A belief is debased when believing is given a basis that is not proper for knowledge, such as wishful thinking or superstition. The possibility of a debasing demon is the possibility of a maximally powerful agent who aims to prevent knowledge by debasing beliefs. Jonathan Schaffer contends that the debasing demon is a threat to all knowledge. Schaffer does not assess the strength of the skeptical challenge from debasing. It is argued here that debasing does not strengthen any case for skepticism. A debasing demon is possible. We should acknowledge that our beliefs could have been debased, and that this could have been done in an introspectively undetectable way. But acknowledging this leaves us in a position to know that our apparent knowledge is genuine. It does not enhance any reason to think that we lack knowledge.

Duncan Pritchard has argued that luck is fundamentally a modal notion: an event is lucky when it occurs in the actual world, but does not occur in more than half of the relevant nearby possible worlds. Jennifer Lackey has provided counterexamples to accounts which, like Pritchard's, only allow for the existence of improbable lucky events. Neil Levy has responded to Lackey by offering a modal account of luck which attempts to respect the intuition that some lucky events occur in more than half of the relevant nearby possible worlds. But his account rejects that events which are as likely as those in Lackey's examples are lucky. Instead, they are merely fortunate. I argue that Levy's argument to this effect fails. I then offer a substitute account of the improbability condition which respects this intuition. This condition says that the relevant notion of probability for luck is epistemic.

Earl Conee and Richard Feldman have recently argued that the evidential support relation should be understood in terms of explanatory coherence: roughly, one's evidence supports a proposition if and only if that proposition is part of the best available explanation of the evidence (2008). Their thesis has been criticized through alleged counterexamples, perhaps the most important of which are cases where a subject has a justified belief about the future (Byerly 2013; Byerly and Martin forthcoming). Kevin McCain has defended the thesis against Byerly's counterexample (2013, 2014a). I argue that McCain's defense is inadequate before pointing toward a more promising solution for explanationism. The Byerly–McCain exchange is important because it casts light on the difficult issues of the standards for justification and the nature of epistemic support. Furthermore, McCain's defense of explanationism about epistemic support represents an important recent development of the burgeoning explanationist program in epistemology and philosophy of science.

According to Selim Berker the prevalence of consequentialism in contemporary epistemology rivals its prevalence in contemporary ethics. Similarly, and more to the point, Berker finds epistemic consequentialism, epitomized by process reliabilism, to be as misguided and problematic as ethical consequentialism. This paper shows how Berker misconstrues process reliabilism and fails to pinpoint any new or substantial defects in it.

Solidarity is supposed to facilitate collective action. We argue that it can also help overcome false consciousness. Groups practice ‘epistemic solidarity’ if they pool information about what is in their true interest and how to vote accordingly. The more numerous ‘Masses’ can in this way overcome the ‘Elites,’ but only if they are minimally confident with whom they share the same interests and only if they are (perhaps only just) better-than-random in voting for the alternative that promotes their interests. Being more cohesive and more competent than the Masses, the Elites can employ the same strategy perhaps all the more effectively. But so long as the Masses practice epistemic solidarity they will almost always win, whether or not the Elites do. By enriching the traditional framework of the Condorcet Jury Theorem with group-specific standards of correctness, we investigate how groups can organize to support the alternatives truly in their interests.

I argue that Alvin Goldman has failed to save process reliabilism from my critique in earlier work of consequentialist or teleological epistemic theories. First, Goldman misconstrues the nature of my challenge: two of the cases he discusses I never claimed to be counterexamples to process reliabilism. Second, Goldman's reply to the type of case I actually claimed to be a counterexample to process reliabilism is unsuccessful. He proposes a variety of responses, but all of them either feature an implausible restriction on process types, or fail to rule out cases with the sort of structure that generates the worry, or both.

According to transmission theories of testimony, a listener's belief in a speaker's testimony can be supported by the speaker's justification for what she says. The most powerful objection to transmission theories is Jennifer Lackey's persistent believer case. I argue that important features about the epistemology of testimony reveal how transmission theories can account for Lackey's case. Specifically, I argue that transmission theorists should hold that transmission happens only if a listener believes a speaker's testimony based on the presumption that the speaker has justification for what she says. If this does not happen in the persistent believer case, then the case is no counterexample to transmission. If this does happen in the persistent believer case, then there is an available framework for rejecting the idea that the listener's belief is in fact justified.

