Believing Well

In this chapter it is argued that knowledge stands to believing the right thing as acting well stands to doing the right thing. Knowledge is believing well along both objective and subjective dimensions. The argument uses artificial games as a control case in order to argue for a general principle about the relationship between acting well and doing the right thing. And it uses this general principle in order to offer both direct and indirect arguments that knowledge is believing well. It is a consequence of the argument of this chapter that knowledge is essentially normative in nature, and so epistemology is a normative discipline, through and through.

Doubts about “Genuinely Normative” Epistemic Reasons

Instrumentalist and teleologist views in metaepistemology hold that epistemic reasons are goal-relative or value-relative. In the face of counterexamples involving apparently pointless or counterproductive beliefs that are nonetheless supported by excellent epistemic reasons, some have retreated to the following view: while even pointless or counterproductive beliefs can be supported by excellent epistemic reasons in a not-genuinely-normative sense, we can have genuinely normative epistemic reasons only for beliefs that do serve some goal or value. In this chapter doubts are raised about the distinction between genuinely normative and not-genuinely-normative epistemic reasons employed here. It is suggested that there’s no real need or intuitive motivation for the distinction, beyond the ad hoc need of salvaging instrumentalist and teleologist views from counterexamples. The sense in which all epistemic reasons—even reasons for apparently pointless or counterproductive beliefs—seem to be equally normative is explained; and the implications for instrumentalists and teleologists are outlined.

Passing the Epistemic Buck

While buck-passing accounts are widely discussed in the literature, there have been surprisingly few attempts to apply buck-passing analyses to specific normative domains such as aesthetics and epistemology. In particular, there have been very few works which have tried to provide complete and detailed buck-passing analyses of epistemic values and norms. These analyses are, however, both interesting and important. On the one hand, they can bring to the surface the advantages and difficulties of extending the buck-passing account to specific normative spheres, either providing further support for the approach or highlighting substantive difficulties. On the other hand, epistemic buck-passing analyses can be beneficial for normative epistemology, providing new perspectives on traditional epistemological problems, and possibly providing fresh approaches to such problems. This chapter aims at partially filling this gap.

Introduction

Epistemology is, at least in part, a normative subject. Normativity raises many questions. Are there facts about what we should do or think? If so, what is the nature of these facts? If there are no such facts, how should we make sense of our practices of making judgments about what people should do and think? Such questions have been most widely discussed in their application to ethics. This volume explores their application to epistemology: it is a collection of metaepistemology. The aim of this introductory chapter is to introduce some of the questions of metaepistemology, the issues they raise, and some important attempts to answer them, and to situate the volume’s chapters in relation to them.

Logical Disagreement

This chapter investigates what we disagree about when we disagree about logic, on the assumption that judgments of logical validity are normative. If logic is normative, then the popular anti-realist thesis that there are no normative facts or properties generalizes—it entails that there are no logical facts or properties. When faced with this anti-realism, it is tempting to endorse a pluralist thesis, according to which two people who disagree about the validity of an argument can both say something true. This chapter explores the limitations of three prominent forms of pluralism: contextualism, relativism, and expressivism. It argues that none of these forms of pluralism gives an adequate account of what we disagree about when we disagree about logic.

Is Epistemology Autonomous?

It’s commonly held that the best metaethical account of our normative thought and language won’t place any significant constraints on our first-order normative theorizing; once we have the right metaethics, we can go on having the same first-order normative debates, and accepting the same first-order normative views. This thesis of the “autonomy of ethics” is particularly popular among writers in the expressivist tradition. This chapter argues, however, that broadly expressivist metanormative commitments have significant consequences in first-order normative epistemological debates. It begins with the example of judgment internalism—a thesis endorsed by expressivists, but many non-expressivists as well. While judgment internalism is generally seen as irrelevant to first-order ethical debates, it is argued here that it has significant consequences for some putatively “first-order” debates in epistemology. It is then argued that accepting more thoroughly expressivist metaepistemological commitments has more far-reaching epistemological consequences, focusing on the internalism/externalism debate.

Epistemic Planning, Epistemic Internalism, and Luminosity

In this chapter, a “doxastic planning model” of epistemic evaluation is employed to argue for a form of epistemic internalism. In doing so, a response is first given to Schoenfield’s recent argument against the author’s previous attempt to develop such an argument. Various ways in which that argument might be understood are distinguished, and it is discussed how both internalists and externalists might make use of the ideas within it. It is then argued that, despite these complexities, the doxastic planning model continues to support a modest form of epistemic internalism. The chapter concludes by showing that, far from conflicting with “anti-luminosity” arguments in epistemology, this form of internalism is best understood as a natural reaction to these arguments.

Thick Epistemic Concepts

In metaethics, there is a range of different views of the nature of thick concepts. At one extreme are views holding that thick concepts are irreducibly thick, and that this has important consequences. At the other, there are views that hold that thick concepts are not inherently evaluative as a matter of content, and that they are of little or no theoretical consequence. In the epistemic case there is an additional element of controversy: whether the thick–thin distinction holds for epistemic concepts at all. In this chapter it is argued that it does. First, it is characterized what thick concepts are, before an outlining of why the question of whether there are any thick epistemic concepts is important. It is then argued that the epistemic domain is the right place to look for thick concepts, and that within that domain we find examples of concepts that are sufficiently like paradigm cases of thick ethical concepts to count as thick.

How to Be an Epistemic Expressivist

Arguments from the practicality of moral judgment are typically taken to be among the more powerful arguments for meta-ethical expressivism. However, analogous arguments seem unpromising in meta-epistemology. Morality is obviously practical in a very direct way, whereas epistemic judgments tell one what to believe, not what to do. It is argued in this chapter that what epistemologists call ‘pragmatic encroachment’ provides the best starting point for an argument from practicality in meta-epistemology. However, the best argument drawing on this idea seems compelling only for judgments about what someone has sufficient reason to believe, not judgments about what there is some reason to believe. But a similar dichotomy can be found in the moral context, and seeing why this does not spell doom there helps in the meta-epistemological context too. The chapter concludes by defending a better strategy for defending epistemic expressivism which relies on a specific conception of non-representational ‘direction of fit’.

Moral and Epistemic Error Theory

Many moral error theorists hold that moral facts are irreducibly normative. They also hold that irreducible normativity is metaphysically queer and conclude that there are no irreducibly normative reasons and consequently no moral facts. A popular response to moral error theory utilizes the so-called ‘companions in guilt’ strategy and argues that if moral reasons are irreducibly normative, then epistemic reasons are too. This is the Parity Premise, on the basis of which critics of moral error theory draw the Parity Conclusion that if there are no irreducibly normative reasons, there are no moral reasons and no epistemic reasons. From the Parity Conclusion and Epistemic Realism (the view that there are epistemic reasons), it follows that it is false that there are no irreducibly normative reasons. In this chapter, it is argued that the Parity Premise and the Parity Conclusion can both plausibly be rejected.