Scott Boykin, "The Ethics of Secession," in David Gordon, ed., Secession, State and Liberty, Transaction Publishers, 1998.


Scott Boykin

The moral status of secession should be a significant issue in contemporary political theory. Various secessionist movements around the globe have had a tremendous impact on world politics, and the fluidity of power relations in many parts of the world suggests that they will continue to have such an effect for some time. Moreover, secession addresses the fundamental problem of political theory: namely, the moral basis for the state's authority.

Modern political thought has produced three main types of argument for the state's legitimacy. One, found in Kant, grounds the state's authority on the purported rightness of its institutions and aims. A right of secession challenges this position if it allows a group to legitimately withdraw from a just state. Another, found in Locke, holds that consent, whether explicit or tacit, is the source of the state's authority. A right of secession challenges this-position in maintaining that consent may be legitimately withdrawn in favor of an alternative political arrangement. The third, found in Hume, bases the state's authority on its usefulness in producing order, which facilitates the individual's pursuit of self-chosen ends. A right of secession challenges this position in negating the general Humean duty to support a historical or conventional state and morally permitting a group to transfer its loyalties to one it expects to find more useful.

A right of secession does not demand that we reject any of these positions; indeed, I employ each in defending secession. If secession is morally justifiable, however, it calls for qualifications to each of these three accounts of political obligation. Given its practical and theoretical import, it is surprising that only one recent work in academic political theory, Allen Buchanan's Secession, has approached the subject at length.1 Buchanan seeks to establish a framework for considering the ethical issues surrounding secession, and some key parts of what follows are written with a view to his arguments.

In this essay, I make a moral case for secession. The argument is concerned solely with secession by groups; while individual secession is an interesting idea, I will not pursue it here.2 In the first section, I lay the ethical grounds for a right of secession. An indeterminate contractualist argument, which eliminates specific rules or principles as elements of a just political order, can justify a right of secession. No state, I argue, can justly prohibit secession under all circumstances. The second section defines the right of secession and specifies the moral limits on the exercise of the right. Though secession as a right is exercised collectively, it is an individual and not a group right. This is important because it renders unnecessary a group claim to have shared ethnic or cultural attributes, and thus removes one obstacle to the right to secede. All that counts is whether the group wishes to define itself as a political unit. The right of secession is a limited one in that secessionists must, first, be able to establish a viable political order, and second, protect private property and the market. I argue that any group of persons that meets these criteria cannot be justly prohibited from seceding. The upshot of this section is that any political system ought to provide for a constitutional right of secession subject only to these limits. Buchanan sets further restrictions on the right of secession, which I criticize and reject in the final section.


In this section, I develop a normative foundation for a right of secession consisting of two basic elements. The first is a treatment of value from the perspective of normative individualism. Under normative individualism, values are exclusively personal in nature, or agent-relative. The agent-relativity of all values places strict constraints on the moral claims individuals can legitimately make on one another, and thus limits the depth of social or political obligation. The second element is a contractualist ethical argument that establishes a moral presumption favoring a constitutional or procedural right of secession, which I define later in the chapter.

Normative Individualism

Under normative individualism, all values are agent-relative. An agent-relative value is a value for a particular person. A value is agent-relative if and only if its description must refer to the valuing agent.3 For agent A1 and states of affairs S1 and S2, S1 possesses agent-relative value for A1 if and only if A1 ranks S1 over S2. With this ranking, A1 has an agent-restive motivation to act so as to realize S1, but no reason to promote S2. An agent-neutral scale of value, as found in cardinal-comparable utilitarianism, yields an ordering that all agents must value and promote -- without regard to their personal disposition -- the highest-ranked state of affairs. If A2 and A3 prefer S2, Ai must value and promote S2 in spite of his preference for S1; Ai's life thus becomes shackled to others' preferences. The individualist theory of value rejects the validity of agent-neutral scales and their moral implications.4

Normative individualism entails that the value of a social and political order, like that of anything else, lies in its usefulness to the individual. Order facilitates the individual's pursuit of his plans and projects, collaboration with other individuals, and the emergence of competitive social processes that serve to coordinate individuals' actions. Rational agents have reason to participate in and observe the rules of a social and political order because, and only because, that order is useful to them. If a political arrangement is advantageous to A1 and disadvantageous to A2, it is not a value for A2, and A2 lacks reason to support it. A political arrangement has value for all the agents concerned, and provides them with motivation to support it, if and only if it is mutually advantageous.

