaning Conference, Gothenburg 2020.
Henderson and McCready 2017, 2018, 2019 build a novel theory of so-called 'dogwhistle' communication by extending the social meaning games of Burnett 2017. This work reports on an ongoing project to build systems to model the evolution of dogwhistle communication in a population based on probability monads (Erwig and Kollmansberger, 2006; Kidd, 2007). The ultimate results will be useful not just for dogwhistles, but modeling the diffusion and evolution of social meaning in populations in general. The initial results presented here is a computational implementation of Henderson and McCready 2018, which will serve as the basis for models with multiple speakers and repeated interactions.
With Jian Gang Ngui
We discuss the effect of the SE sentence-final discourse particlel lo, which can be used to "agree to disagree" (Farkas & Bruce 2010), but also for a variety of other discourse effects. We propose aunified account of these effects, which will show that "agreeing to disagree" is a special case of amore general phenomenon in which interlocutors attempt to move to a previous discourse state by ignoring the public effects of proposed updates.
Chapter for Handbook of North American Languages.
Verbs are the canonical way that languages allow speakers to talk about events, but events are slippery things. Imagine how hard it would be, for instance, to watch a short video of an action scene and try to decide how many distinct events took place. Languages often circumvent this problem by having morphological marking on verbs that makes clear the structure of the events being discussed. In particular, languages often have pluractionality marking that makes clear that a plurality of events are being discussed, and this marking often makes clear how this plurality of events is being individuated for counting (e.g., by taking place in different locations, or with different participants, etc.). Languages also often mark distributivity on verbs, which is a category that disambiguates how events are related to participants. This chapter will provide a crosslinguistic survey of pluractionality and distributivity, focusing especially on Native North American language. It will also present strategies for determining the types of pluractionality and distributivity available in languages for which those categories have not been extensively documented.
With Elin McCready
Proceedings of the Amsterdam Colloquium 2019.
Given that someone is consistently untruthful, why should we ever trust them? The question is not academic. Consider politicians and others who are known to consistently lie, but who are still voted back into office. This talk addresses this puzzle via three mechanisms: (i) a theory of source evaluation based on interactional histories and heuristics for judgments of reliability (McCready, 2015), (ii) a game-theoretic view of how speaker ideologies and political positions are communicated by linguistic acts (Burnett, 2018; Henderson and McCready, 2018), and (iii) a theory of how ideological considerations are valued alongside truth-conditional content. In the process, an analysis of fake news claims is provided.
Invited presentation at the 2017 DGfS workshop on secondary information. Then appeared as a chapter in 2019 in the volume Secondary Content. With Elin McCready
A dogwhistle is language that sends one message to an outgroup while at the same time sending a second (often taboo, controversial, or inflammatory) message to an ingroup. Stanley (2015) argues that dog-whistle language like involves a non-at-issue component. We argue against a CI account of dogwhistles and instead propose alternative, purely pragmatic account combining aspects of McCready 2012, Burnett 2016; Burnett 2017, and which we think better accounts for the core their core properties. In broad strokes, we make the novel proposal that dogwhistles come in two types. The first concerns covert signals that the speaker has a certain persona, which we model by extending the Sociolinguistic Signalling Games of Burnett 2016; Burnett 2017. The second involves sending a message with an enriched meaning whose recovery is contingent on recognizing the speaker's covertly signalled persona.
Semantics & Pragmatics. doi: 10.3765/sp.12.1. With Lucas Champollion and Dylan Bumford.
Donkey sentences have existential and universal readings, but they are not often perceived as ambiguous. We extend the pragmatic theory of non-maximality in plural definites by Kriz (2016) to explain how hearers use Questions under Discussion to fix the interpretation of donkey sentences in context. We propose that the denotations of such sentences involve truth-value gaps---in certain scenarios the sentences are neither true nor false---and demonstrate that Kriz'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, we define a general schema for dynamic quantification that delivers the required truth-value gaps. Given the independently motivated pragmatic theory of Kriz 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.
Linguistic Inquiry. doi. 10.1162/LING_a_00272 With Ryan Bennett and Boris Harizanov.
