Published in the series Founders of Modern Political and Social Theory, Joshua Cohen's Rousseau is a very difficult book. The series is aimed at scholars and students, but you'd be a brave student to take your first steps into the minefield of Rousseau interpretation using this as a guidebook. As aware as any other commentator that Rousseau's thought is riven by tensions, Cohen curtly dismisses those who have suggested that those tensions may be profound and ineliminable.

The book's conclusion is already announced in its subtitle: ‘a free community of equals’. That slogan, repeated frequently in the text, is ambiguous, since a community might be free of dependence on other communities, even though all its members were equally unfree by liberal standards. Cohen quickly clears up that pedantic quibble and explains that it really means a community of free, equal citizens. Having revealed the dénouement at the start, the author devotes the rest of the book to establishing Rousseau's credentials as a founding father of the modern democratic order. To reach this happy ending, Cohen slays many exegetical monsters along the way. He presents a dense sequence of complex arguments, unpacking each one into several sub-arguments.

Cohen undertakes to answer Rousseau's question: ‘How to find a form of association that defends and protects with all the common force the person and goods of each associate and by means of which each one, in uniting himself with all, obeys only himself and remains as free as before?’ and his answer: ‘Each one of us puts his person and power in common under the supreme direction of the general will’ (Social Contract [hereafter SC] I.6). In what follows Cohen goes on to explain the connection between the ideas of general will, common good and equality in Rousseau's theory. His ultimate aim is to show how, when taken together, they provide the framework within which individual freedom and well-being may flourish.

Cohen has three aims: to establish the plausibility of reconciling equality and autonomy within a free community of equals, to show how Rousseau's idea of participatory democracy might be applicable to a modern complex society, and to ‘reconcile two tendencies within Rousseau's political thought. namely an individualist tendency in which self-interested agents seek a form of coexistence with a framework of law, and a communitarian tendency’, in which the key value is ‘shared social and national solidarity’.

In chapter 2 Cohen begins to set Rousseau's texts a number of requirements (‘conditions’) which they must satisfy if they are to play their allotted role in the providential saga leading up to modern constitutional democracy. In the case of the general will the conditions are four-fold. First, it must protect the legitimate particular interests of citizens. Second, it must be the case that ‘citizens share and it is common knowledge that they share a conception of the common good’. Third, ‘the fact that an institution or law promotes the common good provides citizens with [an overriding] reason for supporting that institution or law’. Fourth, ‘citizens have reasonable confidence that the institutions will conform to a shared understanding of the common good’. The first condition is particularly important if Rousseau is to emerge innocent of the charge that in his theory the individual is sacrificed to the corporate whole. In response Cohen correctly points out that parties to the contract put themselves under the supreme rather than the exclusive direction of the general will, while acknowledging that many other passages strongly suggest the total absorption of the individual into the sovereign. The crucial passage for Cohen's reading is SCII.4: ‘in addition to the public person we have to consider the private persons who compose it and whose life and freedom are naturally independent of it’; though Rousseau qualifies even that with the rider: ‘But the sovereign alone is the judge of what is of importance to the community’. Cohen concludes by passing summary judgment on the exclusive direction camp: ‘Rhapsody notwithstanding, the complete civic unity view … gets Rousseau wrong. They [its proponents] exaggerate Rousseau's view or … focus on Rousseau's … own exaggerations …’ One might be unsure why certain passages are more rhapsodic and exaggerated than others.

Much of chapter 3 (Reflections on the General Will's Sovereignty) is devoted to questions of motivation. Cohen concludes that it is wrong to ‘think of the parties to Rousseau's social contract as moved solely by self-love’, and that ‘once we properly understand their motivations, we will see that there is no tension between the idea of legitimacy based on common agreement and the idea that legitimacy resides in the general will’.

Cohen explains the parties' motivations in terms of The Natural Goodness of Humanity (chapter 4); and shows how that natural goodness is everywhere corrupted, but can yet be redeemed in a better ordered society, by means of channelling individuals' amour-propre. Cohen pays generous tribute to Nicholas Dent's eye-opening treatment of amour-propre in 1988, and his own account of it is based on Dent's. For Rousseau, once men and women have entered society, they begin to seek recognition in the eyes of others. Then, depending on circumstances, the universal drive for recognition can bring about mankind's elevation or debasement. Rousseau has more to say about the latter than the former, but Cohen, following Dent, shows how a positive outcome is always possible. The (or a) salient factor in the transition from innocent to vicious amour-propre is the emergence of inequalities of wealth and power between individuals and groups, described by Rousseau in the Second Discourse. That text provides a ‘structural’ explanation of the way in which the powerful come to find their own self-worth to be dependent on its absence in the powerless. Only a new order can realize Rousseau's tentative vision of egalitarian, reciprocal social interaction. Cohen correctly attributes to Rousseau (and to Kant) the view that each member of a well-ordered society should receive equal recognition of his worth as a citizen. He is also right to emphasize Rousseau's particularly clear awareness that a degree of economic equality is needed in order to make that sense of self-worth a reality.

