Thomas Kelly

This chapter addresses the question "How much weight should I give to the opinions of other people in arriving at my own opinions?" More specifically, it evaluates the Thermometer Model, which implies that in two party cases of peer disagreement, substantial conciliation is called for by both parties, ending in states of agnosticism. It subjects this account of the epistemic significance of disagreement to greater scrutiny than it has thus far received. It is particularly concerned with the question of which specific versions of Conciliationism the Thermometer Model supports and which it does not. It argues that such a model does not support the versions of Conciliationism that are most popular among leading Conciliationists, and that the versions that it does support are extremely strong views that we have independent reason to reject.

This chapter explores the challenge that the New Evil Demon Problem poses for so-called “externalist” theories of evidence. It argues that the best view of evidence is a liberalized form of externalism. According to that view, paradigmatic evidence in the sciences and elsewhere includes publicly known facts about the external world, evidence that we would not have in a hypothetical scenario in which we are the victims of radical deception. On the other hand, our evidence is not limited to our knowledge, and for this reason even radically deceived thinkers are often in a position to have justified beliefs about their environment. Given the radical diversity of topics about which we can have justified beliefs, and the extremely wide range of circumstances in which it’s possible for us to hold such beliefs, we should embrace the idea that even radically different things can play the justifying role of evidence.

In contemporary epistemology, a number of different ideas travel under the banner of “evidentialism.” In this paper, I distinguish and offer some critical reflections on three of these ideas: (1) evidentialism as an account of epistemic justification; (2) evidentialism as anti-pragmatism; and (3) evidentialism as ‘evidence-first’ epistemology. I argue that, when evidentialism is offered as an account of epistemic justification, it is best understood as a grounding thesis, as opposed to a thesis about supervenience or as a biconditional, as proposed by Conee and Feldman in their work on the topic.

This article explores the epistemological significance of disagreement in philosophy in the light of some currently prominent theories of disagreement. More specifically, it asks whether the kind of pervasive and intractable disagreement that is characteristic of philosophy warrants a certain kind of skepticism about the subject. Some hold that, given the kind of disagreement found in philosophy, it would be irrational to hold confident views about controversial philosophical questions. According to this line of thought, the rational response to the diversity of opinion within philosophy is that of the philosophical agnostic, who consistently suspends judgment about controversial issues. Against this, it is argued that there is no plausible view about the epistemology of disagreement on which philosophical agnosticism is compelling.

This is a contribution to a book symposium on Bias: A Philosophical Study, in which I respond to commentaries by Gabbrielle Johnson, Daniel Greco, and Selim Berker. In response to Johnson, I argue that many paradigmatic cases of bias are not best understood as involving underdetermination, and I defend my alternative account of bias against the concerns that she raises. In response to Greco, I note some of the ways in which the credibility of my claims depends on further empirical research, and I clarify my claims about introspection in order to show that they are consistent with the possibilities that he raises. In response to Berker, I offer a view about the metaphysical status of “non-pejorative” biases while resisting his suggestion that all non-evaluative uses of the term “bias” are misuses of the term. I defend my proposal that our knowledge that skeptical hypotheses are false is a case of “biased knowing” against a dilemma that he raises for that possibility.

Throughout the history of western philosophy, the Socratic injunction to ‘follow the argument where it leads’ has exerted a powerful attraction. But what is it, exactly, to follow the argument where it leads? I explore this intellectual ideal and offer a modest proposal as to how we should understand it. On my proposal, following the argument where it leaves involves a kind of modalized reasonableness. I then consider the relationship between the ideal and common sense or ‘Moorean’ responses to revisionary philosophical theorizing