This paper deals with a so-far unnoticed phenomenon in prosodic phonology, which we dub prosodic smothering. Prosodic smothering arises when the prosodic status of a clitic or affix varies with the presence or absence of some outer morpheme. We first illustrate prosodic smothering with novel data from two genetically unrelated languages, Macedonian (Slavic) and Kaqchikel (Mayan). We then provide a unified account of prosodic smothering based on a principled extension of the theory of prosodic subcategorization (Inkelas 1990, Peperkamp 1997, Chung 2003, Yu 2003, Paster 2006, Bye 2007, among others). Prosodic subcategorization typically involves requirements placed on items to the left or the right of the selecting morpheme. We show that prosodic smothering naturally emerges in a theory which also allows for subcategorization in the vertical dimension, such that morphemes may select for the prosodic category which immediately dominates them in surface prosodic structure. This extension successfully reduces two apparent cases of non-local prosodic conditioning to the effects of strictly local prosodic selection.
Glossa. doi: 10.5334/gjgl.515
In addition to roots for familiar classes like verb, noun, and adjective, Mayan languages have a class of roots traditionally called ``positional''. Positional roots are distinct from other roots most prominently in terms of requiring derivation into stems of one of the more familiar categories to be used. The goal of this work is to show that the behavior of positionals follows from semantic facts, in particular, the fact that they denote measure functions of type
Language. doi: 10.1353/lan.2019.0014 With Daniel Gutzmann.
This paper investigates a novel use of much in a construction that has not yet been recognized in the literature---Angry, much?---which we dub "expressive much". Our primary proposal is that expressive much is a shunting operator in the sense of McCready 2010, which targets a gradable predicate and adds a speaker's evaluative attitude about the degree to which an individual stands out on the relevant scale. In particular, we argue that it does so in a way that allows it to perform an "expressive question", which can be understood as a counterpart to a polar question, but in the expressive meaning dimension.
LENLS 14. doi: 10.1007/978-3-319-93794-6_16 With E. McCready.
The paper focuses on the semantics and pragmatics of dog-whistles, namely expressions that send one message to an outgroup while at the same time sending a second (often taboo, controversial, or inflammatory) message to an ingroup. There are three questions that need to be resolved to understand the semantics and pragmatics of the phenomenon at hand: (i) What kind of meaning is dogwhistle content-implicature, conventional implicature, etc; (ii) how do (some but not all) hearers recover the dogwhistle content, and (iii) how do expressions become endowed with dogwhistle content? These three questions are interrelated, but previous analyses have emphasized answers to a subset of these questions in ways that provide unsatisfactory answers to the others. The goal for this paper is to take stock of existing accounts, while showing a way forward that reconciles their differences
Natural Language & Linguistic Theory. doi: 10.1007/s11049-016-9334-z.
This paper presents cross-domain evidence that natural language makes use of two types of group entities that differ in terms of how they cohere. The first kind of groups, which I call swarms, are defined in terms of the spatial and temporal configuration of their members. The second, which are the canonical group entities, are defined in terms of non-spatiotemporal notions. To motivate this distinction, I present systematic differences in how these two types of group entities behave linguistically, both in the individual and event domains. These differences support two primary results. First, they are used as tests to isolate a new class of group nouns that denote swarm individuals, both in English, as well as other languages like Romanian. I then consider a crosslinguistically common type of pluractionality, called event-internal in the previous literature (Cusic 1981; Wood 2007), and show that its properties are best explained if the relevant verbs denote swarm events. By reducing event-internal pluractionality to a type of group reference also available for nouns, this work generates a new strong argument that pluractionality involves the same varieties of plural reference in the event domain that are seen in the individual domain.
Natural Language and Linguistic Theory. doi: 10.1007/s11049-017-9370-3 With Jessica Coon
In many languages with ergative morphology, transitive subjects (i.e.ergatives) are unable to undergo A'-extraction. This extraction asymmetry is a common hallmark of "syntactic ergativity", and is found in a range of typologically diverse languages (see e.g. Deal 2016; Polinsky to appear, and works cited there). In Kaqchikel, the A'-extraction of transitive subjects requires a special verb form, known in Mayanist literature as Agent Focus (AF). In a recent paper, Erlewine (2016) argues the restriction on A'-extracting transitive subjects in Kaqchikel is the result of an Anti-Locality effect: transitive subjects are not permitted to extract because they are too close to C0. This analysis relies crucially on Erlewine's proposal that transitive subjects undergo movement to Spec,IP while intransitive subjects remain low. For Erlewine, this derives the fact that transitive (ergative) subjects, but not intransitive (absolutive) subjects are subject to extraction restrictions. Furthermore, it makes the strong prediction that phrasal material intervening between IP and CP should obviate the need for AF in clauses with subject extraction. In this paper, we argue against the Anti-Locality analysis of ergative A'-extraction restrictions along two lines. First, we raise concerns with the proposal that transitive, but not intransitive subjects, move to Spec,IP. Our second, and main goal, is to show that there is variation in whether AF is observed in configurations with intervening phrasal material, with a primary focus on intervening adverbs. We propose an alternative account for the variation in whether AF is observed in the presence of adverbs and discuss consequences for accounts of ergative extraction asymmetries more generally.
