The democratic boundary problem and social contract theory

Democracy means “rule by the people”. It refers to a form of political rule in which “the people” collectively govern themselves by participating as free and equal individuals in the political decision-making process. But how to decide who legitimately make up the group of individuals who are bound together as a people for the purpose of collective self-government?

The practical relevance of this question, which is referred to in the political philosophical literature as the democratic boundary problem, ¹ can hardly be overstated. To see this one only has to take a look at the many disputes that arise concerning (the drawing of peoples’) boundaries. Disputes over secession and migration—phenomena which are becoming increasingly more common in a globally interconnected, shrinking world—essentially constitute articulations of the boundary problem. Processes of globalization have also incited intense debate on the moral desirability and practical possibility of installing transnational or global forms of democratic governance. What is being challenged in this debate is an assumption that has largely been taken for granted since the rise of the modern doctrine of popular sovereignty: that the nation-state constitutes the proper political unit within which democracy is practiced. But to question this assumption is, of course, to raise the boundary problem. ²

The aim of this article is to contribute to settling these disputes concerning the constitution of the people by means of a social contract approach. The longstanding tradition of the social contract offers one of the most dominant normative frameworks for evaluating political structures and acts. Furthermore, since modern democratic theory can be understood as an outgrowth of a tradition of political thought that focuses on consent as the foundation of political authority, it is only natural to engage in the quest for the legitimacy of the people using the conceptual tools of the social contract tradition.

Despite this, however, social contract theorists have altogether ignored the boundary problem. Classical and early modern contract theorists (e.g. Hobbes, Locke, Rousseau, and Kant), as Whelan (1983: 25) correctly notes, have been primarily concerned with answering the question of how political authority can be established among a group of previously unattached individuals, rather than with confronting the logically prior question of “how determinate communities come to be set off from one another in the boundary-less state of nature” (Whelan, 1983: 25). Similarly, modern contract theorists have been, and still are, mainly interested in developing general moral principles (e.g. Gauthier and Scanlon) or principles of justice for already existing political societies, usually democratic nation-states (e.g. Rawls and Buchanan). Essentially, then, all social contract theorists work from the assumption of a given people. They presuppose what requires legitimization, namely the existence of a delimited group of individuals, and subsequently only ask how they could agree among themselves to establish a political and moral order. Developing a social contract theoretical solution to the boundary problem enables me to fill a gap in social contract theory, thereby making it a more complete moral and political theory as well as practically more relevant since it can guide political action where it previously could not. ³

This article is organized as follows. Based on a distinction between Lockean, contractarian (i.e. Hobbesian), and Rawlsian contractualist (i.e. Kantian) contract views—these being the dominant incarnations of social contract theory—I shall argue first that each of these three contract views can be (re)interpreted in such a way that it is capable of solving the democratic boundary problem. Next, I shall present some practical implications of these contract views. Finally, I shall argue that the Rawlsian contractualist solution is morally superior and more deeply democratic than its Lockean and contractarian rivals.

Lockean political voluntarism

Central to Locke’s political philosophy is his natural law-based interpretation of the moral values of freedom and equality. This interpretation starts from the thesis that all individuals are naturally free (Locke, 2000 [1698]: II 4). ⁴ Man’s natural freedom refers to a natural claim right. ⁵ It is a right that is possessed equally by all men in virtue of their humanity and that, as opposed to other moral rights, holds unconditionally. ⁶ According to Locke (2000 [1698]: II 4, 190), individuals are free in the normative sense that they have a natural right “to order their Actions, and dispose of their Possessions, and Persons as they think fit”. This does not mean, however, that individuals have a right to act without any moral constraints at all; it is not a “State of Licence”, i.e. a “Liberty for every Man to do what he list” (Locke, 2000 [1698]: II 6, 57, emphasis in original). Locke is very clear about this when he writes: “we must consider what State all Men are naturally in, and that is, a State of perfect Freedom to order their actions (…) within the bounds of the Law of Nature” (Locke, 2000 [1698]: II 4, emphasis in original). Although individuals are naturally free, their freedom is nevertheless limited by the rules of natural law, which teaches us that we have a natural duty to preserve mankind (Locke, 2000 [1698]: II 6–7, 16, 134–135, 149, 159, 171, 183). According to Locke, mankind is best preserved negatively by securing individuals a private sphere of non-interference in which they can freely (re)formulate and pursue their ends (Locke, 2000 [1698]: II 4, 6–7, 22, 57). Essentially, man’s natural freedom refers to what might be called the natural negative claim right to personal self-determination: the natural right to act without external constraints and a correlated negative duty on others not to harm him in his life, liberty, and estate.

On Locke’s view, an individual can give up his natural freedom and become bound by obligations only by deliberately and voluntarily entering special relationships. These special obligations are distinct from natural duties in that they hold relatively to those individuals only who, by having performed certain voluntary acts (such as consent), have clearly expressed their intention to become so bound. According to Locke, our political bonds are special obligations; they are acquired through the deliberate and voluntary performance of consensual transactions by individuals (Simmons, 1979: 63–64).

