Social and Political Philosophy

Spring-Summer 2004

Lecture 10: Rawls and social justice

1. Social justice

Ordinarily when we talk about justice we have in mind legal justice, i.e. punishing the criminal but not the innocent, punishing people who have committed the same crime in the same way, punishing more serious crimes proportionately more heavily etc. A classical definition of justice that summarises this is ‘giving people their due’. Social justice (or distributive justice) is really an extension of the idea of a fair allocation of punishments in the legal system to a fair allocation of goods (especially of money) in society as a whole. Rawls summarises this idea when he says that concept of ‘justice’ (and social justice is what he has in mind) is that of ‘a proper distribution of the benefits and burdens of social cooperation’.

2. Hypothetical contract, the original position and the veil of ignorance (TJ secs. 1-4)

Through most of the 20th century normative political philosophy was dominated by utilitarianism. Rawls resurrects contract theory as a basis for evaluating states: more accurately, for evaluating the ‘basic structure of society’, i.e. its economic as well as legal and political structure, from the point of view of its ‘justice’. (He does mention that one might evaluate a social structure from other points of view, but he clearly thinks that the point of view of justice is the crucial one.)

He asks us to imagine ourselves as contracting to form a social structure from an ‘original position’ (original position). Here we are behind a ‘veil of ignorance’, i.e. none of us knows what role he or she we will end up by occupying in the social structure. In fact no-one knows anything about him/herself which will enable him to predict what role he/she will end up by occupying: a person’s skin colour, sex, abilities, industriousness, physical healthiness and so on are all behind the veil of ignorance. Also, they do not know their own ‘conception of the good’, i.e. they do not know what they think is ultimately valuable and worth pursuing in life.

A social structure is more ‘just’ than another one, according to Rawls, if in the original position people would form a contract to establish and inhabit that social structure rather than the other. This is the essence of his theory of justice, which he dubs ‘justice as fairness’ (with the implication that other theories of justice do not make it a matter of fairness). In fact the idea of forming a contract is a bit misleading, since according to Rawls anyone rational placed in the original position, as he defines it, will have the same preferences as between possible social structures.

3. The primary goods and the principles of justice (TJ secs. 11, 13-15)

Behind the veil of ignorance people do not know their own ‘conception of the good’, so Rawls argues that they must evaluate basic social structures in terms of what those social structures would provide them by way of goods that will be useful to them as means towards realising their particular conception of the good no matter what it may turn out to be. He argues that his ‘primary goods’ (rights, liberties, opportunities, powers, income, wealth, sense of one’s own worth) are such all-purpose means. So rather than simply assert that the good for humans consists in their self-realisation or in excelling at human activities like art (as perfectionists do), and rather than saying that the good for human beings consists in happiness or preference satisfaction (as utilitarians do), he develops a third theory of the good as consisting in those resources which will enable you to realise any individual’s conception of the good, whatever it may be. He calls this the ‘thin theory of the good’. It is close to a utilitarian theory of the good though it focuses on resources rather than outcomes.

The idea is now that people in the original position will choose between different social structures on the basis of what quantity of primary goods they can expect to get, in each of the different social structures in question, when they step out from behind the veil of ignorance. The theory enables him to say that for him ‘the right’ (i.e. principles of social justice and of individual morality – though in later writings he drops his original ambition to extend his theory to individual morality) is prior to ‘the good’. Justice is defined independently of any particular conception of the good.

The original position could be used directly to generate judgments about whether one social structure is more just than another (e.g. Britain today with or without a particular law). But Rawls uses it to generate general principles which in turn can be used to generate such individual judgements. These are the ‘principles of justice’. They stipulate how primary goods must be distributed in an ideally just social structure:

(1) Principle of equal liberty: Give each person an equal right to the most extensive total system of the ‘basic liberties’ (freedom of speech, assembly, right to vote, run for elections etc.) compatible with an equal system for all. This is an updated version of Kant’s ‘principle of right’.