Because normative individualism indicates that the value of political order lies in its capacity to facilitate the self-chosen plans and projects of individuals, it can play a key justificatory role in political argument. As James Buchanan argues, the principle supplies the

normative premise that individuals are the ultimate sovereigns in matters of social organization, that individuals are the beings who are entitled to choose the organizational-institutional structures under which they will live. In accordance with this premise, the legitimacy of social-organizational structures is to be judged against the voluntary agreement of those who are to live or are living under the arrangements that are judged. The central premise of individuals as sovereigns does allow for the delegation of decision-making authority to agents, so long as it remains understood that individuals remain as principals. The premise denies legitimacy to all social-organizational arrangements that negate the role of individuals as either sovereigns or principals.5

Normative individualism, then, offers a critical perspective from which we can evaluate political institutions. Political institutions possess value only if they serve the interests of the persons subject to them. A model of voluntary agreement enables us to ascertain whether individuals can be expected to refuse assent to a rule. If they would do so, the rule in question cannot be justified.

Contractualism and Secession

A contractualist ethics can appraise political arrangements by considering whether they contain elements that anyone seeking a rational agreement on public institutions could reasonably reject.6 This test is a variant of the principle of universalizabil-ity, and it narrows the range of individual interests considered morally significant. An agent, for example, might value his committing robbery and murder, but other persons could legitimately reject a rule permitting him to do so. An indeterminate contractualism evaluates particular principles and institutions, but does not fully determine the structure and contents of a worthy political order.7 That some rules can be eliminated as alternatives does not indicate which remaining proposals individuals will accept. Contractualism can, and indeed must, take into account the indeterminacy of its conclusions with regard to the rules actual persons will approve. While contractualism cannot specify what people will accept, it can place moral constraints on actual political systems by showing that some rules should always be rejected. A rule that fails the contractualist test cannot be justly imposed on anyone.

With these constraints in mind, we can ask whether anyone could reasonably reject a political order that prohibits secession under any circumstances. Because normative individualism implies that institutions possess value only insofar as individuals ascribe value to them, it invalidates arrangements that benefit some persons at others' expense. When a subset of a group has determined that a new arrangement, which requires an act of secession to establish, would serve themJjetter than the status quo, the subset can claim that forceful opposition to their secession on behalf of the interests of the remainder of the group benefits the latter at the prospective secessionists' expense. A constitutional or a procedural right of secession enables individuals to reveal their preferences over alternative arrangements by offering a means of disclosing whether, in the judgment of a subset of a larger group, present conditions are in fact a disvalue to them. Anyone, then, who suspects that at some point he may become dissatisfied with a set of political institutions could reasonably object to a rule prohibiting secession under any circumstances. Anyone, moreover, who considers it possible that he may, at some future time, judge secession a means by which he could participate in forming a new political unit which would better serve his interests could reasonably reject a rule prohibiting secession. All that is required is an assumption of uncertainty over the long-term consequences of life under a political arrangement.8 Given this uncertainty, it is clear that anyone could reasonably reject a rule prohibiting secession under all circumstances. Contractualism, then, yields an ethical presumption favoring a right of secession.

Whether anyone could legitimately oppose a particular act of secession is another matter. In the following section, I examine the conditions under which secession might be justifiably prohibited, and show that the right of secession is a limited one.


The moral considerations advanced in the previous section support a right of secession that grounds a duty on the part of any state to refrain from prohibiting secession under all circumstances. This specifies the action which the right protects, and who is obliged to respect the right. Any state should provide a legal or constitutional means for secession. A complete right-statement must also define the bearer of the right and detail any limits on its exercise. In this section, I fill in these blanks.