This can be seen clearly when Locke raises the question of how a group of individuals can legitimately set themselves apart from other individuals, and join together in a political community in which they live under a common political authority. “This”, Locke writes (2000 [1698]: II 95), “any number of Men may do, because it injures not the Freedom of the rest; they are left as they were in the Liberty of the State of Nature”. What Locke calls the “Liberty of the State of Nature” refers to the natural negative claim right to personal self-determination. And this state of natural freedom, as the quote shows, extends to political matters as well. In other words, the natural right to personal self-determination includes a right to political self-determination. This is a negative claim right to be free from political bonds. Political relationships cannot be forced upon individuals any more than personal relationships—marital, friendship, work, or otherwise.

The negative natural claim right to personal political self-determination underpins Locke’s endorsement of what is generally known as the thesis of political voluntarism. According to this thesis, “political relationships among persons are morally legitimate only when they are the product of voluntary, willing, morally significant acts by all parties” (Simmons, 1993: 36). The acts Locke has in mind here are of a consensual kind: the establishment of political relationships (and any accompanying political obligations and institutions) among a group of individuals in the state of nature is legitimate if it is grounded in an agreement among all individuals involved (Locke, 2000 [1698]: II 95–97, 99, 117, 171). It follows, in Locke’s view, that the group of individuals that sets itself apart and binds itself together as a people ultimately ought to include, to use Goodin’s words (2007: 41–42), “all and only those persons each of whom is not rejected by any of the others as a member”. Let us call this the None Rejected Principle.

What makes the None Rejected Principle interesting is that it is part of a Lockean voluntarist model of political association that offers an acceptance-based solution to the boundary problem that is fundamentally democratic. This is important because a satisfactory solution to the boundary problem should be democratic. After all, that is what the problem is ultimately about: how to determine democratically who does and who does not belong to the people? (cf. Miller, 2009: 203). Democracy is often depicted as centering around the ideas of (1) a collectively self-governing people and (2) collective self-government as the expression of the people’s common good. In so far as (1) and (2) warrant talk of democracy, the Lockean political voluntarist solution to the boundary problem is democratic as well because it offers a particular interpretation of these ideas. The None Rejected Principle supports a collective decision-making procedure in which all parties involved jointly decide whether they are (un)acceptable to one another as members of the same people. And it is a form of collective self-government aimed at establishing and regulating forms of social interaction—a people—through which (common) goods are created and sustained that are instrumental to the realization of the interests of all individuals participating in it: the preservation of their “Property”, i.e. their “Life, Liberty and Estate” (Locke, 2000 [1698]: II 85, 87).

Still, one might think that this way of demarcating the people raises a problem. Lockean political voluntarism is an actual contract view: the legitimate constitution of the people is the outcome of a democratic procedure in which free and equal individuals actually participate in order to see whether they accept each other as members of the people. However, this requirement of actual agreement, which reveals a commitment to the democratic idea of collective self-government, might seem to cause the Lockean quest for the legitimacy of the people to lapse into an infinite regress. After all, before a collective decision can be made on the substantive issue as to who constitute the people, a prior decision has to be made as to whose consent is required. In order to remain consistent, however, this prior decision requires a collective decision for it to be legitimate as well. But, clearly, this only begs the question as to whose consent is required once more, thus causing an infinite regress of decision-making procedures from which no procedural escape is possible. ⁷

Interestingly, Lockean political voluntarism offers a democratic solution that avoids the aforementioned infinite regress. If we want to solve the boundary problem by appealing to a democratic procedure, then we have to make sure that nobody is a priori excluded from it. Otherwise, the procedure would already presuppose the us/them distinction that it is meant to generate, and this would trigger an infinite regress of collective decision-making procedures. The question, then, is whether the collective decision-making procedure derived from the None Rejected Principle excludes any individuals from the start. Interestingly, this is not the case. The None Rejected Principle assumes a foundational situation in which all individuals (inhabiting the state of nature) are always free to approach one another in order to find out whether they are (un)acceptable to one another as members of the people. In other words, the None Rejected Principle yields a democratic procedure that is both open and serial. Since nobody is a priori excluded from this procedure, it follows that the people are not presupposed by but are instead the genuine result of it, and so the infinite regress never arises because the necessity of legitimizing a presupposed us/them distinction is absent.

Contractarianism

Let us now turn to contractarianism and contractualism and see whether these contract views can solve the boundary problem as well. Any attempt to do so, however, faces an immediate problem: although we are interested in a democratic solution to the boundary problem, contractarian and contractualist views are, on the standard interpretation at least, premised on a liberal instead of democratic logic, and as such engage in a strategy of hypothetical as opposed to actual justification (Abizadeh, 2008: 41). The basic idea is that laws or norms are legitimate if they are the object of a hypothetical agreement that a group of idealized individuals would reach under idealized circumstances (Stark, 2000: 314). So, if contractarianism and contractualism are to offer satisfactory (i.e. democratic) solutions to the boundary problem, then it is necessary to (re)interpret the two fundamental elements underpinning these views: the nature of (1) the contracting parties and (2) the initial choice situation.