Un fait Mooréen, selon l'expression de David Lewis, est « l'une des choses que nous savons mieux que toute prémisse d'une argumentation philosophique visant à é tablir le contraire. » Le sujet des faits Mooréens soulève des questions profondes, à la fois de méthode philosophique et d'épistémologie de premier ordre. Comment devrions-nous répondre aux arguments qui remettent en question des croyances dont nous sommes extrêmement confiants ? Dans quelle mesure ces arguments – ou plutôt ceux qui les avancent – peuvent-ils espérer avoir une emprise ou une traction rationnelle sur nous ? Dans quelle mesure, si même c'est le cas, notre point de départ limite-t-il les types de révisions de nos croyances que la philosophie pourrait légitimement inspirer ? Quand sommes-nous justifiés (si jamais nous le sommes) à refuser de se laisser influencer par un argument qui, pour autant qu'on puisse en juger, est sans faille, et quand un tel refus relève-t-il simplement du dogmatisme ? Ce sont là de vastes questions, auxquelles je ne tenterai pas ici de répondre complétement. Mon ambition, plus modeste, est de tenter de progresser sur ces questions, et d'autres questions connexes, en examinant la notion de fait Mooré que je trouve chez Lewis et chez d'autres.

In this paper, I explore the question of whether the expected consequences of holding a belief can affect the rationality of doing so. Special attention is given to various ways in which one might attempt to exert some measure of control over what one believes and the normative status of the beliefs that result from the successful execution of such projects. I argue that the lessons which emerge from thinking about the case ofbelief have important implications for the way we should think about the rationality of a number of other propositional attitudes,such as regret, desire, and fear. Finally,I suggest that a lack of clarity with respect to the relevant issues has given rise to a number of rather serious philosophical mistakes

In this paper, I clarify and defend some of the central ideas of Bias in response to commentators, with a special focus on the theme of skepticism. In response to Michael Veber, I defend the project of offering a modest as opposed to an ambitious response to the skeptic. In response to Jonathan Matheson, I defend my account of the way in which bias attributions function in contexts of interpersonal disagreement, as well as the claim that an unbiased believer will generally be in a stronger position to resist skeptical pressure from disagreement than a biased believer. In response to Brett Sherman, I clarify the way in which my account of bias accommodates the phenomenon of biased suspension of judgment, and I explore some of the connections between bias, suspension of judgment, and skepticism. In response to Jared Celinker and Nathan Ballantyne, I defend the possibility of emergent biases.

The third and final part of Bias: A Philosophical Study explores the connections between bias and some of the central topics of epistemology, including knowledge, skepticism, and disagreement. It defends the possibility of biased knowing: biased believers can sometimes know, even when they believe in accordance with their biases, and even if those biases guarantee that they would believe as they do even if the truth were otherwise. It argues that the possibility of biased knowing has significant implications for both philosophical methodology and traditional forms of skepticism. Extended consideration is given to the question of whether those of us who robustly favor our ordinary commonsense views over the possibilities that the skeptic calls to our attention are biased against skepticism in any objectionable sense. Finally, it critically examines our tendency to attribute bias to others in response to interpersonal disagreement and the epistemological significance of “the bias blind spot.”

In his book Prejudice, Endre Begby argues that people who hold paradigmatically prejudiced beliefs – for example, the belief that women are less adept at math than men – might be fully rational in holding those beliefs. In this article, I argue that Begby fails to provide compelling examples of beliefs that are both rational and prejudiced. On Begby’s account, whether a belief is prejudiced is determined by its content: it follows that any two token beliefs with the same content will either both be prejudiced, or both unprejudiced, regardless of how they differ in other respects. I sketch an alternative account, on which whether a person’s belief counts as prejudiced might be sensitive to a greater range of factors. I then turn to Bebgy’s discussion of ‘evidential preemption,’ a phenomenon by which certain speech acts seem to inoculate themselves from having their contents disconfirmed or falsified by later counterevidence. I argue for skepticism about evidential preemption. To the extent that there is a genuine normative phenomenon in the neighborhood, it is the familiar one of testimonial defeat, in which testimony from one source is neutralized by conflicting testimony from another source that one has reason to think is more reliable.