SALT 26. doi: 10.3765/salt.v26i0.3786, and as an invited talk at LENLS 2015.
This paper develops a novel formal semantics for ideophones that can account for their meaning and compositional properties. The proposal extends recent work on iconicity in sign languages by Davidson (2015), whose demonstration-based framework provides a formal foundation for the semantics of ideophones that captures the difference between descriptive meaning and depictive meaning, the kind of meaning ideophones traffic in. After providing a demonstration-based account of the basic ideophone construction in the Mayan language Tseltal, the paper then shows how the demonstration-based account can be used to analyze pluractionality in the ideophone domain. In particular, through case studies on Tseltal and Upper Necaxa Totonac (Totonacan), I show that there are two previously unrecognized types of ideophonic pluractionality, and that their properties support the demonstration-based account. The first, which I call "demonstration-external pluractionality", involves a speaker using an ideophone to do a plurality of demonstrations that characterize a plurality of events. The second kind of ideophonic pluractionality, which I call a "demonstration-internal pluractionality", is much more similar to pluractionality in the verbal domain, and involves special morphology that derives ideophone stems that can only be used to demonstrate plural events. Finally, I use the contrast between these two types of pluractionality in the ideophone domain to clarify the line between the iconic and non-iconic aspects of the semantics of ideophones.
Language and Linguistics Compass. 10(10), 551-588. 10.1111/lnc3.12187
This article has two interlocked goals. The first is to highlight the strands of research that have played an important role in shaping our understanding of Mayan language semantics. The second is to acquaint non-Mayanists, and especially semanticists, with empirical phenomena that might prove especially interesting on typological or theoretical grounds. Given its particular dual mandate, this article cannot be an exhaustive survey of Mayan semantics, but it should instead be seen as introduction to those aspects of Mayan languages that have had an impact on the wider field of semantics, and that are current research hotspots. That said, I do want to acquaint the reader with a broad range of phenomena, and so the paper is organized like a grammar would be, first considering lexical categories, then phrase- and clause-level phenomena, and finally issues of discourse and informations structure.
Chapter in a Routledge volume entitle Mayan Languages
Pluractionality is a category that has not been traditionally talked about in grammars of Mayan languages. The goal of this chapter is to get a broad view of pluractionality in Mayan by presenting a series of pluractionals from two languages, Kaqchikel and Tzeltal, which I take as representative of Western- and Eastern-branch languages, respectively. What we find is that Mayan languages instantiate all of the typologically familiar varieties of pluractionality, making them the ideal languages for studying the phenomenon.
Semantics & Pragmatics. doi: 10.3765/sp.7.6
This paper presents an analysis of a new scope puzzle that arises through the interaction of two lesser-studied constructions, dependent indefinites and verbal pluractionality. The result is a novel account of dependent indefinites that correctly predicts their grammaticality with pluractionals by recognizing two ways of establishing the covariation they require: (i) true distributive quantifiers, and (ii) pluractional operators that structure thematic dependencies. The core insight is that both routes, while compositionally different, lead to similar output structures in Dynamic Plural Logic (DPlL) (van den Berg 1996) or its closely related alternatives (Brasoveanu 2008, Nouwen 2003), which is what dependent indefinites constrain. The analysis not only permits a better understanding of dependent indefinites in Kaqchikel, an endangered and understudied Mayan language of highland Guatemala, but it clarifies their place in a crosslinguistic typology of similar expressions (e.g., Balusu 2006, Choe 1987, Farkas 1997a, 2002, Yanovich 2005). Along the way we produce the first description and analysis of these phenomena in Kaqchikel.
Language Documentation & Conservation. vol 8, pp. 75-91. With Brent Henderson and Peter Rohloff.