Starting with contractarianism, there is a distinctive and continuous line of contractarian reasoning in the tradition of the social contract that originates in Greek and Roman philosophy in its rudimentary form, that received its mature form in Hobbes’s (2000 [1651]) philosophical work, and that, in its most sophisticated form to date, has been defended by contemporary contractarians Buchanan (2000 [1975]) and Gauthier (2006 [1986]). Modern contractarian views aim at deriving morality from the non-moral premises of rational choice (Buchanan, 2000 [1975]: 9–12; Gauthier, 2006 [1986]: 2–4, 7–8). This aim underpins the specific way in which contractarians characterize (1) the contracting parties and (2) the initial choice situation. Understanding morality in terms of, and as being derived from, rational choice, presupposes a conception of the person as a homo economicus with a maximizing capacity for practical reason. This means that each contracting party is instrumentally rational in that he seeks to maximize his utility by taking efficient means to his given ends. In a contractarian view, the initial situation, which is normally referred to by contractarians as the “state of nature” or “initial bargaining situation”, constitutes the non-cooperative baseline against which the rationality of engaging in cooperative interaction is evaluated by each of the contracting parties.

If contractarianism is to solve the boundary problem, then its characterization of (1) the contracting parties and (2) the initial choice situation has to be reinterpreted in terms that make it compatible with the democratic logic of actual justification. Consequently, the state of nature has to be interpreted in actual terms. Thus understood, individuals are in the state of nature if they are, as a matter of fact, not engaged in cooperative interaction. Furthermore, the contracting parties should also be interpreted in a way that is consistent with a democratic strategy of actual justification. The basic idea is this: in so far as we, human beings, actually think of ourselves as persons with a maximizing conception of practical reason, we are committed to accepting a specific bargaining procedure as the democratic mechanism through which we legitimize norms and political arrangements. This procedure offers a particular interpretation of a democratic strategy of actual justification: it specifies the conditions under which norms and political arrangements should actually be justified by and to individuals and so count as democratically legitimate.

Let us, then, take a closer look at the nature of this actual bargaining procedure. It is, to begin with, rational for utility-maximizing individuals to engage in, or to remain engaged in, cooperative interaction with each other if (1) there is potential for mutually advantageous cooperation between them (i.e. if they are in the circumstances of justice), and they are, and know of each other that they are, both (2) capable and (3) willing to constrain their behavior toward one another (Morris, 1991a: 59). So, rational utility-maximizers will accept—not as the object of rational agreement but as a necessary presupposition of such agreement—a bargaining procedure that incorporates these conditions.

This contractarian account of rational cooperation has clear implications for the way in which the people are to be understood: as a cooperative venture for mutual advantage. Central to this view is the idea that utility-maximizers collectively determine the terms of their political association by means of a (suitably defined) bargaining procedure. Whether individuals engage in cooperative interaction, and on what terms, depends entirely on the degree to which they are willing and able to make and accept concessions during a bargaining process. If a set of mutually compatible offers is reached, the contracting parties bind themselves to one another through mutual acts of consent, and thereby create a political society in which they subject themselves to a common political authority and abide by a common set of rules (expressing certain moral principles). In a contractarian world, then, the people consist of all individuals for whom it is mutually beneficial to bind themselves together.

It is important to emphasize that utility-maximizing individuals are committed to accepting a bargaining procedure that does not only satisfy conditions (1), (2), and (3) but also two further conditions. Like the Lockean political voluntarist approach, the contractarian bargaining procedure is also (4) open and (5) serial. Since cooperation holds the promise of reduced costs and increased benefits in comparison to what one could expect in a non-cooperative state of nature, it is rational for utility-maximizing individuals to try to engage in it. This means that a rational utility-maximizer at the very least wants to see what another individual, as a potential partner for mutually advantageous cooperative interaction, brings to the bargaining table. So utility-maximizers will opt for a bargaining procedure that is accessible to any individual. Notice, again, that this choice for an open bargaining procedure itself is a necessary presupposition of rational agreement. Since every individual is conceived of as a potential partner for mutually advantageous cooperative interaction, it is rational for each utility-maximizing individual to accept a bargaining procedure that is open for all. Rejection of the open character of this bargaining procedure would be contradictory to the idea of rational cooperation.

Furthermore, the bargaining process can be repeated at any time (e.g. when individuals already cooperating with each other meet individuals they have not met before, or when some or all members of a cooperative venture have reasons to reconsider their cooperative allegiances). Given that utility-maximizing individuals consider each other as potential partners for mutually advantageous cooperation, it follows that they will accept the requirement that the bargaining procedure must be serial—and, once more, the acceptance of this requirement is a necessary presupposition of rational agreement. According to the contractarian view, the constitution of the people is legitimized as the ever temporary result of a series of mutual acts of consent on the part of those individuals for whom it is mutually advantageous to bind themselves to one another.

