In this paper, I propose that the distinction between what is at-issue and what is not can be modeled as a distinction between two components of assertion. These two components affect the common ground in different ways. The at-issue component of an assertion, which is negotiable, is treated as a proposal to update the common ground. The not-at-issue component of an assertion, which is not negotiable, is added directly to the common ground. Evidence for this proposal comes from evidentials, which (...) I argue grammaticize this distinction. It has been observed that sentences with evidentials make both an ‘evidential’ and a ‘propositional’ contribution (Faller 2002, 2006, Matthewson et al. 2008). The evidential contribution is not directly challengeable or up for negotiation. In contrast, the propositional contribution, the ‘main point’ of the sentence, is directly challengeable and up for negotiation. I analyze these two contributions of evidentials as the not-at-issue component of assertion and the at-issue component of assertion, respectively. Supporting data comes from Cheyenne, a language with evidentials that are part of the illocutionary mood paradigm. (shrink)

Do you know who you are? If the question seems unclear, it might owe to the notion of ‘knowing-wh’ (knowing-who, knowing-what, knowing-when, etc.). Such knowledge contrasts with ‘knowing-that’, the more familiar topic of epistemologists. But these days, knowing-wh is receiving more attention than ever, and here we will survey three current debates on the nature of knowing-wh. These debates concern, respectively, (1) whether all knowing-wh is reducible to knowing-that (‘generalized intellectualism’), (2) whether all knowing-wh is relativized to a contrast proposition (...) (‘contrastivism about knowing-wh’), and (3) whether the context-sensitivity of knowing-wh is a semantic or purely pragmatic phenomenon (‘contextualism vs. invariantism about knowing-wh’). (shrink)

In "Knowing the Answer" I argued that knowledge-wh is question-relative. For example, to know when the movie starts is to know the answer p to the question Q of when the movie starts. Berit Brogaard and Jesper Kallestrup have each responded with insightful critiques of my argument, and novel accounts of knowledge-wh. I am grateful to them both for continuing the discussion in so thoughtful a way.

In the logical, philosophical and linguistic literature, a number of theoretical frameworks have been proposed for the meaning of questions (see Ginzburg (1995), Groenendijk & Stokhof (1997) for recent overviews). I will concentrate on two general approaches that figured prominently in linguistic semantics, which I will call the proposition set approach and the structured meaning approach (sometimes called the “propositional” and the “categorial” or “functional” approach). I will show that the proposition set approach runs into three problems: It does not (...) always predict the right focus structure in answers, it is unable to distinguish between polarity (yes/no) and a certain type of alternative questions, and it does not allow to formulate an important condition for a type of multiple constituent questions. On the other hand, I will show that the main argument brought forward against the structured meaning framework, namely that it does not give us an elegant way to account for embedded questions, does not withstand closer scrutiny. In this I will take up an issue raised in von Stechow (1990), namely, that the greater expressive power of the structured meaning approach might be necessary for the proper treatment of semantic phenomena like question formation and focusation. (shrink)

Suppose that Ann says, “Keith knows that the bank will be open tomorrow.” Her audience may well agree. Her knowledge ascription may seem true. But now suppose that Ben—in a different context—also says “Keith knows that the bank will be open tomorrow.” His audience may well disagree. His knowledge ascription may seem false. Indeed, a number of philosophers have claimed that people’s intuitions about knowledge ascriptions are context sensitive, in the sense that the very same knowledge ascription can seem true (...) in one conversational context but false in another. This purported fact about people’s intuitions serves as one of the main pieces of evidence for epistemic contextualism. (shrink)

Noh (1998a, b) analyzes echo questions in terms of metarepresentation and pragmatic enrichment within the framework of Relevance Theory. This paper argues that while the basic idea of metarepresentational analysis seems correct, it is better implemented differently. The alternative analysis proposed in this paper consists of three claims: first, echo questions are metarepresentational with rising intonation, with the rise alone conferring the question status; second, echo questions question the pragmatically enriched attribution; third, the focus of metarepresentation is to be distinguished (...) from the rest of the metarepresentation. This version of metarepresentational analysis reveals why echo questions behave the way they do, both syntactically and semantically/pragmatically. At the same time, this analysis captures not only similarities but also differences between echo questions and interrogatives on a principled basis. (shrink)

