The Democratic Legitimacy of Border Coercion: Freedom of Association, Territorial Dominion, and Self-Defence

Transcription

1 The Democratic Legitimacy of Border Coercion: Freedom of Association, Territorial Dominion, and Self-Defence Arash Abizadeh Department of Political Science McGill University 855 Sherbrooke Street West Montreal, Quebec, Canada H3A 2T7 Tel: x Fax: arash.abizadeh at mcgill.ca For valuable comments, I am grateful to Chris Bertram, Thomas Christiano, Pablo Gilabert, Kinch Hoekstra, Daniel Kofman, Colleen Murphy, Cara Nine, Jonathan Quong, Annie Stilz, Kit Wellman, Leif Wenar, Lea Ypi, and participants at the Workshop of the Territory and Justice Network in Novi Vinodolski, Croatia, October 8-11, 2009; the University of California, Berkeley, February 16, 2010; the Territorial Rights Workshop, Queen s University, June 4 5, 2010; the American Political Science Association, Washington, DC, Sept. 2 5, 2010; the Territorial Rights Conference, Princeton University Center for Human Values, April 1, 2011; the American Philosophical Association, Pacific Meeting, San Diego, April 20-23, 2011; and the Workshop on Immigration, Integration and Membership, University of Zurich, Feb , 2012.

2 Abstract: According to the democratic borders thesis, a state s regime of border control is democratically legitimate only if the laws governing it result from political processes in which both citizens and foreigners can participate. This is because, to be democratically legitimate, the (coercive) exercise of political power must be democratically justified to all subject to it; and both citizens and foreigners are subject to a polity s regime of border control. I defend this thesis against three objections. First, it might be argued that legitimate states have the right to close their borders thanks to a collective right of freedom of association, grounded in selfdetermination. I argue that such an argument, while grounding a negative claim-right against coercively imposed association, fails to establish a liberty-right to coerce others to prevent unwelcome association. Moreover, it misconstrues the proper collective subject of a right of selfdetermination: not only the persons whom state agents recognize as members, but all persons subject to the coercive exercise of political power. Second, one might object that citizens enjoy rights of dominion over the territory of their state, and may thus unilaterally refuse entry to foreigners. I respond that just as property laws, to be democratically legitimate, require democratic justification to those subject to them, so too must democratically legitimate border laws. Finally, one may object that the coercive exercise of political power may sometimes be legitimate even if not democratically legitimate. I concede this, but argue that the lack of democratic legitimacy imposes dynamic duties to enable democratic legitimization in the future.

3 1 In Democratic Theory and Border Coercion, 1 I argued that for a state s regime of border control to be democratically legitimate, the laws governing the border regime must result from political processes in which not only citizens, but also foreigners possess democratic rights of participation. The argument for this apparently radical democratic borders thesis is rather simple: the argument s first, normative premise articulates a widely accepted democratic principle of legitimacy, while its second premise takes note of an undeniable fact. First, for political power to be democratically legitimate, it must be democratically justified to all those over whom it is exercised, i.e., at least to all persons subject to state coercion. 2 Second, a bounded polity s regime of border control subjects both citizens and foreigners to the coercive exercise of political power. Therefore, the democratic justification of a state s regime of border control is owed to both citizens and foreigners, i.e., not merely to those whom its boundaries mark as members. 3 Notice that the democratic borders thesis does not defend open borders or a 1 Arash Abizadeh, "Democratic Theory and Border Coercion: No Right to Unilaterally Control Your Own Borders," Political Theory 35, 1 (2008): I say at least because other (non-coercive) modes of being subject to political power may also trigger the demand for democratic justification. 3 The democratic borders thesis does not assume that all those affected by the exercise of political power are owed a right of democratic say; it assumes, rather, that all those subjected to state coercion are presumptively owed such a right. David Miller has recently argued that, contrary to all appearances, states regimes of border control do not in fact subject foreigners to the coercive exercise of political power. David Miller, "Why Immigration Controls are not Coercive: A Reply to Arash Abizadeh," Political Theory 38, 1 (2010): ; see also David

4 2 right to freedom of movement per se: it is not a substantive moral thesis about how open or closed a liberal democratic polity s entry policy should be. It advances, rather, a procedural, political claim about who has the right to participate in deciding the entry policy. Here I wish to defend and further develop this account by way of considering three potential objections to it, which respectively appeal to a right of freedom of association, of territorial dominion, and of self-defence. According to the first objection, democratic citizens have the right unilaterally to close their state s territorial and/or civic borders to foreigners thanks to a collective right freely to choose with whom to associate. Just as individuals have the right to reject marriage proposals without granting their suitors any democratic say over their choice, democratic citizens may unilaterally refuse entry and membership to foreigners without granting them any democratic say. According to the second objection, democratic citizens collectively enjoy rights of dominion over the territory of their state, analogous to the property rights individuals enjoy over their own homes. Just as individuals may refuse to permit others into their home, democratic citizens may unilaterally refuse entry to foreigners. Finally, according to the third objection, the assumption that a polity must democratically justify its coercive exercise of political power to foreigners has absurd implications: for example, that before waging war against foreign enemies intent on destroying them, democratic polities must grant those enemies Miller, "Democracy's Domain," Philosophy & Public Affairs 37, 3 (2009): I indicate why I find Miller s efforts unsuccessful in Arash Abizadeh, "Democratic Legitimacy and State Coercion: A Reply to David Miller," Political Theory 38, 1 (2010): For a definition of being subject to coercion, see my reply to Miller as well as the appendix to Abizadeh, "Democratic Theory and Border Coercion."

