46
A C D 8
Fig. 3.1. The optimal distribution of income under utilitarianism
efficiency and equity. This approach derives from two strands of thinking: utilitarian analysis and the writing of the philosopher John Rawls.
3.1. Utilitarianism
The utilitarian arguments tret form the basis of much of this book derive from the 'New Liberalism' of the early twentieth century (Chapter 2, Section 2.1), which was itself firmly rooted in the nineteenth-century classical tradition. Thus modern utilitarians have common intellectual roots with empirical libertarians.
the theory. The utilitarian aim is to distribute goods so as to maximize the total utility3 of the members of society. 'Goods' are interpreted broadly to include goods and services, rights, freedoms, and political power. Maximizing total welfare has two aspects: goods must be produced and allocated efficiently (discussed in Chapter 4); and they must be distributed in accordance with equity (though not necessarily equally). The equitable distribution is shown in Figure 3.1. Total income to be distributed is AB. Individual A's marginal utility (read from left to right) is shown by the line aa, and is assumed to diminish as his income rises. Individual B's marginal utility, which declines from right to left, is shown by the line bb. Total utility is maximized when income is shared equally; A's income is AC, and B's is SC.
Utilitarianism can therefore justify redistributive activity by the state in pursuit of an egalitarian outcome, but this result depends crucially on two conditions. First, A and E must have identical marginal utility of income functions.4 If B's marginal utility is shown by h'b', then the distribution that maximizes total welfare is unequal, since A now has an income of AD. Secondly, utilitarianism can fully specify the optimal distribution only where the utility of A and B can be measured cardinally (see the Glossary),
47
criticisms include questions such as: is utility capable of precise definition; does interpersonal comparison of utility have any meaning; and whose utility counts (e.g. future generations, animals, etc.)? These issues are set to one side to focus on two fundamental criticisms.
An unjust outcome. Utilitarianism can sanction injustice by justifying harm to the least well-off if this maximizes total utility. 'The trouble with [utilitarianism] is that maximising the sum of individual utilities is supremely unconcerned with the interpersonal distribution of that sum' (Sen 1973: 16). Formally, suppose that individual B in Figure 3.1 derives less pleasure from life than A because he has major health problems. His marginal utility is shown by the line b'b', and the optimal distribution of goods by point D. Thus B should receive less income than A because of his health problems. This outcome is criticized as being unjust.
The impossibility of a Paretian liberal. Consider two desirable objectives: individual freedom (which includes the idea that an individual is the best judge of her own welfare), and maximizing total welfare. Sen (1970, 1982) (see also Brittan 1995: ch. 3) argues that it is not always possible to achieve both objectives simultaneously—that is, individual freedom may not be compatible with simple utilitarianism. The argument goes as follows.
Suppose that my action imposes a cost on other people, not in economic terms (e.g. polluting their garden with smoke) but because they have views about my actions. They might think it wrong that I have long hair. More generally, they might think it wrong that a wealthy person has a yacht in Monte Carlo, or that people live together before marriage. Thus the action of one person can affect the welfare of another for aesthetic or moral reasons.
What does this imply for public policy? If policy-makers take such interdependen-cies into account, 'people will be penalized for carrying out private personal acts which affect others only because thinking makes it so' (Brittan 1995: 74). Accepting such preferences can make utilitarianism an illiberal doctrine, because they 'are a disguised form of coercion which arise from a desire to regulate the way other people spend their lives ., .' (ibid.).
To avoid this difficulty, policy-makers may choose to ignore the preferences of some people (e.g. those who wish to impose mandatory haircuts). In that case, however, policy is no longer decided only on a utilitarian basis; it will incorporate judgements about which forms of interdependence are allowable and which not. The heated debates about appropriate public policy (if any) about personal appearance, soft drugs, and sexual behaviour illustrate the point.
3.2. Rawls on social justice
Rawls in some ways is Nozick's liberal counterpart. Nozick is a natural-rights defender of liberty. For Rawls the natural right, and hence the primary aim of institutions, is social justice: thus 'each person possesses an inviolability founded on justice that even the welfare of society as a whole cannot override' (1972: 2). Justice, to Rawls, has a twofold purpose: it is desirable for its own sake on moral grounds; also, and importantly, institutions will survive only if they are perceived to be just. Rawls argues that there exists a definition of justice that both is general (i.e. not specific to any particular culture) and can
be derived by a process that everyone can agree is fair. The resulting principles deal with the distribution of goods, interpreted broadly to include also liberty and opportunity.
the original position is Rawls's starting point. He invites us to contemplate a group of rational individuals, each concerned only with his own self-interest, coming together to negotiate principles to determine the distribution of goods. They are free agents in the negotiation, but they must abide by the resulting principles, Rawls thus uses the convention of a nodal contract.
In this situation no discussion between interested parties will yield principles of justice that command universal acceptance. Rawls therefore abstracts the negotiators from their own society by placing them behind a veil of ignorance. They are assumed to be well informed about the general facts of the world—psychology, economics, sociology —but each is deprived of all knowledge alwut himself—that is, of his characteristics or endowments, his position in society, and the country or historical period into which he is born. The negotiators seek to advance their own interests, but are unable to distinguish them from anyone else's.