How should one understand knowledge-wh ascriptions? That is, how should one understand claims such as ‘‘I know where the car is parked,’’ which feature an interrogative complement? The received view is that knowledge-wh reduces to knowledge that p, where p happens to be the answer to the question Q denoted by the wh-clause. I will argue that knowledge-wh includes the question—to know-wh is to know that p, as the answer to Q. I will then argue that knowledge-that includes a contextually (...) implicit question. I will conclude that knowledge is a question-relative state. Knowing is knowing the answer, and whether one knows the answer depends (in part) on the question. (shrink)

Traditional mechanistic accounts of language processing derive almost entirely from the study of monologue. Yet, the most natural and basic form of language use is dialogue. As a result, these accounts may only offer limited theories of the mechanisms that underlie language processing in general. We propose a mechanistic account of dialogue, the interactive alignment account, and use it to derive a number of predictions about basic language processes. The account assumes that, in dialogue, the linguistic representations employed by the (...) interlocutors become aligned at many levels, as a result of a largely automatic process. This process greatly simplifies production and comprehension in dialogue. After considering the evidence for the interactive alignment model, we concentrate on three aspects of processing that follow from it. It makes use of a simple interactive inference mechanism, enables the development of local dialogue routines that greatly simplify language processing, and explains the origins of self-monitoring in production. We consider the need for a grammatical framework that is designed to deal with language in dialogue rather than monologue, and discuss a range of implications of the account. Key Words: common ground; dialogue; dialogue routines; language comprehension; language production; monitoring; perception-behavior link. (shrink)

How must knowledge be formed, if made in the image of assertion? That is, given that knowledge plays the normative role of governing what one may assert, what can be inferred about the structure of the knowledge relation from this role? I will argue that what one may assert is sensitive to the question under discussion, and conclude that what one knows must be relative to a question. In short, knowledge in the image of assertion is question-relative knowledge.

In biscuit conditionals (BCs) such as If you’re hungry, there’s pizza in the fridge, the if clause appears to apply to the illocutionary act performed in uttering the main clause, rather than to its propositional content. Accordingly, previous analyses of BCs have focused on illocutionary acts, and, this, I argue, leads them to yield incorrect paraphrases. I propose, instead, that BCs involve existential quantification over potential literal acts such as assertions, questions, commands, and exclamations, the semantic objects associated with declarative, (...) interrogative, imperative, and exclamative sentences, respectively. Such an existential interpretation of BCs requires only that we add potential literal acts to our inventory of individuals, and it produces reasonable paraphrases in which if has its normal meaning: If you’re hungry,[there’s a (relevant/salient) assertion that] there’s pizza in the fridge. These potential literal act variables are introduced into semantic interpretations and then undergo Existential Closure. Hence, we would expect to see similar interpretations in contexts other than BCs, that is, with other if constructions, with connectives other than if, with potential literal acts other than assertion, and in root sentences. This prediction is borne out, along with the parallel prediction that we cannot quantify over purely illocutionary acts like offers, but only over potential literal acts, those conventionally associated with a particular morphosyntactic shape. (shrink)

The notion of kontrast, or the ability of certain linguistic expressions to generate a set of alternatives, originally proposed by Vallduví and Vilkuna (1998) as a clause-level concept, is re-analyzed here as connecting the level of information packaging in the clause and the level of discourse structure in the following way: kontrast is encoded at the clausal level but has repercussions for discourse structure. This claim is supported by evidence from the distribution properties of three colloquial Russian particles -to, e, (...) andved' which are analyzed as unambiguous markers of kontrast. Both the placement of these particles at the clausal level and their role in discourse are viewed as consequences of the type of the kontrast set and the cognitive status of information marked by each particle. (shrink)

The paper investigates an elliptical construction, Clarification Ellipsis, that occurs in dialogue. We suggest that this provides data that demonstrates that updates resulting from utterances cannot be defined in purely semantic terms, contrary to the prevailing assumptions of existing approaches to dynamic semantics. We offer a computationally oriented analysis of the resolution of ellipsis in certain cases of dialogue clarification. We show that this goes beyond standard techniques used in anaphora and ellipsis resolution and requires operations on highly structured, linguistically (...) heterogeneous representations. We characterize these operations and the representations on which they operate. We offer an analysis couched in a version of Head-Driven Phrase Structure Grammar combined with a theory of information states (IS) in dialogue. We sketch an algorithm for the process of utterance integration in IS which leads to grounding or clarification. The account proposed here has direct applications to the theory of attitude reports, an issue which is explored briefly in the concluding remarks of the paper. (shrink)

This compilation of references includes all references for the knowledge-how chapters included in Bengson & Moffett's edited volume. The volume and the compilation of references may serve as a good starting point for people who are unfamiliar with the philosophical literature on knowledge-how.

