Janice Dowell | Syracuse University (original) (raw)

Papers by Janice Dowell

Inquiry

In Context and Coherence (2021), Una Stojnic defends a conventionalist account of context-sens... more In Context and Coherence (2021), Una Stojnic defends a conventionalist account of context-sensitivity on which interpretations are resolved by linguistic convention alone. Her target is the standard, Gricean account, which accords reasoning about speakers' intentions a central role. Central to her argument is the observation that much interpretation is resolved automatically, a phenomenon that Gricean accounts are ill-suited to explain.
Her arguments for Conventionalism over Gricean Intentionalism are largely persuasive. However, a full defense of Conventionalism requires comparison to non-Gricean, non-conventionalist accounts. Here I sketch and assess the comparative merits of a novel objectivist account. Objectivist accounts accord extra-linguistic, extra-intentional features of a speech situation a role to play in resolving interpretation. After identifying several gaps in Stojnic's account, I show how Objectivism is able to explain the automaticity of interpretation, as well as to provide fuller explanations of several of the cases she discusses.

Oxford Handbook of Applied Philosophy of Language, 2024

Silencing occurs when a conversational participant is not accorded an ordinary speaker’s standing... more Silencing occurs when a conversational participant is not accorded an ordinary speaker’s standing to update the conversational record. Here I argue that silencing poses a distinctive problem for assertion in contexts of inquiry. In the standard model, an assertion met with silence is one that has not been rejected by any participant. Like affirmation, silence adds the asserted content to the Common Ground. When a conversational participant is silent because she has been silenced, however, her silence does not indicate assent. In such cases, assertions cannot succeed in their characteristic perlocutionary effect, mutual audience belief. The remedy for this distinctive problem of assertion requires a remedy for the conditions that give rise to silencing. I close with a proposal for how to revise the standard model to allow it to represent the phenomenon of silencing, as well as the conditions needed for repair.

Reading Parfit, 2017

We argue that Parfit's "Triviality Objection" against some naturalistic views of no... more We argue that Parfit's "Triviality Objection" against some naturalistic views of normativity is not compelling. We think that once one accepts, as one should, that identity statements can be informative in virtue of their pragmatics and not only in virtue of their semantics, Parfit's case against naturalism can be overcome

Speech can be used to change societies in bad ways. It supports institutional oppression, establi... more Speech can be used to change societies in bad ways. It supports institutional oppression, establishes new oppressive norms, silences opponents, spreads disinformation and propagates feelings of hate. Online communities magnify the effects of individual speech acts. We'll look at social norms and institutions, silencing and free speech, social meaning, norm-shifting and disinformation. We'll seek answers to how oppressive speech works and how to defend against it.

The Routledge Handbook of Epistemic Contextualism, 2017

Oxford Studies in Metaethics, 2016

What types of consideration place genuine constraints on an adequate semantics for normative and ... more What types of consideration place genuine constraints on an adequate semantics for normative and evaluative expressions? Data points linguists recognize include facts about ordinary uses of such expressions and competent speakers’ judgments about which such uses are acceptable and appropriate. The contemporary literature in metaethics reflects the widespread assumption that there is an additional source of data linguists don’t rely upon, however, namely, competent speakers’ judgments about the possibility of disagreement with hypothetical speech communities. Here I argue that we have as yet no good grounds for thinking such judgments are probative for the purposes of semantic theorizing. Indeed, we have at least several good reasons to think they are not. For these reasons, we should accord these judgments no probative value for the development of a semantics for our moral terms. If correct, this conclusion has important consequences. First, it means that these judgments can no longer be presumed to put pressure on theories according to which our moral expressions share a semantics with ordinary, descriptive terms. Second, many rivals to pure, descriptivist theories count among their advantages the ability to accommodate these judgments. If we should accord those judgments no probative value, such theories lose an important source of their support.

Res Philosophica, 2017

Recent challenges to Angelika Kratzer’s canonical contextualist semantics for modal expressions a... more Recent challenges to Angelika Kratzer’s canonical contextualist semantics for modal expressions are united by a shared methodological practice: Each requires the assessment of the truth or warrant of a sentence in a scenario. The default evidential status accorded these judgments is a constraining one: It is assumed that, to be plausible, a semantic hypothesis must vindicate the reported judgments. Fully assessing the extent to which these cases do generate data that puts pressure on the canonical semantics, then, requires an understanding of this methodological practice. Here I argue that not all assessments are fit to play this evidential role. To play it, we need reason to think that speakers’ assessments can be reasonably expected to be reliable. Minimally, having such grounds requires that assessments are given against the background of non-defectively characterized points of evaluation. Assessing MacFarlane’s (2014) central challenge case to contextualism about deontic modals in light of this constraint shows that his judgments do not have the needed evidential significance. In addition, new experimental data shows that once the needed scenario is characterized non-defectively, none of the resulting range of cases provides data that cannot be accommodated by a Kratzer-style contextualism.

