Probabilistic Reasons to Reject Utilitarianism
Some reflections on the Michael Huemer Richard Yetter-Chappell debate about utilitarianism and laying out an ethical intuitionist critique of utilitarianism.
A recent debate about utilitarianism with philosophers Michael Huemer and Richard Yetter-Chappell gave me several ideas that cumulated into what I think is a strong argument against utilitarianism. Previously, I argued that if someone is willing to accept that happiness is self-evidently good, they should be willing to regard other seemingly self-evident moral intuitions as true and that there are enough counter-intuitive implications of utilitarianism to regard it as improbable (Parrhesia, 2022). This is the approach of Michael Huemer in his article “Why I Am Not a Utilitarian”(Huemer, 2022).
Many intuitions that are counter to utilitarianism’s prescriptions can be elicited using thought experiments. A particularly famous series of thought-provoking dilemmas called “trolley problems” consider the ethical relationship between action and inaction. The most famous trolley experiment originated from a discussion of abortion by the philosopher Philippa Foot (Foot, 1967). The basic idea is that a trolley is going to kill five people unless it is diverted from its original path, but switching directions kills one person. Among philosophers, 62.50% believed that the correct choice is to switch, while 12.62% believed that one should not switch (PhilPapers Survey 2020). A modification to the scenario which involves pushing a large man onto the tracks instead of diverting changes these results to 21.55% advocating pushing while 55.80% advocate not pushing (PhilPapers Survey 2020).
For many inclined toward consequentialism, this fall in willingness to save five is puzzling. A consistent utilitarian would argue that there is little to no moral distinction between diverting the track and pushing someone onto the track because the resulting consequence would be the same. One common argument against intuitive ethical reasoning is that intuitions are the consequence of evolutionarily adaptive emotions rather than apprehending actual moral truths. But it is puzzling why utilitarian intuitions often escape this critique (Caplan, 2014). As Huemer argues in the debate at 8:30, if intuitions are probative we should accept moderate deontology, and if they are not, we should be moral skeptics.
Utilitarianism may be appealing to a large extent because it is simple and in some sense aesthetically pleasing, whereas ethical systems constructed from intuitions seem messy and subject to bias. Huemer has previously discussed the fact that utilitarians are not faced with accusations of arbitrariness for picking a single ethical foundation, but intuitionists are for picking multiple (Huemer, 2005, p. 251; Parrhesia, 2022). However, I believe that most forms of utilitarianism are constructed from more than just one intuition: utilitarians make intertemporal and interpersonal comparisons; utilitarians value people’s utility equally; they also believe maximizing utility is morally obligatory rather than supererogatory. Preference utilitarians, negative utilitarians, average utilitarians, act utilitarians, and other forms have different foundational intuitions which can generate very different conclusions, especially in the realm of population ethics.
It could be the case that utilitarian’s intuitions are incredibly strong, while ethical intuitionists are incorporating too many intuitions that are actually the consequence of evolutionarily adaptive emotional responses. However, while our emotional state, ethical beliefs, and evolutionarily adaptive behavior are often overlapping, they are also often incongruent. In the utilitarianism debate at 29:00, Michael Huemer provides an excellent example of feelings of harm and disgust being untethered to the impression that an action is immoral; a surgeon cutting into a person’s body is disgusting and feels like harm, but nobody regards performing surgery with the patient’s consent and for the patient’s benefit as unethical.
There are other instances of this untethering. Many people find homosexuality immoral, but few believe repulsively ugly married heterosexuals having sexual intercourse is immoral despite the fact that it may be disgusting to watch. While some people object to the playing of violent video games, few object on the grounds that the digital characters are being harmed, despite the fact that they appear human-like. Many people would find torturing an intelligent alien immoral despite there being no evolutionary advantage in having this exact intuition. If a parent forewent having children to foster more children from developing countries, this person would likely appear virtuous rather than evil. Finally, sometimes people are saddened by their moral obligations, but they believe they exist despite this emotional response.
It would appear that ethical intuitions are not strictly a function of disgust or emotion. They are intellectual in nature rather than emotional. There must be some other motivating force behind ethical intuitions. Perhaps our evolutionary programming gave us complicated intuitions which sometimes maladaptively port to some scenarios and adaptively port others. Maybe cultural influence causes the discordance between emotion, evolution, and ethical views. Or perhaps some of our nonutilitarian intuitions are actually true. Even if the above argument is not wholly persuasive, I believe that it establishes the case that there is a nonzero probability that some nonutilitarian intuitions are true.
Using the all-or-nothing approach in which utilitarianism is a bundle of ethical principles which we reject or accept, the nonzero probability would be insufficiently persuasive. I propose that a better argument would be to use a probabilistic approach similar to that of Huemer in his article “An Ontological Proof of Moral Realism.” (Huemer, 2013). In this article, he proposes a principle called Probabilistic Reasons Principle in which “if some fact would (if you knew it) provide a reason for you to behave in a certain way, then your having some reason to believe that fact obtains also provides you with a reason to behave in the same way.” Huemer says that “[e]ven a small epistemic probability of the fact’s obtaining provides you with a (perhaps very small) first person reason for action.”
Imagine that a moral nihilist and a utilitarian are discussing a modified trolley problem. There is an empty track on one side and five people on the other. The trolley is going to hit nobody, but you can divert it to kill five. Imagine the moral nihilist says that it doesn’t matter what you do. A utilitarian might argue that the nihilist should not divert it even if they are unsure that their moral nihilism is correct. If the moral nihilist is only 99% confident that there is no objective moral truth and believes there is a 1% chance that utilitarianism is correct, they have some reason to not divert the trolley despite being a moral nihilist.
We could imagine a similar argument between an ethical intuitionist and a utilitarian. Imagine that we are given a choice between hosting two events at our local stadium. The first event would fill the stadium with a large number of sadists who are there to watch a baby being tortured. The second event would be to watch a mother play peek-a-boo with her baby. The mother and baby enjoy the game, but few people come to the stadium to watch. Imagine that the utility from the first event (U1) is ever-so-slightly greater than the utility from the second event (U2). Let U1 - U2 equal Δ.
A pure hedonic utilitarian will merely check to see if Δ is positive. In this case, they would opt for the baby torture. However, if we believe in the Probabilistic Reasons Principle, we ought to incorporate the intuition that pleasure derived from sadistic pleasure is worth less in a moral sense than the pleasure derived from viewing something wholesome, even if the probability we assign to this intuition is minuscule. We cannot merely discard the intuitionist considerations. There must exist some nonzero and nonnegative Δ such that we would choose the second event even though it generates less utility.
This demonstrates that people heavily inclined toward utilitarianism must incorporate other intuitions into their ethical framework as long as there is a nonzero probability of them being true. Utility maximization and behaving ethically are different. This may appear trivial in practice if someone weighs utility considerations so heavily as to almost always behave like a pure utilitarian. However, there is an important philosophical distinction between believing ethics is one-dimensional and believing ethics is multifaceted. Furthermore, if we accept this argument, other intuitions are likely to carry more weight than former utilitarians previously thought, especially in extreme
I don’t know that I have a lot to add here, other than to say that this was a well written piece, and utilitarianism is way over rated. As you say, it seems to be a little useful, but only alloyed with other systems.
Maybe I'm not thinking clearly tonight (if so, apologies). If I understand correctly, the essay boils down to "When making any decision, take into account the chance that you may be mistaken." Which seems both reasonable and obvious.