Who Bears the Right of Secession

Allen Buchanan argues that because only groups may secede, secession must be a group right. Group rights, as he puts it,

are ascribed to collections of individuals and can only be exercised collectively or at least on behalf of the collective, usually through some mechanism of political representation whereby a designated individual or subset of the group purports to act for the group as a whole. In addition, the good secured by the right is most often a collective good in the sense that if it is secured it will be available to all or most members of the group. Moreover, if we think of rights as serving certain interests, we may also say that the interests served by group rights are individuals' interests, qua members of the group, in the collective goods of the group -- that is, their interests in participating in the common activities and in pursuing the shared goals of the group.9

There are two distinct flaws in Buchanan's treatment of secession as a group right. First, it unduly restricts the good secured by secession to the benefits of common activities and goals. Clearly, members of a group may wish to secede for the sake of pursuing private goals more effectively, and it is not apparent that such goals are a less legitimate basis for secession than are collective ones. Even if those interests that individuals have only as members of a group are a necessary condition of their collectively possessing a group right, such interests are not necessary to a right of secession.

Second, the fact that collective action is required to exercise a right of secession does not entail that it is a group right. Procedural rights of due process, for example, require collective action to establish reliable judicial and law enforcement institutions, but the goods they secure are inextricably bound to interests individuals possess qua individuals.10 Collective action here is necessary to exercise the individual right to be treated justly by others. While collective action may be a necessary condition for a group right, it may also be a necessary condition for the exercise of some individual rights, hence it does not follow that the requirement of collective action makes secession a group right.

Treating secession as an individual right eliminates a complex of moral and practical problems, such as establishing group rights in heterogeneous populations.11 It becomes unnecessary to identify and defend the group interests to which Buchanan refers. A group need not possess shared cultural or ethnic attributes, or assert historical claims to territory which they could justify only as a recognized group.

The right of secession, as an ethical constraint on any political order, establishes the conditions for actual contracts, because it permits individuals to establish new political arrangements through collective action. The actual contracts made possible by a right of secession promote an ideal of the self-defined political community, which is a variant of the principle of national self-determination. If self-determination applies only to identifiable cultural or ethnic groups, a right of secession that is defined in these terms is a group right, in that individuals may claim to exercise it only by virtue of group membership. If we interpret the right of secession as an individual right, this constraint on its exercise is unnecessary. Where political communities are self-defined by the expressed preferences of their members, their cultural or ethnic characteristics are irrelevant; the value ascribed to existing or proposed institutions by the several individuals is the only significant factor to be taken into account. Self-determination here is the individual's freedom to choose between existing and alternative arrangements. As Mises puts it, "The totality of freedom-minded persons who are intent on forming a state appears as the political union."12

As on Mises's analysis, political unity is the other side of the argument from self-determination.13 A constitutional right of secession is exercised through the expressed preferences of those concerned. If a group opts for secession, its members express their dissatisfaction with existing institutions, and their favorable expectations of the alternative before them. It is quite likely, then, that the seceding group will be highly unified with respect to the procedures and aims of the alternative they choose. Having rid itself of at least some dissidents, moreover, the state that loses some fragment of its population due to secession will likewise be more unified than before. As preferences approach unanimity, of course, the more closely will collective choices on institutions correspond to individual preferences. From this perspective, again, secession promotes self-determination, and is defensible as a value in accordance with normative individualism.

Buchanan suggests that it may be desirable to make secession difficult by imposing special taxes on a seceding group or requiring extremely large majorities in favor of secession, because a society may have good reasons for opposing a particular act of secession.14 I examine and criticize these in the next section. On the basis of the argument so far, however, it is clear that no special difficulties should be placed in the way of a group who wishes to secede. Indeed, the individualist conception of the value of political order and the ideal of the self-defined polity suggest that it would be unjust to hedge about a constitutional provision for secession with obstacles the purpose of which is to make the right difficult to exercise. Procedural barriers to secession should, in principle, be no greater to secession than to other types of constitutional changes or legislative acts. Provisions for initiative to propose an act of secession should, then, be no more difficult than those currently in effect to propose constitutional amendments in the American states that allow for initiative. A referendum in which a three-fifths, two-thirds, or simple majority of citizens vote for secession should be sufficient for enactment.

