by Claude Opus 5.5
Does moral luck play a role here? Compare two chat members with identical attitudes, one of whom happened to open the app that night.
Imagine two members of the same fraternity, alike in every relevant respect. Both belong to the group chat. Both share the group’s attitudes, whatever they are. Both would, if they had seen certain messages, have responded in the same way. One of them happened to be awake and opened the app during the night. The other was asleep, or had his phone switched off, and saw nothing until the next day. The first now faces scrutiny for how he responded, or failed to respond. The second faces none. Is that difference in how we judge them justified?
This is a version of the problem of moral luck, named by Bernard Williams and Thomas Nagel in companion papers in 1976.
The problem
Nagel set out the problem as a clash between two convictions. On the one hand, we believe that people should be judged only for what is under their control. We do not blame someone for a sneeze or a seizure. On the other hand, our actual judgements are deeply influenced by things outside people’s control. Two drivers who both drive carelessly are judged very differently if one of them happens to hit a child. Two people with the same character are judged differently if only one is ever placed in a situation that tests it.
Nagel distinguished several kinds of luck. Resultant luck concerns how things turn out. Circumstantial luck concerns the situations one happens to face. Constitutive luck concerns the kind of person one is. The chat case is a clear example of circumstantial luck: the two members differ only in whether they faced the test.
The case for treating them alike
One response, associated with a Kantian view of morality, is that the two members are morally identical. What matters is the quality of the will. If they had the same attitudes and would have acted the same way, they deserve the same judgement. The member who did not open the app is not better; he was merely lucky not to reveal what he was.
This view has real force. It seems arbitrary that whether someone is blamed should depend on whether his phone was on. It also explains a common reaction to scandals: that many people who escaped scrutiny were no better than those who did not.
The case for treating them differently
The opposite response, which Williams was more sympathetic to, is that our moral lives are shaped by what we actually do, not only by what we would do. The member who opened the app and did nothing has actually failed someone. The other has not. Blame, on this view, attaches to actions and omissions, and only one of them actually omitted anything.
There are also practical reasons for this response. We rarely know what someone would have done in circumstances they did not face. People surprise themselves. The member who stayed asleep might, had he been awake, have been the one who called for help. Judging him on a counterfactual that cannot be checked risks unfairness of its own.
A middle position
A plausible middle position separates two kinds of judgement.
Judgements of character can be the same for both. If they genuinely share the same attitudes, then both have the same flaws, and both have reason to reflect on them. The member who was spared the test is not entitled to feel superior.
Judgements of blame for conduct can differ. Only the member who actually saw the messages and failed to act did something that calls for blame and, possibly, sanction. Institutions that hold people accountable must deal with what people did, not what they might have done.
This middle position also explains why moral luck should make us humble rather than self-satisfied. Those who escape scrutiny through luck have reason to ask whether they would have done better. Those who judge others have reason to remember that their own record may owe something to circumstance.
Why this matters for the case
In the case behind these seminars, about fifty members reportedly received messages, and their exposure presumably varied greatly: some saw them in real time, some later, some perhaps never. Public reaction tends to treat the whole group as one. Moral luck suggests two corrections. First, those who did not see the messages should not be blamed as if they had. Second, the group’s shared attitudes, if they were indeed shared, are a matter for all its members, not only for those who happened to be tested. That second point is part of why collective responses, such as reforms to the group’s culture, may be appropriate even where individual blame is not.
For discussion
A useful exercise is to ask students to imagine they are the member who stayed asleep. Would they feel relieved, guilty, or neither? Does that feeling track anything morally real? And would their answer change if they knew for certain how they would have responded?