5 3 a democratic say over their conduct of war; or, more simply, that democratic polities must grant participatory rights, over their regime of border control, to anti-democratic foreigners intent on entering and destroying the polity s democratic institutions. But just as individuals have a right to self-defence, without being obliged to deliberate with their attackers about whether to defend themselves, states are permitted to subject foreigners to coercion to block their entry or membership, without granting them a democratic say. After clarifying the argument for the democratic borders thesis by briefly summarizing why the democratic principle of selfdetermination supports, rather than undermines, the thesis, I take up each of these three objections in turn. The Democratic Borders Thesis and Self-Determination The thesis that democratic legitimacy requires granting foreigners a right of democratic say over a polity s regime of border control may initially seem at odds with the core democratic ideal of self-determination. After all, many have thought that the ideal of self-determination implies that differentiated democratic polities each have the moral liberty-right to control their own boundaries unilaterally. 4 Yet that thought is fundamentally mistaken: it begs the question of who 4 Michael Walzer goes so far as to assert that the right unilaterally to regulate admission and exclusion to one s own polity suggests the deepest meaning of self-determination. Michael Walzer, Spheres of Justice: A Defence of Pluralism and Equality (Oxford: Blackwell, 1983), p. 62. I use the terms liberty-right, claim-right, immunity, and power in their Hohfeldian senses. Wesley Newcomb Hohfeld, Fundamental Legal Conceptions, as Applied in Judicial Reasoning,

6 4 the relevant self of self-determination is. When a regime of border control is at issue, one cannot take it for granted that the collective self possessing a right of democratic selfdetermination is comprised of the collectivity of individuals contained within a pre-existing (politically or legally defined) boundary. This is because, at its most fundamental level, the democratic ideal of self-determination entails a demand for democratic legitimacy and justification: the demand that the people who are subject to political power themselves collectively determine how such power is exercised over them. 5 New ed. (New Haven, CT: Yale University Press, 1964). For discussion and elaboration, see Leif Wenar, "The Nature of Rights," Philosophy & Public Affairs 33, 3 (2005): On the notion that the democratic ideal of self-determination is fundamentally equivalent to the ideal of collective self-rule or self-government, entailing rights of political participation on terms consistent with the equality and freedom of persons, see Carol C. Gould, "Self-Determination beyond Sovereignty: Relating Transnational Democracy to Local Autonomy," Journal of Social Philosophy 37, 1 (2006): 44-60; Brian Mello, "Recasting the Right to Self-Determination: Group Rights and Political Participation," Social Theory and Practice 30, 2 (2004): I should emphasize that the ideal I am treating here is a democratic and rather than nationalist ideal of self-determination, which is why I speak of the people s, and not the nation s, right of selfdetermination. In twentieth-century international law, by contrast, self-determination was traditionally taken to mean national self-determination: in the Wilsonian period, it referred to the self-determination of ethno-national groups; in the post-war period, it referred to the selfdetermination of the majority population occupying an administratively bounded territorial jurisdiction (akin to what Anthony Smith has called a territorial or civic nation). See Margaret

7 5 More precisely, I take the terms regulating the exercise of political power to be democratically legitimate to the extent, and only to the extent, that they receive a democratic justification. Such terms receive a democratic justification to the extent, and only to the extent, that either (1) they are the outcome of political procedures and processes in which (a) all those subject to the exercise of political power have the opportunity to participate (b) on terms consistent with respect for their equality and freedom; or (2) they are the constitutive conditions of these democratic procedures and processes of justification; and (3) they do not constitutively undermine or negate these political procedures and processes. Thus legislation depriving a religious minority political rights of political participation, even if itself the outcome of a democratically legitimate procedure, would be democratically illegitimate, since it would violate condition (3), while a law establishing such rights, even if itself not the outcome of democratic procedure, would be democratically legitimate, since it meets condition (2). 6 Note that I reject Moore, "The Territorial Dimension of Self-Determination," in National Self-Determination and Secession, ed. Margaret Moore (Oxford: Oxford University Press, 1998); Gould, "Self- Determination beyond Sovereignty."; Anthony D. Smith, National Identity (London: University of Nevada Press, 1991). 6 On the need to recognize that even a procedural principle of legitimacy, like the democratic one, must incorporate the constitutive conditions of the required procedures, see Stephen Holmes, "Precommitment and the Paradox of Democracy," in Constitutionalism and Democracy, ed. Jon Elster and Rune Slagstad (Cambridge: Cambridge University Press, 1988); Joshua Cohen, "Deliberation and Democratic Legitimacy," in The Good Polity: Normative Analysis of the State, ed. Alan Hamlin and Philip Pettit (Oxford: Basil Blackwell, 1989).