Philosophical Studies, 2008

Philosophical Studies, 2007

Philosophical Studies, 2006

Australasian Journal of Philosophy, 2008

One strategy for blocking Chalmers's overall case against physicalism has been to deny his cl... more One strategy for blocking Chalmers's overall case against physicalism has been to deny his claim that showing that phenomenal properties are in some sense physical requires an a priori entailment of the phenomenal truths from the physical ones. Here I avoid this well-trodden ground and argue instead that an a priori entailment of the phenomenal truths from the physical ones does not require an analysis in the Jackson/Chalmers sense. This is to sever the dualist's link between conceptual analysis and a priori entailment by showing that the lack of the former does not imply the absence of the latter. Moreover, given the role of the argument from conceptual analysis in Chalmers's overall case for dualism, undermining that argument effectively undermines that case as a whole in a way that, I'll argue, undermining the conceivability arguments as stand-alone arguments does not.

Oxford Studies in Metaethics Volume 7, 2012

Philosophical Topics, 2007

Inquiry

Oxford Handbook of Applied Philosophy of Language, 2024

Reading Parfit, 2017

The Routledge Handbook of Epistemic Contextualism, 2017

Oxford Studies in Metaethics, 2016

Res Philosophica, 2017

Philosophical Studies, 2008

Philosophical Studies, 2007

Philosophical Studies, 2006

Australasian Journal of Philosophy, 2008

Oxford Studies in Metaethics Volume 7, 2012

Philosophical Topics, 2007

Discourse and Coherence

Discourse coherence relations have been invoked to explain how context-sensitive expressions get... more Discourse coherence relations have been invoked to explain how context-sensitive expressions get their values determined, for example, how anaphoric pronoun reference gets resolved over a discourse (Asher and Lascarides (2003), Stojnic (2021)) and how modal domains get determined (Stone (1997), Stojnic (2021)). An open, more foundational question, though, is how these relations themselves are determined. Here I defend a novel proposal on which what I call “Ritualized Situations” have a central role to play in determining discourse relations and thus in settling the values of context-sensitive expressions.

Ritualized Situations, in the sense reserved here, are stereotypical situations organized or governed by rituals, rules, habits, practices, or laws. I argue that the joint recognition of a Ritualized Situation can play a central role in determining a context’s discourse goals when none have been explicitly mentioned. Roughly, the order of ideas is this: Interlocutors’ joint recognition of a Ritualized Situation brings certain Common Ground propositions to prominence and may add a discourse goal to the Conversational Scoreboard. These together may determine specific coherence relations between utterances and resolve the values of any context-sensitive expressions deployed. Thus, Ritualized Situations have an important role to play in resolving context-sensitivity and explaining successful communication.

One advantage of this account in contrast to Stojnic’s recent conventionalist account is that it explains how contexts can include the elements needed to resolve semantic values prior to any linguistic exchange. To preview: While my account holds, like Stojnic’s, that discourse relations have a central role to play in resolving the semantic values of context-sensitive expressions, it departs from hers in rejecting the claim that discourse coherence relations are near-universally settled as a matter of convention. Instead, it holds that such relations are often determined to a large extent by the recognition of Ritualized Situations.

After describing and defending this account, I illustrate and minimally test it by applying it to the case of modal domain restriction. The result is a novel, objectivist account of how modal domains are determined as a function of contexts of utterance.

