Human social orders are characterized by disagreement. In order to live better together we need to find some set of rules that can help us achieve social cooperation rather than conflict. This very real and very serious problem has been the central concern of social contract theorists, public reason liberals and contractarians alike. Michael Moehler's Minimal Morality: A Multilevel Social Contract Theory (hereafter MM) is the most recent contribution to this rich and broad research tradition.

Moehler begins by noting a shortcoming with public reason approaches to the problem of social cooperation: in searching for rules that all agree to, public reason liberals tend to idealize heavily, focusing on so-called ‘reasonable’ persons who possess certain shared liberal sensibilities. Now Moehler notes that, in some sense, this is all well and fine – if we can find a set of rules to live by grounded in shared moral commitments then all the better. Yet, clearly, we will not always have these shared commitments to fall back on that public reason theorists assume. And when we do not, how do we find a set of rules that helps us achieve social cooperation rather than conflict? Moehler's central contribution is to offer a new, compelling answer to this difficult problem: ‘I develop a moral theory that is justified not only for reasonably morally pluralistic societies, but for ‘deeply morally pluralistic societies’ that may be populated by liberal moral agents, nonliberal moral agents, and nonmoral agents alike’ (6).

In searching for agreement among agents who lack shared liberal sensibilities, Moehler argues that it is best to approach this issue from a contractarian, rather than public reason liberal, framework. Though the contractarian project arguably began with Thomas Hobbes's Leviathan, it certainly reached (until Moehler's new book) its zenith with David Gauthier's Morals by Agreement. Roughly, the contractarian project seeks to derive moral rules for all to follow from strictly non-moral premises. In particular, the goal is to show that it is rational for purely self-interested agents to follow moral rules. As Moehler describes it, ‘the position of moral contractarianism does not rely on substantial moral premises as traditionally conceived as a basis for the justification of its conclusions, and thus methodologically represents an instance of the purely instrumental approach to morality’ (12).

The contractarian project is an enticing one but, as Moehler convincingly argues, has not yet succeeded in its current articulations. In looking to Gauthier's version of contractarianism specifically, the main issue has to do with the derivation of Gauthier's principle of minimax relative concession. In setting up a bargaining procedure to derive his principle of distributive justice, Gauthier smuggles in moral assumptions; he ‘moralizes not only the state of nature of his social contract theory according to the traditional understanding of morality, but also his bargaining theory that justifies his principle of distributive justice’ (52). Given the nature of the assumptions Gauthier employs, it can hardly be said that Gauthier derives moral conclusions from non-moral premises. As Moehler argues, many of the premises are included for strictly moral reasons.

Moehler's version of the contractarian project remedies these failings. He begins with what he calls the homo prudens model of human behaviour. Far from attributing liberal sensibilities to individuals as the public reason liberal's notion of reasonable persons does, the homo prudens model only assumes that a person ‘values her life, together with her expected gains from peaceful long-term cooperation, more than noncooperation’ (97). In deriving a principle of conflict resolution Moehler sets up a bargaining procedure – what he calls the ‘empathetic contractor theory’ (§4.2) – that is free of Gauthier's failings, in that no moral assumptions are included. More specifically, the bargaining situation Moehler develops ‘does not assume substantial moral ideals as traditionally conceived as a basis for its decision-making procedure . . . the agents are assumed to know their identities and consider their empirical reality, including the general empirical conditions of their society’ (114). In this sense Moehler's project is both clearly a success and clearly a leap forward for the contractarian research programme overall.

What is the output of Moehler's bargaining procedure? In what is perhaps his most stunning conclusion Moehler argues that persons in a bargaining situation as he models it would agree to something like Kant's categorical imperative, albeit in a watered-down version. Moehler calls this principle the weak principle of universalization (hereafter WPU), and it runs as follows:

Or, to put the WPU more pithily: ‘in cases of conflict, each according to her basic needs and above this level according to her relative bargaining power’ (125). From here Moehler goes on to show why rational persons would be inclined to comply with the WPU – something Gauthier also fails to do with his own principle of minimax relative concession. He then ends by sketching what sorts of social and political institutions the WPU entails. Moehler argues that the WPU entails some form of the productivist welfare state, where the main idea is for the state 'to offer welfare provisions to its citizens not primarily out of concern for justice, but as a means to advance economic development’ (182).

In many ways Minimal Morality is, I think, a stunning success. Indeed, I taught it in a graduate seminar at the University of Maryland this past spring semester alongside A Theory of Justice, What We Owe to Each Other, and Morals by Agreement, where I think it plausibly belongs. Nonetheless, I did have two overarching concerns which, I think, are related. I will now turn my attention to these two lingering worries.

One broader, more abstract worry concerns what it would take for the contractarian project to succeed. The contractarian, we have already seen, wants to derive moral conclusions from non-moral premises. The non-moral premises part is the easiest bit to understand. The contractarian wants to show that it is rational – using the thinnest sense of rationality possible – for persons to comply with moral rules. But what kind of rules does the contractarian want to show it to be rational for persons to follow? It is not just any rules that she seeks to show the rationality of. Rather, it is a certain kind of rules – moral rules – that the contractarian wants to show are rational to comply with.