8 6 the seemingly parallel disjunctive condition that (2 ) such terms are necessary causally to promote democratic procedures, and the conjunctive condition that (3 ) such terms do not causally erode democratic political institutions in the future (e.g. legislation that destroys the sociological preconditions of stable or well-functioning democratic institutions). I do not incorporate such causal conditions because they conflate what democratic legitimacy consists in with the empirical conditions under which realizing democratic legitimacy is feasible. If it is argued that, under the circumstances, excluding some class of persons from political participation is instrumentally necessary for promoting solidarity amongst participants, and that participant solidarity in turn is necessary for the stable functioning of democratic procedures over time, the justification for exclusion is not a democratic one in my sense. The argument does not establish the democratic legitimacy of the exclusion; it claims merely that, under these circumstances, democratic legitimacy cannot be stably maintained over time, or, more precisely, that only partial democratic legitimacy is feasible under the circumstances. 7 7 It seems entirely meaningful to say that democratically legitimate political outcomes sometimes end up causing, perhaps unforeseeably, the erosion of democratic institutions. Moreover, the exclusion of (2 ) and (3 ) reflects the fact that democratic legitimacy is not the only value relevant even to democratic theory (stability, in particular, seems to me a distinct value). Thus I agree with Miller that the ideal of collective self-determination stands at the heart of democratic theory ; I am also sympathetic to his view that a constituent part of this ideal is a principle of legitimacy according to which people themselves must be able to see political outcomes as legitimate; but whether or not he is right to think that solidarity, interpersonal trust, agreement on ethical principles, or sympathetic identification are indispensable for people to

9 7 On this account, then, if a colonial power subjects a set of persons overseas to its rule without granting them political rights of participation, then it violates their right to selfdetermination and, so, compromises the democratic legitimacy of its rule. Even should the colonial power grant such persons political rights of participation, but only after having forcibly annexed their territory and imposed coercive jurisdiction over them, it would still compromise the democratic legitimacy of its rule: the coercive act of territorial annexation itself proceeded without being democratically justified to them on terms consistent with their equality and freedom. (I take it that territorial annexation does not count as an instance of enacting the constitutive legal conditions of democratic procedures.) Similarly, if a state denies democratic rights to a certain class of individuals over whom it exercises political power, such as women or blacks, it curtails the people s capacity for self-determination and, so, compromises its democratic legitimacy. In other words, the collective self or demos relevant to the democratic ideal of self-determination is not simply whomever the state happens already (politically or legally) to recognize as rights-bearing citizens. When the South African state, under Apartheid, denied full citizenship and political rights to non-whites over whom it nonetheless exercised political power, it curtailed South Africans right to democratic self-determination. To assert that Apartheid was justified as an instance of South African whites self-determination simply misconstrues who, from a democratic point of view, the rightful bearer of a right to selfdetermination is: such a right, understood as a democratic right, does not license the unilateral see outcomes as legitimate (Miller, "Democracy's Domain," ), none of these conditions are constitutive of democratic legitimacy. They are putative causal conditions.

10 8 exercise of political power over persons whom state agents (or legally recognized members) declare to be outside the demos. 8 While respect for the equality of persons by political institutions requires the treatment of each as an equal, that is, equal respect for each in determining how to distribute goods or opportunities, it does not require the equal treatment of each, that is, the same distribution of goods and opportunities to each person. 9 Political equality, in other words, constrains the kind of reasons appropriate for justifying the institutional distribution of political rights, but does not necessarily require that each person have the same political rights in all contexts. This means that democratic legitimacy cannot be equated with simple majoritarianism. In some circumstances, being politically treated as an equal may require that those with higher stakes have a greater 8 Cf. Dahl s criticism of Schumpeter s view, according to which democratic legitimacy does not constitutively include any substantive principle of who ought to be granted rights of participation, on the grounds that his view absurdly implies that democracy cannot be distinguished from a closed-party dictatorship, as long as the party is internally democratic. Robert Dahl, "Procedural Democracy," in Philosophy, Politics, & Society (5th Series), ed. Peter Laslett and James Fishkin (Oxford: Blackwell, 1979); Joseph A. Schumpeter, Capitalism, Socialism and Democracy, 3rd ed. (New York: Harper Torchbooks, 1976), pp On the distinction between equal treatment and treatment as an equal, see Ronald Dworkin, Taking Rights Seriously: New Impression with a Reply to Critics, 2nd ed. (London: Duckworth, 1978), pp. 227, 273.

