On the People’s Terms

Precis

To enjoy a free choice, on the neo-republican approach, is to be able to select any of the options in that choice regardless both of what you prefer to do and of what anyone else prefers you to do; it is to be un-dominated in exercising the choice. To enjoy the status of a free person or citizen is to be publicly resourced and protected in the basic liberties: a range of choices—ideally, as large as possible—that all can exercise and enjoy at the same time as others.

This republican notion of freedom has implications on two domestic political fronts. ¹ It suggests, as a requirement of social justice, that the legal and cultural arrangements determining people’s relations with one another should give people the un-dominated status of free citizens. And it suggests, as a requirement of political legitimacy, that the state that establishes those arrangements should not impose in a dominating manner on its people. The state should allow people the status of free citizens both in their horizontal relationships with one another and in their vertical relationships with government.

Having outlined this ideal of freedom in the first chapter, the book expands on its implications for social justice in the second, and on its implications for political legitimacy—the main topic of the volume—in the final three chapters.

Chapter 3 argues that if the state is to be legitimate or un-dominating, it must be subject to the equally shared control of the people; it must be democratic in a rich, etymologically supported sense of the term. And then the final two chapters elaborate on that abstract theory of legitimacy, by providing a concrete model of the institutions that might deliver equally shared popular control. Chapter 4 sketches a system of popular influence over government that would be equally accessible to each. And chapter 5 provides a case for thinking that, suitably developed, that system would push government in a direction that was equally acceptable to all.

Richard Bellamy, Jacob Levy, and Frank Lovett raise some great issues about my claims in the book and I am most grateful for the attention they have given it. I discuss their contributions in the alphabetical order of their surnames.

Richard Bellamy

The version of neo-republicanism that I defend is closer to the Italian-Atlantic tradition, so I claim, than to that which Rousseau inaugurated. Like the Rousseauian approach, it upholds an image of freedom as non-domination, on the lines presented in the Precis. But unlike that approach, it defends a mixed-constitutional picture of republican institutions and a contestatory image of republican citizens. One the one side, it suggests that the best institutional arrangement will involve “a multi-dimensional, multi-centred system of popular interaction and decision-making” (286), not a system in which any one agency retains ultimate control. And on the other, it proposes that within this arrangement people should assume a contestatory role, not just a role as voters. “Given that the voice of the people is meant to emerge from a process of interaction between different bodies, . . . individuals operating in contestatory, as well as electoral, guise should constitute one of those parts” (228).

Richard Bellamy offers a challenge to these claims from within the republican fold, presenting himself as a republican of a neo-Rousseauian stamp. The challenge is republican insofar as it is premised on the assumption that the first concern of the polity is with the freedom of its citizens and that freedom should be taken to require the absence of domination.

But not only do Bellamy and I agree on the importance of freedom as non-domination. We also agree that even in my account, as he puts it, the ideas of the mixed constitution and the contestatory citizenry “have a different theoretical status to . . . freedom as non-domination.” And we agree that whether or not republican arrangements ought to involve the mixed constitution and a contestatory citizenry, they should certainly provide for a system of influence that is equally accessible to each and that pushes government in a direction that is equally acceptable to all: in short, a system of equally shared popular control.

What Bellamy maintains, however, is that I lose sight of the importance, emphasized by Rousseau, of the “two normative criteria” of popular sovereignty and equal participation. He claims that while “certain sorts of mixture and contestation” may be needed to modify electoral arrangements—he takes these to be acceptable for a neo-Rousseauian—these two criteria are fundamental and set limits to the mixture and contestation that ought to be allowed. And he maintains that it is precisely through being guided by these criteria that we can hope to identify a system that instantiates the equally shared popular control over government that we both advocate.

On the People’s Terms, as the subtitle indicates, presents both a theory and a model of democracy. The upshot of these remarks is that Bellamy and I agree, broadly, on the theory of democracy but divide on the best model of how it may be implemented. We would each like an institutional arrangement that generates a sense of the common interests of citizens and a means whereby those interests are imposed on government. But we differ on how this might be best achieved.

He believes that the optimal arrangement is one in which the people themselves are “directly engaged in deliberating with others,” so far as that is possible, and are prepared “to modify their self-interests to conform to the common interest.” And, breaking with the strict letter of Rousseauian doctrine, he thinks that that is best achieved under “a responsive electoral system based on majority rule that feeds into a representative assembly.” He argues that contestation may often serve only special interests and that mixture may multiply veto points counter-productively. Thus, while he goes some way towards admitting constraints on government, ranging from weak judicial review to a version of bicameralism, he insists on “the final decision resting with a single body.”