The open and serial character of the contractarian bargaining procedure guarantees that it does not fall prey to the aforementioned infinite regress, in the same way as it does in the case of the Lockean approach to the boundary problem. Since every individual is a potential partner for mutually advantageous cooperation, it is rational to engage in a bargaining process for any utility-maximizing individuals who happen to encounter each other. This conjunction of the open and serial character of the contractarian bargaining procedure ensures that nobody is excluded a priori from the decision-making process. It follows that the people are not presupposed by the contractarian bargaining procedure (which, as I have argued above, would trigger an infinite regress) but are instead the genuine outcome of it.

This way of understanding (and modifying) the contractarian framework ensures that the acceptance-based solution it offers to the boundary problem can indeed be understood in democratic terms. The decision whether or not to engage in mutually advantageous cooperation can be understood as a democratic one: first, because it is the result of an actual collective decision-making process which involves bargaining among all potential members of a people who actually happen to have met each other; and second, because this collective decision-making process is aimed at creating and sustaining a common good that is instrumental to the furtherance of the given interests of those individuals actually participating in it. So like its Lockean counterpart, contractarianism is also capable of offering a democratic solution to the boundary problem that does not cause an infinite regress. ⁸

Contractualism

Contractualism, which has its origin in the philosophical works of Kant (2000 [1793], 2000 [1795]) and Rousseau (2002 [1762/1750/1754]), and which, in its most sophisticated form, has been defended by Rawls (2000) and Scanlon (1982, 2000 [1998]), deviates from Lockean political voluntarism and contractarianism right from the start: with a presumption in favor of political organization. According to contractualists, the anarchic freedom of the state of nature that Lockeans and contractarians value is unavailable to us. In their view, individuals have a “natural duty of justice”, i.e. a duty to support and comply with a just framework of legal and political institutions that exists and applies to them as well as to create such a framework if it has not yet been established, because such a framework or, to be more precise, the coercion it exerts is a necessary precondition of autonomous human agency. ⁹ It follows that the state’s coercive exercise of political power should not be removed but instead be justified. According to contractualists, a state is legitimate if it is regulated by a public conception of justice, i.e. a set of principles of justice that all individuals who are subject to its coercive effects could reasonably accept as a basis for informed, unforced general agreement. ¹⁰ Call this the All Coerced Principle.

What we have here, then, is a very different solution to the boundary problem. Unlike the Lockean political voluntarist and contractarian acceptance-based solutions to the boundary problem, which both legitimize the people by identifying the set of individuals who accept each other as members of the same people, contractualism offers an impact-based solution to the boundary problem: it identifies the set of individuals who are subject to state coercion.

Although state coercion is unavoidable according to contractualists, it nevertheless requires justification. The crucial question, however, is what kind of justification is owed to individuals who are subjected to state coercion. Like contractarianism, contractualism is generally understood as endorsing a liberal strategy of hypothetical justification. As far as Scanlonian contractualism is concerned, this is correct. ¹¹ When it comes to Rawlsian contractualism, however, this is incorrect. Despite appearances to the contrary—in his seminal A Theory of Justice, after all, Rawls invites us to engage in the thought experiment of the original position—I shall defend the admittedly counter-intuitive view that Rawlsian contractualism is fundamentally democratic.

Begin by noticing that the original position is part of Rawls’s broader approach to moral justification: the method of reflective equilibrium. Rawls endorses a wide variant of this coherence model of moral justification, according to which (a) moral principles are justified if they cohere with (b) considered moral judgments, (c) background theories, and (d) the original position. According to Rawls (2001: 400–401), one should think of justice as fairness “as working up into idealised conceptions certain fundamental intuitive ideas” that “reflect ideals implicit or latent in the public political culture of a democratic society”. These idealized conceptions are the so-called background theories, and they contain the fundamental ideas of society as a fair system of cooperation, of citizens as free and equal persons, of a well-ordered society (WOS), and of a public conception of justice. In Rawls’s view (2001: 400–401), “[t]he original position is a device of representation that models the force (…) of the essential elements of these fundamental intuitive ideas as identified by the reasons for principles of justice that we accept on due reflection”.

The “on due reflection” clause is crucial here. Starting from the fundamental intuitive ideas implicit in the public political culture of a democratic society, we construct the original position in such a way that it models these ideas, and subsequently see whether the moral principles agreed to by the parties in the original position match our considered judgments on reflection. Essentially, the search for reflective equilibrium involves a dialectical process. It is “[b]y going back and forth, sometimes altering the conditions of the contractual circumstances, at others withdrawing our judgments and conforming them to a principle”, that Rawls thinks we shall eventually find “a description of the initial situation that both expresses reasonable conditions and yields principles which match our considered judgments duly pruned and adjusted” (Rawls, 2000: 18. See also Rawls, 1996: 384). According to Rawls (2000: 15), the original position is “the appropriate initial status quo which insures that the fundamental agreements reached in it are fair”. What makes it so, then, is that the principles it generates are in wide reflective equilibrium with the considered moral judgments of the members of a democratic society and a set of background theories implicit in the public political culture of such a society.