The role of the veil of ignorance is best illustrated by example. To distance ourselves from personal interests we (i.e. citizens through our elected representatives) may decide that aircraft hijackers' demands should never be met, even if innocent lives are lost. We do this in order to save even more lives in the long run; and we establish this doctrine in advance of the event (Le. behind the veil of ignorance) because if it were our personal loved ones who were kidnapped we would be likely to do anything to save them, irrespective of the possible consequences for others in the future.
The negotiators can consider any principle of justice—for example, the just action is that which is in the interests of the stronger, or that which ennobles the species or that which maximizes total utility. According to Rawls, the rational negotiator will reject these because under each he might systematically be underprivileged, the only rational choice is to select principles in terms of what Rawls calls the 'maximin rule', which maximizes the position of the least well-off individual or group. The negotiators do this because 'for all they know they may turn out to be the least privileged inhabitants of a country like [pre-reform] South Africa' ( McCreadie 1976: 117).
The original position, together with the veil of ignorance, plays two distinct roles. It is an analytical device, which 'reduc[es] a relatively complex problem, the social choice of the principles of justice, to a more manageable problem, the rational individual choice of principles' (Daniels 1975: p. xix). Secondly, Rawls sees the procedure as a moral justification of the resulting principles—they will be seen to be fair, he argues, because they are selected in a manner that is both rational and fair; hence his term 'justice as fairness'.
the principles of justice that follow are those that Rawls claims would be chosen rationally and unanimously by the negotiators. Because of the veil of ignorance, they will choose to maximize liberty for everyone. Hence:
The first principle (the 'liberty principle'). 'Each person is to have an equal right to the most extensive basic liberty compatible with a similar liberty for others' (Rawls 1972: 60).
The negotiators then turn to the distribution of goods other than liberty. Each will reject any principle of distribution that could leave him disadvantaged or exploited.
49
The negotiators may consider a principle that mandates a thoroughly equal distribution of goods . ., But they will soon come to realise that they stand to benefit by the introduction of certain inequalities. , . For example, giving a rural [doctor] an airplane would make him relatively advantaged, but even—and perhaps especially—the least advantaged among the rural populace stand to benefit as a result, and thus should sanction such inequality.  (Gorovitz 1975: 281)
Hence:
The second principle (the 'difference principle'). 'Social and economic inequalities are to be arranged so that they are both (a) to the greatest benefit of the least advantaged and (b) attached to offices and positions open to all under conditions of fair equality of opportunity' (Rawls 1972: 83).
Possible conflict between the two principles is ruled out by a priority principle, which gives the first principle absolute priority over the second. A reduction in the liberty of the least well off cannot be justified even if it is to their economic advantage. Subject to these priorities the two principles can be regarded as a special case of a simpler, more general conception of justice, in which 'all social primary goods—liberty and opportunity, income and wealth ... are to be distributed equally unless an unequal distribution of any or all of these goods is to the advantage of the least favoured' (ibid. 303). At its simplest, the distribution of goods between individuals A and B in Figure 3.1 should be that shown by point C unless any other distribution benefits the less advantaged of the two.5 If goods are not so distributed, any policy that improves the position of the less well off is an improvement according to Rawls/'
rawls and utilitarianism. Rawls is an explicit opponent of utilitarianism. He regards it as illogical (inasmuch as it would be rejected by rational negotiators in the original position) and unjust (in that it can sanction injustice in the interests of maximizing total welfare). The two theories can have very different implications. Suppose a given policy change makes at least one person better off without making anyone else worse off. This is an increase in Pareto efficiency,7 and hence desirable to utilitarians even if the individual thus benefited were rich. Rawls's difference principle, in contrast, would oppose the policy unless it were also (though not necessarily only) to the advantage of the least well off. Thus an efficient answer in Paretian terms will not always be a just answer in a Rawlsian sense (though, as argued in Chapter 4, Section 2.2, it may be possible to find a distribution that is both efficient and just).
CRITICISMS oť rawls's THEOttv. It has been argued that the negotiators would be unable to make any decisions behind the veil of ignorance. According to Nisbet (1974:112),
[the negotiators) don't know much of anything—anything, that is, that we are justified by contemporary psychology in deeming requisite to (nought and knowledge of any kind whatever. Nevertheless, Professor Rawls is shortly going to put his happy primitives through feats of
s Under the lexical extension of the difference principle, any policy should benefit the worst off; if he is indifferent, it should benefit the next worse off, and soon. Rawls thus admits a policy that benefits only the hest off, provided that everyone else Is indifferent to it,
r A utilitarian social-welfare function (see Chapter 4, Section 1) does not constrain the way individuals are weighted; a Rawlsian social-we If are function gives infinite weight to the least-advantaged individual or group,
7 See Chapter 4, Section 2.1, and the Appendix to Chapter 4, paras. 2-4.
5C
cerebration that even the gods might envy. Out of the minds of his homunculi, these epistemo-logical zombies who don't know their names, families, races, generation or societies of origin, are going to come principles of justice and society so vast in implication as to throw all present human societies into a philosopher's limbo.