It has been frequently suggested that epistemic contextualists violate the knowledge norm of assertion; by its own lights contextualism cannot be known and hence not be knowingly stated. I have defended contextualists against this objection by showing that it rests on a misunderstanding of their commitments (Freitag 2011, 2012, 2013b). In M. Montminy's and W. Skolits' recent contribution to this journal (2014), their criticism of my solution forms the background against which the authors develop their own. The present reply ventures to demonstrate that their objections are ineffective, since they rest on a confusion of two different ways in which contextualism is unknowable. The precise nature of the original problem will be clarified and my solution briefly restated.

The key concept of this contribution is that of aptness, or success that manifests the agent’s competence. This concept is used in an attempt to gain philosophical understanding not only of knowledge but also of perception and of intentional action.

This paper concerns the relationship between epistemology and radically extended cognition. Radically extended cognition (REC) – as advanced by Andy Clark and David Chalmers – is cognition that is partly located outside the biological boundaries of the cognizing subject. Epistemologists have begun to wonder whether REC has any consequences for theories of knowledge. For instance, while Duncan Pritchard suggests that REC might have implications for which virtue epistemology is acceptable, J. Adam Carter wonders whether REC threatens anti-luck epistemology. In this paper, I argue that the possibility of REC has no systematic consequences for theorizing in epistemology. I suggest an alternative relationship between the two: epistemology can play a role in diagnosing cases of REC. Thus, by establishing that entities partially located outside biological boundaries don't play certain epistemic roles, one can establish that they don't play the related cognitive roles either. I conclude the paper by illustrating this last point.

I outline what I call the ‘deniability problem’, explain why it is problematic, and identify the range of utterances to which it applies (using religious discourse as an example). The problem is as follows: To assign content to many utterances audiences must rely on their contextual knowledge. This generates a lot of scope for error. Thus, speakers are able to make assertions and deny responsibility for the proposition asserted, claiming that the audience made a mistake. I outline the problem (a limited version of which Fricker 2012 discusses), before explaining why it is problematic. Firstly it blocks testimonial knowledge according to assurance views. Secondly it prevents epistemic buck passing (the importance of which is emphasized by Goldberg 2006 and McMyler 2013). Finally, it removes a key disincentive to dishonesty. The recent literature on context sensitivity (particularly Cappelen and Lepore 2004) seems to entail that the problem applies to a very wide range of utterances. I consider a series of responses which fail to provide a solution, but which help us narrow down the scope of the problem.

Pragmatic encroachment offers a picture of knowledge whereby knowledge is unstable. This paper argues that pragmatic encroachment is committed to more instability than has been hitherto noted. One surprising result of the arguments in this paper is that pragmatic encroachment is not merely about changes in stakes. All sorts of practical factors can make for the presence or absence of knowledge on this picture – stakes are just one factor among many that are knowledge-depriving. In this way, the focus in the literature on ‘stakes-sensitivity’ is misleading. Furthermore, insufficient attention has been paid to the variety of ways in which on this view pragmatic factors affect knowledge: pragmatic factors are not merely knowledge-depriving but are also knowledge-inducing.

Offering a solution to the skeptical puzzle is a central aim of Nozick's sensitivity account of knowledge. It is well-known that this account faces serious problems. However, because of its simplicity and its explanatory power, the sensitivity principle has remained attractive and has been subject to numerous modifications, leading to a ‘second wave’ of sensitivity accounts. I will object to these accounts, arguing that sensitivity accounts of knowledge face two problems. First, they deliver a far too heterogeneous picture of higher-level beliefs about the truth or falsity of one's own beliefs. Second, this problem carries over to bootstrapping and Moorean reasoning. Some beliefs formed via bootstrapping or Moorean reasoning are insensitive, but some closely related beliefs in even stronger propositions are sensitive. These heterogeneous results regarding sensitivity do not fit with our intuitions about bootstrapping and Moorean reasoning. Thus, neither Nozick's sensitivity account of knowledge nor any of its modified versions can provide the basis for an argument that bootstrapping and Moorean reasoning are flawed or for an explanation why they seem to be flawed.