This paper considers some puzzling knowledge ascriptions and argues that they present prima facie counterexamples to credence, belief, and justification conditions on knowledge, as well as to many of the standard meta-semantic assumptions about the context-sensitivity of ‘know’. It argues that these ascriptions provide new evidence in favor of contextualist theories of knowledge—in particular those that take the interpretation of ‘know’ to be sensitive to the mechanisms of constraint.

Clarification requests, queries posed in response to a “problematic” (misheard, misunderstood, etc.) utterance, are a challenge to mainstream semantic theories because they call into question notions such as “shared content” or “the context.” Given their strong parallelism requirements, elliptical clarification requests introduce in addition significant complexities concerning the need for long-term maintenance of non-semantic information in context. In this paper, we consider a puzzle concerning the emergence of elliptical clarification requests in child English: Data from the Belfast and Manchester corpora (...) from CHILDES demonstrate that reprise fragments, the highest frequency clarification request construction among adults, emerges with significant delay in comparison with reprise sluices, bare wh-phrases used to request clarification. This is a puzzling finding a priori: first, since reprise fragments are by most plausible measures the most readily available clarification request form in that it involves mere repetition of material primed by the previous utterance and children have the ability to repeat parts of the previous utterance from the earliest stages of speech. Moreover, as we show, reprise fragments predominate—in some cases vastly so, over sluicing in the input of fragmentary clarification requests available to the child—and it would, therefore, be difficult to construct an explanation for order of emergence primarily based on frequency of the construction in the children's input. Our account is based on grammatical explication of the difference between the constructions, how constructional difference is represented in a hierarchy of constructions, and in terms of a notion of semantic complexity that suggests reprise sluicing is, in fact, less complex than reprise fragments. (shrink)

On one view, the point of an assertion is to update the common ground (Stalnaker 1978, Karttunen 1974). On another, the point of an assertion is to propose an update to the com- mon ground (Groenendijk 2009, Mascarenhas 2009, and related work on the structure of discourse, e.g., Ginzburg 1996, Roberts 1996, Gunlogson 2001). In Murray (to appear), I merge these two views. I argue based on evidence from declarative sentences with eviden- tials that assertion has two components: what is (...) at-issue and what is not. The not-at-issue component of assertion is added directly to the common ground while the at-issue compo- nent is proposed to be added to the common ground. Here, I extend this analysis to yes/no questions in Cheyenne and their interaction with evidentials. I propose that the distinction between what is at-issue and what is not is also present in questions, and that it can be modeled in the same way. Specifically, both declarative and interrogative sentences make two contributions: they restrict and structure the common ground. The restriction is based on the not-at-issue component while the structuring relation is based on the at-issue component. (shrink)

The distinction between a merely ‘rigidifying’ dthat and a directly-referential take on dthat-terms is well known, and is explicitly highlighted by Kaplan in Afterthoughts, his 1989 commentary on Demonstratives. What is not equally widely recognized is that Afterthoughts also oscillates between three different directly referential proposals. This essay discusses the semantic and philosophical implications of these different directly-referential interpretations of ‘dthat’, paying particular attention to the relationships between syntactic and propositional structure, the structure and makeup of contexts in the semantics (...) of indexical languages, the significance of context shifting devices and of so-called operators on character, and the aims and scope of propositional semantics. (shrink)

Why can I not appropriately utter ‘It must be raining’ while standing outside in the rain, even though every world consistent with my knowledge is one in which it is raining? The common response to this problem is to hold that epistemic must, in addition to quantifying over epistemic possibilities, carries some additional evidential information concerning the source of one'S evidence. I argue that this is a mistake: epistemic modals are mere quantifiers over epistemic possibilities. My central claim is that (...) the seeming anomaly of the data above arises from a mistaken conception of what a possibility is. Instead of conceiving of possibilities as possible worlds, I argue that we should conceive of possibilities as answers to open questions. (shrink)

We present an implementation of a discourse parsing system for alexicalized Tree-Adjoining Grammar for discourse, specifying the integrationof sentence and discourse level processing. Our system is based on theassumption that the compositional aspects of semantics at thediscourse level parallel those at the sentence level. This coupling isachieved by factoring away inferential semantics and anaphoric features ofdiscourse connectives. Computationally, this parallelism is achievedbecause both the sentence and discourse grammar are LTAG-based and the sameparser works at both levels. The approach to an (...) LTAG for discourse has beendeveloped by Webber and colleagues in some recent papers. Our system takes a discourseas input, parses the sentences individually, extracts the basic discourseconstituent units from the sentence derivations, and reparses the discoursewith reference to the discourse grammar while using the same parser usedat the sentence level. (shrink)