One background theory is of particular importance here, namely that of a WOS. The members of a WOS understand themselves as free and equal moral persons. This means that they have a dual moral capacity: individuals are rational because they seek the most efficient means to achieve their ends, and reasonable because they wish to regulate these ends (as well as the means used to realize these ends) by means of principles that can be justified by reasons which all can reasonably accept (Rawls, 1996: 16, 2001: 386). The members of a WOS share a fundamental collective pre-commitment to, or highest-order desire for, this form of moral justification, which presupposes a particular kind of motivational disposition: a common “desire for free and uncoerced agreement” (Freeman, 2007: 33, 36–39; Rawls, 2001: 312, 395, 2000: 452–453, 491). If this line of reasoning is sound, however, then it has an important implication: there is no need to invoke a counterfactual initial choice situation, such as the original position. According to “justice as fairness”, principles are valid if they can be the object of an agreement made by rational agents under circumstances in which they are subject to certain appropriate conditions expressing “the reasonable”. What this means is that these agents “must be situated reasonably, that is, fairly or symmetrically, with no one having superior bargaining advantages over the rest” (Rawls, 1996: 52–53). Given that the veil of ignorance makes it impossible for the parties in the original position to tailor principles to their own advantage, for instance, by coercing others into an agreement by means of threats of force or violence, the original position ensures that the fundamental agreements reached in it are fair, i.e. free and non-coercive.

In Rawls’s view, however, the original position is a device of representation that models the force of certain considered moral judgments held by the citizens of a democratic society and a set of fundamental intuitive ideas implicit in the public political culture of their society. If these citizens themselves have a highest-order desire to propose and act from principles that they can reasonably accept as a basis for informed, unforced general agreement, then the original position becomes redundant. In that case, their shared moral motivation itself already suffices to guarantee that agreements are free and non-coercive.

Although Rawls conceives of the original position as a special feature of his broader reflective equilibrium-based approach to moral justification, it follows, from what I have said so far, that the original position fails to add any justificatory force to the principles—i.e. justificatory force that the principles do not already have in a reflective equilibrium consisting only of a coherently ordered triple of sets of (a) moral principles, (b) considered moral judgments, and (c) background theories (including the contractualist account of moral motivation). Consequently, we have to apply “Occam’s razor” and remove the original position as a superfluous element from Rawls’s reflective equilibrium approach.

Still, one might wonder whether the quest for reflective equilibrium is democratic. Indeed, a criticism frequently leveled against Rawls by deliberative democratic theorists, such as Benhabib (1992) and Habermas (1990: 66–68, 1995: 116–119), is that his approach to moral justification is monological (as opposed to dialogical). Supposedly, this has the serious fault of leaving it ultimately to “the philosopher” as an expert and not to the citizens of an ongoing society to determine what conception of justice can be the object of reflective equilibrium. In similar vein, Saward (2009) has argued that Rawls’s search for reflective equilibrium is monological, although not in an expertocratic sense. It is, he claims (2009: 117–118), “a solitary thought process, engaged in (ideally) by all citizens on their own, conducted in order to reach a specific conception of an initial situation that accords with our considered convictions about the content of justice”. Despite their different views concerning the kind of monological thought process that is involved in reflective equilibrium, Benhabib, Habermas, and Saward agree that the struggle for reflective equilibrium is a private as opposed to a public matter, and so cannot be understood as offering a democratic strategy of justification.

However, I do not think that this is an adequate characterization of the way in which the method of reflective equilibrium features in Rawlsian contractualism. According to Rawls (1995: 140), the point of view from which reflective equilibrium should be reached is that of civil society, which includes all citizens of a democratic society. Furthermore, he writes (1995: 141 n16, emphasis added):

In such a society, not only is there a public point of view from which all citizens can adjudicate their claims of political justice, but also this point of view is mutually recognized as affirmed by them all in full reflective equilibrium. This equilibrium is fully intersubjective: that is, each citizen has taken into account the reasoning and arguments of every other citizen.

Contrary to what Rawls’s critics contend, then, the quest for reflective equilibrium is not monological; the kind of reflective equilibrium involved in Rawlsian contractualism is neither the result of a solitary thought process performed by the philosopher as an expert, nor is it the result of a series of unilateral thought processes performed by the citizens of a democratic society who independently happen to affirm the same public conception of justice. Instead, the quest for reflective equilibrium is dialogical; it is ideally the result of an actual deliberation among the members of an ongoing society. This, in turn, means that Rawlsian contractualism is fundamentally democratic for it demands a specific deliberative democratic discourse: reflective equilibrium as a public method of moral justification.