In “Normativity and Epistemic Intuitions” (NEI), Weinberg, Nichols and Stich famously argue from empirical data that East Asians and Westerners have different intuitions about Gettier-style cases. We attempted to replicate their study about the Gettier Car Case. Our study used the same methods and case taken verbatim, but sampled an East Asian population 2.5 times greater than NEI's 23 participants. We found no evidence supporting the existence of cross-cultural difference about the intuition concerning the case. Taken together with the failures of both of the existing replication studies (Nagel et al. 2013; Seyedsayamdost 2014), our data provide strong evidence that the purported cross-cultural difference in intuitions about Gettier-style cases does not exist.

A growing number of authors defend putative examples of knowledge from falsehood (KFF), inferential knowledge based in a critical or essential way on false premises, and they argue that KFF has important implications for many areas of epistemology (whether evidence can be false, the Gettier debate, defeasibility theories of knowledge, etc.). I argue, however, that there is no KFF, because in any supposed example either the falsehood does not contribute to the knowledge or the subject lacks knowledge. In particular, I show that if the subject actually has knowledge in putative KFF cases, then there is always a veridical evidential path meeting the basing conditions that accounts for her knowledge; if there is no such path, then the subject is in a type of Gettier case. All the recent arguments that rely on KFF are therefore based on a mistake.

In these comments I discuss the account of action, perception, and knowledge presented in “Mind-World Relations,” with a focus on the implications of this account for the would-be normativity of action, perception, and knowledge.

According to Phenomenal Conservatism (PC), if it seems to a subject S that P, S thereby has some degree of (defeasible) justification for believing P. But what is it for P to seem true? Answering this question is vital for assessing what role (if any) such states can play. Many have appeared to adopt a kind of non-reductionism that construes seemings as intentional states which cannot be reduced to more familiar mental states like beliefs or sensations. In this paper I aim to show that reductive accounts need to be taken more seriously by illustrating the plausibility of identifying seemings and conscious inclinations to form a belief. I briefly close the paper by considering the implications such an analysis might have for views such as PC.

In this paper, I make explicit some implicit commitments to realism and conceptualism in recent work in social epistemology exemplified by Miranda Fricker and Charles Mills. I offer a survey of recent writings at the intersection of social epistemology, feminism, and critical race theory, showing that commitments to realism and conceptualism are at once implied yet undertheorized in the existing literature. I go on to offer an explicit defense of these commitments by drawing from the epistemological framework of John McDowell, demonstrating the relevance of the metaphor of the “space of reasons” for theorizing and criticizing instances of epistemic injustice. I then point out how McDowell’s own view requires expansion and revision in light of Mills' concept of “epistemologies of ignorance.” I conclude that, when their strengths are used to make up for each others' weaknesses, Mills and McDowell’s positions mutually reinforce one another, producing a powerful model for theorizing instances of systematic ignorance and false belief.

In Democratic Reason, Hélène Landemore has built a case for the epistemic virtues of inclusive deliberative democracy based on the cognitive diversity of the group engaged in making collective decisions. She supports her thesis by appealing to the Diversity Trumps Ability Theorem of Lu Hong and Scott Page. This theorem is quite technical and the informal statements of it aimed at democratic theorists are inaccurate, which has resulted in some misguided critiques of the theorem's applicability to democratic politics. This paper provides an exposition of the Diversity Trumps Ability Theorem that does not sacrifice precision for accessibility. It also shows that it is not possible to satisfy the assumptions of this theorem when there are only two options. Thus, the Diversity Trumps Ability Theorem cannot provide support for the epistemic virtues of inclusive democratic deliberation for binary collective decisions.

I propose an account of the supply domain for testimonial knowledge: intentional acts in which a speaker offers her word on a matter to her intended audience. Such acts are effected by speech acts of telling exploiting a shared language. From this account I derive an account of the proper doxastic response to such a telling that P, namely accepting the speaker’s offered word that P. I show that one who takes the speaker’s word is committed to holding that she was epistemically so placed as to properly offer her word – namely, that she knows that P. I use my account of the supply domain for and proper response to testimony, to show what is wrong with a thin reliabilist conception of the good supply domain for testimony. The thin conception has it that any reliable testifier is an apt source. I argue that only those who know what they state properly offer their word, and only from them can knowledge be gained through the Core Mechanism, taking the speaker’s word for what she states.