Equal Moral Opportunity: A Solution to the Problem of Moral Luck

ABSTRACT

Many of our common-sense moral judgments seemingly imply the existence of moral luck. I attempt to avoid moral luck while retaining most of these judgments. I defend a view on which agents have moral equality of opportunity. This allows us to account for our anti-moral-luck intuitions at less cost than has been previously recognized.

KEYWORDS:

• moral luck
• moral responsibility
• equal opportunity

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 This is just an initial characterization, not a final analysis, of moral luck.

2 See Zimmerman [2002] for additional discussion of these issues.

3 This option is endorsed by Zimmerman [2002] and Enoch and Marmor [2007]. See Richards [1986] for a similar view.

4 Hartman [2017] argues that, on this approach, we would be both infinitely praiseworthy and infinitely blameworthy.

5 One might object to the claim that she ran a risk, since matters were settled by her own agency. I would regard taking a bet that involves my being tortured if I ever exercise my agency badly as extremely risky. So, I am not persuaded. But those who don’t think that she literally ran a ‘risk’ could substitute ‘encountered a disadvantageous propensity in the world’ for ‘ran a terrible risk’.

6 An approach along these lines has previously been suggested by Crisp [2017: 17]:

A harder choice may be more praiseworthy, so to this extent the circumstantial bad moral luck of the man who stayed in [Nazi] Germany was counterbalanced by the greater moral opportunities available to him. And as it becomes more difficult to make the correct choice, so it becomes a lesser wrong [and the agent less blameworthy] not to make it.

Crisp rightly suggests that this thought could allow one to defend a sort of equality of opportunity. I aim to lay out a precise view along the lines of Crisp’s suggestion that really does achieve full equal moral opportunity, and to thoroughly investigate the plausibility of that view. I suspect that Crisp would be sympathetic to my approach. But it’s unclear to me whether he would accept the strong claim about equal expected desert levels which my Egalitarian Solution endorses.

7 A difference is ‘due to’ (in my sense) the agents exercising free agency, and nothing else, if (i) you cannot fully explain the two expected desert levels differing in precisely the way that they do without appealing to an exercise of free agency by at least one of the agents. And (ii) the set of all of the agents’ exercisings of free agency entails, given the moral facts, the exact difference between them.

8 I am here concerned only with the sort of desert that is earned. Perhaps everyone is born deserving (in some sense) a good life. This claim should not be taken to conflict with the Non-Comparative Anti-Moral-Luck Principle.

9 Note that some persons never get to make a morally significant choice. Their expected (and actual) desert level will thus be 0. (Recall that I am concerned with the sort of desert that is earned.) So, assigning a positive expected desert level to some choice circumstance will inevitably generate comparative unfairness.

10 We could try to keep their expected desert levels the same, by saying that, because her attempt would have succeeded, Alice is more blameworthy than Bill if they both refrain from trying to rescue the child. I don’t find this claim very plausible. And tricky questions would arise in cases where it is causally undetermined whether one’s attempt would have succeeded. For another interesting approach to resultant luck which claims that our current practice of punishment is equivalent to a penal lottery that treats successful and unsuccessful attempts equally, see Lewis [1989].

11 Smith also discusses a third model on which acting on good desires is required, but neither the presence nor the absence of bad desires is required. Smith credits Henson [1979] for the term ‘battle citation’.

12 The one way to avoid this would be to claim that Gail would be much more blameworthy than Howard if she acted wrongly. But it is hard to see the motivation for this move from within the framework of the moral purity model.

13 I granted that luck renders Gail more virtuous than Howard. Why reject luck when it comes to praise and blame, while accepting that luck plays a role in other forms of moral evaluation (such as whether one is virtuous)? Although any deep explanation of this difference will be controversial, it seems very plausible that certain domains of moral evaluation are more sensitive to luck that others. Suppose that, due entirely to bad luck, evil desires are implanted in my brain. I become in some sense a bad person, but it is plausible that I do not automatically become blameworthy (in the absence of any action on my part). So, we should not find it surprising that different domains of moral evaluation differ with regard to their relationship to luck.

14 See, e.g., Lewis [1973] and van Inwagen [1997]. But see DeRose [1999] for an alternative view.

15 See, e.g., Adams [1977] and van Inwagen [1997].

16 I owe this sort of worry to Michael Zimmerman.

17 It is so-named because of Stannis Baratheon’s treatment of Davos in A Clash of Kings. In response to Davos’s bravery, ‘Lord Stannis had rewarded Davos with choice lands on Cape Wrath, a small keep, and a knight’s honors … but he had also decreed that he lose a joint of each finger on his left hand, to pay for all his years of smuggling’ [Martin 1998: 12].

18 See Hartman [2019] for a worry along these lines.

19 I myself find it quite strange to think that desert could depend on one’s risk preferences in this way. An alternative approach might hold that there is an objectively correct level of risk aversion, and use that level of risk aversion, rather than the agent’s subjective preference, to assign the ‘desert points’. But positing one objectively correct level of risk aversion also seems strange.

20 See, e.g., Pereboom [2014: ch. 3].

21 For a discussion of the notion of ‘on a par’, see Chang [2002].

22 Perhaps willpower should be conceived of as a desire to do what one ought even when it requires great effort.

23 Thanks to D. Black for pointing out this sort of case to me.

24 One might think that agents have an option to do A only if A it occurs to them that they might do A. On this view, the limiting of Orene’s options is less artificial. There will surely be cases where no bad options occur to an agent. (Thanks to a referee for mentioning this.)

25 See Fischer [1986], Zimmerman [2002], and Swenson [2019].

26 Thanks to a referee for raising this worry.

27 See Rubio [2018].

28 The Frankfurt Cases might provide another motivation for no lose scenarios. I respond to Frankfurt Cases elsewhere ([2015, 2016, 2019]; Capes and Swenson [2017]).

29 I benefited from a discussion of this sort of case in Holly Smith’s Spring 2017 graduate seminar at Rutgers University.

30 One issue that I did not address is the relationship between the Egalitarian Solution and debates about the nature of free will. The Egalitarian Solution involves appeal to the notion of probability. Compatibilists about responsibility and determinism say that you can be blameworthy (or praiseworthy) for events that are causally determined. One might worry that Compatibilism is committed to no win scenarios in which you are causally determined to perform a blameworthy action, and thus where the probability that you will do so is 100%. But quite a few philosophers have thought that non-trivial objective probabilities (i.e. probabilities other than 0 or 1) are compatible with determinism (see, e.g., Loewer [2001], Hoefer [2007], and Glynn [2010]). So, we should not assume that compatibilists cannot make use of the Egalitarian Solution.

31 Thanks to Brian Cutter, Ian Cruise, Jon Garthoff, Chris Hauser, Bob Hartman, Beth Henzel, David Limbaugh, Dana Nelkin, John Pittard, Alex Pruss, Pamela Robinson, Amy Seymour, David Shoemaker, Holly Smith, Larry Temkin, Neal Tognazzini, Christopher Tomaszewski, Chris Tucker, Chris Willard-Kyle, Michael Zimmerman, and several journal referees and editors for helpful comments or discussion. Thanks to audiences at the APA, the SCP, RoME, Baylor, Christopher Newport University, VCU, and Rutgers. So many people have helped me to think through these issues over the years and I am sure I am forgetting some of you. So, thanks to those whom I am forgetting. I am especially grateful to D. Black and Patrick Todd for numerous helpful discussions. Work on this paper was funded by summer grants from The College of William & Mary.