Canadian Journal of Philosophy Breaking up: An Essay on Secession Author(s): David Gauthier Source: Canadian Journal of Philosophy, Vol. 24, No. 3 (Sep., 1994), pp. 357-371 Published by: Canadian Journal of Philosophy Stable URL: http://www.jstor.org/stable/40231872 . Accessed: 10/06/2014 18:49 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact
[email protected]. . Canadian Journal of Philosophy is collaborating with JSTOR to digitize, preserve and extend access to Canadian Journal of Philosophy. http://www.jstor.org This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/action/showPublisher?publisherCode=cjp http://www.jstor.org/stable/40231872?origin=JSTOR-pdf http://www.jstor.org/page/info/about/policies/terms.jsp http://www.jstor.org/page/info/about/policies/terms.jsp CANADIAN JOURNAL OF PHILOSOPHY 357 Volume 24, Number 3, September 1994, pp. 357 - 372 Breaking Up: An Essay on Secession DAVID GAUTHIER University of Pittsburgh Pittsburgh, PA 15260 USA I Current discussion of the normative issues surrounding secession is both helped and hindered by the existence of but one philosophic treatment of these issues sufficiently systematic and comprehensive to qualify as a theory of secession - Allen Buchanan's.1 He provides the unique focal point, and so simplifies the task of those who seek to begin from the present state of the art. But in providing the unique focal point, Buchanan complicates the task of those who view, or think they view, secession rather differently than he does. He defends 'a moral right to secede' but a very qualified right, focusing on state-perpetrated injustice, the pres- ervation of group culture and, in extreme cases, the literal survival of group members (152-3). And Buchanan further insists that where pres- ervation of group culture is at stake, 'Neither the state nor any third party 1 See Allen Buchanan, Secession (Boulder, CO: Westview Press 1991). This paper was first presented at a symposium of the American Philosophical Association, Pacific Division, with Buchanan as lead symposiast. Buchanan notes a very few philosophic discussions that touch on secession; perhaps the most important is by Harry Beran, who defends a right to secede in The Consent Theory of Political Obligation (London: Croom Helm 1987), 37-42. One of the referees for this journal drew my attention to what, so far as I know, is the only book-length work on secession other than Buchanan's: Lee C. Buchheit, Secession: The Legitimacy of Self-Determination (New Haven: Yale University Press 1978). Buchheit focuses on the claim to secede within the framework of international law and includes several case studies, but deals only briefly with moral and philosophic issues. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 358 David Gauthier has a valid claim to the seceding territory' - or such a claim is waived so that 'secession in order to preserve a culture is permissible if both parties consent to it' - a condition that he thinks may come to apply to the situation of Quebec. I want to sketch a rather more permissive view of secession. And I want to undermine the idea that territorial claims by existing states are significant blocks to secession. But I have no structure in any way comparable to Buchanan's within which I can defend this more permis- sive view. All that I can hope to do is to sketch; if the sketch has some appeal then it might seem promising to attempt to construct the theory that will be conspicuously lacking from my account. In developing my position, I shall have, as surely any Canadian at this time must have, one particular application always in mind. For Canada's existence is threatened by a secessionist movement. Not all secessionist movements constitute such threats. The success of Baltic secessionism was not what doomed the Soviet Union. Should the secessionist Catho- lics of Northern Ireland succeed in their aims (and I shall argue later that however permissive my view of secession may be, it does not support their cause), the United Kingdom would continue largely as before. But if Quebec secedes, what will remain will be a country very different from the one that I know and love. And so I have constantly to ask myself: 'What are the implications of my sketch of the rights and wrongs of secession for my native land?' In discussing secession, I propose largely to ignore all of those cases in which the existing political order is illegitimate. To be sure, one wants an account of political legitimacy to decide which these cases are, but I shall not produce one. Instead, I shall simply share what I hope will seem some plausible judgments of cases. The Baltic states were not legiti- mately incorporated into the Soviet Union; the effective basis of their incorporation was force and their appearance of consent was fraud. Putting Ireland to one side for later consideration, the United Kingdom, conceived as including England, Scotland, and Wales, is a legitimate political order; the thirty-three states of the United States of America in 1860 comprised a legitimate political order; the ten provinces of Canada today comprise a legitimate political order.2 Each of these three has or had a valid claim on the obedience of the inhabitants of its several parts - a claim that I should want to relate to the recognition it gives or gave to the rights of participation of those inhabitants. But this claim, which 2 I leave also to one side the existence of groups who fall outside the legitimate order - certainly Blacks in the ante-bellum United States, conceivably the first peoples - Amerindians and Inuit - today in Canada. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 359 is at the core of legitimacy, does not in my view undermine the possibility of making a case for secession. Buchanan compares political dissolution to divorce (7, and frequently thereafter). Divorce is not annulment; divorce does not call into question the legitimacy of the marriage that it ends. I want to say that metaphori- cally, Lithuania sought annulment, whereas South Carolina sought di- vorce. But even within the framework of divorce, I want to distinguish situations in which the political union has in effect broken down and lost its legitimacy - a fact that is then recognized in secession - from situations in which the political union loses its legitimacy only through secession. And of course these distinctions - between annulment and divorce, breakdown and dissolution - become matters of degree and interpretation when applied. For example, how shall we interpret seces- sion in Croatia and Slovenia: annulment of a union that was coercively formed after the first World War and was never legitimate, or divorce from a union that had acquired but then lost legitimacy? In restricting my attention to legitimate political orders, I want to exclude situations central to Buchanan's account, both those that involve state-perpetrated injustice, at least of a serious kind, and those in which the existing state is unable or unwilling to assure the literal survival of the prospective secessionists. I take significant injustice to deprive the state of legitimacy in relation to those treated unjustly; hence I deny that Blacks in ante-bellum America had any obligation to accept the political and legal order. (Compulsion is of course another matter.) And if literal survival is at stake, then, as Hobbes (for example) acknowledges, per- sons are free to make whatever arrangements for themselves they can, whatever their prior political obligations may have been.3 So I can agree with Buchanan that Jews in pre-WW II Europe would have been entitled to secede from existing states to create their own if that had been a feasible option for them (66). II In what follows, I shall suppose that neither injustice nor literal survival is involved. I shall suppose a secessionist movement that may have cultural or other aims, but that for whatever reason wants to redraw the political map by creating a new and distinct sovereign community - or less frequently, perhaps, by separating from one political community 3 Thomas Hobbes, Leviathan (London: Andrew Crooke 1651), 114 This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 360 David Gauthier and joining another existing one, as I interpret the case of secessionist Catholics in Northern Ireland. To discuss secession in these contexts, I shall appeal to what might be thought of as weak rights of association and non-association. I shall treat these as strictly individual rights; I have no place for group rights in my account. I begin from the simple idea of interaction. Familiar arguments exhibit the costs of unconstrained interaction in which each person directly seeks to realize his or her concerns; the deep contractarian idea is that each person is morally bound by terms of interaction that it would have been rational for her to agree to in a suitably characterized ex ante situation. It is then natural to suppose that these terms will favor social institutions and practices that provide for an extensive although not unlimited freedom of contract and association. Although each person will be constrained in the unilateral costs she may impose on her fellows in the pursuit of her concerns, decisions to associate or not, to contract or not, will be left to be decided by free consent and so, given rationality, mutual benefit, where such decisions are not associated with significant negative externalities. What I have just said is advanced as a supposition, not a conclusive argument. A contractarian justification of political association, in terms of a rational ex ante agreement, need not in principle justify a contractual society - a society in which the freedom to relate to others by contract is central. For present purposes I am simply assuming that there is such a justification; I do not deny the size and significance of the assumption. Consent applies directly to decisions to contract or not, given the bilateral character of the activity, but it must apply less directly to decisions to associate or not, given the multilateral character of many forms of association. I find it helpful to think of each person as having a weak right to enter into and continue political association with those with whom she wishes to associate and who wish to associate with her, and to avoid or exit from association with those with whom she wishes not to associate. By a 'weak right' I intend one whose exercise must be coordinated with that of other persons in such a way that, other things equal, as many persons as possible will find themselves in mutually desirable association. Of course, other things often are not equal, and I shall have at