Knowledge ascriptions seem context sensitive. Yet it is widely thought that epistemic contextualism does not have a plausible semantic implementation. We aim to overcome this concern by articulating and defending an explicit contextualist semantics for ‘know,’ which integrates a fairly orthodox contextualist conception of knowledge as the elimination of the relevant alternatives, with a fairly orthodox “Amherst” semantics for A-quantification over a contextually variable domain of situations. Whatever problems epistemic contextualism might face, lack of an orthodox semantic implementation is not (...) among them. (shrink)

Many languages grammatically mark evidentiality, i.e., the source of information. In assertions, evidentials indicate the source of information of the speaker while in questions they indicate the expected source of information of the addressee. This dissertation examines the semantics and pragmatics of evidentiality and illocutionary mood, set within formal theories of meaning and discourse. The empirical focus is the evidential system of Cheyenne (Algonquian: Montana), which is analyzed based on several years of fieldwork by the author.

This paper discusses three potential varieties of update: updates to the common ground, structuring updates, and updates that introduce discourse referents. These different types of update are used to model different aspects of natural language phenomena. Not-at-issue information directly updates the common ground. The illocutionary mood of a sentence structures the context. Other updates introduce discourse referents of various types, including propositional discourse referents for at-issue information. Distinguishing these types of update allows a unified treatment of a broad range of (...) phenomena, including the grammatical evidentials found in Cheyenne (Algonquian) as well as English evidential parentheticals, appositives, and mood marking. An update semantics that can formalize all of these varieties of update is given, integrating the different kinds of semantic contributions into a single representation of meaning. (shrink)

How should the contrastivist formulate closure? That is, given that knowledge is a ternary contrastive state Kspq (s knows that p rather than q), how does this state extend under entailment? In what follows, I will identify adequacy conditions for closure, criticize the extant invariantist and contextualist closure schemas, and provide a contrastive schema based on the idea of extending answers. I will conclude that only the contrastivist can adequately formulate closure.

Interaction mining is about discovering and extracting insightful information from digital conversations, namely those human?human information exchanges mediated by digital network technology. We present in this article a computational model of natural arguments and its implementation for the automatic argumentative analysis of digital conversations, which allows us to produce relevant information to build interaction business analytics applications overcoming the limitations of standard text mining and information retrieval technology. Applications include advanced visualisations and abstractive summaries.

We examine a variety of dialogue protocols, taking inspiration from two fields: natural language dialogue modelling and multiagent systems. In communicative interaction, one can identify different features that may increase the complexity of the dialogue structure. This motivates a hierarchy of abstract models for protocols that takes as a starting point protocols based on deterministic finite automata. From there, we proceed by looking at particular examples that justify either an enrichment or a restriction of the initial model.

Within the conceptual framework of comparative discourse linguistics, this article examines the similarities and the differences between Anglo-American and French Usenet newsgroups and builds hypotheses about the implicit rules which govern the respective ethnolinguistic communities. The comparison is focused on cultural perceptions of questions and explanations occurring in the messages posted in these newsgroups. Both discursive operations have different conversational functions in the Anglo-American and the French messages. Their cross-cultural analysis shows that within the Anglo-American newsgroup, the purpose of the (...) discussion is to get a better comprehension of the complexity of issues, whereas the purpose within the French newsgroup is to maintain an ordinary conversation between equals. (shrink)

I show how a conversational process that takes simple, intuitively meaningful steps may be understood as a sophisticated computation that derives the richly detailed, complex representations implicit in our knowledge of language. To develop the account, I argue that natural language is structured in a way that lets us formalize grammatical knowledge precisely in terms of rich primitives of interpretation. Primitives of interpretation can be correctly viewed intentionally, as explanations of our choices of linguistic actions; the model therefore fits our (...) intuitions about meaning in conversation. Nevertheless, interpretations for complex utterances can be built from these primitives by simple operations of grammatical derivation. In bridging analyses of meaning at semantic and symbol‐processing levels, this account underscores the fundamental place for computation in the cognitive science of language use. (shrink)

There are many ways in which language can describe the dependency of one occurrence on another and hence many varieties of conditional construction, including conditionals in ‘if’, ‘when’, ‘since’, and ‘as’, the absolutive conditionals of Stump (1985), and the correlative conditional construction (‘the more, the merrier’) discussed in Fillmore (1986). This paper will be concerned with investigating one species illustrated in (1a) and (1b).