The focus on showing the rationality of moral rules is an important part of Gauthier's contractarian framework that, I think, Moehler unduly neglects. In the deeply important Chapter 8 of Morals by Agreement, Gauthier introduces what he calls the ‘Archimedean point of view’, which is a decision procedure very similar to (but in many important ways different from) John Rawls's original position.Footnote ¹ Writes Gauthier:

When a rule is chosen by an actor occupying this Archimedean perspective it counts as a bona fide moral requirement. For Gauthier's project to succeed, then, it must be that the moral rules derived from his (supposedly) non-moral bargaining procedure would also be chosen by a person inhabiting the Archimedean point of view. In this way the contractarian project also includes certain elements of what Moehler calls the contractualist research programme (Chapter 3), whose primary contributors are Rawls, Scanlon and Harsanyi. In other words, it is not sufficient that we show that certain rules are in the rational self-interest of all; we must also show that these rules would be the object of choice in a suitably idealized choice procedure that reflects our moral ideals of impartiality. We do not want just agreement, but morals by agreement.

This latter feature – that the output of the bargaining procedure must also be the object of choice in some kind of impartial decision procedure such as Gauthier's Archimedean perspective or Rawls's original position – is noticeably absent from Moehler's contractarian framework. By my reading, passing such a test is not a success condition for his contractarian project. That said, Moehler does argue that the WPU specifically possesses attractive moral features. In particular, the WPU ‘expresses the core moral ideals, such as the moral ideals of autonomy, equality, impartiality, and reciprocity, that are expressed by Kant's categorical imperative, although these moral ideals are not presupposed as moral assumptions in the traditional understanding of morality as a basis for the justification of the principle’ (133).

But here is where my second worry comes in: I doubt that the WPU would always be chosen whenever homo prudens enters a bargain. Recall that, in order to avoid the failures of Gauthier's contractarian project, Moehler explicitly includes no idealizing assumptions in his bargaining procedure so as to be sure that no moral assumptions are used in the derivation of moral conclusions: persons know who they are, and know their empirical reality. We want moral conclusions from non-moral premises. Yet if this is the case, why think that the Athenians, when pressing their will against the much weaker Melians, would agree to give them according to their basic needs as the WPU suggests? Indeed, they did not. The Athenians slaughtered the adult men, and sold the women and children into slavery.

Moehler has a response here and it has to do with what he calls the veil of uncertainty. Now as we all know, Rawls's veil of ignorance is introduced as an explicit moral assumption of Rawls's project, for ‘to each according to his threat advantage is not a principle of justice’ (Rawls Reference Rawls1971: 141). Moehler's veil of uncertainty is not introduced for these reasons, but does have a similar (although not identical) effect as Rawls's veil of ignorance. Writes Moehler concerning the origins of the veil of uncertainty:

In other words, the very exercise of choice among the strictly rational homo prudens forces them into something like Rawls's veil of ignorance. They are, after all, picking a principle to govern relations in perpetuity, and perhaps the Athenians will end up as the Melians one day. This key feature of the empathetic contractor situation plays a massive role in the derivation of the WPU. In Moehler's words:

So, it is only because they are behind a veil of uncertainty – not a moral assumption, but a natural epistemic constraint that agents in these kinds of bargains actually face – that they choose the WPU which, indeed, strikes one as a bona fide moral principle.

Since the veil of uncertainty is introduced not as a moral assumption Moehler's claims are subject to empirical testing: does the act of engaging in a long-term bargain necessarily put persons in an epistemic state where they must then choose impartially? Or can they still rationally bargain in a biased manner? Indeed, some have pursued this very line of inquiry. Begin by noting that Moehler borrows the veil of uncertainty (114, n. 36) from James M. Buchanan's and Gordon Tullock's masterpiece, The Calculus of Consent: The Logical Foundations of Constitutional Democracy, where the two economists argue that the very act of drafting a constitution puts one in an impartial epistemic state similar to Rawls's veil of ignorance. But as one study notes after examining the dynamics of constitution-drafting in Greece,

Even the American Revolution and subsequent drafting of the Constitution by the Founders – what has been called a ‘best case scenario so far as aiming to probe the constitution-making process under conditions of deep and extensive uncertainty’ (Imbeau and Jacob Reference Imbeau and Jacob2015: 99) – has been charged as biased and partial. Famously, Charles Beard argued that the drafting of the Constitution reflected the personal financial interests of the Founders, not impartial reflections on good, fair governance (Beard Reference Beard1913). Indeed, it is hard to see how the institutionalization of slavery grants to each according to her basic needs.

The point here is that we might be sceptical that the mere act of bargaining among homo prudens will always induce a state of uncertainty that then forces impartiality. Sometimes, the top dog is reasonably certain that she will still be top dog a hundred years from now – certain enough, at least, so as to not grant the bottom dog according to her basic needs, as the WPU requires. But if this is true then the WPU will not always be the object of rational choice. When bargaining asymmetries are particularly high and parties are reasonably certain that such inequalities will persist then there is no reason to grant to each according to her basic needs.

But then the first worry comes back in: if the bargaining asymmetry between the Athenians and Melians is particularly high then they will likely decide on a principle of conflict resolution decidedly less fair than Moehler's WPU. And if such is the case, then has the contractarian project actually succeeded? Ought we embrace a contractarianism that only guarantees us agreement, only sometimes getting us morals?