least to indicate some of the relevant inequalities. Let me underline the mutuality that is involved in the weak right of association. I do not have a right to enter into or continue in association with those who do not want to associate with me, however much I may think such association desirable. There are of course circumstances in which persons may be held to political associations with those whom they would avoid, but not because those whom they would avoid want them as political associates. Just as secession may be compared to di- vorce, so political association may be compared to marriage. I may have This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 361 the right to marry the woman of my choice who also chooses me, but not the woman of my choice who rejects me. It may be helpful to apply the idea of a weak right of association directly to the Irish situation. We may reasonably suppose that a signifi- cant majority of the Catholic inhabitants of Northern Ireland (the Six Counties) do not want to continue in political association with the United Kingdom, and do want to enter political association with the Republic of Ireland,4 a majority of whose inhabitants reciprocally want to enter into association with the North. But we must also suppose that a sub- stantial majority of the Protestant inhabitants of Northern Ireland do want to continue in political association with the UK, and want not to enter political association with the Irish Republic.5 Given that there are significantly more Protestants in Northern Ireland than Catholics (the ratio is about 5:3), if the Six Counties were to be united with the Republic, more persons would find themselves in association with those with whom they did not want to be in association than at present. And this would be contrary to the prima facie requirements of the weak right - the number of persons whose right of association was effectively exer- cised would diminish. It may well be the case that if a referendum were held in the whole of Ireland, a substantial majority would support unification. But the desire for unification has in itself no weight insofar as it fails to be mutual. Consider the situation prior to the establishment of the then Irish Free State (i.e. what is now the Republic) in 1922. If a referendum had been held in what was then the United Kingdom - i.e. in the British Isles as a whole - almost certainly a majority would have opposed Irish seces- sion. But the desire to maintain the British Isles as a single political association had no normative weight insofar as it failed to be mutual - that Englishmen wanted an association that included the Irish was in itself no reason for continuing such association, given that the Irish did not want association with the English. However, in accepting the Free State, the United Kingdom nevertheless recognized that in the largest part of Ulster, most of the inhabitants preferred continued association with the UK than entry into association with the Free State, and so partitioned Ireland in such a way that more Irishmen were associated 4 But perhaps not an overwhelming majority. A study reported in the British press in the autumn of 1992 suggested only 53% of Northern Irish Catholics preferred a united Ireland, while 38% preferred association with the United Kingdom. 5 Indeed an overwhelming majority. The study referred to in n.4 suggested that 92% of Northern Irish Protestants preferred association with the United Kingdom. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 362 David Gauthier with those who were willing to associate with them, than would have occurred without partition. To be sure, one might argue that the boundary between the Free State and Northern Ireland might have been drawn to accommodate the exercise of right of association by even more persons. Indeed, someone might argue that in principle the boundary could be drawn - by permitting enclaves of one state surrounded by another (compare Llivia, or Campione d'ltalia)6 - so that almost everyone could be accommo- dated in the association of his or her choice. But in the drawing of boundaries there are factors other than the right of association to con- sider, such as the provision of public services and the character of local economic relations. I merely gesture at these matters - my only claim is that some partition of Ireland seems clearly justified by the right of association, and that the actual partition is better justified than no partition. I have dwelt on Ireland partly because I think that the reasons that lead me to reject the claims of the Catholic secessionists require me to accept the secession of Quebec, should it be supported by a significant majority of Quebecois. For even if the overwhelming majority of other Canadians opposed secession, our desire to continue association with Quebec would have no more weight than the desire of the inhabitants of the Irish Republic to incorporate the Six Counties. For in both cases, this would be a desire to impose association on persons the majority of whom did not want it. If the Quebecois come or have come to wish to be, in a fully sovereign sense, 'Maitres chez Nous/ then the secession of Quebec would lead to more persons being in mutually acceptable association, in accordance with the requirements of the weak right. Ill But the weak right is certainly not the only consideration relevant to the justification of secession. By returning to the Irish situation I can intro- duce another fundamentally important factor. Suppose that the majority of the inhabitants of Great Britain, fed up with Irish squabbling, the IRA planting bombs in railway stations, blowing up military and police posts, Protestant extremists gunning down elderly Catholics in betting shops, and so on and on, decide that they no longer want to be in political association with the Six Counties. Then it may seem that the Northern Irish Protestants could no longer plead a weak right of association as 6 Llivia is a Spanish enclave surrounded by France, Campione d'ltalia an Italian enclave surrounded by Switzerland. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 363 justification for the continued incorporation of the Six Counties in the United Kingdom. For they would no longer be seeking to maintain asso- ciation with those who reciprocally wanted to be associated with them. As I have been interpreting the Irish situation, the Protestant majority of the Six Counties has been acting within the rights of its individual members in seeking to maintain association with the United Kingdom, and some members of the Catholic minority have wrongfully been seeking to deny members of the majority this exercise of their rights. (That the Catholics have themselves been wrongfully and unjustly treated in certain ways is no doubt true, but that mistreatment does not, at least in my view, justify them in trying to impose their desire to unite with the Irish Republic on a majority that prefers to be part of the United Kingdom.) To suppose that the inhabitants of Great Britain would be, in these circumstances, entitled to cut off the Six Counties is to suppose them entitled to terminate an existing political association because of the costs of defending it against an initially wrongful attempt to disrupt it. In effect the British would be saying to the Northern Irish, 'Your right to be associated with us depends, not only on your wishing such an association, but also on our wishing it. But because of the costs of maintaining that right, we no longer wish it. And so you no longer have a right to association/ The British would be claiming that they were not acquiescing in the wrongful attempt of the Catholics to deny the Protestants exercise of their right of association, but merely termi- nating the conditions for the existence of the right that the Protestants were seeking to exercise. Were the British to do this, the old label 'Perfidious Albion' would be amply justified. A state that would refuse to defend one of its hitherto accepted and welcome parts against an attempt at dismember- ment because the costs of defense made the part no longer welcome, would clearly not be fulfilling an essential part of the role for which political association exists. Thus I deny that the inhabitants of Great Britain have the right of majority secession, as it were - the right to cut off the Six Counties from the United Kingdom. They may not plead the cost of defending the association against an attempt to disrupt it as adequate normative ground for withdrawing their own consent to it. Underlying my argument here is a more general principle. As I see it, the revised Lockean proviso, in the form I present it in Morals by Agree- ment - the requirement that one not better oneself by worsening another - is the core moral requirement.7 Or, one might say, morality requires 7 David Gauthier, Morals by Agreement (Oxford: Clarendon Press 1986), 201-5 This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 364 David Gauthier that one justify unilaterally imposing costs on or worsening the position of another, and admissible justifications must appeal either to an under- lying expectation that particular unilateral impositions lead overall to mutual bettering, or to the necessity of the imposition in order to avoid worsening one's own position. The key problem in applying the proviso is to determine the background conditions against which bettering and worsening are to be measured. Now I take existing normative relation- ships, including those involved in political association, to be among these conditions. Thus given an existing political association between members of two groups, A and B, on terms that do not incorporate significant injustice, the members of A would violate the proviso were they to secede or exclude the members of B in a way that increased their benefits (or lessened their costs) by reducing the benefits of the members of B (or increasing their costs). But this needs qualification: in particular, the intrinsic benefit derived from the desire to be part of the particular association itself is not to be included in the calculation. (Thus if A and B divorce at A's behest, B may not appeal to the loss of satisfaction of her desire to be married to A as a loss of benefit entitling her to compensation to avoid a proviso violation.) Note that appeal to existing normative relationships must assume the justifiability of those relationships. That Lithuania's secession from the then Soviet Union might benefit Lithuanians at the expense of Russians and others provided no basis for alleging a proviso violation, given the coercive incorporation of Lithuania into the Soviet