Current Projects


That an epistemic authority believes that p is often a good reason to believe it yourself. The preemption view claims that optimal way of responding to epistemic authority is to let the fact that an authority believes that p be your sole reason for believing it. But there are three salient types of epistemically authoritative situation for which preempting is either epistemically inappropriate or conceptually inapplicable. The first arises when you have an independent reason for the authority’s belief, the second when you have a reason for that belief which depends on the authority for its status as a reason. The third arises – perhaps most surprisingly – when you have reasons against the authority’s belief. I provide an alternative account on which the epistemically best course of action is to make use of all of your reasons, including but not limited to the fact that the authority believes that p.
When one person trusts another for information, typically both parties are epistemically vulnerable: the trusting person is vulnerable to false information, and the trusted person puts her reputation as a truth-teller on the line. Both parties are roughly on a par in terms of the risks they undergo. Trusting people for information via certain digital media can unbalance this parity of risk. The trusting party takes a much greater risk than the trusted party: because of our psychological wiring, digital contexts exert greater epistemic and psychological pressure to believe what we read than to hold purveyors of falsehood accountable. This imbalance has detrimental effects on the entire epistemic community.
Trusting a person is often understood in merely decision-theoretic terms: one person trusts another to fulfill some practical aim. In contrast, I advance a moral account of trust, on which trust has the moral function of bringing about good states of affairs involving respectful treatment of persons. Trust stands out among other attitudes and actions in this regard, because it makes the trusting and trusted persons especially vulnerable. Mutual vulnerability increases the potential for respectful treatment, but also the potential for disrespectful treatment, so trust is especially morally laden. And it often fails to fulfill its moral function in our imperfect world. I defend my moral account of trust on the grounds that it explains a variety of properties of trust highlighted across a wide cross section of the trust literature.
Trust is commonly assumed to have the epistemic function of helping the trusting person form true beliefs which she would be unable to form on her own. This chapter argues that it also has a moral function – that of bringing about good interpersonal states of affairs. These functions are not always compatible. When they conflict, how should the conflict be resolved? Partialists argue that the moral function should usually take precedence, whereas impartialists argue that the epistemic function should. This chapter advocates an impartialist account, but one with constraints that accommodate partialist concerns.
Social epistemology is concerned with how we metarepresent the viewpoints of other people for the sake of gaining knowledge and other epistemic goods. Yet in spite of the established discussion of metarepresentations in closely related areas like theory of mind and cognitive science, there is as yet no general account of their epistemology. This paper gives such an account. I employ a variety of philosophical tools, including the notions of grasping, de re attitudes, and knowledge by acquaintance. I then give an example of how this account can be put to work in social epistemology: it provides a framework for gaining epistemic benefits from peer disagreements.

Monograph: Faith in Evidence

I argue for a robust but fair evidentialist account of the epistemic norms accompanying religious faith. Part I spells out the form of evidentialism I advocate, and Part II argues that there are strong reasons, stemming from the nature of religious faith itself, to hold one’s religious attitudes in a way that respects evidentialism so construed.

Part I: An Evidentialism for Religious Beliefs. Evidentialism as understood here says that it is epistemically good for beliefs to respect the believer’s evidence. Yet this is a broad thesis, ranging from the strict demand that religious beliefs meet standards associated with scientific reasoning, to permission to assent to one’s subjective experiences with minimal reflection. Neither of these extremes, I argue, is suitable for promoting religious epistemic goals. The evidentialism I advocate steers a course between them that is intellectually robust yet attainable. One important feature of my account is its attention to religious belief systems instead of just religious beliefs; a second is its insistence that religious disagreement can be an important source of evidence about religious matters. A third is its sensitivity to cognitive-psychological insights (e.g., about the effects of bias on our beliefs).