A Limited Right

Ethical limits on the right of secession are part of the moral input associated with the right's justification, and are established in the same manner as the right itself. The seceding group must be capable of forming a viable state; the contractualist standard would allow anyone to reject an act of secession which would subject individuals to an inviable and thus useless state. The population of a seceding group must be large enough to support institutions that perform the fundamental governing functions of rule-making, enforcement, and adjudication. The newly-created state must be capable of penetrating the entire territory it will claim, otherwise individuals could claim that their property and liberty will not be adequately protected. Whether it must be shown that the new state can defend itself against external aggression is an entirely different matter. International affairs are rather unpredictable, and it is conceivable that almost any state might at some point be threatened by more powerful adversaries. Certainly a state must be able to defend itself to be viable and thus useful to individuals. It is difficult, however, to imagine how one could formulate a general constitutional principle that could be applied to all cases. The relative power and expected intentions of a state's neighbors are not good guides because these are subject to change. Moreover, states show a poor record of defending themselves. Aggressor states have won 69.2 percent of the international wars involving major powers over the last two centuries.15 Given this record, it is unreasonable to demand that secessionists show they can provide an ironclad defense of their territory. Because of the greater unpredictability of international as opposed to domestic affairs, and the unimpressive performance of states in defense matters, the requirements for demonstrating this aspect of viability should not be too stringent. Provided the secessionists are capable of raising revenue to support defensive forces, or maintaining diplomatic institutions to form defense agreements with other states, they can legitimately claim viability.

Proposed or expected economic policies are the basis for a second limit on the right of secession. Socialism is ruled out as an alternative, and this exclusion can be reduced in part to the viability requirement. As Mises demonstrated, rational economic calculation is impossible under socialism.16 In light of this theoretical critique and the subsequent failure of socialist economies, anyone could reasonably reject being made subject to a socialist state on account of its known inviability. The contractualist standard, however, also eliminates many forms of market intervention. Individuals have reason to support a social and political order only if it permits them to pursue their own plans and projects. Anyone could reasonably refuse their assent to rules that prohibit or otherwise render impossible the pursuit of goals that would not be excluded by contractualism. Interventionist economic policies that upset individual planning, or prohibit actions that are morally permissible by the contractualist standard, could be reasonably objected to by anyone.17 Anything other than the unhampered market is an unjust imposition by the state, secessionist or not.18

In summary, the right of secession may be defined as follows. First, it is an individual right to engage in collective action for the purpose of secession. There is no need for the seceding group to show that it possesses common ethnic or cultural characteristics. As long as the group follows constitutional procedures, such as those suggested here, it may justly secede. Second, the seceding group must be able to erect a viable political order. Finally, the secessionists must refrain from engaging in unjust forms of market intervention. These are the demands of a just constitutional right of secession. In the next section, I examine and criticize additional limits on the right proposed by Allen Buchanan.


The limits on the right of secession give citizens of a prospective secessionist state grounds for objection. The following objections would be lodged by the remainder state or its citizens. Buchanan discusses each of these and grants them a moral weight that justifies limiting the right of secession further than I have thus far. I shall argue that these objections are without merit, and that the only limits on the right of secession are those discussed in the previous section. Though the right of secession is a limited one, it is much less limited than Buchanan believes.

Secession and State Property

Buchanan argues that secessionists ought to compensate the remainder state for its loss of property and investment in the seceding territory, and that failure to make such an offer provides grounds for resisting an act of secession.19 This is indeed a tangled problem, but it is by no means clear that secessionists must compensate anyone. If public expenditure is universally beneficial prior to an act of secession, it need not be the case that citizens of the remainder state are to be deprived of its benefits afterwards. A highway system, for example, may facilitate trade in both pre- and post-secession periods. If a public facility or construction project benefits an entire group prior to an act of secession, there is no obvious reason why it will fail to do so afterward. The justification for such appropriations is that they promote productivity and economic growth. The remainder state, then, would be deprived of benefits only if the secessionists close their market to the former. Since the right of secession is limited by the claims of property and the market, the secessionists are precluded from cutting off the free exchange which, in this case, would enable citizens of the remainder state to benefit from these public expenditures.

Actual states, of course, do not limit their fiscal decisions to the generally beneficial. Political institutions empowered to appropriate funds for purposes that are not universally beneficial are demonstrably unjust by the contractualist standard; anyone may reasonably reject rules which allow some to benefit at others' expense. Pork-barrel projects are not only wasteful, but unjust as well. Unless it can be shown that the secessionists themselves are responsible for unjust public expenditures from which they have benefited, no compensation is necessary. Pork and logrolling are games played by politicians subject to the pressures of a political system that establishes the game. In the politicians' view, the game is positive-sum, since they benefit through vote trading and claiming to get their constituents a "fair share" of public booty. From the citizens' perspective, it is a prisoner's-dilemma game in which there is no rational incentive to refuse exploitative benefits, and all are worse off as a result. Unless the secessionists themselves can be assigned responsibility for this situation, they owe no compensation.