Union. But the proviso does have clear application to the prospect of the British seeking to exclude Northern Ireland from the United Kingdom. For it is evident that such exclusion would be intended to benefit the British - no more IRA bombings in London, no more defense forces in Northern Ireland - while making it more difficult for the Protestant Irish to protect themselves against the use of force by Catholics in the endeavor to create a single Irish state. And the inhabitants of Great Britain clearly may not complain that their existing political association with Northern Ireland has been coercively imposed on them, or that its terms discriminate against them. I began with an appeal to a weak right of association that might seem to provide a much broader justification for secession than Buchanan is willing to accept. But now I have invoked a proviso that may seem to narrow that justification even more than he might wish. For will not almost any proposed secession benefit the seceding party and impose costs on the other, so that in the absence of injustice in the existing association that would nullify appeal to the proviso, the secession will be unjustifiable? In the case of Quebec, is it not the case that the rest of Canada would be left poorer - culturally if not materially - were This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 365 Quebec to secede, so that the proviso could be invoked as a normative ground against it? To answer this, I must argue that the scope of the proviso should be limited in a non-arbitrary, non-ad hoc way. I shall propose two limita- tions. The first relates directly to what I have just said about the Canadian situation. When different groups unite in political association, the mem- bers of each may expect to benefit from characteristics peculiar to and distinctive of the others, in ways that they would not benefit either from uniting with equal numbers of persons but with characteristics similar to their own, or from the existence of the others outside of the association. Secession necessarily ends those benefits. But I want to insist that their loss cannot justify any claim either against secession or for compensa- tion. These benefits are not necessary to achieve the fundamental pur- poses of political association. They are best thought of as akin to the goods achieved through friendship, which depend on the truly volun- tary and affectionate character of the relationship. To suppose that when friendship ceases, the parties may have claims against each other for the withdrawal of the goods that arose because of their friendly relationship, would be to misunderstand that relationship and its distinctive goods. I take the same to be true in the case of political association. We may see a parallel between the withdrawal of personal affection and that of political affection. And so if the Quebecois lose political affection for us, their fellow Canadians, they owe us no compensation and violate no obligation if we no longer enjoy the goods of that affection. But, it may be urged, the secession of Quebec will deprive the rest of Canada of other goods. Perhaps there need be no direct material cost, but there may be, and there will indeed be a political cost in that the Canada that remains will have a diminished capacity to advance its interests, particularly in relation to that nation which is at once our historic enemy and closest friend.8 Here then I come to the second limitation on the scope of the proviso in relation to secession. I want to distinguish the productive conse- quences of secession from the distributive. A political community may achieve certain economies of scale - not only in the production of material goods, but also as I have suggested in advancing the interests of its members vis-a-vis the members of other political communities - that are lost if it divides. This loss is a productive consequence of 8 The second stanza of /rThe Maple Leaf/ which in my youth rivalled 'O Canada' as the national song, begins 'At Queenston Heights and Lundy's Lane / Our brave fathers fought and died' - it was the United States they fought. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 366 David Gauthier secession. A political community also realizes a certain distribution of goods among its parts, and this distribution may be altered if it divides. Now I want to place the latter, but not the former, within the scope of the proviso. In the absence of existing injustice, the members of a group are not entitled to secede from an existing political community in a way that redistributes the goods achieved in that community in their favor, and at the expense of the other members. On the other hand, they are entitled to secede even if this eliminates certain economies of scale and therefore reduces the capacity of each successor or residual community to produce or provide certain goods for its members, so long as this reduction is shared among the successor or residual communities. Pre- sumably those seceding expect other benefits - perhaps cultural - that outweigh the productive losses; my claim then is that they may secede to secure such goods without compensating the others for a shared productive loss. Before attempting to defend this distinction, let me suggest some of the applications of the redistributive prohibition to our