11 9 political say over some range of outcomes. 10 (Therefore, the kinds of participatory rights that the democratic borders thesis demands may be different between citizens and foreigners, or between different kinds of foreigners, depending on the stakes involved for each.) In other circumstances, where there is an entrenched minority who would be permanently dominated within simple majoritarian political institutions, political equality may require differentiated political jurisdictions, that is, constituted boundaries (whether municipal, provincial, or state) that identify some as members (citizens) with participatory rights within the jurisdiction, and others as nonmembers without the same such rights. This is what I call the minority-protection justification for political boundaries. 11 But given politically differentiated jurisdictions, democratic political equality may also require that political rights of participation over some areas of jurisdiction be detached from the formal status of citizenship, and participatory rights be granted to non- 10 See Charles R. Beitz, Political Equality: An Essay in Democratic Theory (Princeton, NJ: Princeton University Press, 1989), pp , 24. As Beitz notes, political equality may in some circumstances actually require procedural inequalities. 11 See Abizadeh, "Democratic Theory and Border Coercion."

12 10 citizens. 12 According to the democratic borders thesis, this is precisely what is required in the case of regimes of border control. 13 The moral core of the democratic ideal of self-determination, then, just is the right to have the exercise of political power be democratically legitimate. To ensure that the people can exercise self-determination over their political affairs, the democratic ideal of self-determination, via the democratic principle of legitimacy corresponding to it, thus itself demands that the people enjoy political rights of participation on terms consistent with their equality and freedom. Which people? The very people subject to the coercive exercise of political power. This 12 In other words, the package of rights that modern states have traditionally bundled together and tied exclusively to the status of citizenship must be disaggregated. For an argument for disaggregating political rights of participation from the formal status of citizenship, Sarah Song, "Democracy and Noncitizen Voting Rights," Citizenship Studies 13, 6 (2009): An exception to this may arise in the minority-protection case: when a group would become an entrenched and dominated minority unless it could unilaterally control and close its own borders (because it would be swamped by newcomers from a different group). This is because individuals of an entrenched minority would not be able to participate under conditions that respect their equality and freedom. See Abizadeh, "Democratic Theory and Border Coercion," p. 53. I am sympathetic, however, to the argument, advanced by Michael Dummett, On Immigration and Refugees (London: Routledge, 2001), pp , that high levels of immigration might submerge a cultural group in this way only when a political society is colonized or subjected to the rule of oppressive invaders and never simply because of open borders.

13 11 is precisely the first premise of the argument for the democratic borders thesis. The second premise articulates a peculiar conceptual feature of political boundaries of membership: that constituting them politically is always an instance of exercising political power over not only those whom the boundary marks as members, but also those whom it marks as non-members. 14 Freedom of Association Even if the value of democratic legitimacy speaks against the unilateral control of one s own legally defined borders, it might be thought that the central liberal value of freedom of association speaks decisively in its favour. Thus Christopher Wellman has recently argued that states have a right to restrict immigration unilaterally thanks to a collective right of freedom of association (hereafter, right of association). Wellman s argument proceeds in four steps. He begins, first, by claiming that the human interest in freely choosing one s associates is weighty enough to ground a right of association, and that this right includes a right to reject a potential association and (often) a right to disassociate. This is the right that protects, for example, the 14 For a more extensive discussion of the democratic principle of self-determination, and why it supports, rather than undermines, the democratic borders thesis, see Abizadeh, "Democratic Theory and Border Coercion." For why borders do not constitute an exception to the requirement of democratic legitimacy (e.g. as suggested by Mathias Risse, "'Imagine There's No Countries:' A Reply to John Lennon," Harvard John F. Kennedy School of Government, Faculty Research Working Paper No. RWP (2008), p. 21, note 18; available from see Sofia Näsström, "The Legitimacy of the People," Political Theory 35, 5 (2007):

14 12 individual s choice of marital partner and the associates with whom he or she practices his or her religion. Second, not only individuals but also groups collectively may enjoy such a right. Third, the groups possessing this right include political communities organized as legitimate states. And, finally, a political community s right of association implies the state s right unilaterally to restrict immigration: just as an individual has the right to determine whom (if anyone) he or she would like to marry, a group of fellow citizens has a right to determine whom (if anyone) it would like to invite into its political community. 15 Wellman is aware, of course, of the potentially profound moral difference between individuals and groups; he nonetheless shores up his second claim by appeal to the commonly held conviction according to which many groups, and not merely individuals, have a presumptive right to freedom of association. The reason why some groups Wellman s examples are the Boy Scouts of America and the Augusta National Golf Club enjoy such a presumptive right is the weight of the interests at stake. And here is Wellman s argument for his third claim, which imputes a collective right of association to legitimate states in particular: 15 Christopher Heath Wellman, "Immigration and Freedom of Association," Ethics 119, October (2008): , pp Wellman does not, it seems to me, adequately distinguish between the territorial boundaries regulating movement and the civic boundaries regulating membership. For ease of exposition, I maintain this ambiguity in my discussion of Wellman, by using the term immigration to refer both to territorial migration into a state s territory, and to the acquisition of a legal status en route to naturalization or citizenship. As Sarah Fine rightly points out, Wellman s concern is primarily with the acquisition of membership status. Sarah Fine, "Freedom of Association Is Not the Answer," Ethics 120, 2:

15 13 denying such a right leads to unpalatable implications. This collective right is said to be the only way to explain, for example, why a legitimate state has the unilateral right to decide whether to join another state or regional association; without such a collective right, one cannot explain the wrongness of the USA forcibly annexing Canada, or the EU coercing Slovenia to join it. And if legitimate states possess a collective right of association, Wellman concludes, then they must have the right unilaterally to restrict immigration. 16 It is worth asking, however, to what extent the analogies with marriage, private clubs, religious associations, or even state annexation, illuminate the case of immigration. The analogy with marriage in particular deserves attention, since, as one may gather from its frequent appearance in Wellman s exposition, it packs much of the rhetorical punch of his argument. To examine the merits of the argument, we need first to spell out, in much greater detail than Wellman himself does, the content of the right in question. Although we often speak of a right of association, it is in fact a composite bundle of several rights, concerning at least two aspects of human association: first, the fact of contact or interaction in physical space and time and, second, the set of mutual special obligations acquired as a result of interacting with, or standing in a particular relation to, others. (Thus to be married typically involves both interacting with one s spouse for example, in a shared home and acquiring a set of special obligations towards him or her for example, of care or financial support.) The typical function of a right of association is to regulate the use of coercion over these two aspects of association. As the marriage example indicates, the right of association, whether conceived as a moral right or a legal right, normally always includes (a) a negative claim-right against being coerced to interact, or to refrain from 16 Wellman, "Immigration and Freedom of Association," pp , 115.

16 14 interacting, with potential associates; and (b) an immunity against acquiring special obligations towards coercively imposed associates. These components of the right of association protect against coercively imposed or prevented associations. Beyond this coercion-regulating function, the right of association often also has a project-enabling function: to enable individuals to choose, on the basis of their own subjective reasons or personal projects, 17 which associative interactions and obligations to undertake. As the marriage example suggests, the right of association often includes (c) a power to acquire special obligations towards, and rights against, potential associates of one s own choice; and (d) a (paired or bi-lateral) liberty-right (not) to interact with, and/or (not) to take on special obligations towards, potential associates. 18 The power enables individuals to pursue personal projects involving associates of their own choosing, while the liberty-right protects such pursuits by permitting a range of associative choices in spite of other (moral or legal) considerations that might otherwise forbid or require some choice (e.g. despite the fact that one s choice of association may not be one that maximizes the good) I use the term personal in the sense given by Bernard Williams, Moral Luck: Philosophical Papers (Cambridge: Cambridge University Press, 1981), chapter A paired or bi-lateral liberty-right is what Wenar calls a paired privilege. Wenar, "The Nature of Rights," pp Another way to put this is to say that the liberty-right (d) provides a kind of exclusionary reason that blocks other reasons that might otherwise determine that one s association with another is required or forbidden.

17 15 (That regulating coercion is the primary function of the right of association is reflected in the fact that, construed as a moral right, it normally always includes the coercion-regulating rights (a) and (b), but does not always include the project-enabling ones (c) and (d). This is because the grounds of the former rights precisely because they regulate coercion are more robust than those of the latter: the latter are more easily defeasible. 20 There are many cases in which a moral right of association implies rights (a) and (b), for example, but not (d). One may have a moral duty to refrain from joining racist organizations, or a moral duty to carry out a promise of marriage even if, thanks to a right of association, no one has the moral right coercively to enforce those duties. Likewise, the fact that it is wrong for the EU to coerce Slovenia to join it does not itself imply that Slovenians have no bare moral duty to join. A state may or may not, under the circumstances, have a moral duty to join certain regional associations, but whether others have a right to coerce it to join is a distinct moral question. It is one thing to have a bare moral duty (not) to do something, and quite another for someone else to have the moral right coercively to hold one to it which is why the right of association, construed as a moral right, does not always include the liberty-right (d). Nor does it always include (c): the fact (if it is true) that others do not have the moral right coercively to prevent one from joining a racist organization is perfectly compatible with the fact that, having joined, one may still be unable to acquire genuinely moral obligations towards it. It is true that construed as a legal right, by contrast, the right of association always includes a paired liberty-right (d) to associate or refrain from associating, i.e., it always implies the absence of a legal obligation (not) to 20 In fact I have listed these rights in what I take to be their order of robustness, from greatest to least.