(2) Distribute economic and social advantages (income, wealth, status) so that they are: (2a) attached to positions which are open to all under fair equality of opportunity (‘fair equality of opportunity principle’) and (2b) to the greatest advantage of the least advantaged (the ‘difference principle’)

The principles are meant to be applied ‘lexicographically’, i.e. in ranking a number of different social structures according to how just they are we should first rank them by how well they satisfy principle (1), then break any ties by seeing how well they satisfy principle 2(a) then break any remaining ties by seeing how well they satisfy principle 2(b). This is like ordering words in a dictionary.

Some comments:

(a) The difference principle ranks one social structure as more just than another if the people occupying the worst-off role in that social structure are better off than the people occupying the worst-off role in the other social structure. Prima facie this suggests that more egalitarian societies are more just. But if it is the case that e.g. instituting free-market capitalism will make everyone better off than instituting a pure communism (because of the incentives to get rich etc.) then this system will count as more just. So the difference problem avoids the ‘levelling down’ problem faced by straightforward egalitarianism: it has to advocate the impoverishment of everyone if that is what is needed to bring about economic equality.

(b) The principles of justice have a moral character to them in that they demand either that everyone be equally well off (in terms of the basic liberties) or that the worst off be as well off as possible (in terms of other benefits), but this is not a result of any moral instincts on the part of the contractors, who are assumed to act purely on self-interest. The veil of ignorance does the job of translating self-interested decisions into principles that have a moral character, because from behind it I have to think of how everyone will do in a given social structure, since I might turn out to be anyone.

4. Would people in the original position choose the principles of justice? (TJ secs. 25-26)

Rawls assumes that the parties in the original position will adopt a ‘maximin’ or ‘disaster avoidance’ form of rationality, i.e. if they have to choose between several options in each of which there are various different chances of different outcomes for them, they will not choose the option under which the average utility of those outcomes (weighted by the chance of each) is highest, but the option for which the worst outcome is better than for any other option. However it has been argued that a rational self-interested person would not adopt a maximin policy (Harsanyi). For example, every time we take a plane trip we flout the maximin principle.

5. Justifying the principles of justice (TJ secs. 4, 20, 24)

Rawls might respond to Harsanyi that the original position is an artificial construct and he is entitled to define the kind of rationality the parties in it use, as well as features like what is behind the veil of ignorance, so as to best serve the theory. This raises the question of what justifies the principles that would be chosen in the original position as he constructs it. He gives several answers in sec. 4 of A Theory of Justice:

(a) The principles that would be chosen produce judgements about which social structures are more/less just which correspond to our strongest intuitive moral judgements about the relative justice of pairs of different societies in individual cases. Here he allows for a process of ‘reflective equilibrium’ between original position and our intuitive judgements whereby we can tinker with the definition of the original position to make it generate principles whose rankings of social structures correspond to our strongest intuitive judgements, while being willing to sacrifice weaker intuitive judgments for the sake of a match between intuitive judgements and what is generated by an original position. This is a ‘coherentist’ justification of the principles.

But Rawls also says they would be chosen in an original position that is ‘philosophically favoured’. Here he seems to have in mind either:

(b) The original position expresses some abstract intuitions that we have about justice (e.g. the intuition that an argument for a principle of justice should not be based on people’s colour is modelled by placing colour behind the veil of ignorance in the original position).

(c) Some even more abstract intuitions we have about the nature of persons as free and equal which he associates with Kant (TJ sec. 40) but is also associated with the French Revolution and can be traced back to Rousseau and Locke. The original position is supposed to implement the idea of free and equal persons: free perhaps because people do not know their own conception of the good (expressing the idea that this is something it is up to each individual to choose; Mulhall and Swift) (NB here the idea of freedom looks like that of self-authority which I have mentioned before), and equal because people do not know what their own skin colour, sex, abilities etc. are so they are ‘blind’ to those things (like the figure on the Old Bailey). Freedom and equality in turn come out in the first and second principles of justice respectively. So the original position would be a way of ‘modelling’ the basic idea of persons as free and equal.