Canadian situ- ation. Suppose Quebec were to propose to secede, leaving the remainder with the entire national debt (this, I should emphasize, is not what Quebecois separatists do propose). This would clearly be redistributive, and would obviously violate the proviso - per capita public debt would diminish to $0 in Quebec and increase elsewhere. On the other hand, suppose the rest of Canada were to demand that Quebec were to pay for all federal property in the province that would come into its possession, even though such property, in Quebec and elsewhere, had been paid for by tax monies collected equitably in all parts of Canada, and the location of such property had been equitably and reasonably located in the various parts of Canada. This would then also be redistributive, and would be a proviso violation on the part of those opposing secession. Redistributive violations may occur in both directions, making the terms of secession either too permissive or too restrictive. Quebec should not be permitted to secede without taking its fair share of the national debt, but it should be permitted to secede taking its fair share of what is presently federal property. Or consider a very different type of situation. Suppose that the United Kingdom were to propose to cut St Helena loose, not in response to a demand for independence from St Helenans, but simply to save the annual subsidy. The subsidy is in itself redistributive; St Helena con- sumes more than it produces. It might then be alleged that the subsidy constitutes an existing proviso violation that would be eliminated were the UK to 'secede' from it. But this, according to my account, is mistaken. The UK chose to acquire and subsidize St Helena; it can have then no grievance in justice against the existing arrangement. The subsidy is part of the background normative considerations against which violations of This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 367 the proviso are to be determined; were the UK to seek to cut St Helena loose it would therefore be bettering the lot of its inhabitants (by elimi- nating the subsidy) and worsening the lot of the St Helenans. A state is not entitled to divest itself of colonies that it has ceased to want if such divestiture is not wanted by and would be at the expense of the inhabi- tants of the colony. We are, I think, disposed to reject redistributive secession (in the absence of existing injustice). But precisely what is the case against it? And is there a principled distinction between distributive and produc- tive considerations, such that a secession that diminishes production by eliminating certain economies of scale is nevertheless justifiable? Before proceeding, let me qualify the case for what might be called unproduc- tive secession by setting aside the possibility of eliminating economies of scale on such a level that one of the successor communities would not be politically or economically viable. It seems to me that loss of viability would in practice involve a redistributive violation of the proviso; in any event, I think the possibility of loss of viability without such a violation sufficiently unlikely that I shall not attempt to deal with it.9 The proviso is intended to rule out the taking of advantage. It may seem evident that redistributive secession involves such taking. The secessionist party seeks to alter the character of an ongoing relationship, in which the several participants have coordinated their activities on the basis of shared expectations about the distribution of benefits each will receive from the outcome. And the secessionist party seeks to do this, not because the relationship has ceased to be advantageous or the distribution has ceased to conform to agreed terms, but merely in order to improve its relative position. We may suppose that the participants have invested in their relationship, expecting a continuing return on this investment, which the secessionist party proposes to deny them. A particular allocation of resources and pattern of mutual complementar- ity in economic development has been rationalized in part by the exist- ence of boundaries that the secessionist party proposes to alter. One might deny that any of this involves the taking of advantage by arguing that if each participant envisages the possibility of unilateral secession, it will rationally take advance precautions to prevent being disadvan- taged. Thus advantage would be taken only if some participant irration- ally failed to anticipate this possibility, and one should not appeal to such a failure to argue that secession is unjustified. But precautions need not always be feasible, and in any event constitute unproductive costs; the 9 Butseen.ll. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 368 David Gauthier potential value of association is therefore diminished by the prospect of redistributive secession. And so the character and form of political association will be affected adversely if each party thinks of itself as entitled to terminate it by secession merely in order to improve its relative position vis-a-vis the others. But if the value and effectiveness of political association would be undermined by the use of secession as an essentially strategic tool, it would be undermined in a very different way were secession to be unavailable to the members of those groups that seek to express and realize shared values through political association. Now I want to be very careful in the endorsement I offer here. I want to endorse only groups whose members themselves endorse both the weak right of association and the proviso. If you like, I want to legitimate secession only by those groups whose members accept the standards I am propos- ing for legitimating secession. But I want to hold that acceptance of the weak right of association is incompatible with applying the proviso to constrain unproductive secession. The weak right of association ex- presses the value we place on an individual being free to associate with whom he wishes to associate provided they wish to associate with him, and being equally free not to associate with those with whom he wishes not to associate, whether or not they wish to associate with him. Al- though, as I have noted, the effective exercise of this right must be limited in several ways, it would be totally undermined were persons not free to withdraw from existing political associations even when economies of scale would thereby be sacrificed. And so I must be prepared to acquiesce, should the Quebecois express a clear wish to secede from Canada. But there are terms and conditions - and boundaries. It will, I hope, prove illuminating to explore some of these, if only to apply and illustrate the more abstract ideas about secession that I have sketched. IV I want very briefly to introduce three particular issues. First, if Quebec were to secede, then Canada would no longer be geographically contiguous; there would be the problem of access between Atlantic Canada and Ontario. Second, if the large majority of francophone Quebecois were to wish to secede, there would be anglophone regions, especially the western part of the island of Montreal, whose inhabitants would be opposed to secession. And third, there would also be the enormous region of northern Quebec, not part of the original French colony or of the province at the time of Confederation, much of which is inhabited largely by Cree and Inuit who may well not favor secession, This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 369 and also the site of the power projects that many Quebecois see as an essential part their future well-being. All of these issues have a territorial dimension. And I have said nothing, or almost nothing, about territoriality . The technology of politi- cal association is strongly territorial; whether this gives territoriality the appearance of being a value, or whether it reveals territoriality as an underlying human value, I shall not speculate. But let me say that I regard the territorial claims of political communities as strictly deriva- tive from what might be called claims of habitation by individuals. If most of the persons actually inhabiting a particular territory wish to establish a political community among and restricted to themselves, then their claims of habitation provide a basis for the territorial claim of the community they establish - and a basis that normally overrides any other territorial claims.10 If Quebec were to secede, then the terms of secession should guarantee free access between what would become non-contiguous regions of Canada. If the Atlantic provinces, Quebec, and, say, Ontario formed three separate states, and if the Atlantic provinces and Ontario were to wish to unite, they would have no special right of mutual access that Quebec (or alternatively the United States) would have a duty to recog- nize. But the right to secede from an existing state is not the right to disrupt what remains, or to prevent it from functioning as a single community. Access between the parts of a legitimate political commu- nity constitutes part of the background conditions against which benefits and costs are to be measured. And so failure to ensure effective access and communication between the newly non-contiguous parts of a state from which one seceded would constitute a unilateral imposition of cost on the remaining inhabitants of that state. The seceding party should be expected to internalize the cost of providing such access in determining the benefits and costs of secession.11 10 I say 'normally' because, for example, it might be the case that the actual inhabitants had recently and unjustly expelled other persons. 11 One of the referees for this journal suggested that the secession of Quebec might undermine the economic viability of the Atlantic provinces even if their access to the rest of Canada were assured. Since secession would not be overtly redistributive, this questions my claim that any loss of viability resulting from secession would in practice involve a redistributive violation of the proviso. However, the referee suggests, plausibly, that the most adverse effects for the Atlantic region would arise should the secession of Quebec precipitate the break-up of the remainder of Canada. Certainly were that to occur, Ontario and the Western provinces, seeking to better their own situation, would have the obligation to avoid any redistribution costly to the Atlantic region. If honoring this obligation would permit the continuing viability This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 370 David Gauthier My second issue addresses the situation of the anglophone