The "Threat" of Strategic Bargaining

Buchanan suggests that the threat of secession as an instrument of strategic bargaining justifies limiting the right by requiring extremely large majorities (e.g., three-fourths) or a special secession tax.

In conditions in which the majority views secession by a group G as a prohibitive cost, G's threat to secede can in effect serve as a veto. G can use the threat of secession to ensure that the majority's will does not prevail, even when the majority's decision would respect constitutional limits.20

In light of normative individualism, majority rule is valuable only insofar as it provides a useful means for collective decision; it has no intrinsic value whatever. If a minority is so intensely opposed to a majority view that they are willing to secede, there is no legitimate interest in majority rule that justifies limiting the right of secession. The fact that a majority wishes to impose its position on a recalcitrant minority does not morally privilege the majority. Here again, the right to secede appears as a way to limit public power, and since anyone could expect to be in the minority at some point, anyone could reasonably reject restrictions of the right to secede which impose high costs on those wishing to escape a majority hostile to their interests.

Territorial Sovereignty

The violation of a state's territorial sovereignty is a prima facie objection to an act of secession. Buchanan defines territorial sovereignty as a

relationship among the state (the agent), the territory, and the people (the principal), with the state acting on the people's behalf to preserve the territory not only for the present but for future generations as well. Territorial sovereignty is best understood as a set of jurisdictional powers over territory, conferred upon the state.21

Buchanan's definition suggests that the state's jurisdictional powers are conferred upon it by the people. He argues, then, that a withdrawal of consent can "demonstrate the conditions under which the state no longer has authority over people," but that it "cannot show when the state no longer has control over territory."22 It is plain, though, that the state's territorial claim must be dependent on persons, rather than on territory. Its jurisdictional powers obviously cannot be conferred by the territory itself. Buchanan continues:

A sound justification for secession has two territorial components: an argument to show that the state either never had or had but has lost territorial sovereignty over the seceding land, and an argument to show that the seceding group either has had or ought now to have territorial sovereignty.23

As Buchanan recognizes, the simple fact that nearly all actual states were created through conquest puts their territorial claims on shaky moral ground, though he suggests that a "moral statute of limitations" seems to "favor adopting a convention that accords substantial weight to existing boundaries."24 Even if, however, we absolve a state of past injustices committed by officials no longer among us, it is not clear why we should adopt Buchanan's convention as a reason to limit the right of secession. An act of secession is an explicit rejection of the state's jurisdictional powers. Once most of the people in some portion of a state's territory (who meet the criteria previously enumerated) have shown, by wishing to secede, that they disvalue the state, it has lost its claim over that piece of territory. The state's territorial claim has no meaning, and hence no validity, apart from the value ascribed to it by citizens.

Buchanan muddies the waters considerably by invoking other moral questions concerning whether the seceding group can lay claim to the territory. He argues that they must show that they have been targeted by the state for unjust treatment not directed against other groups, and that the other groups have not come to the aid of the group discriminated against. If everyone is unjustly treated, or if other groups try to help the victimized one, secession is not legitimate because it is government officials, not the people at large, who are responsible for the acts which justify secession.25 Secession, however, is not an act of punishment; it falls in the category of procedural rather than rectificatory justice. Those who secede are punishing neither their former state nor its citizens; they are simply expressing their desire to govern themselves. This preference for self-government is the only significant issue, once the limitations discussed earlier are taken into account. That other groups are treated unjustly, or come to their aid, need not give the secessionists reason to ascribe value to the state they are trying to leave. No matter the disposition of other groups, the right of self-government under liberty, that is, the right of secession, cannot be justly opposed.