Oxford Studies in Metaethics

Here is a familiar way of framing an ongoing debate about moral language which has long dominated... more Here is a familiar way of framing an ongoing debate about moral language which has long dominated the literature in metaethics. On the one side, we have the idea that moral judgments can be true or false and that our ordinary use of moral language reflects this. When someone tells you “you shouldn’t lie”, they are representing the moral landscape as being a certain way, namely, as prohibiting lying. On the other side, there is the idea that moral (and more broadly, evaluative) language is distinctively practical or action-guiding. When someone tells you “you shouldn’t lie”, they are directing you to refrain from lying. According to the familiar framing, these two features are in tension with one another. The first idea suggests that moral language is descriptive or representational. The second idea suggests that moral language is prescriptive or non-representational. This is where the tension is seen to arise. Ordinary representational language is not action-guiding. So, how could moral language be both?
The first idea is seen as motivating Moral Realism about moral language. Moral Realism as understood here is the view that the content of an utterance of a declarative sentence containing moral vocabulary is representational. The second idea is seen as motivating Expressivism about moral language. Expressivist theories are non-representationalist. According to such theories, moral statements express non-cognitive states of mind.
This familiar framing amounts to thinking in a box. In particular, this framing suggests that Representationalist theories of moral statements can’t account for their action-guidingness, while non-Representationalist theories cannot account for their shared behavioral with profile ordinary, declarative sentences. The central aim of this paper is to show how extant work in natural language semantics and pragmatics can offer avenues for further progress on these long-standing debates by lifting us out of that box, thereby opening up new options that better capture both these phenomena.
To show this, I will narrow our focus in two ways. First, I will focus on Realism and Expressivism about deontic modal expressions in English. Second, I will focus on two of the most prominent challenges to Realism and Expressivism found in the metaethics literature. Each challenge aims to support one side of the debate. The first challenge rests on the acceptance of Judgment Internalism. Roughly, Judgment Internalism is the thesis that sincerely made moral statements require some motivation to comply. This suggests that the use of deontic modal sentences is prescriptive and so, according to the familiar framing, tells against Realism. Here I’ll show that Judgment Internalism is not best thought of as placing a constraint on a semantics for deontic modal sentences and so does not tell against Representationalism. Rather, the real linguistic phenomenon that proponents of Judgment Internalism have drawn our attention to is a feature of discourses, rather than individual sentences. I call the thesis that better captures the relevant phenomena overall “Deliberative Discourse Internalism”. As we’ll see, the reasons for preferring Deliberative Discourse Internalism are themselves linguistic and so quite independent of the metaethical debate between Representationalists and Expressivists. With that new constraint in hand, we’ll see how a plausible Representational semantics may account for the action-guidingness of moral language.
The second challenges rests on the acceptance of the principle of compositionality. This is the principle that the meanings of complex expressions in a natural language are composed of the meanings of the simpler expressions they contain. Some metaethicists have argued that Expressivist theories of moral language are not compositional, undercutting their plausibility. This is widely known as “the Frege-Geach problem” for Expressivism. Here I’ll show how one prominent Expressivist theory for deontic modals avoids this problem. This will show how a compositional semantics for moral statements may be neutral between Representationalism and Expressivism, while being fully Expressivist overall when combined with an Expressivist pragmatics.
That said, there is an adjacent puzzle for Expressivists to account for the observed conversational update effects of embedded uses of deontic modal sentences. Below I’ll show why an extant, compositional Expressivist theory is unable to explain such effects in its current form. I’ll then propose a modification to that account that is explanatory. The result is an Expressivist theory that is both compositional and can account for the observed update effects of embedded deontic modal sentences.

Oxford Handbook of Contemporary Philosophy of Language

Over the last fifteen years, linguists and philosophers of language have reexamined the canoni... more Over the last fifteen years, linguists and philosophers of language have reexamined the canonical, Kratzerian semantics for modal expressions, with special attention paid to their epistemic and deontic uses. This article is an overview of the literature on deontic modal expressions. Section 1 provides an overview of the canonical semantics, noting some of its main advantages. Section 2 introduces a set of desiderata that have achieved the status of fixed points in the debates about whether the canonical semantics is correct. These include the observations that deontic modal sentences have both deliberative and evaluative readings and both information-sensitive and -insensitive readings. Adequate resolutions of certain puzzles in deontic logic and resolving the Frege-Geach problem for Expressivism have also achieved this status.
The third section provides an opinionated overview of some of the main extent rivals to the canonical semantics, including Cariani, Kaufmann, and Kaufmann’s complex contextualism, Yalcin’s Expressivism, Willer’s dynamic semantics, and Starr’s dynamic Expressivism. Section 4 provides an assessment of each of the views discussed in terms of the desiderata introduced in section 2. Section 5 is an overview of remaining issues that require more attention in the literature.

Recently, our judgments about disagreement cases have been accorded a significant evidential role... more Recently, our judgments about disagreement cases have been accorded a significant evidential role in assessing rival semantic theories for deontic modal expressions. In particular, such judgments have been used to motivate rivals to the canonical, contextualist semantics on the grounds that it is insufficiently complex to explain them. This challenge rests on the assumption that a semantics should be able to represent our sense that two parties genuinely disagree at the level of what is said. If that assumption is correct, such judgments have a constraining role to play in assessing the comparative merits of the canonical contextualist versus revisionary semantic proposals.
Here I examine whether that methodological assumption is correct for a particular type of disagreement case, one that prima facie is animated by a dispute over what morality primarily requires. To that end, I consider a variety of disagreements, identifying their hallmarks and using those hallmarks to devise tests to distinguish them. With those tests in hand, I'll show that our disagreement case is one for this methodological assumption is most plausible. A central conclusion here is that rivals to the canonical view that add a parameter for information to the points of evaluation deontic modals are sensitive to undercuts the ability of the resulting views to explain disagreement between those who hold that our primary moral requirements are information-sensitive and those that don’t. If successful, the argument here turns an alleged pro-revisionary challenge on its head, by showing that some cases of disagreement undermine, rather than support, revisions to the canonical view.

On the classical, Kratzer-style semantics for deontic modals, what one ought to do or how th... more On the classical, Kratzer-style semantics for deontic modals, what one ought to do or how things ought to be is tied directly to what one does or how things are in some set of “best” worlds. A number of recent challenges to that semantics have focused on the logical principles it validates. Among such principles are Inheritance, Agglomeration, and Deontic Detachment. Another challenge stems from Sloman’s Principle, which it allegedly does not validate, but, it is argued, a plausible semantics should. Finally, further challenges stem from the claim that there are possibilities, such as the possibility of deontic conflicts and of supererogation, the classical semantics is alleged to foreclose.
Here I will consider some of the cases that motivate these challenges, showing that they do not have their prima facie force.

MAP Session, Eastern APA, 2021