(d) He also implies that principles chosen in the original position are justified because they are ‘justifiable to individual members of society’. But this begs the question of what form the justification is meant to take. If it is an argument to the effect that ‘If you were behind the veil of ignorance, you would choose these principles too’, the obvious response is ‘Maybe, but I am not behind the veil of ignorance, so what is it to me as I am now that if I was then I would choose them?’ Maybe the idea is that they can be justified from an idea of ourselves as free and equal which we all share, but this reduces to (c). Or maybe the idea is that they can be justified from some principles which are intrinsic to the bare idea of justification (O’Neill ‘Constructivism in Rawls and Kant’ attributes this line of thought to Kant, though not to Rawls.)

6. Some criticisms

(a) One criticism is that Rawls mixes up four different kinds of justification for his original position, where he should have stuck to one. In later writings he tends to emphasise (c) above.

(b) Another criticism is that the concept of persons as free and equal appealed to in (c) is an untenable metaphysical postulate (Sandel). In later writings, Rawls responds to this by saying that the original position simply implements a conception of the person that we in modern Western societies share (on which we all have an ‘overlapping consensus’ despite our different religious and moral beliefs), rather than one which is absolutely true. The implication is that his theory is only applicable to such societies. This looks like a major concession to ‘communitarianism’, the view that the only basis on which political structures for a given society can be justified is the shared cultural values of that society. (See O’Neill, ‘Constructivism in Rawls and Kant’ for a critique of this position from a cosmopolitan-Kantian perspective). Also, does it make the whole theory practically useless, since it presupposes that we all agree to start with? Rawls would say no: there are deep agreements about the freedom and equality of persons which are shared across the political spectrum in the West, and he can show that his principles of justice follow from those deep agreements. But to make this work he has to show not just that his theory gives one possible interpretation of freedom and equality of persons, but that it gives the only possible one.

(c) The list of primary goods is skewed towards individualism. A deeply unequal society (such as may be justified by Rawls) cuts people off from each other, and is worse for human beings than a more equal one. Maybe this could be rectified by adding another primary good, ‘sociability’, but this looks at odds with the thin theory of the good as the means of generating primary goods. (See Kymlicka)

(d) The very way the principles are chosen on the basis of self-interest reflects an underlying individualism. Rawls might say this is correct and implements the idea of persons as free. (See Nagel)

(e) Rawls says nothing about desert, but most people intuitively feel that achievements deserve some kind of reward. (See Rogers 2002)

Supplementary readings:

Rawls, J. (1985) ‘Justice as fairness: political not metaphysical’, Philosophy and Public Affairs 14, reprinted in Rawls’s Collected Papers and revised as lecture 1 secs. 1-5 of his Political Liberalism

Rawls, J. (2001) Justice as Fairness: A Restatement

Rogers, B. (2002) ‘Just deserts’, Guardian 19 April 2002

(http://www.guardian.co.uk/analysis/story/0,3604,686917,00.html)

Seminar questions:

(NB section 4 of A Theory of Justice is crucial. It should be read several times.)

1. What is the difference between the concept of justice and a conception of justice? (TJ sec. 1)

2. Rawls claims that principles chosen in the original position can be publicly accepted. Why should this be? (sec. 3)

3. How can Rawls define the original position both in ‘the most philosophically favoured way’ and so that it is in reflective equilibrium with our intuitive judgements about individual cases? (secs. 4, 20)

4. What is the difference between pure and perfect procedural justice, and why does it matter for Rawls? (sec. 14)

5. How does Rawls justify his ‘formal constraints on the concept of right’? (sec. 23)

6. How does the original position ‘model’ the idea that persons are free and equal? (sec. 40)

AC 9.3.04