Social Choice and Institutions for Building a Fair Society

In Section 3.1 we showed that value judgments - or, better, a theory of justice - are necessary to define a social ordering out of individual preferences and how institutions shaping the desired social rules can be properly implemented.

For example, consider that the various types of democ­racy (as well as other voting procedures or, more generally, canons of social relations) have different consequences for the social order in terms of distribution. We read the link between theory of justice and the related institutions through Sen's analysis of the different types of institutions implied by the different requirements of theories of justice. Then we focus on the types of institutions that are closer to practical applications of some theories of justice and issues, specifically as far as inequality and redistribution are concerned.

Sen (2009) sees two types of institutions, ‘transcendental- focused institutions’ and ‘realisation-focused institutions’. The former aim at devising a unique set of principles that can ideally define just institutions. The latter aim equally at the fundamental goal of any theory of justice, attainment of a better world.

The transcendental theories aim to identify an ideal of a perfectly just society. These theories want to devise a unique set of principles to select specific institutions that satisfy these principles and are then perfectly just. Their most recent and forceful expression is offered by Rawls (1971). Before Rawls, there were other attempts in Western philosophy to pursue the same end of stating propositions that define the archetype of a perfectly ideal society, starting with the Greek philosophers, mainly Plato and Aristotle, moving to modern social thinkers such as Hobbes, Locke, Rousseau and Kant and, finally, to contemporaries such as Dworkin, Gauthier and Nozick. However, as an outcome of an even unitary enterprise for defining a perfect social contract, the bricks and architecture for describing the features of an ideal society differ among these authors depending on the differ­ent foundations or postulates that support their analysis.

To exemplify, Plato in his ‘Form of the Good’ identifies this as the superlative to guide philosopher-kings in ruling the Republic. For Aristotle’s Metaphysics, the concepts trans­cending the categories are ‘being’, ‘unity’, ‘truth’ and ‘good­ness’, but he discusses unity (‘One’) alone in an explicit way because this is the only transcendental intrinsically related to being, whereas truth and goodness relate to rational creatures. In the Nicomachean Ethics, Aristotle offers possibly the first version of justice as fairness, with a requirement of correction of what is inequitable (‘corrective’ justice). Jumping to mod­ern times, Hobbes (1651) starts from the postulate of the nega­tive aspects of the situation that is likely to be an alternative to any kind of government, the ‘war of all against all’.

Among contemporaries, Rawls prefers to speak, instead of utilities, of ‘expectations’ of individuals as members of a society with shared values. His yardstick for inequality is the quantity of ‘primary’ social goods available to indivi­duals: rights and liberties, opportunities and powers, income and wealth (Rawls 1971: 90ff).

In his theory, a given distribution is just when it is fair, in the sense that it offers everyone the same opportunities. Rawls arrives at this position by way of a thought experiment in which individuals find themselves in an ‘original’ posi­tion in which they decide as free and equal persons the structure (rules) of a just society (in other words, the basis of their social contract). Each individual ignores how the various alternatives can influence himself or herself and decides on the basis of general considerations only, since he or she is covered by a veil of ignorance with regard to

1. His or her current and future position in society, parti­cularly his or her class and social status;

2. The distribution of endowments, natural abilities such as strength, intelligence and so on;

3. His or her personal preferences, especially aversion to risk; and

4. Other circumstances such as the political and economic situation, his or her age group and so on.

In such conditions, the members of this society would unanimously accept the following two principles of justice:

1. ‘Each person is to have an equal right to the most exten­sive total system of equal basic liberties compatible with a similar system of liberty for all.’

2.

The second of these could be of particular importance to our discussion, as Rawls’s position differs from the one that would result from other theories of justice, in being markedly - if not perfectly - egalitarian or egalitarian in a space different from that of utilities. The reasons for this conclusion lie in the underlying assumption that people share values with the other members of their society. This results in the need to ensure availability of primary goods for all people to become a rule, and this rule turns into an institutional trait of the society (Schiattarella 2013).

