The non-problem of moral luck

The Good Place - trolley dilemma

The Good Place is an unusual comedy on NBC, featuring a professor of moral philosophy among its main characters. My friend Skye Cleary has interviewed the real life philosopher who consults for the show, Todd May of Clemson University, for the blog of the American Philosophical Association. The exchange is definitely worth a read. In this post I will make an argument that one can learn more about moral philosophy from watching a single episode of the show than by listening to a technical talk in that same field while attending the APA’s own annual meeting.

Episode five of the second season of TGP features a sophisticated discussion of the infamous trolley problem, a thought experiment in ethics that has by now generated a cottage industry among both philosophers and neuroscientists. I will not explain for the n-th time what the problem consists of, you can look it up on Wikipedia. Suffice to say that the more I study virtue ethics, the more I become skeptical of the value of much modern moral philosophy, with its indulging in more and more convoluted hypothetical situations that seem to be designed more to show off the cleverness of the people working in the field than to actually help the rest of us live an ethical life. It is no coincidence that the dilemma is always framed in terms of what a deontologist or a utilitarian would do, those two frameworks having gotten further and further away from any relevance to real life, contra to what either Immanuel Kant or John Stuart Mill surely intended.

At any rate, the episode in question features a theoretical lecture on trolleys by the resident philosophical character, Chidi (played by the excellent William Jackson Harper). One of those on the receiving end of the lecture is the demon-turning-good-guy Michael (played by the awesome Ted Danson). During the lecture, Michael becomes impatient with the theory, so he snaps his fingers and transports Chidi, his friend Eleanor (played by Kristen Bell) and himself aboard an actual trolley, about to kill what appear to be real people. Michael then asks Chidi for a real-life demonstration: what is the philosopher going to do when suddenly faced with the dilemma, in the field, so to speak? Hilarity (and mayhem) quickly ensue. The episode is so good that I made my students watch it and comment on it.

Michael’s point is well taken: ethics is not (or ought not to be!) a theoretical exercise in cleverness, but a guide to navigating life’s real situations, and Chidi the philosopher — while very good in theory — fails spectacularly at it. I was thinking of that sit-com imparted lesson while attending a talk at the Eastern APA meeting last January, delivered by Philip Swenson of the College of William and Mary. In the following I will pick on Swenson a bit, not because his talk was bad (it wasn’t), but because it is an example of a way of doing philosophy that I increasingly object to, on ground of indulgence in irrelevant logic chopping.

Swenson set out to propose a solution to the “problem” of moral luck. He began, of course, with a couple of hypothetical situations:

Resultant luck case. Alice and Bill both go on walks along a riverbank. Both encounter a drowning child and attempt a rescue. They make the same choices and attempt the same actions. Alice’s rescue succeeds, but a sudden current prevents Bill’s attempt from succeeding, and the child drowns.

Circumstantial luck case. Alice goes for a walk along a riverbank and encounters a drowning child. She rescues the child. On a separate occasion, Claire goes for a walk along the riverbank. She does not encounter a drowning child. If Claire had encountered a drowning child she would have rescued the child.

What’s the problem? I mean, other than for the (fortunately hypothetical) child who occasionally drowns? Swenson is bothered by the fact that, in the first case, if we say that Alice is more praiseworthy than Bill, it looks as though we accept something apparently horrible called “resultant moral luck.” In the second case, if we say that Alice is more praiseworthy than Claire, then we accept something equally objectionable, called “circumstantial moral luck.” As Swenson puts it:

“Rejecting circumstantial moral luck appears to require a very significant revision to everyday moral judgment. Consider the plausible claim that a great many people all over the world are not so different from those who went along with the rise of the Nazis. Many people would have done similar things under similar circumstances. If we accept this and reject circumstantial luck then it looks as though some radical claim or other will follow.”