I suggest that the best constitutional arrangement is one in which ordinary people are involved in two ways: at the input end, in electing some officials in a multi-centered network of government—and in their selection of others—and, at the output end, in invigilating and checking officials at any of a number of nodes in the network and in any of a variety of ways. Deliberation also plays a role in this picture as it is the need to find mutually acceptable, network-wide criteria of decision making that generates an evolving body of commonly validated considerations and that allows these, over the longer term, to dictate both the processes and the products of policy making.

Looking at the different institutional pictures projected in our rival models, it is hard to resist the thought that Bellamy and I are at different points on a spectrum rather than on either side of a fissure. We both endorse extra-political constraints on districting, on campaign finance, and I suspect on interest-rate policy and the processing of ombudsman complaints. We both prefer the parliamentary system that guards against gridlock but maintains some bicameral constraints on legislation. And we both allow for an independent judiciary with powers of review over legislation. What divides us, it appears, is whether the constraints introduced should be strict enough to deny the right of final decision to the legislature and, ultimately, to the electorate.

That dividing line, however, is not bright red. For what looks like a dilution of the legislature’s power from one perspective—a sharing of power between different legislative chambers or with other agencies—may well look from another like a reversible delegation of power by an omnipotent legislature. The line would be red only if it turned on specific issues like whether government ought to be organized in the manner of Washington or Westminster; whether legislative power should be shared in a strict way between different chambers—as in Australia’s parliamentary democracy, for example—or reside primarily in a single house, as in Britain’s; or whether independent officials should be immune to being sacked for the period of their tenure in office. But those sorts of issues are surely ones that we would want to resolve pragmatically, not on philosophical grounds.

My suspicion is that whereas there is a principled divide between Rousseau and traditional republicans, there is no such divide between Bellamy and me. Rousseau did argue passionately for popular sovereignty but two claims were central to his case: first, a Hobbesian proposition, that the people could only have power if they were incorporated as a group agent; and second, an anti-Hobbesian claim, that such incorporation could only serve freedom if every citizen was a participant in that corporate body. Under Rousseau’s dispensation, then, unlike that which I associate with the Italian-Atlantic tradition, people would not have any contestatory rights, only rights of legislation, and no other body could share in the sovereignty of the popular assembly.

The difference between Bellamy’s model and mine is of a very different kind. Given that we both espouse an electoral representative system, neither of us can think of popular sovereignty in the same way as Rousseau. When each of us endorses the sovereignty of the people, then, we do so in a way that need not put us at philosophical loggerheads. We both think that the people—the plural constituting people, as I would say—are sovereign insofar as they have electoral and contestatory rights that they use in the maintenance and monitoring of government. While I would add that they enjoy a secondary sort of sovereignty as a singular agent—as the people incorporated or constituted via the state system—this need not be a dividing point. The popular sovereignty we both affirm commits neither of us to the Rousseauian sovereignty of a participatory, incorporated assembly.

This is not to say that the issues that divide us are of little significance. On the contrary, questions of institutional design are where political theory hits the ground and begins to gain some serious traction. I applaud his efforts on this front, particularly in thinking about republican lessons for the European Union. Would that more political theorists would address such institutional questions and not let a concern for the abstract theory of justice monopolize their attention.

Jacob Levy

While Jacob Levy is very generous about many aspects of the neo-republican project, he voices two doubts and raises three main objections. His first doubt is whether liberty consists only in non-domination and not also in non-interference; and his second is about my attempt “to distinguish Rousseau’s theory of liberty” from that of “the broader republican tradition.” I will first comment on these doubts and then turn to his objections.

Under my account of free choice and free status—summarized in the Precis—non-domination is necessary for liberty, so that the dominating master who doesn’t actually interfere still reduces a subject’s freedom. ² But it is wholly consistent with this to hold that domination with interference is worse than domination without and, if you want to keep the books that way, that it is worse in terms of freedom itself. ³ Levy’s doubt as to whether liberty consists only in non-domination, then, may not mark a difference between us. For the view I defend certainly leaves room for acknowledging that someone interested in freedom may have a well-placed concern for seeking to ensure that even if there is domination in place—even if in that sense there is a loss of freedom—still that domination does not take the form of actual interference.

