5 terrible objections to utilitarianism

There are a ton of terrible objections to utilitarianism that sound plausible when given at 300 words per minute, but can ultimately be debunked with a basic understanding of philosophy. In this article, I’m going to harvest some of that low-hanging fruit in the hopes of lowering the usage of these arguments.

If you can’t prove induction works, then you have no idea whether or not any desired consequences of events will happen. This must mean all naturalist moral theories can’t resolve skepticism and fail!

First off, nobody in their right mind believes induction actually fails. I think most skeptics would be willing to wager that their next drink of water won’t kill them, or that the next time they jump they won’t be sent flying into space. Induction as a mechanism for predicting future events has a good chance of giving us an extremely useful heuristic for making decisions—accepting that it’s true is therefore still beneficial; you have nothing to lose and everything to gain from escaping universal skepticism of future events.

Claiming induction works because of the past instances that it has worked is circular, but that goes to show that proving a mechanism we use to prove all future actions shouldn’t be possible, not that we ultimately fail to prove induction works. I can’t test the accuracy of a ruler by measuring it against itself; though, as long as the ruler has successfully measured every object I’ve encountered so far, I have a good reason to continue using said ruler.

Proving that induction works deductively is also self-evidently stupid. If I were to present a deductive syllogism like this one:

P1: All humans are mortal

P2: Raymond is a human

C: Raymond is mortal

How do I know that P1 and P2 are true? I’m not an omniscient being that can peer into the future and observe every single person alive dying—I simply have inductively tested the proposition that “all humans are mortal”, and the success of my tests hints that the underlying fact of all men being mortal is probably true. Put simply, I observe every person dying at a certain age range, so I can assume that any person reaching that age range will likely die.

I also don’t know for certain whether or not Raymond is a man. There could be some alien species that look exactly like most of the people I know to be men, with the only difference being that they’re significantly smarter. But we know that the statistical likelihood of such a scenario is much lower than Raymond simply being a man; this can be a result of me observing traits that I’ve only seen humans have, and Raymond continuously acting in ways I’ve only seen humans act in.

Both of the premises I need to perform the deductive syllogism are contingent on inductive reasoning! If a delivery pipe only ever exists to deliver water from a pump, I cannot prove that the pump always exists using the existence of the pipe—the pipe’s existence is logically implied from the pump’s existence, making it impossible to prove the pump always exists without being self-referential (circular). Raising the problem of induction as a catch-all for naturalist theories doesn’t work.

If good equates to pleasure, they’re basically the same thing. So the statement that “pleasure is good” translates to “pleasure is pleasure”, which is a barren tautology.

Consequentialists can, despite this, easily ground their understanding of what’s morally good.

A response like this completely neglects that utilitarians point to a physical element of happiness and pleasurable experience, elements grounded in our physical reality, to refer to what is good. In this case, the dictionary definitions of the two words can technically be equivalent, but that doesn’t mean pleasure isn’t grounded in something we can experience physically.

The naturalistic fallacy response is probably the least logical objection to utilitarianism you can bring up. When Henry Cavendish discovered that water was actually H2O in the 18th century, did he simply bring up a barren tautology that proves nothing, or did he make a pivotal scientific discovery? Obviously the latter. You can’t just say, “water is then just H2O, creating a tautology”. Water being H2O provides valuable information on what it’s physically grounded in—1 oxygen and 2 hydrogen atoms.

No it isn’t.

Nobody can ever take every single action that increases utility into account; that requires supercomputational levels of logic! Even worse, nobody can predict the future consequences of an action; an art school rejection can cause a genocide, along with tens of millions of military deaths (nobody could’ve seen that coming).

This objection also ignores how utilitarians make decisions in the first place. When solving any problem, there can technically be an infinite number of better solutions nobody can think of—that doesn’t mean solving problems we know how to solve is futile. There is a possibility of a better math system that is far superior for calculations than ours, but none of that proves we don’t have to use our existing understanding of mathematics to solve math-related problems.

You can’t neglect moral permissibility; we are supposed to choose the morally best option that we are acquainted with. An ant that chooses option A, one that increases the total utility of the colony, is still acting morally if said ant couldn’t have possibly conceived of a better option—even if we can easily think of a better option, the ant couldn’t have acted in a superior way.