That would be, in case the reasoning isn’t clear, the radical claim that most of us are not as good as we think, and that if we had lived under the Nazi we would have been just as culpable as the majority of the German population of the time for the Holocaust. But it doesn’t end here, there is a third case to consider:

Constitutive luck case. Alice goes for a walk along a riverbank and encounters a drowning child. she rescues the child. On a separate occasion Daniel goes for a walk along the riverbank and also encounters a drowning child. Because Daniel is — through no previous fault of his own — cruel and uncaring, he refrains from rescuing the child. However, if he had possessed Alice’s naturally caring disposition, he would have rescued the child.

Swenson went on to remind the audience of the two classical “solutions” found in the philosophical literature for the problem of moral luck: “responsibility skepticism” (deny that anyone is ever praiseworthy or blameworthy at all), and the “responsibility explosion” (say that people are praiseworthy or blameworthy in virtue of what they would have done in various circumstances they never actually faced, equipped with character traits they never had).

He then goes on to present his own solution to the problem, which involves a strange calculation of moral desert levels, beginning with the assumption that the “expected desert level” for an agent is zero, and continuing with the notion that we can then assign points to different ethical situations according to a variety of criteria. I will not go into the details because they are irrelevant to my argument here. Which is that we should reject this whole approach to moral philosophy, period.

To begin with, I find bizarre the very idea that we should engage in some sort of morality ledger construction, keeping score of the praiseworthiness or blameworthiness of people. Why? What the heck is the point? Are we some sort of god who has to decide on where to send people in the afterlife? (That, incidentally, is the premise of TGP show. And it’s very funny.) Take the first scenario, the case of resultant luck. It wouldn’t cross my mind for a second to say that Alice is more praiseworthy than Bill just because Bill did not succeed in his attempt at rescuing the drowning child. On the contrary, I am in awe of anyone who would attempt the rescue, regardless of whether s/he succeeds or not.

The circumstantial luck case is even more out there: there is no reason for us to consider Claire at all. If the circumstances were such as not to test her moral fiber, fine, why should that be an issue of any sort? Alice is to be praised for her attempted (and successful) rescue, the question of what Claire would have done simply did not arise, and that’s the end of that.

The last scenario, that of constitutive luck, is interesting, but only academically. To begin with, my view — contra Swenson’s stated hypothesis — is that adult human beings are morally responsible by simple virtue of being adults. That’s what it means to be an adult, regardless of the circumstances of one’s childhood. But if Daniel has an aberrant character because, say, of some developmental abnormality in his brain, or perhaps because a tumor is interfering with his moral decision making brain network, then fine, he is not to be blamed for his inaction. That’s no skin off of Alice’s nose, because moral desert is not (or should not be) a competition! Again, why the karmic obsession with keeping scores?

What about the choice between responsibility skepticism and the responsibility explosion? It seems to me that a society cannot function without a reasonable attribution of responsibility for the actions of its (adult, normally functioning) members. But one shouldn’t be carried away and start thinking of all possible hypothetical scenarios. Ethics should be concerned with what actually happens to real people, not with how hypothetical individuals would behave under (infinite) hypothetical circumstances. If you care about the latter, I suggest you’ve got your priorities seriously screwed up.

In the end, the “problem” of moral luck is not a problem at all. When Thomas Nagel wrote his now classical paper by that title, back in 1979, I took it to call our attention to the humbling fact that we may be far less moral than we like to think, and that that observation ought to make us more sympathetic toward the above mentioned ordinary Germans under the Nazi. To cure us of moral hubris, as it were. That is a very good practical lesson, nudging us toward being both less complacent about our own abilities and more charitable toward the shortcomings of others. But if the whole thing degenerates into an entirely impractical mathematical exercise in the assignment of praise and blame we have lost track of what ethics should be about. As the Stoic philosopher Epictetus put it 19 centuries ago:

“If you didn’t learn these things in order to demonstrate them in practice, what did you learn them for?” (Discourses I, 29.35)

106 thoughts on “The non-problem of moral luck

  1. davidlduffy

    Nothing to do with moral luck, I’m afraid:

    We have a whole realm of science that deals with how humans reason (in the sense of operating in the space of reasons) about “gut feelings” about value, viz decision theory, economics and psychology. Bunge hates decision theory (arguing that utility is plucked out of the air), but says:

    Evaluation is a kind of cognition (Lewis 1946). A rat that presses a lever for more food pellets of a certain kind knows that the food is good for him – but he will revise this evaluation if the last time he sampled food laced with poison. A person who has learned that keeping promises is necessary to stay in good social standing knows something about certain values, and she has come to know it on the strength of experience and reason. And a person who realizes that she has made mistakes in value attributions will correct the latter much in the same way as she corrects perceptual or calculation errors.