Levy’s doubt about Rousseau deserves comment, because my claim is that Rousseau does actually hold the republican view of freedom as non-domination—he does not differ in the theory of liberty, strictly taken—and breaks from the older tradition only in rejecting the mixed constitution and the contestatory role of citizens. ⁴ The great antonym of freedom for Rousseau, as for traditional republicans, was dependency on the wills of others and his institutional design in The Social Contract was meant to guard against this, ensuring that “every citizen be perfectly independent of all the others.” ⁵ The dependence to be prevented is the condition in which you may find yourself, as he puts it elsewhere, “at the other’s discretion.” ⁶ And this is a dominating variety of dependence, distinct from the universal dependency that obtains ‘to one or another degree’ in all sorts of relations. ⁷

Moving from doubts to criticisms, Levy’s first main objection bears on my claim that the existence of a coercive state in our lives does not necessarily involve the domination of its members; and that whether they are dominated or not depends entirely on how the state exercises its power: in particular, on how far the exercise of that power is subject to the equally shared control of the citizenry.

Simplifying considerably, the crux of my claim is that while states may have emerged as a result of the dominating ventures of some in suppressing others, the worldwide system of states is now locked-in as a matter of more or less inevitable social necessity. It is not by virtue of the unconstrained will of any individual or body that there is no territory on earth where you can live without subjection to a state or something like a state. And so the mere fact that you have to live under local coercive laws does not mean that the coercive government under which you live dominates you. It could not cease to coerce you without ceasing to discharge roles it is constrained to play, first, in protecting you and your fellows against other states and, second, in providing for law and order in a way that treats you and others as equals: it does not reserve coercion only for those deemed likely to breach the law.

As against this line of thought, Levy suggests that still, as a matter of fact, we may be “governed in ways that are not wholly compatible with our liberty and not wholly legitimate.” I agree fully, of course, if this is taken to mean that the state under which you live, however inevitable its existence, may not exercise power in a way that is legitimate and friendly to freedom: it may be despotic in character. But I can also agree with the remark if the message is that the system of states we have is less friendly to freedom than it might have been—or than some other international arrangements might have been. As I say in the book, the world of states “may be considerably less attractive than a world in which there are no states or . . . only one state or federation” (164).

Levy’s second objection starts from agreement that in a plenary assembly of all citizens, majority voting—or indeed any mechanical form of voting—would be likely to produce inconsistent judgments over connected issues, even if the members were individually consistent; and that in order to guard against this problem, such an assembly would have to arrange for feedback on any inconsistencies and for deliberation about how best to resolve them. The assembly would have “to be able to conduct second-stage, post-feedback deliberation” as well as more regular deliberation (OPT 194). I argue that the plenary assembly envisaged by Rousseau cannot work effectively in a large-scale society, since it will be unable to practice this second-stage sort of deliberation.

Levy asks, however, why I should be sanguine about the avoidance of inconsistency in the sort of arrangement I propose, which involves a representative assembly or assemblies operating in the context of relatively independent bodies and a contestatory citizenry. Three points in reply. First, it is at least possible to avoid the problem raised if we don’t rely just on mechanical voting, as Christian List and I have argued elsewhere. ⁸ Second, it is a desideratum on any constitutional arrangement, no matter how far it conforms to the mixed constitution, that it should realize this possibility. And third, the sort of model proposed can be designed to meet the desideratum, as I argue later in the book (OPT 284–85). In illustrating that possibility I say that the Washington and Westminster models represent different ways for satisfying the desideratum, although it should be clear that I think the Westminster model is rather more effective on that front, even when bicameralism takes a strict shared-power form, as in Australia.

Levy’s third objection bears on my historical conjecture that it may have been their reformist commitment to expanding the constituency of the state’s concern that led Bentham and Paley to replace the ideal of freedom as non-domination with the less demanding ideal of freedom as non-interference: an ideal that could even be reconciled with eighteenth-century family and master-servant law. He suggests that perhaps they took the pre-existing republican conception of freedom to be essentially hierarchical, identifying freedom with a security against the interference of others that only an elite could enjoy. Perhaps they opted for freedom as non-interference precisely on the grounds that it looked like a suitably egalitarian ideal.