18 16 associate. But this is simply because what distinguishes legal liberty-rights from moral ones is that the former, unlike the latter, by definition govern the use of coercion; thus, to say that one has a legal liberty-right to associate or not is to say that others, including state agents, may not legally coerce one to associate or not. A legal right of association therefore always includes a paired liberty-right (d) to associate or not simply because, unlike the case of moral rights, such a liberty-right is equivalent to the legal rights (a) and (b) against being coerced in one s choice of association. The discrepancy between a moral and legal right of association reflects the fact that regulating coercion is the primary function of the right of association: the right creates a presumption that whatever moral duties one may have to join or refrain from joining in association with others are not coercible duties. 21 ) 21 Thus, to take up one of Wellman s examples: one may consistently hold that the Augusta National Golf Club has (i) a (presumptive) moral right to freedom of association, such that others do not have the moral right coercively to force them to admit female members; that they consequently ought to enjoy (ii) a legal right of freedom of association, such that no one may legally coerce them to admit women, i.e., that they have a legal liberty-right to refrain from admitting women; but that they nonetheless (iii) have a moral duty, themselves, to admit women, i.e., that they lack a moral liberty-right to refuse women admittance. The moral right of association here simply does not include a moral liberty-right: we can admit a right of association while also acknowledging moral duties. The same point holds about the right to free speech: I may have a moral claim-right that others not coercively prevent me from saying nasty racist things, but nonetheless have a moral duty not to say them. When it comes to legal rights,

19 17 Of course it is not simply coercion that determines whether one is in fact able to associate or avoid associating with others. Sometimes one may end up interacting with, and even acquiring obligations towards, persons one prefers to avoid precisely because of the absence of coercion. Consider, for example, three possible scenarios involving a religious group whose members deem anyone not joining them in worship, and in the practice of their religion, as ritually impure or spiritually corrosive. First, the group may avoid association with non-worshippers altogether, for example by having non-worshippers coercively prevented from entering the group s neighbourhood. Alternatively, the group (or the state on their behalf) may coercively compel others to join them in worship and religious practice. Or, finally, absent the use of coercion on their behalf, the group s members may find friendly non-worshipping neighbours interacting with them in public; indeed, they may even find themselves acquiring mutual neighbourly obligations towards these neighbours. 22 The first scenario is of coercively prevented association, the second of coerced association, and the third of (unwelcome yet nonetheless) uncoerced association. A further question, therefore, is whether one s right of association includes another set of coercion-regulating rights, namely, (e) a liberty-right to use coercion against others to prevent others from associating with one, and/or (f) a liberty-right to use coercion to compel others to by contrast, I might not have a legal duty not to say them; that is, my legal right to free speech might include the legal liberty-right to say them. 22 Conversely, friendly agnostics in a predominantly religious society may find that whenever they take up residence in a neighbourhood, others flee, non-coercively thwarting the association they hoped for.

20 18 associate 23 (or alternatively, in each case (e) or (f), a positive claim-right that a third party use coercion on one s behalf). Notice, then, that while rights (a), (b), (e), and (f) all regulate coercion, the former two regulate coercion against the right-bearer, while the latter two regulate coercion by the right-bearer. It should therefore be clear that any putative rights (e) and (f) would be directly in tension with others rights (a) and (b), and by consequence curtail these others ability to exercise their right (c) (and, if it exists, (d)). Thus, under the Soviet regime, coercive measures to compel individuals to join the Communist Party curtailed their rights (a) and (b); in the era of desegregation in the USA, the state s use of coercion to compel whites to associate with blacks in shared public schools might also be thought to limit whites right (a) and (b); conversely, under South African Apartheid, state coercion on behalf of whites to prevent blacks from entering white areas, or from joining white clubs, curtailed blacks right (a) and, as a consequence, curtailed their ability to exercise right (c). The relevant point here is not to settle the question of whether the right of association includes (e) or (f) and, if so, how to balance them with others rights (a)-(d). The relevant point is, rather, that distinguishing between these component rights shows how Wellman s argument for his conclusion that legitimate states have the right unilaterally and coercively to restrict immigration simply conflates rights (a) and (b), which regulate coercion against the rightholder, on the one hand, with right (e), which regulates coercion by the right-holder, on the other. 23 The liberty-right (f) refers to the use of coercion in forcing interaction, such as the coercively imposed worship common in medieval and early-modern Europe. With respect to the second aspect of association, namely the set of mutual, special obligations, one would need to speak of a power, rather than liberty-right, to impose such obligations.