Our basic decisions are often extremely simple, yes/no actions or actions where we know of an existing, morally good way to act. You can also accept that there are spillovers, butterfly effects, and inadvertent consequences of our actions that can be catastrophic, but this does little to cast doubt on the efficacy of expected utility calculations. The chance of a pen dropping and causing nuclear war is going to be the same as the chance of a pen dropping and preventing nuclear war. In any chain reaction model, the predictive power of an event descends to 0 as time goes on.

There’s even a simple solution to all of this—we can build supercomputers that can increase our consciousness, simulate morally perfect actors, and transcend biological barriers, all with the goal of being able to conceive of the actions that create the most utility.

If there were a being that experienced 1 billion times more utility than every human alive for equal actions, utilitarians should sacrifice themselves to the utility monster—they should even have a moral duty to sacrifice their friends to the monster to create more utility.

This scenario makes people who aren’t at all acquainted with utilitarianism reject utilitarianism. You would need a utility monster that somehow avoids rapidly diminishing marginal utility to make this scenario work. Presumably, the monster should also be millions of times more conscious than humans in order to be experiencing astronomically higher amounts of utility for the same actions.

In this case, the utility monster is basically a god compared to humans. A great analogy can be made here. Humans are roughly 1 million times more conscious than fruit flies and experience much higher utility. If a community of 1,000 fruit flies needed to eat an apple to survive, but your friend is mildly hungry, I doubt your friend would even hesitate before grabbing the apple and eating it. To the utility monster, humans are barely alive, barely conscious, and don’t really seem like beings with equivalent interests—the relationship here is analogous to the one between humans and fruit flies.

Cows, pigs, and chickens have, on average, 3 billion, 2 billion, and 200 million neurons, respectively; compared to the 86 billion neurons humans typically have, these animals are only several times less conscious than us. Most humans don’t really care about the welfare of these animals. 9 out of 10 people in the U.S. eat meat—we can technically be viewed as utility monsters who gain pleasure from sacrificing less conscious moral agents and eating them.

Once you detach yourself from only thinking in terms of your own perspective, the utility monster scenario seems to prove practically nothing.

Share

Consequentialist theories can describe biological truths—such as humans seeking pleasure and avoiding pain—but this “is” can’t translate into a normative “ought” on what we’re supposed to do!

This argument doesn’t specifically take out utilitarianism; similar to the argument that induction fails, accepting this argument as true makes you default into moral skepticism or nihilism. Other normative ethical theories need to cross the is-ought gap too. Deontology assumes you ought not to violate perfect duties, and that you ought to prioritize perfect duties over imperfect ones.

How is this responded to? If simply proving that people don’t like pain but like pleasure doesn’t mean you are obligated to create pleasure and reduce pain, or if we can’t deduce normative obligations in the first place, is moral nihilism true?

In the sense that we can’t universally extrapolate ethical duties, nihilism is valid. We don’t get to discover hidden, universal moral obligations written in the stars. We do, however, have moral obligations that shift based on our biological constraints and moral interactions—ethics can be viewed as built around society, not vice versa. It’s a massive societal negative to reject your obligations to increase pleasure and reduce pain, or to abandon any sense of morality whatsoever and act like a psychopath.

There is no universal, relational worth for the U.S. dollar; $1 million in 100 dollar bills is equivalent to a pile of cotton and linen to an alien species that has only ever used bitcoin for their transactional needs. Humans, however, will find immense value in those 100 dollar bills based on how our society operates. A similar view can be applied to ethics, as our obligations don’t have to be universally grounded to exist.

Yes, I really did go for the low-hanging fruit here. The majority of these objections make it easy and intuitive for people who don’t understand utilitarianism to reject utilitarianism. So long as you’ve asked any person (or chatbot) how utilitarians deal with these objections, they stop sounding very plausible. I see these responses being fired off in rapid succession during “philosophy debates” where proponents of utilitarian arguments simply don’t have the time to cover them all on the spot; hopefully this changes!

(If you read my last article you know what debates I’m referring to)