Jan Narveson is well known for his defence of right-libertarianism on contractarian grounds (e.g. Narveson Reference Narveson1988). In this volume, a Festschrift in his honour, friends and students of Narveson critically evaluate Narveson's theory.

Narveson's position can be summed up in three fundamental claims. First, the justification of a political philosophy or indeed any normative ethical theory, requires contractarian foundations. All contractarians consider morality as the outcome of an agreement among relevant parties. More precisely, moral norms are those rules that are agreed upon by agents in a suitably characterized bargaining situation. Contractarians share this starting point with other social contract theorists. However, contractarians differ from other social contract theories, like that of John Rawls, in that the latter treat such an agreement among rational agents as a heuristic instrument for identifying the content of morality. That is, authors like Rawls claim that moral norms are binding for reasons other than that they are agreed upon by agents in the original position. Narveson, like other contractarians, believes that agreement of some sort is necessary and sufficient for the normativity of such norms. (‘Of some sort’ because closer reading reveals that this social contract is not an actual agreement. Rather, it is ‘an agreement in the sense of a co-ordinated set of conditional dispositions’, see Narveson Reference Narveson1994.)

The type of contractarianism that Narveson endorses is Hobbesian. Hobbesian contractarians hold that rational agents are primarily motivated to maximize what they regard as valuable. This could include many things, but among these, self-interest figures prominently. Hobbesian contractarians regard morality as an answer to a problem. The problem is posed by what would happen under conditions of moral anarchy to rational creatures who are disposed to maximize their self-interest. Under such conditions, rational agents, who aim to maximize what they value, will compete with all means at their disposal for the scarce resources needed to realize this aim. Other agents will appear as actual or potential competitors and it is best to eliminate such competition as efficiently and effectively as possible. The result is a situation best modelled as an n-person prisoner's dilemma, where a non-optimal equilibrium is realized. In such a situation, rational agents will realize that they can benefit each other. As Jan Narveson puts it, ‘first because we are vulnerable to the depredations of others, and second because we can all benefit from cooperation with others’ (Reference Narveson1988: 148). This will motivate the agents to start bargaining with the aim of arriving at an agreement to constrain this maximizing behaviour and coordinate actions so as to benefit each other. Morality, for the Hobbesian contractarian, is a form of self-imposed constraint – a rational constraint – on the pursuit of the maximization of value. Unlike Hobbes, Hobbesian contractarians do not regard morality as something that is enforced by an authoritarian state. Instead, the restrictions that morality poses on the unfettered pursuit of what one values are restrictions that rational agents can agree to in a rational bargaining process that aims to bring about an optimal mutually cooperative outcome. Moral constraints are those constraints it is rational to adopt provided others do so as well.

The second fundamental claim of Narveson's philosophy is that such Hobbesian contractarian starting points inevitably lead to a restricted list of rights and corresponding obligations that emphasize individual freedom. The corresponding political conclusion is that a legitimate state necessarily is a libertarian state. Narveson is a so-called right-libertarian (as opposed to egalitarian libertarianism). Such libertarians typically argue for a small, non-authoritarian state in which basic liberties are rigorously respected, but nothing beyond this. As a consequence, right-libertarians do not believe that the state has any business requiring citizens to support others beyond respecting the negative claim rights of others.

The third claim of Narveson is typical for all right-libertarian political philosophy. In order to guarantee individual freedom, a legitimate state respects strong property rights and corresponding institutions (especially the market). That is to say, Narveson believes that individual freedom necessitates a robust respect for private property and the market.

Each of the contributions in this volume takes issue with one or more of these claims. Thus, there are several essays questioning the contractarian foundations of Narveson's philosophy (the contributions of Bond, Sanders, Machan, Groarke, and Tucker). Others have taken issue with the claim that contractarianism inevitably leads to such strict right-libertarian principles as Narveson endorses (the contributions of Dimock, Danielson, Murray, and Viminitz). Finally, there are a couple of essays focusing on Narveson's enthusiasm for the market and private property (the contributions of Levey, MacIntosh, Vallentyne, Brown, and Wein). In the final contribution to this volume Jan Narveson responds to each of these essays. In this review I concentrate on the discussion of the first of these claims: the contractarian foundations of Narveson's theory.

