![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
According to a familiar (alleged) requirement on practical reason, one must believe a proposition if one is to take it for granted in reasoning about what to do. This paper explores a related requirement, not on thinking but on acting—that one must accept a goal if one is to count as acting for its sake. This is the acceptance requirement. Although it is endorsed by writers as diverse as Christine Korsgaard, Donald Davidson, and Talbot Brewer, I argue that it is vulnerable to counterexamples, in which agents act in light of ends that they do not accept but are still merely considering. For instance, a young professional may keep a job option open not because she definitely wants or intends to take it, but just because she is considering taking it. I try to show (1) that such examples are not easily resisted; (2) that they present challenges specifically for Brewer, Davidson, and especially Korsgaard; and (3) that the examples also raise fresh, non-partisan questions in action theory. What is considering, exactly? How could it fall short of acceptance while still guiding behaviour? How can we act for an end before thinking it through?
Notes
1 For now, I ignore responsive actions, which are explained by features of situations, not by goals.
2 Philosophers do, however, deny requirements involving more specific forms of acceptance. Stocker [Citation1979], for instance, denies that agents must always see something good about their actions.
3 A quick qualification. For Korsgaard's Kant, I cannot act unless I ‘make a law for myself with universal normative force, a law that applies to my conduct over a range of cases’, but this force needs only to be ‘provisionally universal’ (Korsgaard [Citation2009: 73]; see also Kant [Citation1785 (1998): 56–8]). This makes it sound like Korsgaard stops short of the acceptance requirement—one counts as acting so long as one has at least provisionally willed an end (provisional willing is not acceptance, since it indicates a less-than-determinately-positive outlook on the end). But what is provisional, according to Korsgaard, is not whether an end is willed, but how far the willing ranges over as-yet-unconsidered cases.
4 The parenthetical remarks are a courtesy to psychologists about motivating reasons (section 5).
5 The analogy with belief is hard to make precise (without saying controversial things), but here's the sort of case that I have in mind. A considered agnostic, mulling over a metaphysical view, realizes that it entails atheism, and on this basis suspends judgment on the view.
6 Bratman [Citation1987: 38–9] and Holton [Citation2008: 28–9] feature a pair of related examples. Here is Holton's version of the first: you want to move a tree from your car and are deciding between removing it yourself and paying someone to do it. In the end, you make a cancellable appointment with a yard-worker and also load up your wheelbarrow with tools. You fully intend to move the tree, but only partially intend each means. This case is similar to Leona's: you load the wheelbarrow because you're considering removing the tree yourself. But there is a dissimilarity: whereas you have accepted the end of having the tree moved and are merely undecided on your choice of means, Leona is still making up her mind about her end itself. Although the case of Leona is a natural extension of Holton's and Bratman's cases, hers alone flouts the acceptance requirement.
7 I won't insist on this point, but I doubt that ‘One is doing A in order to do B’ really does entail a desire or intention to do B, although I think it entails that one is at least considering B.
8 This might be false on some senses of ‘desire’, involving affect or a relation to our psychological reward system [Arpaly and Schroeder Citation2014]. Perhaps Leona feels cold and unenticed by the prospect of keeping options open. But she does desire options in the technical sense I have in mind, which involves motivation: roughly, desiring that p entails being motivated to bring it about that p if one can—whereas one who is considering doing something might not be motivated to do it. (This is consistent with, but weaker than, Dancy's [Citation2000] claim that desire is motivation.) Thanks to a referee for pressing me to say more here.
9 An exception: sometimes, having options is desired finally because a lack of options would undermine personal freedom. But freedom isn't essential to Leona's case, and it certainly isn't relevant to the others.
10 A final end motivates in its own right, whereas an instrumental end motivates only because it conduces toward something else [Korsgaard Citation1983].
11 My distinction is inspired by Raz [Citation2011: 46], who claims that telic (his term is ‘practical’) reasons are grounded in value, whereas responsive (‘adaptive’) reasons are grounded in fittingness (see also Müller [Citation2011]).
12 Notice that this first feature (suspension) might come apart from the others (working toward resolution). One might suspend judgment on p, for example, without wondering whether it's true, or without trying to make up one's mind about it. More generally, it would be a good idea to ask which combinations of these features (and the four from Bratman that I discuss below) would form interesting, natural, or familiar ‘clumps’. I owe these points to conversations with Kieran Setiya and Richard Holton.
13 I hope it's clear that I'm not assuming that there is only one sense of ‘acceptance’, on which we ‘accept’ both ends and propositions. We have two technical uses, introduced for different reasons, although I think they are importantly similar, in that both signal non-provisionality.
14 See also Ullman-Margalit [Citation1983: 162] on ‘presumption rules’, as well as Cohen [Citation1992].
15 Two potential objections (adapted from Holton's [Citation2008] case against the psychological reality of credences): (i) graded intentions don't show up in ordinary thought; and (ii) they are not easily squared with all-out intentions, which do.
16 If my intention to get the coffee strengthens, one would not say that I thereby warm up to anything. To warm up to x is to feel more determinately (not more strongly) pro-x. (My thanks to an anonymous referee for convincing me that my earlier thoughts on ‘warming up’ were half-baked.)
17 For some recent work regarding imprecise doxastic states, see Bradley [Citation2015] and Schoenfield [Citationforthcoming].
18 One clue might lie in recent work on inclination (e.g. Schapiro [Citation2009]).
19 Another ideal: an action's value is transparent to its agent and specific to one episode. Brewer [Citation2009: 39] shows that this idealization is not always met: certain actions (‘dialectical activities’) have only ‘opaque’ value, which may ‘unveil itself incrementally’ over the course of many episodes (see also Arpaly [Citation2003] and Brewer [Citation2009: 73–5, 96]). (One of Brewer's examples is that of doing philosophy.) My point is different, since ‘opaque’ values are still non-provisional. The phenomenology of unveiling is thus unlike that of warming up. In warming up to an end, your mind undergoes a change from considering to accepting; when your end is ‘unveiled’, you see that this was what you wanted all along.
20 Thank you to audiences at MIT's Ethics Reading Group, MIT's Work in Progress Seminar, and the University of Texas at Austin's 2017 Graduate Conference, where I was fortunate to receive comments from Caroline Christoff. Thanks as well to Austin Baker, David Balcarras, Brendan de Kenessey, Branden Fitelson, Kelly Gaus, Elliot Goodine, Caspar Hare, Abby Jacques, Rose Lenehan, Yael Loewenstein, Zelda Mariet, Alex Prescott-Couch, Brad Skow, Jack Spencer, and Steve Yablo. I owe special thanks to Nathaniel Baron-Schmitt, Jonathan Drake, Richard Holton, Leonard Katz, Sarah Paul, Tamar Schapiro, Patrick Quinn White, and two anonymous referees at AJP. Kari Rosenfeld's insights and encouragement during the process of revision were invaluable. I should also like to thank the editors of AJP for their suggestions, which inspired many salutary deletions. What remains of the paper is dedicated to Kieran Setiya. I could not ask for a more thoughtful teacher—or a more considerate advisor.