This is an interesting and possibly correct observation; I leave the issue for specialists in the history of the period. But if Bentham and Paley took the republican ideal to be essentially elitist, then I would say they were wrong. There is no reason in principle, at least in our advanced, affluent societies, why we should not be able to create the material, legal, and cultural conditions under which people are guarded effectively against private and public domination and are able to live up to the traditional image of the free republican citizen. Am I too sanguine, as Levy suggests, in thinking that people could generally be happy to live in such equality? It’s hard to say. But even if Machiavelli is right that the rich want to dominate, and the poor only to avoid domination, there is still some ground for hope. The poor have the numbers, after all, and in a properly organized democracy they ought to be able to use those numbers to establish an egalitarian republic.

Frank Lovett

Frank Lovett begins with a useful recasting of my view of freedom, under which it requires undominated ability. The stress on ability is continuous, as he notes, with a theme in my Republicanism but it receives much greater emphasis here. ⁹ I still hold that your lack of resources and capacity are likely to restrict the choices in which you enjoy non-domination and to facilitate domination by others. But I have also come to recognize that unless ability is included, the linkage between freedom and responsibility fails (OPT 44). If you believe that you are free to do X or Y, then you must believe that you are fit to be held responsible for which option you choose. But you cannot self-ascribe fitness to be held responsible in the choice if you believe that you lack the capacity to choose one of the options. So if you believe that you are free to do X or Y, you must believe that you have the capacity or ability to choose either option.

Lovett is happy enough to treat freedom as non-domination as a value and, on the basis of that assumption, he asks how we should respond in our political arrangements to that value, a question he has addressed with great insight in his own work. ¹⁰ By my account, social justice in people’s relationships with one another requires that they each enjoy the same sort of free status; and the political legitimacy of the state that imposes that arrangement requires that they each share equally in controlling how the state provides for that status. While not directly addressing the question here, I believe that this creates a tension between the demands of justice and legitimacy, which has escaped the attention it deserves. The state that satisfies your criterion of justice might not satisfy your criterion of legitimacy, and vice versa; for example, the democratic, legitimate state might not promote justice, as you understand it. ¹¹

Does Lovett disagree about this tension between justice and legitimacy? He may seem to do so, since he says that “there exist feasible socio-political orders satisfying both.” But the disagreement is superficial only. I hold that it is not inevitably the case either that a legitimate regime will produce a just order or that a just order will emanate only from a legitimate regime. He observes that nonetheless a given order may be both legitimate and just. The two ideals may not necessarily converge, as I say; but neither, as he says, do they necessarily diverge.

Lovett raises two objections to the claim that social justice requires equal free status or, more specifically, the enjoyment of freedom as non-domination in the same choices—the basic liberties—under the same public regime of protection and resourcing. I say in defense of this account that it would give life to the image or heuristic of the free republican citizen, able to stand tall and deal as an equal with others. In particular, it would enable each to satisfy the eyeball test of being able to “look others in the eye without reason for the fear or deference that a power of interference might inspire” (OPT 84). The two objections are, first, that this account sits uneasily with my further claim that the basic liberties must also be a maximal class of choices that all can exercise and enjoy at the same time as others; and second, that it does not resolve “the classic problem of conflict-of-liberties.”

The first objection is that the requirement that the basic liberties be such as to satisfy the eyeball test may diverge from the requirement that they be co-exercisable and co-satisfying. Lovett usefully reviews various ways of harmonizing them but I would defend a distinct resolution; I do not address the issue explicitly in the text. The eyeball test identifies a role that the basic liberties must fulfill and the maximal co-exercisable, co-satisfying criterion identifies at a more specific level the sorts of choices required to fill that role.

Thus, in order to enable each to look others in the eye without reason for fear or deference, the choices to be publicly protected—and, if necessary, resourced—must clearly be the same for each; at least that is so on my assumption that it is not possible to give each a customized set of choices that no one else would envy (OPT 92). That explains why the basic liberties must be co-exercisable and co-satisfying. And, to make a point I do not explicitly register in the text, in order to enable each to look on his or her community or state as a non-dominating agency, the choices in question must not be “unnecessarily restricted” (OPT 93). That explains why they must be not only the same for all but also as large a set as possible.

Lovett’s second objection is that I leave various conflicts of liberties in place but I am not persuaded that I do this. I conclude with a number of observations in defense of that claim, which are designed to cover various doubts that he raises.