E.J. Bond, in his contribution, shares Narveson's commitment to a rational justification of (libertarian) morality which connects morality to wellbeing. Narveson, as a good Hobbesian contractarian, identifies well-being with self-interest. As a result, the moral life looks like a second-best kind of life, Bond argues. Respect for other people's rights and autonomy is ensured by coercive state institutions that curb the individual pursuit of self-interest. This assumes that individuals are characterized as Robinson Crusoes – creatures who would be perfectly happy and flourishing regardless of others. Bond takes issue with this implication of Narveson's contractarianism and argues that a proper defence of libertarianism takes seriously the idea of eudaimonia.

John Sanders discusses whether contractarianism can provide a foundational justification for the principles of morality for everyone. He agrees with Bond that one of the reasons contractarianism may not do so, is precisely because it seems attractive only to those who believe, like Narveson, that the point of morality is the protection and advancement of self-interest. Sanders has other arguments for his doubts about contractarianism. First, he shows that the type of ‘contract’ endorsed by Narveson and other contractarians could not be a contract at all. After all, as Hume has shown, that would turn the foundation of moral principles to a moral principle itself (i.e. the principle that requires agents to observe their contracts), thus leading to a regress. Of course, Narveson accepts this explicitly. A social contract is a ‘contract’ only in the sense that it involves the mutual restraint of interacting agents. Sanders is quick to object that this holds for non-human animals as well. Even ants have a social contract in this sense. And while ants may not perceive their restraint as necessary for the promotion of their interests, most of the time human beings do not have this perception either. Therefore, there are serious conceptual problems with the idea of ‘contract’. Still, it might be the case that contractarianism is superior to its alternatives. Sanders briefly discusses Narveson's dismissal of Kant's theory and Rawls’ famous Veil of Ignorance, and mentions Narveson's opposition to intuitionism before turning to a comparison with utilitarianism. He shows that the superiority of contractarianism over utilitarianism is not straightforward. Many of the problems that Narveson identifies for utilitarian foundations of morality are shared in some form by contractarianism.

Tibor Machan, the well-known libertarian, echoes many of Sanders’ concerns about the regress signalled above. Contractarianism, so he argues, begs the question, because it rests on the (moral) commitment or inclination of agents to keep their word. In a version of Euthyphro's dilemma, Machan denies that the principles of morality are correct because they would be the outcome of an agreement. Rather, we agree to them because they are moral. Machan, therefore, defends a version of intuitionism. I suspect that this is how most libertarians think of the foundations of their theory.

Leo Groarke is perhaps the most sympathetic to Narveson's contractarianism of the contributions in this first part of the Festschrift. He argues that Narveson's contractarianism is a response to a deep problem in political and moral philosophy – the problem of scepticism about moral principles. How could there be such principles, the sceptic argues, in the face of such widespread deep divergence and disagreement. The social contract theory of John Rawls tried to solve this disagreement by stripping people from their cherished moral intuitions and indeed their personal identity and inquiring if agreement is possible. However, it seems the wrong path to ask of people to bracket their deepest commitments, moral and otherwise, in order to secure agreement. Narveson, on the other hand, tried to reach agreement by ignoring these commitments and focusing on economic self-interest as the basis of morality. Groarke investigates whether a more full-bodied contractarianism is feasible. Such a contractarianism does not try to abstract from people's commitments in either the Rawlsian or the Narvesonian way. Instead, it seeks to arrive at a ‘fair’ compromise that everybody can live with, or tries to allow for multiple alternatives. While this may be a viable contractarianism – Groarke really does not say enough to determine whether it works or what he has in mind exactly – it is unlikely that it would lead to libertarian political morality.

Chris Tucker signals the very same problem that Bond, Sanders, Machan, and Groarke were grappling with in their contribution: the implicit regress in contractarianism. What reason is there for rational agents to comply with the social contract? Tucker notices that contractarians typically believe that such compliance should be universally rational to all in every situation. It is this latter belief that makes the regress problem so intractable. Tucker's solution is to relax the insistence on full and complete compliance with the contract. Instead, we should aim for an efficient level of compliance. Tucker argues that this means that the solution to the compliance problem primarily will be an educational enterprise. How should we educate one another and, especially, future generations (i.e. shape their preferences and beliefs) such that it will be rational for them to comply with the terms of rational agreement.

