Mill argues as follows: the sole evidence that something is visible is that people actually see it; the sole evidence that something is audible is that people hear it; analogously, the sole evidence that happiness is desirable is that people actually desire it. Since each person desires their own happiness, happiness is a good to each person, and therefore the general happiness is a good to the aggregate of all persons. Critics have objected that 'desirable' does not mean 'capable of being desired' in the way that 'visible' means 'capable of being seen' — it means 'worthy of being desired,' which is a normative claim requiring normative argument, not mere observation.
Whether or not the proof succeeds, its strategy is significant. Mill aims to ground ethics in observable human psychology rather than in a priori intuitions, divine commands, or social contracts. The ultimate moral facts, he believes, must connect with what humans actually are and actually want. The proof is not a deductive proof in the logical sense — Mill explicitly says that ultimate ends are not susceptible to proof in the ordinary sense — but an appeal to the best evidence available: our deepest and most reflective desires, once examined and refined.
Mill further argues that the only things desired as ends — rather than as means to happiness — are virtue, money, power, and fame, and that these are desired because they have become constitutive parts of happiness for those who value them. A person who truly desires virtue does not desire it as a tool but as something whose exercise is itself a component of their happiness. This move allows Mill to accommodate the intuition that virtue is intrinsically valuable without abandoning the claim that happiness is the ultimate end.
Chapter 4 of Utilitarianism. G.E. Moore's criticism of the 'naturalistic fallacy' in Principia Ethica (1903) is largely directed at Mill's proof; subsequent defences have argued Moore misread what Mill was attempting.