By contrast, Nozick (1974), starting from the Kantian idea that people should be treated as individual ends (because of ‘separateness of persons’), argues in favour of a ‘minimal state’ designed only to protect them against force, theft and fraud and provide the norms to enforce contracts in order not to violate people’s rights. Nozick (1974) develops the so-called entitlement theory, which has numerous forebears (among the most recent ones, see von Hayek 1960). His the­ory assesses distributive justice with regard not to results (e.g. the quantity of goods available to individuals or the utility each enjoys) but rather to procedures. Any distribu­tion is considered just if individuals’ fundamental rights have been respected. Regardless of levels and differences in individual utilities, these rights include the right to life and to enjoy the products of one’s labour and freedom of choice. They are inalienable rights, being independent of, and ante­cedent to, all forms of social organisation. They are also absolute, in that they are not subject to any constraint other than the duty to respect the fundamental rights of others.

This sort of theory of justice aims only to guarantee liberty and the exercise of rights, not to satisfy preferences. The yardstick of justice is not individual utility, i.e. the conse­quences of a certain social state for an individual, but rather the exercise and respect of rights. For Nozick, individual utilities are not the appropriate yardstick for assessing whether the justice criterion (safeguarding rights) has been satisfied. Thus, apart from the common reference to a space different from that of utility, Nozick challenges the princi­ples of justice derived by Rawls (1971), setting himself almost at the opposite end of the spectrum.

Realisation-focused institutions have been suggested by a number of authors, mainly economists and philosophers, starting with Smith, the Marquis de Condorcet, Bentham, Marx and John Stuart Mill to Sen and Nussbaum. Most of these authors have different ideas of how a just society should look and of the ways of making social comparisons. However, in referring to them, Sen says:

They were all involved in comparisons of societies that exist or could emerge, rather than confining their analyses to transcen­dental searches for a perfectly just society. Based on realization- focused comparisons, they were often primarily interested in the removal of manifest injustices they saw in the world, such as slavery, or policy-induced poverty, or cruel and counterpro­ductive penal codes, or rampant exploitation, or the subjugation of women. (Sen 2010: 58)

According to Sen, to design the contour of a perfectly just society is neither necessary nor sufficient to advance justice in the real world, where hunger, illiteracy, torture, arbitrary incarceration and medical exclusion are sources of iniquities and injustice. For getting rid of these, we just need to make comparative judgments of justice across different social sys­tems. This would avoid being engaged in a more difficult exercise also because - as discussed earlier - very different conceptions of the perfectly just society emerge from differ­ent postulates.

Some authors are inclined to think that Sen’s distinction has some flaws, since realisation-oriented comparisons need some idea of transcendental superiority (Valentini 2011; Ege, Igersheim and Le Chapelain 2012). However, Sen (2009) illustrates the reasons why transcendental theories are neither necessary nor sufficient for comparative judgments of justice.

The relative merits of each type of institution could be beyond our interest, as both positions would imply a relationship between social-choice theory and institutions. However, reference to realisation-focused institutions seems to be more pregnant to us, as this concept insists on the implementation of institutions and their effects. In addi­tion, it is simpler for people to accept its conclusions, since it is not based on an ultimate conception of what justice is. It thus can offer us a practical counterpart to social goals.^[130]

Think in particular of the procedural counterpart of the axiomatic approach to social welfare and choice followed by Arrow (1950,1951). If we take an example of this approach, we easily understand why a realisation-focused theory of justice can better assess institutions. Let us take some form of democratic voting. This would be both a social-choice procedure and a (fundamental) institution of a society, tightly connected with social justice. Sen (1999a) insists on this tight connection, as democracy would enhance capabil­ities for both its instrumental value in supporting people’s claims and rights and its constructive value for participation to public life (‘democracy as public reason'). These conclu­sions can be easily accepted by any other theory of justice, either realisation focused or transcendental focused, as all could share the ability to support people's claims and rights and the dynamic value of democracy as a way of enlarging the potential for participation in public life and thus achieve­ment of ‘democracy as public reason'.

The central task of a theory of social justice should be - at least from our point of view - to suggest policy changes to the social and institutional setting of some country. Suggestions should be useful and feasible. From the former point of view, they should start from the social situation existing in the country in question and the most basic or urgent needs to be satisfied there; reference to the best possible solution - as derived from an ideal optimum - may not be useful and can even be dangerous. From the latter, from a dynamic perspective, suggestions of changes should account for the potential improvements that can be derived from their implementation, given each country's initial historical and institutional situation. Transcendental-focused theories, aiming at the best possible institutions, disregard both sug­gestions of these requirements. For this reason, it is easier and more useful to start from realisation-focused theories. The following sections will articulate the issues arising in democratic institutions and their possible content.

7.3