Author: Jonathan Spelman
Word Count: 1000
Resultant Moral Luck
While my brother and I were growing up, our father would tell us stories from his time as a police officer. One of those stories was about a teenager who fell asleep at the wheel, crossed over the center line, and hit an oncoming vehicle containing two passengers, an elderly couple, both of whom were killed in the crash.
Years later, a friend of mine told me a similar story. Driving home one night, he fell asleep at the wheel, crossed over the center line, and hit an oncoming vehicle containing one passenger, a middle-aged woman. Though seriously injured, she survived.
Given these details, you might be tempted to think that the teenager is morally worse than my friend because the results of his actions were worse than the results of my friend’s actions. If this is what you think, then you believe in the existence of moral luck, for you believe that how good a person is can depend on factors beyond one’s control (e.g., the safety features of the vehicle one hits, the physical condition of that vehicle’s passengers, whether those passengers are wearing seatbelts, etc.). In particular, you believe in the existence of resultant moral luck because you believe that how good one is can depend on the results of one’s actions, even when those results are beyond one’s control.
The Control Principle
While you may initially have been tempted to regard the teenager as morally worse than my friend, after carefully considering that the results of my friend’s actions were better due to purely lucky factors, you might be willing to reevaluate the moral situation. You might then conclude that the teenager and my friend are moral equals, neither any relevantly worse than the other. Why are they moral equals? Because the only difference between them is attributable to factors beyond their control. If this is what you think, then you do not, in fact, believe in the existence of resultant moral luck.
In “Moral Luck,” Thomas Nagel describes the motivation for denying the existence of moral luck. He writes, “Prior to reflection it is intuitively plausible that people cannot be morally assessed for what is not their fault, or for what is due to factors beyond their control.”1 We’ll call this principle, that how good one is cannot depend on factors beyond one’s control, the control principle.
As Nagel admits, the control principle is quite plausible. We do not think that how good one is depends on factors beyond one’s control. There is a worry, however, that everything about a person depends on factors beyond her control, in which case moral assessment turns out to be impossible. To see how one might arrive at that conclusion, we’ll look at three other potential species of moral luck, namely circumstantial, constitutive, and causal moral luck.
Circumstantial Moral Luck
Whereas resultant moral luck exists if how good one is can depend on the results of one’s actions, even when those results are due to factors beyond one’s control, circumstantial moral luck exists if how good one is can depend on how one acts, even when how one acts is due to factors beyond one’s control. For example, imagine that two married women are drinking at a bar. Both find the bartender attractive, and each is willing to cheat on her husband with the bartender. The bartender, however, is only interested in one of them. Thus, only one of the two women cheats.
Intuitively, those who cheat are worse than those who don’t, but in cases like this one, whether an individual cheats depends on factors beyond his or her control. Thus, if those who cheat are worse than those who don’t, circumstantial moral luck exists.
Constitutive Moral Luck
Constitutive moral luck exists if how good one is can depend on one’s character traits, even when one’s character traits are due to factors beyond one’s control.
For example, although violent criminals seem worse than upstanding citizens, it’s plausible to think that whether one is a violent criminal or an upstanding citizen depends on one’s genes and the environment in which one is raised. Moreover, if that’s correct, then whether someone is a violent criminal or an upstanding citizen is due to factors beyond his or her control. Thus, if violent criminals are, in fact, worse than upstanding citizens, then constitutive moral luck exists.
Causal Moral Luck
Causal moral luck exists if how good an individual is can depend on anything about that individual despite the fact that everything about him or her is ultimately attributable to the laws of nature and antecedent circumstances.
The position that everything about us is attributable to the laws of nature and antecedent circumstances is known as determinism. There is philosophical controversy regarding the compatibility (or incompatibility) of determinism and free will. Regardless of this philosophical controversy, it does seem as though if determinism is true, everything about us is ultimately beyond our control.
The Problem of Moral Luck and Potential Solutions
By this point, it should be clear that the control principle is incompatible with the way we practice moral assessment. This is the problem of moral luck.
There are two popular responses to this problem. First, there are those who would have us reject the control principle in order to preserve the way we practice moral assessment.2 Second, there are those who would have us alter the way we practice moral assessment in order to preserve the control principle.3
If neither of those options seems plausible, one could give up on the possibility of moral assessment altogether or embrace a revisionist solution. Bernard Williams, for example, recommends that we draw a distinction between two kinds of assessment, moral and ethical. While affirming the truth of the control principle on which people cannot be morally assessed for what is due to factors beyond their control, Williams argues that people can yet be ethically assessed for what is due to factors beyond their control. Furthermore, he contends that ethical assessment is the more important of the two kinds of assessment.4
Although none of these solutions seems entirely satisfactory, one of them must be correct.
1Nagel 1976, 138.
2For examples of this response, see Adams 1985, Walker 1991, Wolf 1993, and Fischer and Ravizza 2000.
3For examples of this response, see Richards 1986, Sverdlik 1988, Zimmerman 2002, and Enoch and Marmor 2007.
Jonathan is a Ph.D. candidate in philosophy at the University of Colorado Boulder. He holds a B.A. in philosophy and political science from Ashland University and an M.A. in philosophy from of the University of Missouri—St. Louis. His primary areas of interest include normative and applied ethics, and when he’s not doing philosophy, he enjoys playing golf, taking photos, and spending time with his wife and daughter. Website: spot.colorado.edu/~josp8471