Part II: Why Evidentialism? Why should religious beliefs be answerable to the evidentialism I have proposed? One might think that religious faith comes with norms of its own that conflict with it, such as that faith demands going beyond our evidence or at least ceasing to search for more. If so, then it would be natural for religious believers to think that the norms of faith trump those of epistemology. I argue, however, that evidence-respecting belief is a good-making feature of religious faith. One reason is that love and trust, two attitudes associated with a flourishing faith, are good to the extent that the beliefs accompanying them respect the believer’s evidence. A second reason is that evidence-seeking can (as in the case of Job, I argue) be an expression of great faith indeed. A final reason is that a fledgling evidentialism can be found in key Christian and Jewish sources.

PART I: An Evidentialism for Religious Beliefs

The first ingredient in religious evidentialism is an evidence-weighting policy: an account of how different sorts of evidence should feature in our reasoning. I argue that evidence-weighting policies, in order to nudge us toward epistemic ideals, must be sensitive to cognitive-psychological facts about how we form beliefs. One such fact is that we are subject to a variety of cognitive biases which are especially strong in our beliefs about religion. The best course of action might seem to be for an evidence-weighting policy to minimize the effects of bias altogether. But things are not so straightforward. For although some biases are epistemically unreliable, others are reliable – indeed, some philosophers argue that God communicates via the deliverances of reliable religious biases.

As it happens, the evidence-weighting polices in the philosophy of religion can be understood as taking two opposed approaches to bias. One view (associated e.g. with Schellenberg and Locke) tries to minimize the effects of all biases by giving the most weight to evidence that meets criteria associated with scientific reasoning. A second view (associated e.g. with Plantinga, Alston, and Moser) tries to maximize the effects of reliable biases by allowing the most weight to be given to evidence that fail to meet these criteria, but in so doing licenses the output of unreliable biases too. I advocate a third, middle-ground, view (defended also by Swinburne, Wainwright, and Greco) that gives approximately equal weight to both kinds of evidence. This policy, which I call egalitarianism, puts us in the best position to achieve epistemic goals in the religious domain.

The second ingredient in religious evidentialism is an account of where and how we should seek evidence. For even properly weighted evidence will fail to be truth-conducive if it is an unrepresentative sample. Traditional sources of evidence about religious matters include arguments, religious experience, and testimony. Religious disagreement is a source of evidence too, but the literature tends to focus on the defensive question of whether the fact of religious disagreement is a threat to one’s own religious view. This chapter, by contrast, focuses on what I call the positive question, of whether and how religious disagreement can further religious epistemic aims – that is, can yield important evidence about ultimate reality that would be difficult to obtain otherwise. I argue that religious disagreement has the strong potential to do just this.

One might think that this argument can be made by drawing parallels with scientific research, where disagreement is welcomed as a source of epistemic benefit. But such an argument would be rejected by many religious believers themselves, who take religious epistemic aims to be fundamentally different from scientific ones. Whereas scientists aim to confirm or disconfirm hypotheses whose truth or falsehood is up for grabs, religious believers aim to preserve beliefs assumed to be true. I aim to convince such religious believers of the (possibly counterintuitive) conclusion that religious disagreement, far from threatening their religious beliefs, can help preserve them. In the context of this argument I give an account of religious belief systems (as opposed to just religious beliefs), many aspects of which can be improved by disagreement.

I then address the following objections that might worry a committed religious believer: first, that religious disagreement can threaten the epistemic justification of true religious beliefs; second, that it can psychologically shake believers even if their beliefs are justified; and third, that it is otiose because the Holy Spirit ensures that religious believers achieve epistemic goals already.

This chapter examines one kind of disagreement that can produce particularly important evidence about religious matters: disagreement between religious authorities and people who are marginalized in religious communities.

Every social position in a community comes with epistemic advantages as well as epistemic limitations, and religious authority is no exception. This is so even if the religious authorities in question are divinely epistemically guided and the religious belief system largely true. On the one hand, religious authorities have the advantages of training and experience in the community belief system; on the other hand, they have the disadvantage of tending to see the world through the lens of their privileged position in the religious community. This makes it difficult for them to appreciate certain weaknesses that even a largely accurate religious belief system might contain. Religious authorities moreover face pressure to uphold the belief system as it stands, warts and all, insofar as their livelihoods and status depend on their so doing.

