Utilitarianism sounds like a good idea. After all, it makes sense to seek the greatest good!

Well, no. Here's a thought experiment that shows how utilitarianism would make you perform an obviously unethical act. I've no attribution for this, but it certainly isn't mine. The subject matter of the paper is, I think, the difference between killing and letting die.

You're the head of a hospital in some remote area. In 5 separate wards, you have 5 different patients waiting for transplants of 5 different organs. If you don't do a transplant tonight, each patient will die. If you do the transplant, they'll live (you can make it more realistic by giving each a 95% chance of survival; it doesn't change the argument). The 5 patients are 5/6ths of an identical sextuplet, between which you can transplant (no, this is not a very realistic scenario). And you have no organs.

Until sextuple #6 walks in, complaining of a slight headache, and asks for an aspirin.


  1. Give sextuple #6 an aspirin. Result: 1 person lives, the other 5 die.
  2. Take the opportunity to "cannibalise" sextuple #6 for "spare parts" (organs for the transplants). Result: 1 person dies, the other 5 live (if you're still with the 95% chance of survival, that's an expected utility of 1 dead person and 5.7 live ones).
Which do you pick?

Option #1, I sincerely hope. And before you go about complaining the whole scenario is impossible, consider that if you pick #2, you're willing to kill a person whenever their organs are compatible with 2 recipients.

What gives? Evidently you make ethical judgements based on principles other than utilitarianism.

In his most famously fallacious argument for happiness as the sole good, John Stuart Mill offers that as an object's being viewed makes it visible, so does a thing’s being desired make it desirable. The argument, taken straightforwardly, is heavily flawed, because Mill equivocates when using the word desirable.

The word desirable has three senses: the first is consistent with Mill's literal claim—that a thing is desirable if and when it is in fact desired, especially if it is commonly desired, (e.g., "Wealth is desirable"); the second sense means that a thing is merely capable of garnering desire, (e.g., "Personally, I consider broccoli to be quite desirable"); and the third imbues desire with a sense of moral correctness, goodness, and duty, (e.g., "Feeding starving children is desirable.") I will refer to these as desirable1, desirable2, and desirable3, respectively.

Mill is engaging in an obvious semantic game. He uses desirable1 in his 'proof', while hoping we will interpret it as the moralistic desirable3. But desirable3 is a property independent of the amount of actual desire people experience. That is, it is presumably a "desirable" thing to feed starving children, even if everyone in the world fails to desire it. Hence, Mill has offered no proof that happiness is desirable3, but rather, only that it is desirable1, which is morally irrelevant.

On its face, the most sensible way to improve Mill's claim is to reinterpret it as a truism, which would render the following:

Happiness is universally desired, and all desirable things are desired because of their ability to cause us happiness.

Now, it doesn't logically follow from the premise that "happiness is universally desired," that, "happiness is the only valued good." One might argue that, say, broccoli is valued by some people. But Mill could quite reasonably respond that regardless of the specific object of desire—i.e., whether we desire broccoli, wealth, or to feed starving children—our ultimate desire is for happiness. Broccoli, wealth, and altruism serve as means to the end of happiness.

Unfortunately, though, even this improved restatement will not prevent Mill's next, unforgivably silly step. He will proceed to claim that whereas "each person's happiness is a good to that person . . . the general happiness, therefore, is a good to the aggregate of all persons." Consider his claim alongside an equivalent one, to elucidate Mill’s illogic:

  1. I voted for a Presidential candidate in the November, 2004 Presidential election.
    cf. "Each person’s happiness is a good to that person."
  2. Millions of other Americans also voted for a Presidential candidate in that election.
    cf. "Whereas one person desires happiness for himself, a million people desire happiness for themselves."
  3. All American voters voted for the same Presidential candidate in the November, 2004 Presidential election.
    cf. "Each individual desires the happiness of the group."

Sadly, neither of these conclusions follow from the premises. While it is perfectly easy to accept as a premise the truism that every person desires his own happiness, this does not support the conclusion that a collection of persons desires a collective happiness.

It seems that Mill was trying to build a moral theory from a fact that David Hume had brought attention to, in the preceding century: Even when people make moralistic or altruistic decisions, they are ultimately guided by their passions—by their desires—and moreover, it could be no other way. Human behavior, Mill and Hume agree, is and ought to be passion-driven.

Mill's project was to show how we can use this innate, seemingly selfish pursuit of desire-fulfillment, ("happiness"), to render morally correct behavior. And though he successfully argues that people do, and ought to, desire their own happiness, he has offered nothing useful to support the claim that people do, or ought to, desire the total happiness of everyone. Why ought I use my money to feed starving children if it brings me no pleasure? What if I can get more pleasure out of buying a new plasma-screen TV instead? Utilitarianism compels the behavior which brings the most happiness to the most people, but supposing I don't get a 'warm feeling inside' for having helped others altruistically, what interest do I have in the happiness of people I may never meet? Absent an answer to such a crucial question, Mill's argument for utilitarianism is not compelling.

Node your homework

Log in or registerto write something here or to contact authors.