Miller (1976) (discussed shortly) similarly argues that removing all cultural knowledge will immobilize the negotiators; but failure to do so, though permitting them to make a decision, will result in a culture-bound definition of justice.
The first principle is criticized" because Rawls's list of liberties may be too narrow, because the principle of toleration (e.g. of diversity of goals) inherent in Rawls's definition of liberty may reflect class bias, and because some issues are left unresolved—for example, what liberty should be accorded racists? Additionally, Barry (1973: 6) and Hart in Daniels (1975: p. xxx) dispute the priority given to liberty. Poor people might well be willing to trade some liberty for greater social or economic advantage. The second principle is criticized for its crucial dependence on maximin, which, it is argued (Arrow 1973; Letwin 1983: 22-9), is the optimal outcome only under very restrictive assumptions.
miller's analysis of social justice. A final criticism of Rawls is that he developed not a general theory of justice but a liberal theory. Miller (1976) argues that a completely general theory of justice is logically impossible, and that in this respect Rawls was bound to fail. According to Miller, social justice has three distinct elements:
• rights—e.g. political liberty, equality before the law;
• deserts—i.e. the recognition of each person's actions and qualities;
• needs—i.e. the prerequisites for fulfilling individual plans of life.
The 'deserts' aspect implies that someone who works longer hours should receive more pay, and the 'needs' aspect that an individual incapable of work should not be allowed to starve. Though admitting the difficulty of precise theoretical definition, Miller argues that each element is a logically distinct principle embodying a particular type of moral claim.
It is easy to see that rights and deserts can be reconciled (e.g. a person should have the right to keep all her income if she has earned it legally); similarly, rights and needs can be compatible (e.g. a person should be entitled to health care if she is ill). But conflict can arise between desert and need: if 1 am rich and healthy and you are poor and ill, then either I am taxed (and do not receive my deserts) to pay for your medical treatment, or you receive no treatment (hence your need is not met) so as to protect my deserts.
The core of Miller's argument is that the definition of social justice depends crucially on the type of society being discussed. In a pure market economy, justice will be defined in terms of rights and deserts. A collectivist defines justice as distribution according to need.
Miller thus argues that the different principles of justice are connected to wider views of society. He criticizes utilitarians and Rawls because they take no explicit account of the conflicting claims of rights, deserts, and needs, but blur them into a single, indistinct
* See Daniels {1975: pp. xxviii-xxix) and Ihe chapters therein by Hart, Scanion, Daniels, and Hsk.
51
whole. Miller also criticizes the view implicit in Rawls that there is a single conception of justice upon which everyone's definition will converge, arguing instead that justice comprises conflicting principles, the relative weights attached to which may vary sharply between different societies. 'The whole enterprise of constructing a theory of justice on the basis of choice hypothetically made by individuals abstracted from society is mistaken, because these abstract ciphers lack the prerequisites for developing conceptions of justice' (Miller 1976: 341). Or, if they do manage to make choices, it must be in terms of culturally acquired attitudes. In short, the negotiators in the original position will be immobilized unless they have some knowledge of the nature of the society for which they are choosing rules of justice. Finally, 'Rawls's individuals are given the attitudes and beliefs of men in modern market societies, and it is therefore not surprising that the conception of justice they .. . adopt should approximate to the conception .. . dominant in those societies' (ibid. 342). Hence, he argues, Rawls fails to develop a general theory of social justice; such generality is not possible.
4, Coliectivist views
4.1. Democratic socialism
Coliectivist writers agree on the importance of equality. They regard resources as available for collective use, and consequently favour government action; but historically they have disagreed about whether socialist goals could be achieved within a market order. Some writers advocate a mixed economy that blends private enterprise and state intervention; Marxists (discussed in Section 4.2) argue that this is not possible, that capitalism is inherently unjust; and that socialism is possible only where the state controls the allocation and distribution of most resources.
sociaijst aims vary widely, but three—equality, freedom, and fraternity—are central. Equality is a variant of vertical equity discussed in Chapter 1, Section 2.2, and fraternity of ,he social solidarity aim. it is recognized that these aims can clash; and different writers accord them different weight; but together they make up the socialist definition of justice. In Miller's terms, the dominant themes are rights and needs, with deserts assigned a smaller role.
There is a measure of agreement (Tawney 1953, 1964; Crosland 1956) that the crucial element of justice is equality, which to socialists is an active concept. Kquatity of opportunity on its own may be insufficient (Tawney 1964; Laski 1967: ch. 4; Hattersley 1987), since substantial inequality of outcome may persist. Positive equalizing measures are needed, though not necessarily complete equality of outcome (see Daniel 1997).
Such emphasis on equality bears closely on Miller's concept of need. Weale points out that 'in some political arguments . . . the assumption is made that to distribute according to need is to satisfy the claims of equality' (1978: 67), but suggests (ch. 5) that the relationship is rather more complicated. For present purposes we need note only that equality and meeting need are closely related concepts, though not logically equivalent.