Gauthier and justice without virtue

This is where the central figure of his version of contractualism comes in: the 'constrained maximiser'. Not an altruist, not a righteous by vocation, but a rational and self-interested individual capable, however, of accepting limits to his own action on condition that others do the same, that these limits are reciprocal. Morality, in this way, is not represented as a code of superimposed rules, but as a rational strategy that emerges in the long run.

Justice as an agreement

On this point Gauthier clearly distances himself from Rawls. No veil of ignorance, no impersonal choice. Individuals know who they are, they know their endowments, their advantages, their vulnerabilities. Justice does not arise from an abstract moral principle, but from a rational ex ante agreement on the terms of cooperation. As he explicitly writes, 'The contractualist finds in rational agreement the proof of moral and political validity' (p. 15).

Justice thus becomes not so much a property of intentions as a criterion for the validity of social practices. Institutions, rules, markets are just if and only if they could be accepted by rational individuals who recognise each other as interdependent. The guiding criterion is not equality of outcome, but 'non-exploitative reciprocity' as the philosopher calls it.

In order to preserve the assumption of self-interested behaviour and at the same time prevent it from degenerating into predatory actions, such as mutual defection in a Prisoner's Dilemma, Gauthier introduces the constraint of the 'Lockean proviso' that we have repeatedly discussed (Mind the Economy of 07/07/2024 and 28/07/2024). "The clause prohibits the improvement of one's own situation through interactions that worsen someone else's situation" (p. 218). Justice does not demand that everyone gains equally, but implies that no one can improve his or her own position at the expense of that of others. It is a minimal and at the same time demanding criterion of justice. It does not impose active solidarity, but prohibits all forms of exploitation.

Because selfishness is incomplete

For Gauthier, therefore, the decisive philosophical point is that egoism is an incomplete principle of action. It works well in individual choices but fails in strategic interactions. A point that Amartya Sen had made clear a decade earlier when he stated that 'the purely economic man, in effect, is close to being a social idiot' ('Rational Fools: A Critique of the Behavioral Foundations of Economic Theory', Philosophy & Public Affairs 6(4), pp. 317-344, 1977). In a world of interdependent agents, the pure egoist becomes predictable, excludable, inefficient. Not because others are morally better, but because they have good reasons not to trust him. Cooperation requires a stable disposition, the ability to commit, to keep agreements, to sometimes give up the best immediate response. All these things are alien to homo economicus.

Being fair, for Gauthier, does not mean always cooperating. It means cooperating under fair conditions. Because a wider openness would expose one to exploitation just as a narrower one would turn the agent into a recognisable free-rider, who would be legitimately excluded from the benefits of social cooperation.

The political implications

The contractualist view of justice in the version Gauthier offers us may seem minimalist and perhaps a little cynical, but it is certainly a version that sheds light on a wide range of concrete situations. Markets, for example, do not function efficiently because individuals are virtuous, but because interest is channelled within shared rules. Without these there would be no trust, no enforcement of contracts, and without expectations of reciprocity even the most efficient markets would collapse under the ineradicable risk of opportunism. The rules that one imposes upon oneself and decides to abide by are therefore, in this sense, not external limits to efficiency but rather the conditions of its possibility.

Let us then consider collective bargaining, an area that represents an almost paradigmatic case of justice à la Gauthier. Companies and workers who, despite conflicting interests, choose to cooperate with each other not out of benevolence or altruism, but because permanent conflict would be destructive for all. It is only when bargaining power is too asymmetrical that the 'Lockean clause' is violated, and the eventual agreement will reflect domination instead of justice. This is why 'justice as agreement' requires fair terms of negotiation, which often require the intervention of a public arbitrator.

A third example is the tax system. Choosing whether or not to pay taxes is for Gauthier a problem of coordination, not virtue. If many pay, it will be convenient to evade. But if many evade, the system will collapse. The state with its norms and control agencies is not, in this sense, a moral preceptor, but rather a guarantor of reciprocity. Rules and sanctions serve to make cooperation rational. This is why rules that instead make violation rational, such as amnesties, scrapping and all the amnesties with which we are familiar, favouring another kind of equilibrium, one in which widespread evasion becomes the rational choice, are harmful and open to criticism. They benefit those who propose them but are against the general interest.

By the same logic we can interpret the problem of environmental protection. A great Prisoner's Dilemma in which continuing to pollute is the individually best strategy that will, however, collectively lead us to disaster. Incomplete selfishness requires external intervention, binding agreements that allow us to coordinate our choices and policies towards optimal outcomes that individually we would not be able to coordinate on.

