The notion of common ground is important for the production of referring expressions: In order for a referring expression to be felicitous, it has to be based on shared information. But determining what information is shared and what information is privileged may require gathering information from multiple sources, and constantly coordinating and updating them, which might be computationally too intensive to affect the earliest moments of production. Previous work has found that speakers produce overinformative referring expressions, which include privileged names, (...) violating Grice’s Maxims, and concluded that this is because they do not mark the distinction between shared and privileged information. We demonstrate that speakers are in fact quite effective in marking this distinction in the form of their utterances. Nonetheless, under certain circumstances, speakers choose to overspecify privileged names. (shrink)

While speakers have been shown to adapt to the knowledge state of their addressee in choosing referring expressions, they often also show some egocentric tendencies. The current paper aims to provide an explanation for this “mixed” behavior by presenting a model that derives such patterns from the probabilistic combination of both the speaker's and the addressee's perspectives. To test our model, we conducted a language production experiment, in which participants had to refer to objects in a context that also included (...) a visually misleading object whose function was either known to both partners, or known just to the speaker but not the addressee. Modeling results indicate that the experimental findings cannot be explained by assuming that speakers tailor a referring expression solely to their own perspective or to the perspective of their addressee. Instead, accounting for the behavioral pattern requires an approach where both perspectives influence the choice of referring expressions. Nevertheless, in our situation, speakers consider their partner's perspective more than their own. (shrink)

From Higgins (1973) to Iatridou and Varlokosta (1998), connectivity has been considered in the literature to be a defining characteristic of specificational pseudoclefts. This paper argues against this view based on an analysis of specificational pseudoclefts in Hebrew. Pseudoclefts in Hebrew are interesting in two ways. First, predicational and specificational pseudoclefts are distinguished lexically in the choice of the copula. Second, specificational pseudoclefts fall into two classes, each exhibiting a different set of connectivity effects. The connectivity pattern in Hebrew is (...) accounted for following Jacobson (1994) and Sharvit (1999), who assume that connectivity effects are independent of each other and derive them as a by-product of semantic equation. It is shown that this pattern cannot be obtained using any single grammatical operation, such as reconstruction or coping, which would necessarily derive all connectivity effects at once. (shrink)

We investigate the processing of wh questions in Niuean, a VSO ergative–absolutive Polynesian language. We use visual‐world eye tracking to examine how preference for subject or object dependencies is affected (a) by case marking of the subject (ergative vs. absolutive) and object (absolutive vs. oblique), and (b) by the transitivity of the verb (whether the object is obligatory). We find that Niuean exhibits (a) an effect of case, whereby dependencies of arguments with absolutive case (whether subjects or objects) are preferred (...) over dependencies of arguments with ergative or oblique case, and (b) an effect of transitivity, whereby dependencies of obligatory objects (i.e., of transitive verbs) are preferred over dependencies of optional objects (i.e., of intransitive verbs). These results constitute evidence against theories that appeal to a universal subject advantage, or to the linear distance between filler and gap. Instead, the effect of case is consistent with a frequency‐based account: Because absolutive case has a wider syntactic distribution than ergative or oblique, absolutive dependencies are easier to process. The effect of transitivity reflects sensitivity of the parser to whether or not an argument is obligatory. We propose that these two strategies could be unified if the parser prefers dependencies with arguments that are more likely to materialize. (shrink)

Inspired by early proposals in philosophy, dominant accounts of language posit a central role for mutual knowledge, either encoded directly in common ground, or approximated through other cognitive mechanisms. Using existing empirical evidence from language and memory, we challenge this tradition, arguing that mutual knowledge captures only a subset of the mental states needed to support communication. In a novel theoretical proposal, we argue for a cognitive architecture that includes separate, distinct representations of the self and other, and a cognitive (...) process that compares these representations continuously during conversation, outputting both similarities and differences in perspective. Our theory accounts for existing data, interfaces with findings from other cognitive domains, and makes novel predictions about the role of perspective in language use. We term this new account the Multiple Perspectives Theory of mental states in communication. (shrink)

This article discusses two puzzles regarding identity questions: (i) certain definites cannot occur in the post-copular position of identity questions; and (ii) the same definites are the only possible answers to identity questions with post-copular names. We demonstrate that the range of these definites crucially depends on interlocutors' shared assumptions about how entities in the physical surroundings are perceived and categorized. We propose that these definites are directly referential in the sense of Kaplan (1989a,b), and only contribute the referent itself (...) to the semantic composition. To explain the asymmetry between perceptually grounded descriptions and proper names, we draw on Gupta's (1980) framework of relative identity. This analysis suggests that direct reference is not always determined lexically, but is—at least in part—a pragmatic phenomenon. More generally, this phenomenon shows that natural language is sensitive to the source of information in the common ground. (shrink)

An -ever free relative is felicitous only when the speaker doesn’t know, or doesn’t care about, the identity of the entity denoted. In this paper we investigate what it means to identify an entity by examining the non-identification condition on -ever free relatives. Following Dayal (In A. Lawson (Ed.), Proceedings of SALT VII, 1997 ), we analyze -ever free relatives as definites with a modal dimension. We show that the variation in the identity of the entity across the possible worlds (...) in the modal dimension cannot be captured in a model where transworld identity is expressed using a single trivial principle of identity, and present an analysis within a model where transworld identity is relativized to noun meanings, which has been proposed in the philosophical literature for other reasons (Geach 1968 ; Gupta, The logic of common nouns: an investigation in quantified modal logic, 1980 ). The analysis thus shows that natural language semantics is sensitive to relative identity in the sense of Geach and Gupta; furthermore, it sets the stage for a new typology of referring expressions based on which expression types contribute principles of transworld identity. (shrink)