The viewpoint of those who are marginalized in religious communities is diametrically opposed and so an important complement to that of religious authorities. I give an account of religious marginalization, and argue that the religiously marginalized are apt to have significant epistemic advantages over religious authorities. One arises from the fact that the religiously marginalized are unlikely to endorse the community’s belief system wholeheartedly, yet are apt to understand it deeply. This dual perspective offers a valuable basis for critical reflection that is less likely to be available to religious authorities. Another advantage arises from religiously and epistemically significant forms of suffering that attend religious marginalization. Religious communities can promote their epistemic goals by engaging seriously with the viewpoints of their marginalized members.

PART II: Why Evidentialism?

Having presented and argued for my robust but fair religious evidentialism in Part I, in Part II I argue that not only is this evidentialism compatible with religious faith, but abiding by it can make a person’s faith better than it might have been otherwise.

Chapter 5 begins by addressing an objection by Lara Buchak, to the effect that faith entails foregoing evidence, at least when this evidence would influence your decision to act on the proposition in which you have faith. I argue that Buchak’s account is mistaken. It may pinpoint a regularity that applies across many cases of faith, but it falls short of being a necessary or sufficient condition.

My argument against the necessary condition takes the form of a counterexample that elaborates on the story of Job. I argue for the (perhaps initially counterintuitive) claim that searching for evidence for the sake of practical reasoning can constitute an act of faith. But one might still think that, even if declining to seek evidence is not necessary for faith, it still makes for a more praiseworthy faith. I argue that this normative claim is false too. Once more I draw on an example from the book of Job. A particularly laudable form of faith, I argue, combines loyalty with integrity, where the integrity is both moral and epistemic. Seeking evidence for the purpose of practical deliberation can make for a praiseworthy faith. I then argue, with an illustration of the biblical character Balaam, that Buchak’s account also fails to present a sufficient condition for faith.

I offer some remarks to explain the intuitions behind the intuitions driving Buchak’s account. An alternative account of faith emerges from these considerations: far from being defined in terms of one’s epistemic activity (or inactivity), what faith essentially amounts to is having a certain sort of motivational profile.

This chapter argues that respect for one’s evidence about the person in whom one has faith is a good-making property of faith. I draw on the nature of love, an attitude that – at least ideally – accompanies certain sorts of faith, including faith in God.

I use as my foil a common literary trope according to which love is “blind”, in the sense that lovers ignore or downplay evidence pointing to negative characteristics of their beloveds, or at least interpret it more charitably than other people would. In some cases such partiality can be explained by the fact that lovers’ have additional background evidence about their beloveds, which third parties lack and in the light of which such charitable assessments are epistemically appropriate. But additional background evidence cannot always explain the epistemic partiality of love. So it would seem, if such cases represent praiseworthy or ideal love, that departures from evidence can be a feature of such love. This chapter argues, by contrast, that love is at its best – at least along one important dimension – when the lover’s beliefs about her beloved respect her evidence.

Two features of love, I argue, are better realized when the lover’s beliefs about her beloved respect her evidence. The first is the lover’s ability to value her beloved for who he really is. A love that ignores negative characteristics of the beloved (in the sense of not even acknowledging on the basis of evidence that he possesses them) misses its true mark and therefore risks being idolatrous. It also fails to create the proper sense of security within which loving relationships flourish: there will always be space for the beloved to worry that the lover would reject him if she learned what he is really like. Second, a lover whose beliefs about her beloved respect her evidence is better able to ­show love to her beloved, and to care for him, than she would otherwise, because she is apt to be better informed about his needs. I apply these considerations both to a faithful person’s love of God and to the love of a God who has faith in human beings.

Another component of a flourishing faith is trust in God. This chapter argues that trust makes for better faith, all else equal, to the extent that the trusting person’s beliefs about the trusted party respect her evidence. The idea is that trusting someone (as I understand it here) is an action, and the function of actions is to achieve the actor’s goals. Here is how respect for evidence fits in: evidence provides information about what the world is like. So by ensuring that your beliefs about the other party respect your evidence, you increase the likelihood that you will achieve the goal that you aim to achieve in trusting.