    …Value judgments can be justified or criticized, rather than accepted or rejected dogmatically, when they are rooted to basic needs or legitimate wants. In this case they can be shown to be true or false.

    Related to GD’s comments: “Many epistemologists have thought there cannot be any doxastic norms because we do not or cannot have voluntary control over what we believe (Alston 1989). But this contradicts the social facts as Social Epistemology (Coady 1992; Goldman 1999) or an Economic Theory of Knowledge (Hardin 1997; 2002) make clear.” [my emphasis; unfortunately I can’t remember who wrote this ;)]

    As to the ought-is gap, this;rgn=main

    discusses Searle and SInger and what they thought about dissolving the gap.


  2. Massimo Post author


    to be fair to Garth, he surely doesn’t derive his conclusions from knowing himself, but from general principles. And if he is right, you are an automaton regardless of what your introspection tells you…


  3. Massimo Post author


    “if you are right on hard determinism, then it follows that any response to your argument, positive or negative, is also pre-determined, so why bother arguing? I’ve seen the response to this, from Jerry Coyne and others, that argumentation effectively changes the electrochemical nature of the brain, but this will not do: If the responses of the initial brain-state are pre-determined, then the response to changes in the brain-state itself will be pre-determined”

    Again my chemicals this mornig are predisposing me to be fair to Garth (and to Coyne!). That position is not strictly incoherent, your critique simply highlights that there is no point to any conversation because we are all just going through motions that were set in stone since the Big Bang. We can’t avoid it. Such nihilism ought to convince us to commit suicide, except most of us won’t, because… the Big Bang.

    Liked by 2 people

  4. Massimo Post author


    why do you think that hard physical reductionists promise a perfect human (whatever that means)? For that matter, neither does Marxism, though the latter does promise a better human society.


  5. labnut

    to be fair to Garth

    To be fair to me, I am indulging in some gentle humour. I hope you noticed the play on words! I should have used an emoticon.

    We routinely flog this matter to death but never conduct a funeral, so I thought I would lighten the mood with some ironic humour.

    Liked by 2 people

  6. SocraticGadfly

    EJ: I again stress that Coyne et al are really engaging in a tautology, in my opinion. They think that hard determinism is a necessary philosophical stance to support ontological non-dualistic naturalism. Such is not true, of course.

    That’s the most charitable defense I can offer, and again, why they need to learn more philosophy.

    Garth allegedly knows more philosophy than Coyne …

    Liked by 1 person

  7. Philip Thrift

    On determinism: I’m not quite sure what some physicists or others mean by that term, but from a programmer’s point of view, there is a lab at the University of Bristol that is making a chip (see that could be part of the hardware of like a MacBook or iPhone to generate absolutely random numbers, which some programmers need for some applications.


  8. brodix

    Whether personal morality, communal, cultural ethics, or civil laws, they are defining structures and strictures of society, against which emotional, organic, biological impulses and energies are going to be pushing on and against.
    The effect is either a positive relationship, where the structure grows, heals, is maintained, as it provides a skeleton and support for the people and community.
    Or it is a negative relationship, where the structure is unsuited, rigid, contradictory, corrupted, weak, etc and that organic energy breaks it down, seeks weaknesses and cracks to exploit, drains off to other directions, etc and the structure declines.
    The notion of will is emergent from and descriptive of this process and relationship.