Judging Alternatives to Secession

A final issue to consider is whether the alternative constitutional provisions of nullification or group veto (Calhoun's concurrent majorities principle) are suitable replacements for a right of secession. If they are, a right of secession is not an essential element of a just constitution, provided one of these alternatives is in place. Buchanan suggests three criteria for weighing the alternatives: the independence the right offers its bearer, the strategic bargaining value of the right, and the disruption which the exercise of the right presents to others. On the first two criteria, secession ranks higher than group veto and nullification, because it grants greater independence and strategic value to its bearer. On the third criterion, the rights of nullification and group veto are ranked over secession because secession causes greater disruption in the remainder state's affairs. Buchanan argues that the choice among the three rights is indeterminate because differing circumstances affect how heavily we should weight the three criteria (e.g., how disruptive an act of secession will prove). Consequently, there is no sure means of deciding whether a right of secession must be part of a just constitution.26

If we can weight the criteria, however, they yield a determinate ranking. The criteria of independence and strategic value are lexically prior to that of disruption, because normative individualism morally permits individuals to afford greater weight to their own plans and projects than to those of others. A ranking that failed to grant priority to independence and strategic value would lock individuals into a political order they disvalue, and this is what contractualism enables them to reject. Since the criteria of independence and strategic value are prior to that of disruption, the ranking they yield is determinate. Secession is ranked over group veto and nullification, so the latter are not suitable replacements for a constitutional right of secession.


1 Allen Buchanan, Secession: The Morality of Political Divorce from Fort Sumter to Lithuania and Quebec (Boulder, Colo.: Westview, 1991).

2 But see Lysander Spooner, No Treason: The Constitution of No Authority, in The Lysander Spooner Reader (San Francisco: Fox and Wilkes, 1992), pp. 49-111. 3Thomas Nagel, The View from Nowhere (Oxford: Oxford University Press, 1986), pp. 152-53.

4 For criticism of agent-neutrality, see Eric Mack, "Moral Individualism: Agent-relativity and Deontic Restraints," Social Philosophy & Policy 7 (Autumn 1989): 81-111; Loren Lomasky, Persons, Rights, and the Moral Community (New York: Oxford University Press, 1987), pp. 16-55; J.J.C. Smart and Bernard Williams, Utilitarianism: For and Against (Cambridge: Cambridge University Press, 1973), pp. 108-18.

5 James M. Buchanan, The Economics and the Ethics of Constitutional Order (Ann Arbor University of Michigan Press, 1991), p. 227, emphasis in original.

6 Here I am following T.M. Scanlon, "Contractualism and Utilitarianism," in Utilitarianism and Beyond, Amartya Sen and Bernard Williams, eds. (Cambridge: Cambridge University Press, 1982), pp. 103-28.

7 See John Gray, Post-Liberalism (New York: Routledge, 1993), pp. 48-50.

8 See Geoffrey Brennan and James M. Buchanan, The Reason of Rules (Cambridge: Cambridge University Press, 1985), pp. 28-31.

9 Buchanan, Secession, p. 75. He discusses this idea further on pp. 75-80.

10 See Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), pp. 96-101.

11 See Buchanan's comments, Secession, pp. 139-43.

12 Ludwig von Mises, Nation, State, and Economy, Leland B. Yeager, trans. (New York: New York University Press, 1983), p. 34.

13 Ibid., pp. 36-37.

14 Buchanan, Secession, pp. 132-39.

15 Kevin Wong and James Lee Ray, "The Initiation and Outcome of International Wars Involving Great Powers" (paper presented to the International Studies Association Convention, Washington, D.C., April 1990).

16 Ludwig von Mises, Socialism, J. Kahane, trans. (Indianapolis: Liberty Fund, 1981).

17 For an account of interventionism and its effects, see Ludwig von Mises, Human Action, 3rd rev. ed. (New York: Henry Regnery, 1966), pp. 716-861.

18 This is a stronger version of an argument presented by John Gray, who maintains that the liberal neutrality implied by contractualism excludes a socialist economy but does not require laissez faire. See John Gray, Liberalism (New York: Routledge, 1989), pp. 161-98. Surely, though, interventionism must fail to be neutral with respect to individuals' plans and projects, even if as an unintended consequence.

19 Buchanan, Secession, pp. 104-6.

20 Ibid., p. 100.

21 Ibid., p. 108, emphasis in original.

22 Ibid., p. 73.

23 Ibid., p. 113.

24 Ibid., p. 110.

25 Ibid., pp. 111-14.

26 Ibid., pp. 143-48.