Ethical limits of utilitarianism
The principle of aggregation has been criticized for being a mere sum of utilities which rules out any distributive considerations. Indeed, in a crude version of utilitarianism, it is considered socially equivalent to give a large amount of money either to one rich person or to a myriad of poor people who would be able to change their quality of life thanks to it.
This example runs against moral intuitions and should induce the rejectionof the theory. Yet it seems implausible to claim that equality has never been a concern of utilitarian thinkers. In an old and standard assumption that goes back not only to the precursors of the marginalists, but also to the first utilitarians, the marginal utility of money is decreasing. In order to maximize the sum of individual utility, it is therefore socially better to give an extra euro to a poor person than to give the same extra euro to a rich one: for the former act is a bigger producer of utility than the latter. A way of representing the egalitarian functions of social welfare is specifically to retain a utilitarian utility function with concave individual utility functions. In this sense, the criticism over the issue of distribution holds only for crude versions of utilitarianism where 1 euro is supposed to provide the same amount of utility to any person; yet this case is more important than we might think at first sight, in view of Robbins’s claim that the assumption of decreasing marginal utility was a normative assumption and should consequently be kept away from economic science. Nevertheless, even with some given concave utility function, the choice of the criterion of sum (or average) may have consequences that collide with distributive concerns. Utilitarianism can judge certain inequalities as socially better in so far as they contribute to increasing the sum of utilities. For example, it may be appropriate to give subsidies to a rich businessperson rather than using the same amount of money to improve the fate of worse-off children if the extra utility associated with the grant is higher than that associated with assistance.
Some utilitarians are comfortable with this criticism since it implies, after all, an increase in the sum of utilities: the decisions of the businessperson, they say, may provide more jobs and wealth in a wide area, increasing the welfare of numerous families in a way likely to compensate for the low quality of the children’s lives. Opponents of utilitarianism then return with another example, where the unquestioned new increase of welfare only accrues to families that were already well-off. While the choice of examples in this vigorous debate seems crucial to determine which side is more convincing, it remains that there exist cases where the utilitarian criteria favour the rich over the poor, the well-off to the needy. In the end, it remains that distributive issues may be important in utilitarianism, yet only on instrumental bases. This dependency on utility as the sole intrinsic value still generates unquestionable violations of equality concerns.Beyond the issue of equality, the utility principle is likely to justify the sacrifice of minorities. In a famous example, utilitarianism underwrites the Roman spectacle of the circus. In the arena, a few Christian martyrs suffer a great deal, while each spectator derives a little pleasure from observing their suffering. If the spectators are sufficiently numerous, the sum of their low positive utilities shall compensate for the pain of the few Christians.
More generally, the focus on utility implies possible violations of any liberal rights. According to an interpretation of Sen’s (1970) theorem of the Paretian liberal, it is impossible to attribute an intrinsic importance to utility while simultaneously endorsing a specific concept of freedom. This interpretation has been generalized on the basis of other results pertaining to utility and other values.
Sen has called “welfarism” the doctrine according to which social welfare only depends on utility and on no other values or information (Sen 1979b, 1979c).
All versions of utilitarianism are fundamentally welfarist. However, welfarism is a problem, first, because it implies the rejection of any other values (or at least it subsumes them). Some alternative theories consider that a plurality of values should be at stake for welfare issues: they make a case for a multidimensional account of welfare, such as quality of life or capabilities (Nussbaum and Sen 1993) and complex equality (Walzer 1983). Secondly, utility may not be the suitable information for gauging social welfare, because it entails counterintuitive consequences. After Sen’s famous paper “Equality of what?” (Sen 1979a), it has been common to consider alternative informational bases of justice such as primary goods (Rawls 1971), resources (Dworkin 1981a, 1981b), access to advantages (Cohen 1989), welfare opportunities (Arnerson 1989), opportunities (Roemer 1999), etc.Utilitarianism considers that utility, which is important for individuals, is also the relevant information for resolving issues of justice. Some regret the confusion of justice and morality in utilitarianism, where goodness is prior to fairness. Conversely, in other traditions - mainly political liberalism - morality focuses on every individual’s judgement of what is good or bad; justice concerns all society and tackles the issue of living together, no matter what the individuals’ differences, and even considering divergences in their moral views. Consider the usual case of conflict of interests: utility, even if perfect for one individual, is hardly suitable for improving social welfare. The redefinition of welfare imposed in utilitarianism in order to circumvent the absence of a natural conjunction of interests raises fundamental problems regarding the integrity and the autonomy of persons. First, utilitarianism may force individuals to sacrifice something to benefit others, including total strangers. For instance, utilitarianism is likely to justify the execution of innocent people to prevent rioting and consequent numerous deaths, where raising the general welfare is presented as a serious justification for the innocents’ deaths.
Secondly, utilitarianism imposes negative responsibilities. Individuals are as responsible for what they do not do as for what they do: people have the responsibility to pursue better social consequences, and not just better consequences for themselves (Williams 1973). It is debatable, however, on what basis utilitarianism can impose upon people a goal that is not a priori supposed to be theirs. Thirdly, Rawls (1971: 26-7) claims that utilitarian neutrality implies the loss of the essential separateness of individuals, such that some human beings may be used as a means to the welfare of others.On [a utilitarian] conception of society separate individuals are thought of as so many different lines along which rights and duties are to be assigned and scarce means of satisfaction allocated in accordance with rules so as to give the greatest fullfilment of wants.... The correct decision is essentially a question of efficient administration........................................................................................ Utilitarianism does not take seriously
the distinction between persons. (Rawls 1971: 27)
This argument quickly become famous within the debates over utilitarianism, and has been greatly discussed and refined (see, for example, Parfit 1984). Rawlsian political liberalism, in which fairness is prior to goodness, constitutes the most famous alternative to utilitarianism. Fourthly, another family of ethical criticisms concerns the democratic failures of utilitarianism when its implementation is at stake.