21 19 Recall that Wellman s conclusion is supposed to follow directly from his third claim, that legitimate states possess a right of association. And recall that the argument for this third claim is that only a right of association could explain the wrongness of the USA annexing Canada, or the EU forcing Slovenia to join. And indeed the wrongness of annexation is straightforwardly explained by the rights (a) and (b) against coercively imposed association. But this is completely tendentious to Wellman s conclusion. At best, Wellman s argument for his third premise establishes the wrongness of coercion against legitimate states in matters of association: it establishes that states have the rights of association (a) and (b). What it does not establish is that the state s right of association includes the (e) liberty-right to coerce others. Yet this latter right is what his conclusion asserts: the claim that legitimate states may coercively prevent immigration does not concern coercion against the state, but coercion by the state, against potentially unwelcome immigrants. 24 And unfortunately for Wellman s overall argument, neither right (a) nor (b) nor, indeed, (c) or (d) entails (e) the right to use coercion against others to prevent unwelcome association. Are there independent reasons for thinking that a state s right of association includes (e) a liberty-right to subject others to coercion? It is true that in many contexts the right of association 24 Critics of a state s right unilaterally to restrict immigration, as it sees fit, are not ipso facto defending the right of others to deploy coercion against the state (to force it to open its borders, for example); they are disputing the state s (e) moral liberty-right unilaterally to deploy coercion against others to keep its borders closed. They are attributing bare (not necessarily coercible) moral duties to the state in its use of coercion for the purpose of avoiding unwelcome associations.

22 20 does include a liberty-right to coerce others in order to prevent unwelcome association; indeed, the individual s right of association often includes a positive claim-right against the state requiring it coercively to prevent others from interacting with one against one s wishes. To bring out this feature of the right of association is presumably one of the functions of Wellman s marriage analogy: an individual typically possesses a positive claim-right against the state obligating it coercively to prevent others from coercively imposing marriage on her. Nor is such a right always restricted to preventing coerced association; it is frequently also geared towards preventing uncoerced yet unwelcome association. Consider again, for example, the case of the exclusivist religious group. While few liberal states would recognize that it is their duty coercively to prevent non-worshippers from entering a neighbourhood populated by such a group, they often do recognize the right of worshippers to acquire private property in which to worship, along with the state s duty to deploy coercion on behalf of the group, if need be, to exclude unwelcome outsiders. But it is precisely here that the analogy between marriage, private clubs, and religious associations, on the one hand, and immigration, on the other, breaks down. It is true that one s ability to choose one s associates in a marriage, private club, or religious association is often legitimately protected by the state, coercively if need be both against coerced and against uncoerced but unwelcome interaction. Yet there is a fundamental disanalogy between the legal regime in a liberal democratic state that coercively regulates individuals choices in marriage, or private clubs choice of members, on the one hand, and the regime of border control that coercively regulates a foreigner s admission into its state territory, or as a member of the political community, on the other. In the former cases, the use of state coercion against some persons, on behalf of the individual or group in possession of right (e), is governed by a legal system that is

23 21 presumptively justified democratically to both the bearer of the right and those subject to coercion as a result. Persons are subject to democratically legitimate state coercion, in other words, on terms presumptively justified democratically to both those on behalf of whom it is exercised and those against whom it is exercised. But in the case of a unilaterally imposed regime of border control, the state subjects foreigners to coercion to protect the state s (or citizens ) putative right of association, without the regime that regulates coercion having been democratically justified to the foreigners subject to it. 25 And this returns us to the heart of the matter. It is true that, within the domestic context of a liberal democratic state, individuals have the right to choose their marriage partner without justifying their choice to rejected suitors; it is also true that individuals normally have a claimright obligating the state to protect these choices, coercively if need be, from unwelcome intruders. But the state s use of coercion here is democratically legitimate only to the extent that the laws that recognize the individuals claim-right and that regulate the state s use of coercion have been democratically justified to both the bearers of the right and those consequently subjected to coercion. 26 The coercive laws that protect individuals from having to justify their 25 It is true, of course, that contemporary states frequently negotiate their bilateral regimes of border control, and often within the confines of international law. These constraints are not nothing, but they obviously fail to meet the conditions of democratic justification (requiring (a) participation (b) on terms respecting equality/freedom). 26 For example, property laws that, however much justified on independent grounds, are enacted via a political process that excludes some persons subject to them, are not democratically

24 22 choice in marriage to rejected suitors are themselves supposed to have been democratically justified to both the protected individual and the rejected suitors. My thesis about borders simply makes the parallel point about a state s coercive restrictions on immigration: for the state s regime of border control to be democratically legitimate, it must be democratically justified to all those subject to it. On behalf of Wellman one might respond here that the right of association and democratic legitimacy are grounded in distinct and potentially competing values and that, where they conflict, a state s right of association simply defeats the requirement of democratic legitimacy. But this line of argument is not straightforwardly viable in the case at hand, because the right of association possessed by a legitimate state (or its people) does not compete with the requirement of democratic legitimacy, but is ultimately grounded in and conditioned by that requirement. Recall that the state s right of association, as Wellman conceives it, is not a right held by an individual but is a collective right held by the state itself (or by citizens collectively as a group). 27 Recall further that Wellman s argument for attributing such a right to legitimate states legitimate even if they enjoy a distinct, non-democratic kind of legitimacy (see paper s final section). 27 Wellman seems to equivocate on who the rightful bearer of the right of association is: the state, or its citizens collectively. The conflation of the categories of state and the people finds its venerable pedigree in Hobbes s momentous and ideologically motivated collapse of the latter into the former; the conflation is also of course part of the ideological scaffolding of all modern democratic states. But the distinction remains of capital importance for the dispute, in the history of political thought, between the fundamentally opposed doctrines of state sovereignty,

