Browse by Volume

A synchronic probability is the probability at a time that an outcome occurs at that very time. Common sense invokes synchronic probabilities with values between 0 and 1 (e.g., the probability right now that the top card of this deck is presently the ace of spades is 1/52), as do scientific theories such as classical statistical mechanics. Recently, philosophers have argued about whether any synchronic probabilities are best interpreted as objective chances. I add to this debate an underappreciated reason we might have to believe in synchronic chance; it might turn out that the best interpretation of our common sense and scientific theories is one in which the macrophysical properties of physical systems are partly determined by synchronic chance distributions over microphysical properties of those systems. Additionally, I argue against the common charge that synchronic probability fails to satisfy various platitudes about chance—most notably Lewis’s (1986) Principal Principle.

For Kant right and ethics are two formally distinct departments of a single morality of reason and freedom. Unlike ethics, right involves an authorization to coerce, and this coercion serves as a pathological incentive. I argue that for Kant the distinctive character of right flows from the fact that juridical obligation has a different relational structure than ethical obligation. I argue that this relational structure explains the connection of right to coercion, and also explains how a categorical imperative can be known a priori to issue in both a pathological and non-pathological incentive. Thus the justification of coercion and its special role as incentive are rooted in the relational character of juridical obligations, and so ultimately in categorical imperatives of reason. Since this pathological incentive has a moral basis in the structure of juridical obligations, and so ultimately in a representation of moral laws, I argue that Kant’s discussion of the juridical incentive of coercion is more continuous with his main discussion of moral incentives in the Critique of Practical Reason than it might at first appear. I illustrate the consequences of this reading by discussing the propensity to injustice, as Kant understands it, and the unique way in which the juridical incentive undermines it.

We ought to treat others’ moral views with respect, even when we disagree. But what does that mean? This paper articulates a moral obligation to make ourselves open to sincere moral persuasion by others. Doing so allows us to participate in valuable relationships of reciprocal respect for agency. Yet this proposal can sound tritely agreeable. To explore its full implications, the paper applies the general obligation to one of the most challenging topics of moral disagreement: the morality of abortion. I consider and reject arguments that abortion decisions have special features exempting them from the obligation to be open to moral persuasion. Further, I argue that viewing fetal ultrasound images can accomplish morally persuasion. Accordingly, in at least some cases a woman seeking abortion has an obligation to view fetal ultrasound images as a means of being open to moral persuasion. However, this conclusion does not support recent laws compelling women seeking abortion to view ultrasound images; such laws are in fact incompatible with the respect for agency that underwrites the obligation to be open to persuasion.

Everyone agrees that not all norms that govern belief and assertion are epistemic. But not enough attention has been paid to distinguishing epistemic norms from others. Norms in general differ from merely evaluative standards in virtue of the fact that it is fitting to hold subjects accountable for violating them, provided they lack an excuse. Different kinds of norm are most readily distinguished by their distinctive mode of accountability. My thesis is roughly that a norm is epistemic if and only if its violation makes it fitting to reduce epistemic trust in the subject, even if there is no doubt about their sincerity, honesty, or other moral virtues. That is, violations of epistemic norms don’t merit resentment or other forms of blame, but rather (other things being equal) deduction of credibility points in internal scorekeeping and related attitudinal and behavioral changes. As Fricker’s work on epistemic injustice shows, such distrust is undesirable from the point of view of an epistemic agent. Consequently, when one manifests epistemic distrust towards a subject in suitable circumstances, it amounts a way of holding her accountable. Since this form of accountability involves no opprobrium, there is good reason to think it is not linked to voluntary control in the same way as moral accountability. Finally, I make use of this account of what makes epistemic norms distinctive to point out some faulty diagnostics in debates about norms of assertion. My aim is not to defend any substantive view, however, but only to offer tools for identifying the right kind of evidence for epistemic norms.

The Univalent Foundations of mathematics take the point of view that all of mathematics can be encoded in terms of spatial notions like "point" and "path". We will argue that this new point of view has important implications for philosophy, and especially for those parts of analytic philosophy that take set theory and first-order logic as their benchmark of rigor. To do so, we will explore the connection between foundations and philosophy, outline what is distinctive about the logic of the Univalent Foundations, and then describe new philosophical theses one can express in terms of this new logic.

The paper presents a kind of normative anti-realist view of epistemology, in the same ballpark as recent versions of expressivism. But the primary focus of the paper is less on this meta-epistemological view itself than on how it should affect ground-level issues in epistemology: for instance, how it should deal with certain forms of skepticism, and how it allows for fundamental revision in epistemic practices (deductive, inductive and perceptual). It is hoped that these methodological consequences will seem attractive independent of the normative anti-realism. Indeed, some normative realists seem to embrace the view on skepticism, but it is argued that their position is unstable: the realism undermines the methodology. The general theme of the paper is that the issue of normative realism is deeply entwined with issues of methodology, in strong contrast to the common claim that meta-epistemological views in the tradition of expressivism have no first order impact.