  9. Philosopher Eric

    To continue on with my last thought (found here:, apparently for two and a half millennia western philosophers have been exploring the nature of value by means of aesthetics and ethics, but without developing effective definitions for the “value” term. This is good news! It means that perhaps once we are able to develop effective value definitions, then it will be possible for the field of philosophy to develop experts that have their own generally accepted understandings in this regard to share with humanity!

    I’ve provided a consciousness based definition from which to potentially found axiology. Would anyone like to submit a different one? Or indeed, would anyone like to argue that philosophy doesn’t need effective definitions for value, in order to effectively explore value based ideas?


  10. synred

    we do not or cannot have voluntary control over what we believe

    The protestant idea of salvation by faith never made sense to me. At d leave. At least I never felt I had choice about what I believe or don’t.


  11. synred

    Most physicist to not think the world is deterministic. There is an element of randomness though things are not completely random either. Probabilities are determined.

    Again I don’t see how this helps with ‘radical freedom’. Of course it seems to us we make decisions and we have to cope. We have no choice but to chose — existentialism?


  12. Robin Herbert

    The philosophers that Garth was citing, Pereboom, Caruso, are not pushing any kind of naive determinism, in fact Pereboom stipulates that the Humean argument is not sufficient to defeat even libertarian free will, which is an unusual concession for an incompatibilist. I have not read their stuff in sufficient detail and, in any case, I don’t think I am best qualified to evaluate their arguments.

    But the arguments do seem to me to be pretty abstruse grounds to make me doubt that the feeling I have of control over events is, in fact, genuine control (albeit very imperfect control).


  13. ejwinner

    hard-line Marxism offers a critique of everyday consciousness within the capitalist ‘bourgeois’ society as a “false consciousness,” generating a false sense of subjectivity. In other words, all of my judgments, desires, standards sense of community are tainted with a false sense of self-hood and self possession. In the wake of a global revolution, and the achievement of a communist society, however, we will realize our true subjectivity – each will know who she or he truly is, what he or she wishes to to do, to be, in proper relations with others, each will know exactly his or her proper place in the community, and what he or she may properly claim from the community, which enlightened community willingly grants. in short the realization of human hope and human possibility.

    Physicalist-reductionist hard determinism, simply as a metaphysical position, of course doesn’t necessarily offer us anything social. But when it is presented as a a grounds for rewriting our laws or justice system, or restructuring our language in an enlightened scientistic manner, when it suggests an enlightened economics or social environment predicated on the understanding that our hopes and desires are driven by the molecules of our being, rather than by persons or other folk explanation constructs – well, then we’re getting directed toward utopia. A blameless/praiseless society would, presumably, be one in which each could discover a ‘true’ self, and realize one’s true potential.

    As sidebar comment, I should note, as a Buddhist, that while Buddhism offers an enlightenment predicated on release from the Self as source of desire and anxiety, this is achieved only through individual discipline, and in no way releases one from the burden of being a person in a social environment. In other words, everyday consciousness is just as it is and must be engaged to interact responsibly with others.

    BTW, it was a bit of interpretive hubris to suggest that the virtue ethic credo I offered is somehow the one true such credo possible. It is really just how I understand it; I hope I wasn’t stepping on any Stoic toes!

    Liked by 1 person

  14. brodix


    Value is the degree of benefit. The problem is when different benefits conflict. Then we get to the trolley problem.


  15. ejwinner

    Evidence on Marxism: It should be remembered that Firestone’s Dialectic of Sex, remembered as a text in radical Feminism, uses as its foundation the kind of hard-line Marxism of which I write, and offers us a future free of pregnancy, of work, of discrimination, of having to make difficult moral judgments – etc. I use the term ‘hard-line’ specifically to isolate such utopianism from the variants of Marxism that are primarily about economics, social justice or politics, or simply about achieving a ‘better’ rather than perfected society.

    Utopian Marxism is largely forgotten today, but it had its place among Left intellectuals for some decades – The Dialectic of Sex appeared in 1970.