The excluded

Markets, taxes, environment. Gauthier's morals by agreement explains the rationale for the operation of these cases, and many others, with parsimony and theoretical elegance. There is, however, a point at which the theory of justice as mutual benefit encounters a limitation that cannot be circumvented. If, as argued by all contractualists, from Hobbes to Locke, from Rousseau to Rawls, justice arises from rules that rational individuals would accept as terms of a mutually beneficial cooperation, what happens to those who cannot participate in that cooperative and mutually beneficial exchange? To those who are not in a position to bring a concrete benefit and thus improve the outcome of the agreement? People with severe disabilities, young children, non-productive individuals, those without bargaining power. All those who in the logic of mutual benefit have nothing to give but everything to take. 'These people,' writes Gauthier with great honesty, 'are not part of the moral relations based on contractualist theory. The problem here is not care for the elderly, who have paid for their benefits with their previous productive activity (...) The main problem is care for the disabled' (p. 19).

Gauthier cannot get around this difficulty. And unlike many other contractualists, he chooses to expose it with rare frankness. In Morals by Agreement he explicitly acknowledges that this point represents 'a problem that, understandably, no one wants to talk about' (ibid.). Understandably, because this is an issue that touches the raw nerve of any theory of justice that is based on the interaction between rational and independent individuals; 'autonomous throughout their lives', as Rawls writes in A Theory of Justice.

Gauthier's answer does not consist in a strong solution, but in a strict delimitation of the field. Justice as mutual benefit,' he clarifies, 'is not an all-encompassing theory of morality, but a theory of justice understood as the regulation of cooperation between equals. It concerns those who can take part in a system of mutual benefit. It does not claim to cover the entire spectrum of moral duties, nor to found a comprehensive theory of social justice. It is a conscious theoretical choice, not an oversight.

The 'Lockean clause' plays an essential but limited function here. By prohibiting agreements that improve the position of some while worsening that of others, it ensures that the excluded cannot be exploited. But it is a negative protection, not a redistributive one. It is too little. The clause prevents harm but does not ensure positive benefits. It protects against oppression and exploitation, but does not guarantee inclusion and equality. Those who cannot cooperate remain protected, but are still excluded, relegated to the margins. As Martha Nussbaum would state years later, in this perspective the relationship with vulnerability becomes "a matter of charity and not justice" (Beyond the Social Contract, Clare Hall, University of Cambridge., p. 438, 2003).

And this is precisely the most controversial and at the same time most honest point of Gauthier's theoretical framework. Individuals who cannot offer any advantage in a cooperative exchange remain to a large extent 'excluded'. Not because they do not count morally, but because they do not count strategically.

This is why Gauthier's theory is extraordinarily powerful when applied to markets, contracts, institutions, rules of the game between actors endowed with similar capacities. And for the same reason, it appears fragile and at least incomplete when confronted with the issues of disability, care, dependency and radical vulnerability. This is not an internal contradiction, but a structural boundary.

There was, of course, no shortage of criticism on this point. Brian Barry argued that a theory that leaves out the weakest is not a theory of justice, but of cooperation between the strong. John Rawls saw this dependence on bargaining power as a decisive reason for introducing the veil of ignorance, precisely to include those who might be at a disadvantage. Thomas Scanlon showed how what cannot be reasonably rejected includes considerations that cannot be reduced to strategic advantage, such as vulnerability and need. Martha Nussbaum, finally, argues that any theory of justice that aspires to universalisability must instead start from people's vulnerability and basic capacities and not from their bargaining power.

But reading Gauthier only in the light of this criticism risks forgetting the essential point of his construction. For his aim is not so much to say all that justice should be, but to show how far instrumental rationality alone can go. To show that it is capable of generating real, stable and cooperative constraints.

The problem of outliers thus does not reveal an error in the theory, but its inner limit, a real stress-test. It marks the exact point at which justice as an arrangement ceases to suffice, and it is precisely there that, inevitably, politics must begin to make its voice heard.

Where agreement ends, justice begins

Gauthier's great ambition was to show that justice can arise among us even without appealing to external ideals, without having to recruit altruistic individuals or those willing to sacrifice. To show that being just in most cases pays off. And it is an ambition that has largely found satisfaction. Cooperation can emerge from individual interest and the social order does not necessarily need heroes to function. But precisely this linearity, paradoxically, makes the theory fragile. Those who cannot bargain, in fact, because they are perhaps not, even if only apparently, able to give something, do not weigh in the agreement. And therefore does not count. On this point Gauthier does not give us a definitive answer but a threshold. He shows us, that is, how far the rationality of interest can ground justice. But also the point at which agreement can no longer include everyone. It is here that we can see that a social contract can be rational without being, for that matter, fully just.

*Professor of Economics C-BASS, Center for Behavioral and Statistical Sciences, Director Department of Economics and Business,
University of Cagliari