I argue for a conciliationist treatment of peer disagreement, on the grounds that the evidence that non-conciliatory theorists point to--the evidence that conciliatory-friendly independence principles would rule out--bears a troubling relation to accuracy. Namely, we can anticipate that trying to respond to it is a bad deal with respect to our expected accuracy. I consequently argue that we shouldn't try to respond to it. Instead we should ignore it, and be conciliationists.

Does it count against a normative theory in political philosophy that it is in some important sense infeasible, that its prescriptions are unlikely to be complied with? Though a positive answer seems plausible, it has proved hard to defend against the claim (most forcefully made by David Estlund) that this is not how normative theories work - noncompliance shows a problem with the noncomplying agents, not with the normative theory. I think that this line of thought - this defense of Utopianism - wins the battle but loses the war. It’s right about what does and what does not refute a normative theory. It’s wrong in misidentifying the problem. The right way to think about the feasibility worry is as essentially involving multiple agents, and how expected noncompliance by one agent may refute a normative claim addressed at another. Thus understood, feasibility problems may very well refute a theory in political philosophy. In this paper I develop this understanding of the feasibility worry, tie it to more general discussions in normative ethics (about the morally right way to take into account expected violations by others), and in political philosophy (about ideal and non-ideal theory; a long appendix engages that debate in detail).

We propose a novel account of the distinction between innate and acquired biological traits: biological traits are innate to the degree that they are caused by factors intrinsic to the organism at the time of its origin; they are acquired to the degree that they are caused by factors extrinsic to the organism. This account borrows from recent work on causation in order to make rigorous the notion of quantitative contributions to traits by different factors in development. We avoid the pitfalls of previous accounts and argue that the distinction between innate and acquired traits is scientifically useful. We therefore address not only previous accounts of innateness but also skeptics about any account. The two are linked, in that a better account of innateness also enables us better to address the skeptics.

Being able to ask others to do things, and thereby giving them reasons to do those things, is a prominent feature of our interpersonal lives. In this paper, I discuss the distinctive normative status of requests – what makes them different from commands and demands. I argue for a theory of this normative phenomenon which explains the sense in which the reasons presented in requests are a matter of discretion. This discretionary quality, I argue, is something that other theories cannot accommodate, though it is a significant aspect of the relations that people stand in to one another, and the kinds of practical reasons that flow from those relations.

This paper explores the possibility that Zhuangzi (a pre-Qin Chinese philosopher) can be fruitfully interpreted as a fictionalist. It proceeds in four parts. Part one discusses two distinct and very general types of fictionalism—force and content—that might prove useful for an interpreter of the Zhuangzi. The former type of view would have it that the expressions in question—that is, the expressions that Zhuangzi is held to advocate using and interpreting non-literally—are not best seen as used in a way that aims at, e.g., truth, whereas the latter type of view would have it that the expressions in question are best seen as used in a way that aims at truth, if in a non-literal fashion. Part two surveys evidence in favor of the claim that Zhuangzi can be interpreted in terms of one or the other of these two types of fictionalism and argues that he is better characterized as endorsing a version of the former. Part three explains how interpreting Zhuangzi as a fictionalist can help to resolve notable tensions in the text and briefly explores a few additional merits of this reading of the Zhuangzi: namely, that it can give us a clearer idea of what Zhuangzi’s positive project is, unify seemingly disparate scholarly interpretations of it, and reconcile objectivist and non-objectivist strands in his work. Finally, part four concludes by gesturing toward how the interpretation proposed here might bring the Zhuangzi into productive dialogue with two longstanding philosophical questions: specifically, the question of how we should respond to skeptical (and similar) arguments, and the question of how aesthetic features of works of art—and in particular, literature—might be related to their cognitive or epistemic value (insofar as they have cognitive or epistemic value of an interesting sort).

Just as theory of representation is deficient if it can’t explain how misrepresentation is possible, a theory of computation is deficient if it can’t explain how miscomputation is possible. Nonetheless, philosophers have generally ignored miscomputation. My primary goal is to clarify both what miscomputation is and how to adequately explain it. Miscomputation is a special kind of malfunction: a system miscomputes when it computes in a way that it shouldn’t. To miscompute is to compute. To explain miscomputation, then, you need an adequate account of what the system is computing while malfunctioning. A secondary goal of this paper is to defend an (a quasi-)individualist, mechanistic theory of miscomputation. Computational behavior is narrowly individuated. Computational norms are widely individuated. A system miscomputes when its behavior manifests a narrow computational structure that the widely individuated norms say that it should not have.