One of the enduring problems in democratic theory is its inability to specify who should belong to the demos. Democrats seemingly do not have conceptual resources to determine who should comprise the people that are to govern itself democratically. Once a demos exists, it can employ majoritarian procedures to alter its own composition. A democratic majority can approve minority secession, for example, or define criteria through which nonmembers might acquire citizenship. Yet a “majority” can only exist as part of an already-constituted group, which triggers the question of who should belong to that group. For any group we pick, the question arises which prior group defined it as part of the demos, and how that prior group obtained its own right to belong to the demos or decide on its composition. Given the danger of infinite regress, democratic theory cannot directly answer this question. “Democracies,” as Seyla Benhabib notes, “cannot choose the boundaries of their own membership democratically.”Footnote 1 This is the famous “boundary problem” in democratic theory.Footnote 2
The traditional response to the problem consisted largely of embarrassment.Footnote 3 First, scholars have been embarrassed to take a pretheoretical demos as a given. As Robert Dahl remarked, democratic theory simply assumes that “a people has already constituted itself. … This mysterious transformation is therefore treated as a purely hypothetical event that has already occurred in prehistory or in a state of nature. The polis is what it is; the nation-state is what history has made it. Athenians are Athenians. Corinthians are Corinthians, and Greeks are Greeks.”Footnote 4 Alternatively, scholars have argued that democratic theory must admit an unsolvable paradox at the heart of the demos's “founding”: the phrase “we the people” can only be both the outcome and the origin of a democratic founding.Footnote 5 If not outright embarrassment, scholars have argued that the boundary problem should at least generate humility. According to Fredrick Whelan, for instance, acknowledging democracy's inability to define the demos should “have the beneficial effect of moderating the sometimes excessive claims that are made in its name.”Footnote 6
In recent years, however, a more ambitious strategy has gained momentum. Several scholars have been arguing that democratic theory should employ the “productive gap”Footnote 7 of the boundary problem's infinite regress to argue for global reform leading to more inclusionary demoi. As Gustaf Arrhenius, for example, puts it, “there is an obvious end to the regress: when everybody is included.”Footnote 8 Hans Agné provides another version of this thought:
The paradox of founding may be a serious problem as long as our reasoning is informed by nationalist or republican assumptions that some people should or must be excluded at the moment of founding new political orders or communities. The truth, however, is that we need not exclude anyone. The founding of political orders or communities is democratic if and only if decided in procedures that give equal opportunities of participation to people who will live within as well as beyond the boundaries inherent in the foundation. In short, the paradox of founding does not exist in global democracy.Footnote 9
This statement exemplifies the kind of argument that I wish to take up in this essay—an argument moving from the boundary problem to global reform. What exactly this move entails, and how it is achieved, varies with different theorists who entertain it. Thus for example, Robert Goodin, whose approach to the boundary problem has gained considerable attention, argues from a principle of enfranchising affected interests that he considers central to democracy, to the conclusion of a single global polity, as an ideal to be approximated as much as possible.Footnote 10 Arash Abizadeh, whose argument has been equally prominent, emphasizes that the demos of a democratic polity is “in principle unbounded,”Footnote 11 and argues that because the regime of control over membership in each demos coerces citizens and noncitizens alike, the latter too ought to be able to participate in shaping this regime—meaning that there need to be global democratic decisions with regard to the composition of all demoi, even if not necessarily a single global political community.Footnote 12 However, notwithstanding their internal variations, Goodin, Abizadeh, and other theorists like Agné all share the basic analytical move from the boundary problem to some kind of global reform. I will accordingly refer to this idea as boundary problem reform, or, to make things less cumbersome, BPR.
BPR arguments are important for practical, scholarly, and philosophical reasons. At the practical level, they seem to have the potential to influence how we think about global reform on issues ranging from immigration to a world state. At the scholarly level, BPR arguments are integral to the increasingly “heated debate”Footnote 13 on the boundary problem and its myriad aspects. Finally, less often noted but at least equally significant, is the philosophical level, where BPR arguments stand out for their minimalism.
The move from the boundary problem to global reform is minimalist in that it seems to reach a very significant result without appealing to any contested theory. This move does not invoke, for example, familiar claims about a right to free movement that separate demoi curtail through their border controls.Footnote 14 The BPR move similarly does not invoke concerns about global economic inequality, such as a duty to remove unequal life prospects that result from birth into different demoi with different levels of wealth. Nor does it invoke any complex theory of historical injustice that calls for the removal of existing borders separating current demoi, tainted as they are with historical violence. Without appealing to any of this philosophical machinery, the BPR move proceeds from an extremely modest starting point, merely seeking to make domestic democracy internally coherent, to radical global reform.
However, my thesis will be that, notwithstanding its appeal, this minimalist strategy fails. The move from the boundary problem to global reform does not work. Moreover, I shall argue that whenever the boundary problem is invoked (directly or indirectly) to justify global reform, the real grounds for reform actually come from another argument that is unrelated to the problem. Therefore, rather than being key to global reform, the boundary problem turns out to be an analytical distraction in thinking about such reform.
This thesis should matter to multiple audiences. Most obviously, this thesis will matter to those who currently do believe that the democratic boundary problem has a key role to play in thinking about global reform. But, more broadly, my thesis should also matter to the larger audience that is interested in the relationship between the commitments of democratic theory and other commitments essential to modern political morality. Once we put this relationship center stage, the democratic boundary problem turns out to be far less promising as the basis for any global reform. Or so, at least, I will argue.