  16. garthdaisy

    Okay first off I am not a hard determinist. I’m not even a determinist. I am agnostic on physical determinism because it’s also possible the universe is indeterminist. I am a hard incompatibilist because the notion of “free will” is incompatible with both a determinist and indeterminist universe. It is also incompatible with psychology, biology, and even the manifest image if one just thinks about it for a minute. It is easy to notice in the manifest image that one can do what want wills but one can not will what one wills.

    Next, “gut feelings” and “chemicals” = same thing. Chemical reactions in your body are what generate “gut feelings.” It is simply referring to the FACT that our choices are made by our subconscious rather than our conscious mind. The thing we think of as “the deciding self” is actually a lawyer after the fact. Psychology and biology make this quite clear. Your thinking brain does not decide your actions but rather justifies them a microsecond after they have occurred. So soon after in fact that most people think their thinking mind made the decision. Denial of this fact helps no one.


    Yes you can rethink current information but I think you will discover that this rethinking of current information changes your belief about the way the world actually is, then your unconscious feeling spit out a new ought based on the new “is”.

    From your example:
    “When I thought about it I realised that the opposite was the case and I was surrounded by giving people”

    That’s a new “is”. (surrounded by giving people rather than not surrounded by giving people). Based on this new “is” your feelings spit out a new ought. So to answer your query, rethinking current information can bring you to a new idea about the way the world “is.” The new ought is your feelings reacting to the new “is”.


    I despise Marxism, thought Marx was not a Marxist and had some pretty good ideas. The idea that anti-capitalists are therefore Marxists or Communists is a false dichotomy and a pathetic ploy.


  17. Philosopher Eric

    Value is the degree of benefit. The problem is when different benefits conflict. Then we get to the trolley problem.”

    The thing about that Brodix, is you’re thinking from a moral context here, or a social construct. I’m talking about something that is beyond morality, which is to say neither moral nor immoral. I’m theorizing an effective definition for what’s valuable and anti valuable from which to potentially found axiology. (From this amoral perspective all trolley problems questions actually get straightened out in terms of the “valence” or “happiness” of a given defined subject. That there are winners and losers is naturally to be expected.)

    If you now understand that I’m talking about something that’s more fundamental than ethics and aesthetics, there are a few ways that you could explore my position. You could submit that people already accept the theory of value that I’ve provided above (in which case I’d like some evidence). Or you could submit a different definition of value for me to consider. Or you could disagree with me that in order to have effective value based disciplines, it’s necessary to have effective definitions for the “value” term.


  18. Robin Herbert


    “Again I don’t see how this helps with ‘radical freedom’. ”

    I could only answer that if someone could tell me what the problem is supposed to be in the first place.

    There are all these people lining up to claim that some aspect of our mental processes is a “illusion” but they can’t say just exactly what this is.

    If they can’t be specific then I can’t even know what it is that I am supposed to be sceptical about.

    Without any good reason to feel otherwise it seems to me that the most rational course of action is to carry on believing that my mental processes work more or less the way they seem to work.


  19. Robin Herbert

    On the other hand the most useful take home lesson about freedom and morality is “moral luck” in the Nagel sense as Massimo describes it in the article.

    Yes, I would have likely been a Nazi had I grown up in Nazi Germany, I would have likely approved of rape if I had grown up in a society where rape is routinely used for social control.

    Fortunately I come from a background where decency and fair play were valued and where rationalism was encouraged.

    I think, though, that praise and blame still apply.

    But they apply, not to the cards we are dealt, but how we play those cards. And, yes I think we can employ our reason and volition to play those cards, however imperfectly.

    Liked by 2 people

  20. synred

    I could only answer that if someone could tell me what the problem is supposed to be in the first place.

    If you could only tell us what ‘radical freedom’ meant, perhaps we could tell you what is wrong with it. Be specific.

    To me it appears to be mini word salad.


  21. synred

    Without any good reason to feel otherwise it seems to me that the most rational course of action is to carry on believing that my mental processes work more or less the way they seem to work.

    Nobody doubts you can make decisions, have volition and so forth. The question is how that works — and nobody knows.

    What is your alternative to determined or not determined (a.k.a., random)?


Comments are closed.