Thus each of us, if we are to play our part as citizens, must be sustained by a sense that we are of equal worth in the eyes of our fellow citizens. But it does not follow from this that esteem must always be equally distributed if amour-propre is to remain innocent: ‘To be loved one has to make oneself lovable. To be preferred, one has to make oneself more lovable than every other, at least in the eyes of the beloved object’, Rousseau tells us in Emile Book IV.

That observation does not constitute an objection to Cohen's argument, since his aim is to extract from Rousseau's texts a theory of an acceptable political order, not one of passionate love. But even in the public domain Rousseau did not ban all inequality in the distribution of deserved esteem and the competitive striving for it. That is what motivates the scholars and scientists working for the public good in the First Discourse and the patriotic citizens striving for acclaim in the Considerations on the Government of Poland. Cohen's response to this latter point might be that we are entitled to equal respect only as citizens, as bearers of the equal basic liberties listed by Rawls at A Theory of Justice (11). Beyond that, whether in our private lives or in the public arena of ongoing politics, equal distribution of esteem is not always required or even desirable.

Towards the end of chapter 4 Cohen acknowledges that although Rousseau produces a plausible story in the Second Discourse to explain how, if man was originally good, he has come to be corrupted, he ‘does not show – positively – that the doctrine [of natural goodness] is true’. That further step depends, according to Cohen, on the principle of reasonable hope espoused in the Savoyard Vicar's Profession of Faith. Assuming that principle, ‘endorsement of the doctrine of natural goodness … is recommended … on moral-practical grounds’. We are permitted ‘to hope, woth reason, for a society that answers to the demands of self-love and freedom’.

Those who reach the last chapter of Cohen's book will by now be familiar with his strategy. Once again he sets Rousseau a set of Herculean labours from which we know in advance he will emerge victorious. This time Rousseau has the apparently impossible task of winning the title of the, or a, forerunner of modern representative democracy. The seemingly insuperable obstacle is Rousseau's categorical declaration that ‘the deputies of the people are not and cannot be its representatives; they are only its agents (commissaires) …’ (SC.III.15). Admittedly, Rousseau then bids us ‘understand how the government does represent the sovereign’, in the sense that the sovereign, in exercising its will, requires the services of its agents, the executive, to carry out that will. But since it is executive rather than legislative power that is represented, this does not contradict Rousseau's initial declaration and it cannot be read as an anticipation of modern representative democracy, which looks more like Rousseau's contemptuous picture of ‘the English people who mistakenly thinks it is free, which it is only during the election of its MPs; once they are elected, the people is enslaved, it is nothing’. Cohen's response is that Rousseau dismisses representation in the context of his concern that members of the government should not usurp sovereign power. It is therefore a more or less technical question of fashioning institutions to prevent that from happening, rather than a matter of principle. But is this the most natural reading of Rousseau's words, even when taken in context? Many would question it. Elsewhere, in the Government of Poland, Rousseau does allow for representation, but it is evidently a pis-aller, a system designed to improve an already existing, imperfect system. So could the model of popular sovereignty be transposed from the ancient to the modern world, from a micro to a macro scale? Rousseau himself recognized how difficult it would be and entertained various possible solutions without definitively endorsing any of them.

In the last pages of the book Cohen addresses the objection that ‘Rousseau's theory of popular sovereignty is authoritarian’ since it allots formal power to the people, but real power to the government. Cohen agrees that in Rousseau's system the government or executive, understood as a separate institution, is entitled by the constitution to elect its members to office, but, he maintains, this should not threaten popular sovereignty provided that suitable institutions are in place to ensure that the commissaires are answerable to the people. The question of ‘legislative initiative’ is more difficult since a cornerstone of democracy is the right of the people to be involved in making and changing the law on the basis of free, informed discussion. Yet at SC.IV.1 Rousseau explicitly denies it that right, telling us of his reflections on ‘the simple right to vote [exercised in] every act of sovereignty, a right which nothing can take away from citizens, and the right to express opinions, propose, divide and discuss, which the government has always taken great care to reserve for its own members …’ Interestingly, Rousseau expresses neither approval nor disapproval of this restriction of popular sovereignty, but even in ‘reflecting’ on it, he betrays, as Cohen suggests, ‘ambivalence’ about the political competence of ordinary citizens.

If I have expressed unease about Cohen's strategy at certain points, it is because I think the ‘ambivalences’ in Rousseau's thought lie deeper than Cohen does and because I doubt whether they can be fully resolved in a drive to assign Rousseau to a preordained place in the history of political theory.