This argument might superficially seem committed to an instrumental account of trust, on which trusting a person for an object or a goal amounts simply to calculating that that person will come through, and letting someone trust you amounts to calculating that you might sometime need him to return the favor. On the instrumental account it is hard to see what makes trust – and indeed, trustworthiness – good. But my argument is not committed to this instrumental account; indeed, I reject it. On my view, trust can have any number of goals, including but not limited to that of securing the object that the trusted person commits to provide. The goal might for instance be to deepen your relationship with the trusted person, or even to encourage her to grow in trustworthiness. Indeed, I argue that trust has the important moral function of bringing about good interpersonal states of affairs – and that it serves this function best when the trusting person’s beliefs about the other’s trustworthiness are evidence-respecting. I apply this account to the relationship of trust between human beings and God. In addition to supporting evidentialism for our beliefs about God, it helps to make sense of the idea that God (whose evidence about our trustworthiness consists trivially in his omniscience) can trust us.

Religious writings and tradition are an important source of information about the notion of religious faith. This is trivially so for those who regard such texts as divinely revealed, but it also applies to those who regard them simply as data concerning deep-rooted aspects of human thought. This chapter argues that Jewish and Christian scriptures contain a proto-evidentialism that resembles the account I have developed.

To make this case I show first that the Hebrew and Greek words for faith, respectively ’emuna and pistis, mean roughly to rely on something that is reliable and seen to be reliable – as opposed to something that one harbors any doubts about. But why should this reliance and perception of reliance involve beliefs that are evidence-respecting? To answer this question I draw on the biblical texts themselves. Narrative portrayals of faith in God are typically, and normatively, accompanied by God’s providing evidence on which this faith can rest (for example, the God of the Jewish scriptures does mighty works expressly so that his people may know that he is the LORD, a tradition that Jesus takes himself in the New Testament to be continuing with his miracles). Prophets continually exhort the faithful to trust God on the basis of his character revealed in his past works (think of Moses and Isaiah), and New Testament authors exhort faith on the basis of the history they have meticulously documented (the gospels of John and Matthew).

There are of course apparent counterexamples, such as Abram, who to all appearances leaves his homeland on the basis of a command out of the blue and then keeps his faith against the evidence that God wants him to murder his son; the letter to the Hebrews, which calls faith the “conviction of things not seen”; Job, who is held up as an exemplar of faith for maintaining his faith in God’s goodness and justice in the face of disastrous counterevidence; and of course Doubting Thomas, whom Jesus chastises for refusing to believe in Jesus’ resurrection without tangible proof. I address these and other apparent counterexamples.

On the basis of these and other considerations from religious sources, I argue that a robust yet fair evidentialism is the norm in at least the Judeo-Christian religious tradition.

Edited Volume: Trust in Epistemology

Trust is fundamental to epistemology. Many discussions, particularly recent ones, rely on the notion of trust as a theoretical bedrock – witness the recent birth of social epistemology and discussions of epistemic self-trust. Yet trust, as a recent surge of discussion is increasingly indicating, is a philosophically problematic notion to say the least. In a way that is reminiscent of attempts to analyse knowledge, it is surprisingly difficult to pin down. It is time for epistemology to draw on recent general discussions of trust for the sake of digging deeper into the fundamental role that trust plays in a broad cross-section of epistemological issues.

This volume aims to cross-fertilize cutting-edge epistemology with nascent systematic discussions of trust. The eleven contributions by well known philosophers will prompt philosophers of trust to explore the epistemological implications of their theories, and epistemologists to tease out both the difficulties and the potential of appealing to trust. Although trust is presupposed in epistemology and occasionally discussed for this or that specific purpose, as yet there has been no comprehensive discussion of this vital notion across different areas of epistemology. Trust in Epistemology aims to fill this lacuna.