If this is correct, then it follows that Rawlsian contractualism offers a democratic solution to the boundary problem. In a Rawlsian contractualist view, a society is well-ordered if it is based on a public conception of justice, i.e. if it is regulated by a set of principles of justice that individuals can reasonably accept as the object of a wide reflective equilibrium. The scope of this democratic justification, i.e. the set of individuals who jointly constitute the people and as such should be able to understand themselves as the authors of the rules regulating their political relationships, is limited by the All Coerced Principle to all individuals upon whom a scheme of political and legal institutions coercively imposes laws. ¹²

Practical implications

I now want to spell out some practical implications of the three social contract theories discussed. Let us start with the Lockean and contractarian acceptance-based solutions to the boundary problem. Thus far I have argued that both approaches are able to explain how “in the beginning” free and equal individuals can legitimately set themselves apart from other individuals in the state of nature, and join together in a political community in which they live under a common political authority. However, the boundary problem is not solved by focusing exclusively on the original constitution of the people. On the contrary, the problem will inevitably resurface once a political association has been founded (e.g. because of migration flows). So if an acceptance-based approach is to be not only of historical interest but of practical relevance to us nowadays as well, then it should also be capable of providing a solution to those specific articulations of the boundary problem that arise in an already existing world of states. So the Lockean and contractarian approaches should be able to address the problem of reconstituting the people, i.e. the question of how the people can continue to exist over time as a legitimately bounded association. Is this possible?

Notice, to begin with, that the Lockean and contractarian state of nature constitutes a heuristic device that models individuals’ natural freedom and helps us to think through its implications for our political bonds. It describes individuals prior to their having acted in ways which politically bind them to one another. Thus understood, the state of nature does not refer to a pre-political period in the history of mankind. Rather, it is a relational concept: individuals are in the state of nature with respect to one another if they have not performed any acts which create political relationships between them (cf. Simmons, 1979: 64, 1993: 16). Thus understood the state of nature has a distinctive advantage when applied to the boundary problem: it is a helpful device for analyzing the problem of the original constitution of the people as well as that of reconstituting the people.

According to Lockean political voluntarism, individuals who do not want to acquire politically bonds should be left as they were in the liberty of the state of nature. It follows that a state has the moral duty to give any dissenting individuals the option of becoming non-citizen residents who have neither political obligations nor rights (cf. Klosko, 1991; Simmons, 1993: 242–244). Furthermore, a state has the moral duty to allow individuals to emigrate. If the (re)constitution of the people is legitimate to the extent that it is consistent with the None Rejected Principle, then an individual can, if he rejects any and all of the others as members of a people of which he is part, unilaterally decide to exclude himself from the people. Since, in an already existing world of states, assuming the status of non-citizen resident or performing the act of “voting with one’s feet” offer two possible ways of expressing one’s refusal to become (or remain) politically bound to others, it follows that the None Rejected Principle implies a negative claim right both to non-citizen residency and to emigration. Finally, the None Rejected Principle supports a political practice in which individuals have a positive liberty right to immigration. An individual who applies for citizenship in another state is in the state of nature with respect to the citizens of that state. If these citizens accept the would-be immigrant, they bind themselves to one another through mutual acts of consent and so reconstitute the people.

Contractarianism considers political relationships morally legitimate if they are based on an agreement among all individuals for whom it is mutually advantageous to bind themselves to one another. If there is no (longer any) potential for mutually beneficial cooperation, then individuals remain in, or return to, the state of nature. However, although the Lockean state of nature is a moral state regulated by the “Law of Nature”, morality is silent in the contractarian state of nature. Consequently, individuals have no claim right to personal self-determination, as exemplified by the rights to non-citizen residency and emigration, but instead only a liberty right to everything. Essentially, then, the status of individuals for whom it is not beneficial to engage in cooperative interaction is that of “outlaws”. Of course, they can apply for citizenship in another state. In that case, the would-be immigrant and the members of the state in which he would like to settle are in the state of nature with respect to each other. As a result, they will start bargaining in order to determine whether it is rational for them to cooperate. If cooperation is rational, the people are reconstituted in such a way that they now include the would-be immigrant.

Now consider, by contrast, Rawlsian contractualism. According to this impact-based approach, the people consist of the set of individuals upon whom a coercive framework of political and legal institutions imposes laws. The proper political unit within which democracy is practiced depends on the coercive scope of laws. Since some laws claim universal jurisdiction whereas others only regulate actions of citizens within a state (and thus claim local jurisdiction), Rawlsian contractualism supports a system of multi-level governance, according to which the democratic decision-making process should sometimes be limited to a state’s domestic sphere and at other times be extended to the global sphere (and thus requires global democratic institutions).

Applied to the issue of immigration, we can see that the All Coerced Principle demands that the relevant constituency for deciding on the issue of immigration should be global in scope, for a state’s immigration laws claim universal jurisdiction and thus subject all individuals, both insiders and outsiders, to coercion. What makes immigration laws coercive is that they impose a conditional threat of the form “Do not take course of action x; if you do, y will be imposed on you” (i.e. “Do not attempt to enter this country without explicit permission of its public officials; if you do, you will be apprehended, put in detention, and evicted”). Such a coercive threat invades an individual’s autonomy because, as Abizadeh argues (2008: 59), “it threatens to interfere with the setting and pursuit of her own ends by using her body for purposes that are not her own”. Moreover, the threat has universal scope because it is communicated to all foreigners. ¹³