Existing models for language revitalization focus almost exclusively on language learning and use. While recognizing the value of these models, we argue that their effective application is largely limited to situations in which languages have low numbers of speakers. For languages that are rapidly undergoing language shift, but which still maintain large vital communities of speakers, a model for revitalization is currently lacking. We offer the beginnings of such a model here, arguing that in these communities doing language revitalization must primarily mean addressing the causes of language shift, a task that we argue can be undertaken in collaborative efforts with social development organizations. The model contrasts strongly (though complementarily) with existing models in that it focuses on work in which explicitly language---focused activities are undertaken only as intentional support for social development projects. Where successful, we argue this approach achieves language revitalization goals in organic and sustainable ways that are much more difficult for language-focused programs to achieve. It therefore has the potential to stop and potentially reverse language shift in specific ways. We offer our experiences with Wuqu' Kawoq | Maya Health Alliance, a healthcare NGO in Guatemala, which attempts to follow this model, as evidence for the model's viability.
Semantics and Linguistic Theory 23. doi: 10.3765/salt
This paper provides a new analysis of N-BY-N adverbials that captures their previously unrecognized close connection to verbs of scalar change. After providing a series of arguments that N-BY-N modification requires the VP to provide a scalar interval it can measure, we use this as novel evidence that incremental theme verbs, as well as inherently directed motion and change of state verbs, must make reference to scales. The analysis thus supports a unified scalar account of verbs of variable telicity (e.g., Hay, Kennedy & Levin 1999; Kennedy & Levin 2008; Kennedy 2012). Finally, we show that our analysis avoids empirical problems for previous approaches to these adverbials in both English (Beck & von Stechow 2007; Brasoveanu & Henderson 2009) and Russian (Braginsky & Rothstein 2008).
Journal of Semantics. doi: 10.1093/jos/fft014. With Scott AnderBois and Adrian Brasoveanu.
Potts (2005) and many subsequent works have argued that the semantic content of appositive (non-restrictive) relative clauses, e.g., the underlined material in John, who nearly killed a woman with his car, visited her in the hospital, must be in some way separate from the content of the rest of the sentence, i.e., from at-issue content. At the same time, there is mounting evidence from various anaphoric processes that the two kinds of content must be integrated into a single, incrementally evolving semantic representation. The challenge is how to reconcile this informational separation with these pervasive anaphoric connections. We propose a dynamic semantic account that accomplishes this by taking appositive and at-issue content to involve two different kinds of updates to the Context Set (CS). Treating the context set as a distinguished propositional variable, pcs, we argue that appositives directly impose their content on the CS by eliminating possible values assigned to pcs. In contrast, we treat at-issue assertions as introducing a new propositional dref and proposing that pcs be updated with its content, subject to addressee's response. In addition to capturing the behavior of appositives in discourse, we show that the account can be extended to capture the projection of appositive content past various sentential operators.
Natural Language and Linguistic Theory. doi: 10.1007/s11049-013-9196-6. With Ryan Bennett
Uspanteko (Guatemala; ~2000 speakers) is an endangered K'ichean-branch Mayan language. It is unique among the K'ichean languages in that, along with obligatory right-edge stress, Uspanteko has innovated a system of contrastive pitch accent. Word-level accent in Uspanteko is of theoretical interest for several reasons. First, it has a mixed accentual system with both stress and lexical pitch accent. Second, lexical pitch has striking effects on prosodic and segmental structure, interacting with stress shift, vowel length, vowel quality, and two deletion processes. Third, pitch accent is closely tied to morphology (especially possessive marking) even though the location of morphologically-derived tone is entirely a matter of surface phonology. Fourth, interactions between tone and vowel length provide evidence for lexical strata within the accentual system of Uspanteko. In this paper we develop a novel analysis of Uspanteko accent, using data drawn from previous research as well as our own recent fieldwork. We propose that the location of pitch accent and stress in Uspanteko can be straightforwardly captured under three assumptions: (i) Uspanteko words contain a single right-aligned iamb; (ii) pitch accent must dock to the head of a foot; and (iii) pitch accent cannot dock to a word-final mora. These assumptions account for default word-final stress, as well as penultimate stress in [CVCV] words bearing pitch accent, which we treat as an iambic-trochaic foot form reversal. Interactions between prosody and segmental structure in Uspanteko are analyzed as the result of further constraints on foot shape, stress assignment, and tone non-finality. A surprising finding of this paper is that there is robust evidence for foot structure in Uspanteko, despite the fact that accent in Uspanteko could easily be described in non-metrical terms. Finally, we model accentual cophonologies in Uspanteko using partially-ordered prosodic constraints.
