![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
A good account of normative reasons should explain not only what makes practical and epistemic reasons a unified kind of thing, but also why practical and epistemic reasons are substantively different kinds of reasons that underlie significant categories of normative assessment and exhibit different weighing behaviours. I argue that a disjunctive account of normative reasons, according to which practical and epistemic reasons have very different grounds (what I call the Different Source View), can do both of these jobs, unlike some prominent unified alternative accounts. And the viability of this view has significant implications for metanormative theorizing: it implies that the answer to certain metanormative questions may differ between the practical and epistemic domains.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
Notes
1 It’s controversial whether having a reason (in the sense relevant to rationality) amounts to there being a reason and one’s possessing this reason in some way (see Schroeder [Citation2008] and Lord [Citation2010]). But nothing that I say here hangs on this issue.
2 Thanks to Selim Berker for this point.
3 I’m only claiming that epistemic reasons for belief balance in this way. One might think that there are epistemic reasons for action and other attitudes, and that such reasons balance out like practical reasons. My desiderata in this section are neutral about this, but I will revisit the issue in sections 4 and 5.
4 Thanks to Selim Berker again for discussion of this sort of case. One might think, instead, that in such cases one ought to have one or other of these conflicting desires, but not both. But this, too, implies that practical reasons for desire balance out differently than do epistemic reasons for belief.
5 One might take ‘facts’ here to be states of affairs [Dancy Citation2000] or true propositions [Schroeder Citation2007], and one might also take the relata of these relations to include a circumstance and a time [Scanlon Citation2014].
6 Parfit [Citation2011] argues precisely that, but I argue elsewhere [2017] that this is not a promising strategy.
7 I assume that the most plausible desire that can ground all epistemic reasons is a two-pronged desire to believe the truth and to avoid error, for reasons originating with James [Citation1896]: if our epistemic goals included only believing the truth, this would imply that we ought to believe every proposition, but if our epistemic goals included only avoiding believing falsehoods, this would imply that we ought to believe nothing. I also assume that this desire must be relativized to a particular proposition, for reasons originating with Fumerton [Citation2001]: if epistemic reasons were grounded in a general desire to believe the truth and to avoid error, then the fact that believing some known falsehood will cause one to have many more true beliefs would be an epistemic reason to believe that known falsehood.
8 Schroeder [Citation2007] himself rejects this account of the weights of reasons, though, and argues that a better account is available to a desire-based theory of normative reasons.
9 This is inspired by Hurka [Citation2001] and Sylvan [Citation2012]. Hurka argues that showing respect or commitment to some final value v may itself be derivatively valuable, even though showing respect or commitment to v does not promote v and is thus not instrumentally valuable. Similarly, Sylvan argues that, while epistemic justification may not always promote believing the truth, and may thus fail to be instrumentally valuable, its value may nonetheless be derivative of the value of truth because it shows respect or commitment to the truth. I am not making any claims about the value of justification or the relation between values and reasons; instead, I am making a similar point about reasons—that some epistemic reasons may be reasons not because they promote having true beliefs, but because they show respect or commitment to the truth.
10 This view is different from a desire-based view of epistemic reasons in two crucial ways. First, on this view, epistemic reasons do not depend on the agent’s actually having any desire to believe the truth and avoid error with respect to whether p, and thus does not face the sorts of counterexamples provided by Kelly [Citation2003]. Second, the Truth-Commitment ViewE does not claim that all epistemic reasons to φ with respect to p are reasons because they promote believing the truth and avoiding error with respect to p. The Truth-Commitment View is thus not a teleological view of epistemic reasons, and thus avoids Berker’s [Citation2013] criticisms.
11 Indeed, the Different Source View is compatible with adopting an analysis of the reason relation itself along the lines of Broome [Citation2004] or Kearns and Star [Citation2008, Citation2009].
12 Thanks to participants of the 2016 Language and Metaphysics of Normativity Conference at Uppsala University, Sweden, the graduate students in David Plunkett’s 2016 metaethics seminar at Stanford University, and audiences at the University of Notre Dame and McGill University for their feedback. And special thanks to Bob Beddor, Selim Berker, Ruth Chang, Andy Egan, Debbie Roberts, Geoffrey Sayre-McCord, Jonathan Schaffer, and Daniel Wodak for their many helpful comments on earlier drafts.