In principle, any token belief can be evaluated practically as well as epistemically . My thesis is an exploration of what I take to be philosophically interesting issues which arise from reflection on this commonplace. In the first part of the thesis, I explore the question of whether the fact that beliefs can be evaluated practically can have normative upshot: that is, whether practical evaluations can make a difference to what one should believe, or to how one should act. I argue that, although practical considerations cannot rationalize beliefs, they can rationalize acting in ways that lead to the acquisition of practically advantageous beliefs---including beliefs which it would be irrational to hold. In the course of the inquiry, I pursue questions about the extent to which our beliefs are under our voluntary control as well as questions about the distinctive value of being a rational believer, and I attempt to apply the lessons which emerge from my discussion of belief to other propositional attitudes such as regret, fear, and desire. In the second part of the thesis, I develop the hypothesis that the oft-noted 'epistemic unscrupulousness' displayed by great scientists at various key historical junctures is an artifact of the practical advantageousness of holding certain kinds of beliefs in the absence of evidence for those beliefs. I argue that this hypothesis has important implications for contemporary debates concerning the rationality and objectivity of scientific progress. In the course of the discussion, I examine norms of conservatism and inference to the best explanation and suggest that, in order to appreciate the distinctive virtues of these norms, one must take practical considerations into account. In the third and final part of thesis, I argue against the reductionist view that epistemic goodness is simply a special case of practical goodness, viz. practical goodness in the service of one's cognitive or epistemic aims. I criticize the assumption, common among epistemologists, that there is some general, global cognitive goal relative to which our cognitive practices and efforts are to be assessed. In the more constructive segment of the third part, I contend that although there is a deep and irreducible distinction between epistemic and practical rationality, theoretical reasoning of any significant degree of complexity requires responsiveness to both epistemic and practical reasons, and that theoretical rationality is thus best thought of as a hybrid virtue

This book is a philosophical exploration of bias and our practices of attributing it. It develops and defends the norm-theoretic account of bias, according to which objectionable biases involve systematic departures from objective norms or standards of correctness. It explores the perspectival character of bias attributions, or the ways in which our views about which people and sources of information are biased about a topic are influenced and constrained, both rationally and psychologically, by our views about the topic itself. The book defends a robust pluralism about bias, according to which a radical diversity of things are genuinely biased, with none of these more fundamental than all of the rest. Biases of people are understood as multiply realizable dispositions to depart from objective norms. It offers a novel account of the bias blind spot, or our tendency to fail to see bias in ourselves in a way that we see it in others. It explores the connections between bias and central topics in the theory of knowledge, including truth, knowledge, rationality, reliability, introspection, skepticism, and disagreement. A number of radical conclusions are defended: that both rationality and morality sometimes require us to be biased; that in many cases of disagreement, we are rationally required to view those who disagree with us as biased, even if we know nothing about how they arrived at their views or why they hold them; and that even God could not have made us reliable detectors of our own biases through introspection.

According to one view about the rationality of belief, such rationality is ultimately nothing other than the rationality that one exhibits in taking the means to one’s ends. On this view, epistemic rationality is really a species or special case of instrumental rationality. In particular, epistemic rationality is instrumental rationality in the service of one’s distinctively cognitive or epistemic goals (perhaps: one’s goal of holding true rather than false beliefs). In my (2003), I dubbed this view the instrumentalist conception of epistemic rationality.

In this paper, I explore the distinction between historical and current time slice theories of epistemic justification. I argue that the distinction is best understood in terms of metaphysical grounding, an approach that improves upon more familiar characterizations that appeal to the notion of supervenience. I then take up the question of which kind of theory we should accept. Goldman has long maintained that no current time slice theory can do justice to the epistemic role of preservative memory. I argue that a principle governing preservative memory on which Goldman’s case depends is false, but that Conee and Feldman’s recent attempt to provide a current time slice reconstruction of the relevant phenomena is unsuccessful. Finally, I develop a different argument against current time slice theories, based on cases involving temporally extended reasoning. These cases suggest that facts about a subject’s past can play an indispensable role in determining the status of her present beliefs and thus provide independent grounds for favoring a historical conception of epistemic justification.

According to Peter van Inwagen, there are no successful philosophical arguments for substantive conclusions. He argues for this thesis in two steps. First, he puts forward and defends a “criterion of philosophical success,” according to which a philosophical argument is a success just in case it has the power to convert any ideally rational agnostic to its conclusion. He then argues that, given the kind of disagreement we find among philosophers, we have good reason to think that no philosophical arguments for substantive conclusions satisfy that criterion. This paper contests van Inwagen’s case at both steps: first, it offers objections to his proposed criterion of philosophical success; second, it argues that even if we are wrong and his criterion is correct, the kind of disagreement that we find among actual philosophers does not provide a good reason to think that no philosophical arguments meet the relevant standard.