What this shows is how different the Lockean and contractarian solutions to the boundary problem are from that of Rawlsian contractualism. The latter differs from the former two, both in terms of its approach (which is impact-based instead of acceptance-based) and outcome. According to the Lockean and contractarian approaches, a legitimate regime of border control is one that is the outcome of a democratic decision-making process in which both the would-be immigrant and the citizens of the state he wants to enter have participatory standing. By contrast, Rawlsian contractualism demands that the demos be global. ¹⁴

In defense of Rawlsian contractualism

Each of the social contract theories discussed offers a democratic solution to the boundary problem. This raises the question as to which solution should be preferred. I have defined democracy as centering around the ideas of a collectively self-governing people, and collective self-government aimed at the realization of the people’s common good. I shall now reflect on three aspects of this idea of democracy: (1) the relationship between collective self-government and the common good, (2) the conditions of legitimate collective self-government, and (3) the value of individual autonomy. By defining each of these aspects of the idea of democracy in increasingly more demanding terms, I shall argue that it becomes possible to explain that the Rawlsian contractualist approach to the boundary problem should ultimately be preferred.

Consider, first, the relationship between collective self-government and the common good. According to the democratic logic, in which the doctrine of popular sovereignty takes center stage, the common good tracks the will of the people. Consequently, the notions of collective self-government and the common good are intrinsically connected. That is to say, it is by means of a collective decision-making procedure that the people jointly determine the nature of their common good. A Lockean political voluntarist framework, however, operates mostly from a liberal logic in which the doctrine of the rule of law takes center stage. Indeed, the democratic moment in the Lockean political voluntarist account of the constitution of the people is rather limited. Here individuals should be able to understand themselves as the authors of their political relationships but not of the laws regulating their political ties. The common good does not track the will of the people but instead natural law. Although individuals can agree to become politically bound, natural law specifies the nature of their political relationship: if they decide to bind themselves to one another as a people, however, then the fundamental natural law demands that all individuals involved (conditionally) give up their natural right to execute the natural law, and transfer it to a common political authority to which they subject themselves as the supreme interpreter, judge, and enforcer of natural law. Although this political authority can be of a democratic kind in a Lockean political voluntarist framework, natural law is still sovereign; a government, even when it is democratic, has to act in accordance with natural law (Locke, 2000 [1698] II, 135). In a Lockean framework, then, the relationship between collective self-government and the common good becomes disentangled: the people rule themselves in a way consistent with natural law, which provides the ultimate standard for the general regulation of political society.

Unlike Lockean political voluntarism, however, contractarianism and Rawlsian contractualism do preserve the intrinsic relationship between collective self-government and the common good. Contractarianism does so because it offers a democratic bargaining procedure that aims at establishing and regulating forms of cooperation through which common goods are created and sustained that are instrumental to the realization of the interests of utility-maximizing individuals. Rawlsian contractualism does so because it offers a democratic method of public justification—namely, the reflective equilibrium approach—that aims at articulating the common good of a public conception of justice for the basic structure of a democratic society, the substance of which is responsive to the highest-order desire of free and equal individuals to interact with others in ways that all can reasonably accept. So, if we define democracy as collective self-government aimed at the realization of a self-imposed common good (as opposed to a common good externally imposed by natural law), then Lockean political voluntarism is insufficiently democratic and should, consequently, be rejected in favor of its contractarian and Rawlsian contractualist rivals.

Consider next the second aspect of the idea of democracy: the conditions of legitimate collective self-government. Central to democracy is the value of individual autonomy. In order to preserve it, individuals should be able to understand themselves as the authors of the coercive laws to which they are subjected. Consequently, the outcomes of collective decision-making procedures are legitimate if they track the will of those individuals who (ought to) participate in it. This democratic requirement is reflected by the contractarian idea of a rational compromise and the Rawlsian contractualist idea of a reasonable consensus.

It might be argued, however, that the idea of democracy requires more than mere agreement. After all, the fact that a collective decision is based on agreement does not necessarily guarantee its non-coerciveness. An agreement can still be exploitative. The nature of an exploitative agreement is such that it is coercively extracted by the stronger from the weaker party. The coercion involved in this kind of agreement does not stem from threats of violence or force, but rather from the unequal bargaining position of the parties to the contract itself and the unfair advantage taken of that inequality by one of the parties in contract. Although all parties signed the contract, their agreement is nonetheless democratically illegitimate because it violates the autonomy of the weaker party.

Here an important difference between contractarian and Rawlsian contractualist views comes to surface. In a Rawlsian contractualist view, individuals have a highest-order desire to regulate the objects of their desires as well as the ways in which these are pursued on the basis of principles that no one can reasonably reject as a basis for informed, unforced general agreement. It follows that individuals, thus motivated, are neither willing to coerce others into an agreement by means of threats of force or violence nor prepared to exploit others by taking unfair advantage of their relatively weak bargaining position in contract. This means that Rawlsian contractualism unconditionally tracks the democratic value of individual autonomy.