majority in the western part of the island of Montreal, assuming that most of these persons would be strongly opposed to secession from Canada. The territorial claim - that the island of Montreal was part of the original French colony, and of the province of Quebec as recognized at the time of Confederation - carries little weight in my argument. At most I can allow an appeal to historical considerations in seeking to establish salient boundaries in areas that might reasonably be disputed between different political communities. Of greater importance, in my view, is the extent to which the economic and social relations among parts of the island of Montreal and between Montreal and other parts of Quebec, make the political division of the island infeasible. To be sure, one can divide a single urban community if one is sufficiently determined; the example of Berlin is fresh in memory. But it is hardly an example to be emulated. My third and final issue addresses the situation of northern Quebec. Let us assume that the Cree and Inuit inhabitants of that region do not wish to be part of an independent Quebec, but prefer to remain related to Canada.12 The territory in question was not part of the original province of Quebec, but was acquired by Canada from the Hudson's Bay Company in 1869, and was later transferred, part in 1898 and the remain- der in 1912, by the Canadian government to Quebec, reserving the federal government's rights and responsibilities in relation to the native peoples. Here considerations of economic and social feasibility do not speak strongly against the exercise of the weak right of association by the Cree and Inuit. To be sure, given the investment by the province of Quebec in the development of hydroelectric power in the James Bay region, compensation for what would otherwise be a loss of capital, and use at favorable rates of much of the electricity generated, would have to be provided by Canada. But there is a further issue that does illustrate one way in which territorial considerations may be relevant to secession. Suppose that the land was uninhabited, or inhabited only by workers on the power projects. Would a seceding Quebec have the right to this territory, given that it had come to jurisdiction over it only by transfer from the Canadian government? I think not. We may assume that the territory was trans- of the Atlantic provinces, then the referee's real worry should be that the obligation would be ignored. While this is a legitimate political concern, it would not tell against my claim. 12 I say 'related to' because there are complications here arising from the demands of first peoples for some form of autonomy. This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Breaking Up: An Essay on Secession 371 ferred to the jurisdiction of Quebec in the general interests of Canada. For Quebec now to claim the territory would be to claim a benefit that the Quebecois would not have attained on their own, since had Quebec not been part of Canada it would not have gained jurisdiction over the territory of an English company, and to impose a cost on the inhabitants of the rest of Canada, who would no longer be members of the overall political community sharing in the products and wealth of that region. Against the background constituted by the way in which northern Quebec was acquired by Canada and later placed under the jurisdiction of Quebec, for an independent Quebec to claim the area would be for it to seek to benefit itself at the expense of the remainder of Canada, in violation of the proviso. The weak right to associate with those with whom you want to associate provided they want to associate with you, and the proviso against bettering yourself by worsening your fellows, have provided what normative framework there has been to my discussion. I have argued that these ground a broad right for the members of a group to bring about the secession of that group from the political community of which it has been a part, even if that community has been fully legitimate. Just as there is a moral case for no-fault divorce, so there is a moral case for no-fault secession. And just as no-fault divorce may be emotionally agonizing for those who experience it, so may no-fault secession. I do not want that agony, and so I can only conclude with the hope that my country will not prove to be one of the first states in which the theory of no-fault secession is put into political practice. Received: September, 1993 Revised: February, 1994 This content downloaded from 195.78.108.140 on Tue, 10 Jun 2014 18:49:18 PM All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Article Contents p. 357 p. 358 p. 359 p. 360 p. 361 p. 362 p. 363 p. 364 p. 365 p. 366 p. 367 p. 368 p. 369 p. 370 p. 371 Issue Table of Contents Canadian Journal of Philosophy, Vol. 24, No. 3 (Sep., 1994), pp. 357-526 Front Matter Breaking up: An Essay on Secession [pp. 357-371] Marxism, 'Ideology' and Moral Objectivism [pp. 373-393] Hacking's Experimental Realism [pp. 395-411] Lewis on 'Might' and 'Would' Counterfactual Conditionals [pp. 413-418] Paternalism and Rights [pp. 419-439] Feminism, Fundamentalism, and Liberal Legitimacy [pp. 441-463] Nonreductive Materialism and Mental Causation [pp. 465-493] Critical Notice Review: untitled [pp. 495-509] Review: untitled [pp. 511-524] Back Matter