Wh-the-hell as a polarity-insensitive, speaker-oriented domain restrictor

I argue for a reanalysis of "wh-the-hell" questions as restrictions on the domain of alternatives, rather than polarity items or domain extensions. This analysis returns to their original characterization as "aggressively non-D-linked" by specifying what it means to be D-linked, and in doing so captures a number of otherwise unconnected properties of such questions without positing extraneous features. Consequences of the proposal include a new notion of question-specific contextuality, evidence for an inquisitive semantic condition on sluicing, and a potential general discourse restriction on wh-movement.

A revisit to the [Num-Cl-Modifier-de-N]/ [Modifier-de-Num-Cl‑N] variation in Mandarin Chinese

This paper investigates the licensing condition of the [Num(eral)-Cl(assifier)-Mod(ifier)-de-N(oun)] / [Mod-de-Num-Cl‑N] variation in Mandarin Chinese. It is observed that this variation represents a complex interface phenomenon in the nominal domain, which is subject to the semantic condition concerning the i(ndividual)-level/​s(tage)-level nature of the modifier contained on the one hand, and the discourse-related condition concerning contrastive topic (ct) on the other. Based on this, at the syntax-semantics interface level, this paper proposes a division of the syntactic domain of adnominal modification to account for the discrepancy between i‑level and s‑level modifiers in terms of their capability in forming [Mod-de-Num-Cl‑N] in the neutral context. In the meanwhile, at the syntax-discourse interface level, in light of the interface-induced analysis pursued by Neeleman & Van de Koot (2008) and Horvath (2010), it is claimed that the word order of [Mod-de-Num-Cl‑N] could be adopted as a linguistic device to encode ct within the nominal domain in Mandarin Chinese, in which case the ordering of [Mod-de-Num-Cl‑N] is licensed for the purpose of establishing a transparent mapping between syntactic configuration and information structure.

Ways of being and expressivity

In this paper, I present a hermeneutic version of ontological pluralism, addressing the question of the discursive articulation of ways of being. The first section presents the notion of a pluralism of ways of being as a restriction of an ontological monism. The second section puts forward a criticism of Kris McDaniel’s proposal of understanding ways of being as kinds of quantifiers. The third section analyses the notion of way of being as a modal concept, explaining ways of being as internal possibilities endowed with a normative force regarding the identity-conditions of entities. The fourth one is a statement about the need of developing a pluralist account of the propositional reference to entities based on ontological pluralism. The fifth section deals with the issue of the discursive articulation of ways of being. The two last sections present a hypothesis concerning a semantic condition for an adequate articulation of ways of being. I argue for a kind of finitude-sensitivity in the semantics of the discursive articulation of internal possibilities, which implies the requirement of developing a hermeneutic notion of silence that may properly work in the discursive articulation of ways of being.

REFLEXIVE-INSENSITIVE MODAL LOGICS

We analyze a class of modal logics rendered insensitive to reflexivity by way of a modification to the semantic definition of the modal operator. We explore the extent to which these logics can be characterized, and prove a general completeness theorem on the basis of a translation between normal modal logics and their reflexive-insensitive counterparts. Lastly, we provide a sufficient semantic condition describing when a similarly general soundness result is also available.

Sluicing as anaphora to issues

Since Merchant 2001, it has been widely agreed that the licensing condition on Sluicing is at least partly semantic in nature. This paper argues that the relevant semantic condition is one of symmetric entailment over a semantics which includes not only truth-conditional information, but also issues in the sense of Groenendijk & Roelofsen 2009. One kind of evidence for the proposal comes from expressions like doubly-negated inde?nites and implicit passive agents which do not license Sluicing despite truth-conditional equivalence to overt inde?nites. In addition to these facts, the paper examines novel data which show that Sluicing is not licensed by even overt inde?nites inside of appositive relative clauses, arguing that these facts (and related facts regarding VP-Ellipsis) follow from the account together with an independently motivated semantics for appositives.

Harmonic sluicing: Which remnant/correlate pairs work and why

In this paper, I claim that two generalizations about sluicing in the recent literature are epiphenomenal, emerging from the interaction of independent properties of sluicing and answerhood. The primary focus is on the generalization in Dayal & Schwarzschild 2010, dubbed Antecedent-Correlate Harmony. The empirical pattern is shown to follow from how the semantics of Wh-phrases and correlates interact with Barker’s (in press) Answer Ban generalization, where the antecedent cannot be an answer to the sluiced question. In the latter part of the paper, I shift focus from ACH to the Answer Ban and show how the Answer Ban fills a previously unnoticed gap in Romero’s (1998) empirical coverage of inheritance of content effects in sluicing. The Answer Ban is, in turn, shown to follow straightforwardly from the recent semantic condition on sluicing proposed in AnderBois 2011.

Harmonic sluicing: Which remnant/correlate pairs work and why

In this paper, I claim that two generalizations about sluicing in the recent literature are epiphenomenal, emerging from the interaction of independent properties of sluicing and answerhood. The primary focus is on the generalization in Dayal & Schwarzschild 2010, dubbed Antecedent-Correlate Harmony. The empirical pattern is shown to follow from how the semantics of Wh-phrases and correlates interact with Barker’s (in press) Answer Ban generalization, where the antecedent cannot be an answer to the sluiced question. In the latter part of the paper, I shift focus from ACH to the Answer Ban and show how the Answer Ban fills a previously unnoticed gap in Romero’s (1998) empirical coverage of inheritance of content effects in sluicing. The Answer Ban is, in turn, shown to follow straightforwardly from the recent semantic condition on sluicing proposed in AnderBois 2011.

Voice Mismatch and Syntactic Identity

On the basis of an asymmetry between VP-deletion and pseudogapping, Merchant (2008a) concludes that ellipsis is conditioned by syntax. This article demonstrates that both pseudogapping and VP-deletion potentially allow voice mismatch. The unacceptable cases of voice mismatch in these constructions are attributed to a discourse factor (Kehler 2000, 2002). Nevertheless, since sluicing does not allow voice mismatch (Merchant 2001, 2007) even in the same context that allows voice mismatch in VP-ellipsis, Merchant's (2007, 2008a) conclusion is still valid. A syntactic condition on ellipsis is proposed, based on a semantic condition from Takahashi and Fox 2005. Kehler's discourse explanation is phrased in a revised format.

Sluicing as anaphora to issues

Since Merchant 2001, it has been widely agreed that the licensing condition on Sluicing is at least partly semantic in nature. This paper argues that the relevant semantic condition is one of symmetric entailment over a semantics which includes not only truth-conditional information, but also issues in the sense of Groenendijk & Roelofsen 2009. One kind of evidence for the proposal comes from expressions like doubly-negated inde?nites and implicit passive agents which do not license Sluicing despite truth-conditional equivalence to overt inde?nites. In addition to these facts, the paper examines novel data which show that Sluicing is not licensed by even overt inde?nites inside of appositive relative clauses, arguing that these facts (and related facts regarding VP-Ellipsis) follow from the account together with an independently motivated semantics for appositives.