The contributions in this first part taken together illustrate some general and fundamental concerns for the viability of any form of contractarianism – not just the libertarian version that Narveson endorses. Contractarianism is a form of naturalist constructivism: moral obligations are constructed out of non-moral starting points. It combines this naturalist constructivism with a form of rationalism, the idea that moral norms are if not reducible, then at least compatible with our best theory of rational choice as it is expressed in game- and decision theory. This combination is what gives rise to both the regress problem and the compliance problem. The standard version of the regress problem, which is present in virtually all textbooks, is to argue that a social contract cannot be binding for those concerned, unless it is assumed that there is a pre-existing moral obligation to honour one's contracts. Obviously, that would give away the game, for the point of the whole contractarian exercise is to construct our moral obligations from non-moral materials. So contractarians, like Narveson, insist that the binding nature of the social contract lies in the rational pursuit of self-interest (or, more generally, whatever one values). However, we can simply repeat the challenge of the regress problem: what is it about self-interest, or value in general, that makes acting in accordance with it binding? We can agree that it would be a good thing if one's self-interest was promoted, but why is it obligatory to comply with a scheme designed to further one's interests (or values) in an environment composed of other agents similarly disposed? Contractarians will now say that this is what the contract does: the fact that one has contracted makes such compliance obligatory. However, that raises the questions as to how a contract can do this. Contractarians simply assert that this is the case, but it is hard to see how an act of agreement could create obligations for rational agents. In his response to Chris Tucker, Narveson argues that he takes his contractarianism to be such that it can justify claims of obligation, by showing that agents have good reason to do as their obligation requires of them. Note that this already moves away from the sort of strong constructivism implicit in contractarianism, for he no longer maintains that obligations arise from the social contract.

Suppose that we grant that rational agreement does create moral obligations for rational agents. Then we face the compliance problem for while it is one thing to agree to certain obligations, it is quite another to actually act in agreement with these obligations if this is not in one's interest. And if it is not always rational to comply with the social contract, it is – again – an open question what binds agents to the social contract. Therefore, the regress and the compliance problem are two sides of the same coin.

One could escape these problems if one gives up on one or more of the key ingredients of the social contract. First, one could give up on the assumption that agents in the state of nature have values that pit them against one another (more or less like E. J. Bond suggests). Perhaps people are deep down naturally altruistic and social and inclined to act in ways that conform to what their contractual obligations would be. That would take care of both the compliance problem – we can appeal to what people are motivated to do naturally – and it would if not solve, at least make the regress problem less threatening (only a hard core Kantian would insist that acting on natural motives is not the same thing as acting on an obligation even if these two types of actions are always identical). Narveson does not find this plausible.

Secondly, one could give up on the idea that agents are rational. Then, it seems that the compliance problem is neutralized. Morality then is not something that agents will need reasons for performing their part of the agreement. However, if this is the route to be taken, then it seems that there is no a priori reason to assume that the social contract will have a certain content, for if agents do not need reasons to act as they are obliged to do, they could be obliged to do anything. Given that there is already a lot of debate (see virtually all contributions in this volume) about Narveson's claim that strong libertarian principles will be the result of a contractarian justification, this is not a fruitful way out of the regress/compliance problem.

I suppose it is not a coincidence that many of us working with a rational choice approach to morality have abandoned Narveson-style contractarianism. Evolutionary game theory is the new game in town (see, for example, the contribution of Grant Brown to this volume). Furthermore, games other than the prisoner's dilemma are being analysed (e.g. Peter Danielson and Malcolm Murray's contributions to this volume). Several of the contributions to this volume take this route to see what can be salvaged from Narveson's project.

There is much more in this volume that merits mentioning and discussion. The contributions of Peter Vallentyne and Duncan MacIntosh especially pose really interesting challenges to the political part of Narveson's project: his insistence on pre-political (but not pre-moral) property rights and his enthusiasm for the free market. I cannot discuss these in this review. Suffice it to say that this volume truly honours Narveson in the best way a philosopher can be honoured by his friends, students, and admirers. It takes his work and develops it in ways that are often surprising and even partly convincing to their master, who graciously acknowledges each in his response at the end of the volume. The essays are well-written and taken together provide a very nice set of commentaries on Narveson's work. However, as I have tried to show above, the merit of this volume goes further than just supplying some interpretations and readings of Narveson, it addresses some very basic concerns about the foundations and implications of the contractarian enterprise.