Moral Luck and Messy Ethics: An Argument against Moral Realism

 

  

    Lately I've been thinking about how much of a mess our moral intuitions and normative ethics in general are, and whether this fact could be developed into an argument against moral realism. While I am not sure enough about the thoughts I have on this topic to claim that they constitute a successful argument, I want to share them and see how they might be best developed here. First, let me expand on what I mean when I say that morality is a 'mess':

    Consider moral luck: One of the fundamental principles of morality that we all take for granted in our ordinary moral practice is that factors completely outside of our control should not be relevant to our moral standing. People cannot be assessed for what is not their fault and is instead just a matter of pure luck (Call this the 'control principle'). And, indeed, this moral judgement is clearly reflected in our criminal justice system: we treat murder worse than attempted murder, which is worth than manslaughter, etc. Yet, the problem is that if we were to consistently apply this principle across normal ethical judgements that we make, it would completely erode our ordinary judgements about morality. This was famously pointed out in an essay by Thomas Nagel, and I will just summarize some of the examples of moral luck that he gives.

    First, consider resultant luck. Often, we are judged for the way things turn out. For instance, consider someone who drives recklessly and then hits and kills a child, or someone who forgets to check his car break and then hits and kills a child because they cannot break on the road. Someone in this scenario would feel immense guilt and blame themselves for their negligence. Yet, imagine the same scenarios yet in which the person's negligence went unchecked: though they drove recklessly, no one was killed, and though they forgot to check their breaks on time, no one was killed. Legally and morally we treat this latter person as far less blameworthy, despite the fact that the difference between these cases is purely a matter of luck. Whether a child happens to walk out in front of your car is a factor beyond your control. Thus, the outcome of situations influences people's culpability in ways that are purely the result of luck, in violation of the control principle.

    Second, consider circumstantial luck. Often, we are blamed and praised for actions that are the result of circumstances we happen to be subject to via luck. For instance, consider that we took someone living in modern America and put them in Nazi Germany. Such a person would have a much greater opportunity for courage or cowardliness. Imagine that this person, if they had lived in such a time and place, had went on to commit horrible crimes. We clearly do not judge them to be as culpable for this than the people who actually did happen to be in such circumstances.  People who happen to be lucky or unlucky enough to face certain moral tests are treated differently than those who never have such an opportunity Yet, the difference between those actually born in Nazi Germany and those who were not is as matter of luck. 

    Third, consider constitutive luck. The kind of person that you are in character and inclination is something that you are praised and blamed for. For instance, consider that one day you learn that I give much of my salary to charity. You would likely think: ‘Wow, Sebastian is really a good person. He must have a truly empathetic and giving character’. However, now suppose that you learn that though I give to charity, I am actually inherently greedy in disposition. I internally want to keep all my money, and I do not enjoy giving. I only do it by controlling my greedy impulses. You may think that it is praiseworthy that I can overcome my greedy impulses. However, you would likely blame me for being greedy. You might say: “If you secretly hate giving away your money and are greedy, that indicates a vice. You are not giving away the money for genuine reasons. In fact, your character is a fundamentally bad one, you should enjoy giving to charity.” In this case, you would be blaming me from the way I’m constituted. However, there is something strange about your moral evaluation in this case. The reason I attained this inherent greediness is a result of my genes and the teachings of my parents, and that I had these genes and this family environment is a result of luck. If you had gotten those same genes and teachings, you too would have been a greedy person. We do not think you should judge people differently based on factors that are completely outside their control. Yet, the fact that I am constituted to be greedy is outside my control since I did not choose the background forces that determined my character.

    Moral luck leaves us with a puzzle. We want to praise people for their actions and their character, yet the chance to exercise these actions, their results, and people's inherent character is all the result of luck. One kind of response to these concerns about moral luck is to abandon the control principle in order to preserve our ordinary judgements. Another is to keep the control principle, and reform all of our ordinary judgements. Yet, neither of these moves is satisfying, since they require giving up something that seems essential to morality. What I want to consider is a third possibility: that we can embrace these counterintuitive results all at once, by admitting that moral discourse does not generate true moral judgements, in the stance-independent sense. This possibility prevents us from having to make ad hoc, implausible moves to save moral discourse. Only if you are a realist do you need to make these moves as a matter of preserving the integrity of your theory.

    Put another way, the fact that we have such deeply engrained, conflicting moral judgements is better explained via the theory that there is no fact of the matter about the correct moral principle. Suppose moral anti-realism is true. If anti-realism is true, then plausibly the reason we come to form the moral judgements we do is because it was beneficial for our survival and because of cultural conditioning. There is no reason, however, that such surface level principles or reflections on them should lead to consistency. It may very well be that what is adaptive to believe in one situation is not adaptive in another situation, or that what is adaptive in one situation happens to conflict with what we want to think upon sustained reflection on a topic. While we might want to reconcile incompatible judgements for practical reasons or for our own psychological satisfaction, doing so is not a requirement of our metaethical theory. However, if moral realism is true, and we think that when we make moral judgements or reflect on moral claims we are actually getting at what is true, it is deeply surprising that fundamental moral principles would lead to such contradictory results.  After all, on the assumption that almost all moral realists want to make that we can actually access these moral truths, it is surprising that what seem to be the most essential parts of these truths are paradoxical. In a weaker sense, what moral luck at least suggests is that our deepest, most important intuitions and conceptual analysis of morality lead to moral principles that are contradictory, which challenges the epistemological reliability of these sources of moral reasoning. Note that I am not claiming that moral realism is logically inconsistent with the counterintuitive results of moral luck, just that moral antirealism is a better explanation for the bizarreness of this result than moral realism is. 

    While this argument has similarities to the argument from moral disagreement, I think it is relevantly different, and much stronger. What this argument appeals to is the fact that deeply held, shared moral judgements that seem to be at the essence of morality yield counterintuitive, contradictory results for the ordinary practice of morality. The nature of ordinary morality is that it is paradoxical, our judgements to not cleanly submit to any moral system. This is different from citing the sundry disagreements that pervades our normal ethical lives, which are usually about applied moral principles but still often appeal to a base of fundamental axioms most people agree on. It is not surprising that we might disagree and get things wrong about matters for which there are true, stance-independent facts. Yet, it is surprising that for true claims that we think are essential to a particular matter, that those claims would be paradoxical and even inconsistent, at least insofar as we think we have a reliable means to access that matter.

    One thing someone might want to say in response is that we might expect, or at least not be surprised, that basic folk intuition about morality is deeply counterintuitive on realism, but that sustained reflection about morality and discourse in philosophy leads us to resolve these initially absurd results. Yet, moral luck still infects modern ethical theories and has been criminally underdiscussed in philosophy. Additionally, normative ethics is a completely split field: philosophers are basically divided into thirds about what stance to adopt, there have been no signs of converging ethical consensus. 

    In this post, the data I focused on was moral luck. This is just one way in which ethics strikes me as a highly messy area. Other data that I might have appealed to is that our moral intuitions about paradigm moral cases diverge drastically in contradictory directions. And, resolving the dispute between which system to erect based on these moral intuitions strikes me a nigh-irresolvable. There are many challenges for the realist to face when it comes to moral messiness. 

    This argument can also be seen as a way in which arguments from moral progress and moral convergence rely on understated evidence. Though there may be evidence that we are progressing morally as a species, there are still deep paradoxes at the root of our moral thinking that pervade our everyday lives. Whether these will eventually resolved is an empirical matter, one that could be at the heart of future debates in metaethics.