Natural Language and Linguistic Theory. doi: 10.1007/s11049-012-9170-8
This article develops an analysis of a pair of morphological alternations in K'ichee' (Mayan) that are conditioned at the right edge of intonational phrase boundaries. I propose a syntax-prosody mapping algorithm that derives intonational phrase boundaries from the surface syntax, and then argue that each alternation can be understood in terms of output optimization. The important fact is that a prominence peak is always rightmost in the intonational phrase, and so the morphological alternations occur in order to ensure an optimal host for this prominence peak. Finally, I consider the wider implications of the analysis for the architecture of the syntax-phonology interface, especially as it concerns late-insertion theories of morphology. Warning! Could you please reference the preprint hosted here when citing this paper. I've standardized some of the spelling in the examples (that thankfully do not change empirical or theoretical conclusions.)
Sinn und Bedeutung 16. With Scott AnderBois and Adrian Brasoveanu
Most investigations of quantifier scope are concerned with the range of possible scopes for sentences with multiple quantifiers. Instead, this study examines the actual scopes, i.e., the pragmatics of quantifier scope disambiguation. The three main findings of our investigation are as follows. First, we confirmed the results in the previous literature that linear order and grammatical function have an effect on scope-taking preferences. Second, we discovered that lexical effects on scoping preferences are at least as important as linear order or grammatical function. Third, the relational aspect of these lexical effects, i.e., the lexical realizations of the other quantifiers in the sentence, is also important. The present investigation opens the way towards a broader research program of identifying scoping-behavior patterns that should ultimately enable us to group quantifiers into classes depending on the type of scopal behavior they exhibit. Identifying such classes could provide an empirical basis for semantic theories that assign different kinds of semantic representations to these classes and / or for psycholinguistic theories that hypothesize different processing strategies for different classes.
In "Representing Language: Essays in Honor of Judith Aissen". With Jessica Coon
This paper examines binding puzzles in two Mayan languages and proposes an analysis which unifies two otherwise different-looking constructions: the Chol applicative and the K'ichee' agent focus (AF). In both the Chol applicative and the K'ichee' AF, subjects are banned from binding object possessors. That is, the equivalents of English Maria bought her own tortillas or It was Juan who burned his own foot are impossible in the relevant constructions (though they are possible under a reading in which the subject and object's possessor are not coreferential). We propose that in both types of construction, binding of the object's possessor by the subject is blocked by an intervening v head. In the Chol (low) applicative, this is the head added to introduce the applied argument. In the K'ichee' AF, this is the head needed to introduce the subject; we may think of this as a type of high applicative. In this paper we show that the similar binding restrictions in these two different languages are easily accounted for under a theory which ties the availability of binding to locality with domains defined by v heads, such as the minimal pronoun approach of Kratzer (2009).
West Coast Conference on Formal Linguistics 28.
Based on the previously unnoticed contrast between standard counterfactuals and the non-canonical counterfactual construction "if not for P, Q" (hereafter NC, for `Not' Counterfactuals), this paper argues for the emerging proposal that two distinct routes to counterfactuality are available in natural language (see e.g., Schulz 2007): (i) global revision of a belief state, and (ii) local revision of a world. Schulz's (2007) observation is that the classic epistemic inferences are precisely those that disappear under local update in a causal model, allowing for a treatment of the ontic-epistemic distinction in counterfactuals as an ambiguity between local and global belief revision. We show that NCs systematically reject epistemic readings, supporting the position that different ways of evaluating the antecedent permit different types of counterfactual inferences, where NCs only allow local revision due to their distinct morphology. The core proposal is that models for interpreting counterfactuals must be enriched with causal laws, and that the antecedents of NCs make local changes to a world to remove a fact contributed by the antecedent, potentially violating the model's causal structure. This accounts for the fact that NCs systematically reject non-causal epistemic inferences, while otherwise retaining their paraphrasability with standard counterfactuals.
Semantics and Linguistic Theory 19. doi: 10.3765/salt.v19i0.2538. With Adrian Brasoveanu
The main goal of the paper is to argue that distributive quantificational dependencies in natural language can be established in two different ways: (i) by encapsulating quantification into functions storing quantificational dependencies as a whole, needed to account for one by one-based distributive sentences like The boys recited "The Raven" one by one -- and (ii) by decomposing quantification in such a way that each n-tuple of quantificationally dependent entities is individually stored in a variable assignment and quantifiers are interpreted relative to the entire set of variables assignments that stores quantificational dependencies in this pointwise, assignment-wise manner needed to account for each-based distributive sentences like The boys each recited a different poem.