25 23 in particular is that not doing so would lead to unpalatable implications, such as an inability to explain the wrongness of forced annexation. Yet if we dig a little deeper for the grounds explaining the wrongness of forced annexation, and the consequent attribution of a collective right of association to the state (or its citizenry), those grounds turn out to be the more fundamental ideal of self-determination: the state s (or citizenry s) right of association is merely a constituent part of a people s democratic right to self-determination. The justification for imputing a right of association to a state (or its citizenry) in particular i.e., the justification for the third claim in Wellman s argument is that possessing such a right is necessary for the collective exercise of self-determination. As Wellman puts it, freedom of association is...an integral part of the self-determination to which some groups (including legitimate states) are entitled. 28 Yet as we have already seen, at its most fundamental level, the democratic right of self-determination itself amounts to a demand for democratic legitimacy: the demand that the people subject to political power themselves collectively determine how such power is exercised over them. The right of association cannot defeat the requirement of democratic legitimacy because the former is merely a constituent part of the democratic right of self-determination, championed by Hobbes and his heirs, and popular sovereignty, championed by democrats or republicans. The argument about self-determination that I provide obviously requires that the distinction between the state and the people be kept clearly in view. On the competing doctrines of state sovereignty and popular sovereignty, see Quentin Skinner, "From the State of Princes to the Person of the State," in Visions of Politics, Volume II: Renaissance Virtues (Cambridge: Cambridge University Press, 2002). 28 Wellman, "Immigration and Freedom of Association," p. 115.

26 24 whose most fundamental demand is that of democratic legitimacy. As a result, when the component rights of association are those licensing coercion against others (such as rights (e) or (f)), then their appropriate bearer is not comprised only of those licensed to exercise coercive political power over others: the collective self that is the proper bearer of any such right includes those subject to coercive political power. Thus, even though democratic legitimacy and selfdetermination may also require a collective right of association, and not merely individual rights of political participation, the former does not compete with, and cannot defeat, the latter requirement. When what is at stake is the regime of border control, the rightful bearer of a collective right of self-determination is not restricted to those whom the politically constituted and enforced border picks out as members only A potential exception to this, once again, is the case of minority protection (see note 13); and the right of association can be used to express this exception. Recall my claim that democratic legitimacy requires that political power be exercised according to democratically justified terms, and that democratic justification in turn requires (a) allowing the participation of all those subject to the coercive exercise of political power (b) on terms respecting their equality and freedom. The question then is how to analyze circumstances, if they exist, in which the (a) participation condition is in conflict or tension with the (b) equality/freedom condition (in which the right of association is grounded) circumstances in which, for example, allowing for the participation of all those subject to the coercive exercise of political power would ipso facto make respecting some participants equality and freedom impossible. And a circumstance that potentially fits this description is the one faced by individuals forming an entrenched minority; this is precisely why such circumstances can yield a minority-protection justification for political boundaries. Now

27 25 This consideration also blocks another potential reply on Wellman s behalf. One might argue that precisely because the right of association is a constituent condition intrinsic to democratic procedures and processes of justification that without recognizing a right of association, political procedures could not allow for truly democratic participation the democratic legitimacy of such a right does not depend on being the outcome of a democratic procedure but derives, rather, from being a constitutive condition of democracy. 30 Recall, however, that what is in question in particular are rights (e) or (f). Even if these rights are rightly considered to be constitutive of democratic procedures, the question at stake here is who which collectivity is their rightful collective bearer. To repeat: it is not an intrinsic condition of democracy that the rightful bearer of such rights be restricted to those whom the politically constituted and enforced border picks out as members only. Quite the contrary. insofar as the right of association is grounded in a right of self-determination qua demand for democratic legitimacy, it is arguably grounded in the (b) equality/freedom condition: a right of association is necessary to ensure that individuals political participation is on terms respecting their equality and freedom. As such, part of the justification for a collective right of association would be to serve minority protection. In that case, it is not that a right of association competes with democratic legitimacy; rather, once again, the point is that circumstances justifying minority protection may thereby justify a collective right of association restricting the scope of the participation that democratic legitimacy requires. 30 For an argument like this about the right of association, see Thomas Christiano, The Constitution of Equality: Democratic Authority and its Limits (Oxford: Oxford University Press, 2008).