Justice

5. Justice as fairness

Theories of justice as fairness start, like those of justice as mutual advantage, from the premise that the role of justice is to provide a framework within which people with competing ideas of the good can live together without conflict. However, justice as fairness seeks a framework that can be said to be fair to all the parties. It aspires to allocate to each person a fair set of opportunities to pursue their idea of the good life. The problem is that if what is due to each person is not determined by convention, by some overarching theory of the good, or by mutual advantage, how is it to be determined? On this question there is widespread disagreement. What is agreed is that there is an initial equal claim to consideration (see Kymlicka 1990). The root of disagreement lies in determining the scope, grounds and nature of this equality.

The range of accounts of the content of justice which is compatible with justice as fairness is very great: Robert Nozick’s entitlement theory and John Rawls’ theory of justice, for example, come to very different conclusions about what justice demands, even though both start with the basic idea that justice is to regulate the interactions of free and equal persons. Nozick believes that ‘individuals have rights, and there are things no person or group may do to them (without violating their rights)’ (1974: ix). From this starting point he generates what he calls an entitlement theory of justice (see Nozick, R. §2). The just pattern of distribution is that which would result from voluntary transfers, given that holdings were justly acquired in the first place (by just transfer or by an appropriation that makes no one else worse off). Nozick, then, regards the claim to an equal set of absolute rights as defining the limits of justice: any actions which interfere with those rights (such as redistribution) are unjust no matter what the pattern or outcome of the entitlement theory. Nozick does not, however, offer any account of the existence of such robust rights, and his arguments from intuition to show that any interference with individual rights is unjust are unconvincing. It is plausible that injustice may result from a large number of individual transactions each of which taken separately seems just.

John Rawls (1971) has argued that justice requires the provision of equal basic liberties and fair opportunities for all, and that social and economic inequality can only be justified where it is to the benefit of the least advantaged. These principles are derived by arguing that they would be chosen by free persons in an ‘original position’, the specifications of which prevent people from making unfair use of their natural and social advantages (see Contractarianism §7; Rawls, J. §1). Rawls then, in contrast to Nozick, believes that justice requires us to do much more for each agent than merely provide them with absolute property rights. Rather, the pattern of distribution is set at that which will maximally benefit the worst off. The question is one of determining the content of justice: when is it appropriate to move away from the concept of equality? ‘when have we done enough’ for a person to be able to say that variations in outcome are ‘deserved’? (Scanlon 1988: 187) (see Equality; Desert and merit)

Recent work in Anglo-American political theory has been dominated by the development of theories between these positions. Ronald Dworkin (1981), for example, has argued that Rawls goes too far in treating all characteristics that make people more or less productive as ‘morally arbitrary’, and thus fails to allow for the justice of rewarding enterprise and ambition. Dworkin, in his own work, seeks to accommodate differences in ambition while retaining the feature of endowment insensitivity (see Dworkin, R.).

A different line of argument is that Rawls’ original position is not a bargaining environment because the veil of ignorance entails that the participants are identical. Moreover, Rawls secures the two principles only by building in a number of ad hoc requirements (for example, that the participants are risk averse). The alternative proposed is to posit participants who are aware of their identities and morally motivated to seek agreement. Such a position has been proposed by the philosopher Thomas Scanlon (1982) and developed into a theory of justice by Brian Barry (1995) (see Contractarianism §9).

All theories of justice as fairness face the problem of grounding the commitment to the fundamental equality of persons and giving an account of each agent’s motivation to behave justly. Rawls offers two justifications for his principles. The first is that the principles match our ‘considered moral judgements’ about justice: we come to a ‘reflective equilibrium’ in which the principles reflect and organize our moral intuitions. The second justification is offered as a Kantian interpretation. On this account the original position provides a ‘procedural interpretation’ of Kant’s realm of the ‘kingdom of ends’ (see Kant, I. §9). The original position and the choice of the principles are viewed as an attempt to replicate Kant’s reduction of morality to autonomy and autonomy to rationality. Thus, by living in accordance with justice we realize our true natures as autonomous beings. This provides the motivation required.

In later papers and in his second book, Political Liberalism (1993), Rawls has moved away from the Kantian interpretation on the grounds that it requires a controversial metaphysics and that it commits him to a particular view of the good life as autonomy. Instead Rawls now emphasizes that his theory relies on nothing more than ideas ‘latent in the public political culture’ of modern Western democratic states. This move aligns him with the conventionalist position discussed in §2, and is open to the criticisms raised there. Without the Kantian interpretation it is also difficult for Rawls to give an account of moral motivation. What is left is a theory that claims to give a content to the idea of justice but which cannot, when the demands of justice conflict with the self-interest of the agent, provide the agent with a reason to be moral (Nagel 1991).

Scanlon’s and Barry’s accounts assume morally motivated individuals who are concerned to find terms of agreement which reflect the equality of persons. Barry has argued that the demands of fairness can be generated from the lack of an authoritative account of the good; but this argument still relies on a prior commitment to freedom and equality. There is, however, no reason why such theories should not claim that such a commitment has universalist implications: an appeal to what we believe does not mean that what we believe applies only to us. Similarly, theorists of justice as fairness need not make the illegitimate claim (much more often attributed to them than made by them) that justice as fairness is neutral between all ways of life (no matter how illiberal) (see Neutrality, political).

Justice as fairness can claim to give a content to the idea of justice by telling us what is justly due to whom, and can do so in a way that matches our fundamental intuitions about the nature not only of justice but of morality generally – namely, that each agent is a locus of equal value. That they cannot provide the agent with a decisive reason to behave morally when to do so conflicts with the agent’s self-interest or conception of the good may not reflect a defect of justice as fairness. Rather, it may be an inevitable aspect of moral theorizing without recourse to a divine order or a single, comprehensive, idea of the good (see Moral motivation).