Contractarianism, however, tracks this democratic value in a significantly different way. Indeed, it is commonplace among philosophers to argue that contractarianism is compatible with exploitation of the weak by the strong—the notorious “slave contract” comes to mind. However, whether the weak will always be exploited by the strong in a contractarian world depends on the nature of individuals’ preferences. In a world inhabited by mutually unconcerned or non-tuistic utility-maximizers (who either take no interest in the interests of any others at all or take no interest in the interests of those with whom they interact) an unequal bargaining position makes it rational for the stronger party to exploit the position of the weaker party in contract if he can. If, however, individuals take an interest in the interests of those with whom they interact, then they will not enslave others. This scenario is possible if we make the motivational assumption that an individual seeks to maximize the satisfaction of his coherent preferences, whatever the objects of these preferences are. Consequently, contractarians can claim that it may be rational for utility-maximizing individuals to grant moral standing to the weak, and that this may mean that the weak are agreeable as parties to a non-exploitative social contract, the conclusion of which makes them full (as opposed to second-class) members of a people (cf. Morris, 1991b). I deliberately say “may”, however, because the weak will only acquire moral standing if their interests are the object of the preferences of the strong. This is a theoretical possibility but not necessarily reality. Consequently, contractarianism contingently tracks the democratic value of individual autonomy.

If we are committed to the democratic view that non-coercive agreement is a condition of legitimate collective self-government, then it seems that Rawlsian contractualism should be preferred to contractarianism. Contractarians could reply that it is not a weakness of their theory that it only contingently tracks the democratic value of individual autonomy. Indeed, they could argue that morality is impossible if individuals are not equals by nature and if they do not take any interest in the interests of others (see Verschoor (2014) for a more extensive discussion of this issue). But although this element of contingency is essential to a coherent contractarian view, it does not follow that contractarianism is necessarily a correct (moral) view. In order to see this, it is worthwhile to reflect on the nature of contractarianism’s maximizing conception of practical rationality.

Contractarianism’s aim is to derive morality from the non-moral premises of rational choice. It follows that the contractarian account of practical reason itself cannot be characterized in normative terms. In that case, the entire contractarian enterprise would be compromised from the outset. If individuals ought to act instrumentally rational, then morality is no longer derived from non-moral premises. If the contractarian enterprise is to make any sense, then it seems that the maximizing conception of practical rationality on which it is built should be taken to refer to an underlying empirical, or more precisely, philosophical anthropological account of the nature of human beings and their interpersonal relationships. In that case, individuals are conceived of as beings who are maximizers of coherent and agent-relative preference only and who do interact with others solely on the basis of economically rational considerations. However, this way of characterizing the contractarian conception of practical reason is also problematic because the underlying philosophical anthropology is empirically false. An individual’s capacity for practical reason is not limited to means-ends reasoning. In fact, contractualists are right in claiming that individuals’ capacity for practical reason contains both a rational and reasonable element. However, if we acknowledge the fact that individuals have a dual capacity for practical reason, then, as Freeman correctly claims (2007: 44), we have no choice but to ask ourselves the following practical, indeed moral, question: “What kind of person can we most closely identify with and do we finally want to come to be?”. Do we conceive of ourselves as individuals who merely wish to maximize the satisfaction of their given desires (without regard to their substance or origin)? Or, alternatively, do we conceive of ourselves as individuals who want to regulate the objects of their desires and their pursuit in ways that others can reasonably accept? This is a moral question because of the significant implications it has for the way in which political relationships between individuals should be regulated. The answer to this question determines, for instance, the substantive issue concerning the legitimate demarcation of the people.

If we want to be as autonomous as possible, and now I come to the third aspect of the idea of democracy, we should adopt the Rawlsian contractualist disposition, the reason being that an individual acting from that disposition is more autonomous than an individual acting from the contractarian disposition. Unlike the latter, the former individual does not merely want his life and interpersonal relationships to be regulated by whatever principles that “happen” to be in accordance with his contingently held desires. Rather, he wants to control his desires by regulating the objects of his desires as well as their pursuit in ways that other similarly motivated individuals can reasonably accept. In other words, the scope of his autonomy is wider: he is rational and reasonable.

In conclusion, then, I have developed three social contract theoretical solutions to the boundary problem that are democratic in character. Now, (1) if we accept the democratic idea that the relationship between collective self-government and the common good should be intrinsic; (2) if we do not wish to live in a society in which legitimate, i.e. non-coercive, self-government is a mere theoretical possibility but (always) actuality; and, finally, (3) if we want to be fully autonomous moral agents; (4) then we are committed to Rawlsian contractualism as well as its solution to the boundary problem. It follows that Rawlsian contractualism offers the best way to close the aforementioned gap in social contract theory. Moreover, this conclusion also has a broader implication for democratic theory. After all, if Rawlsian contractualism is morally superior and more deeply democratic than its Lockean and contractarian rivals, then this suggests that democratic theorists, in so far as they wish to offer a genuinely democratic solution to the boundary problem, should focus on impact-based (as opposed to acceptance-based) approaches. And despite appearances to the contrary—indeed, against the virtually universally accepted view—I have demonstrated that Rawlsian contractualism offers a coherent democratic framework for addressing the boundary problem.