I advance this argument as follows. I begin with general analytical doubts, as to whether it makes sense to demand of democratic theory itself that it specify the composition of the demos that is to be self-governing (section 1). Building upon these doubts, I join the developing conversation on Abizadeh's and Goodin's versions of the BPR move, seeking to offer new reasons for why both versions encounter fundamental problems (section 2). I conclude by anticipating the concern that my argument does not take the need for global reform seriously enough.
1. Can the Democratic Boundary Problem Ground Global Reform? General Doubts
I want to start with some general doubts as to whether the democratic boundary problem can motivate an argument for global reform. At the root of these doubts is the following question: why is there a moral duty to have democratic theory itself specify the composition of the demos?Footnote 15 The familiar goal of democratic political theory has been to align a commitment to popular sovereignty—to collective self-rule by “the people”—with a commitment to the moral equality of each individual citizen. The traditional inquiries of democratic theory accordingly concern what political procedures and/or outcomes an egalitarian conception of popular sovereignty requires, prohibits, or permits.Footnote 16 But why is it morally necessary to specify the group of individuals that is to comprise the citizenry through democratic theory itself?
BPR arguments depend on an answer to this question. After all, if it is merely permissible rather than obligatory to specify the composition of the demos through democratic theory, then there cannot be a moral duty to pursue any global reform based on such specification. However, I wish to show that proponents of BPR arguments cannot establish a moral duty to specify the demos democratically. More precisely, I wish to show that proponents of BPR cannot defend this duty in a way that will retain the relevance of their move from the democratic boundary problem to global reform. It is worthwhile to see in some detail why this is the case, by considering several defenses which proponents of BPR arguments might offer.
We can begin by considering one kind of defense that is arguably implicit in many BPR arguments. According to this defense, we ought to specify the composition of the demos democratically, because unless we do so, the democratic commitments essential to modern political morality remain in some fundamental sense incomplete.Footnote 17
It may initially seem intuitive to think that democratic theory remains incomplete unless it specifies the demos that is to be self-governing. Yet, upon reflection, this thought turns out to be quite problematic. The main reason is that there is actually something odd about demanding of democratic theory itself that it specify the composition of the demos whose collective conduct it seeks to guide. We can see this once we try to generalize this demand, so as to extend it to other bodies of normative theories. Expressed in more general terms, the view would arguably have to say something like this: “in order for any normative theory to be complete, we must try to address, through its own internal resources, higher-order questions about its core concepts.” This demand, however, makes little sense, as we can see by considering what it would do to other established kinds of normative theories. Consider, for example, any kind of ethical theory attempting to articulate the principles that individuals must follow if they are to be moral—whether consequentialist, deontological, virtue ethics, or anything else. We do not typically think that these theories of individual ethics are incomplete unless we can use their own internal resources to answer higher-order questions about why morality matters at all. Similarly, it seems odd to say that an egalitarian theory of distributive justice remains incomplete, unless the theory itself provides tools for addressing higher-order questions about why equality matters.Footnote 18 To be sure, any normative theory, whether referring to political justice or individual ethics, depends on higher-order judgments about its core concepts. But it does not follow that these judgments have to be internal to the theory itself, in order for the theory to be considered complete. Specifically in our case, it is obvious that any democratic theory depends on judgments about the demos, as the complex agent whose behavior it aims to guide, including judgments about its composition. But it does not follow from this that democratic theory itself remains incomplete unless we employ its own internal principles to specify the composition of the demos.
The idea that democratic commitments remain incomplete unless democratic theory steps in to define the demos thus turns out to be more contentious than it may at first seem. But there is another idea that lies in the vicinity, and that also seems to lie in the background of certain BPR arguments. According to this idea, it is morally essential to have democratic theory itself define the demos, because democratic theory provides the most reliable way to rule out compositions of the demos that are clearly morally untenable. The thought here is that we must have a systematic basis for ruling out certain compositions of the demos that are obviously morally unacceptable, and that democratic theory provides the best hope for such a systematic foundation.
The problem with this thought once again lies at the metatheoretical level. It is very much disputable whether a systematic theory of any kind (democratic or otherwise) is needed in order to vindicate moral judgments that we clearly take as axiomatic. It is more intuitive to see these “fixed points” in our moral thinking as preexisting checks against which we measure any systematic theory, rather than as a result of such a theory.Footnote 19 Take a few paradigmatic examples: we clearly reject racist criteria for the composition of any demos. We think it morally unacceptable to compose any demos in ways that involuntarily strip individuals of citizenship. We think that there is, at the very least, a strong moral presumption in favor of extending citizenship to stateless persons and refugees who suffer the most extreme forms of oppression and destitution.Footnote 20 And we also think it unacceptable for any state to perpetually refuse to grant citizenship to individuals who were born and raised within its borders. But these deep-seated normative convictions precede rather than follow from any systematic normative theory.Footnote 21 We would have ample reason to be suspicious of any theory (democratic or otherwise) that turns out to be incompatible with these specific judgments. But we do not appeal to such theories to ground these judgments. Rather, we actually see such specific judgments as preceding systematic theories.Footnote 22
This claim, in turn, leads to an additional point. If it is the case that our condemnation of certain compositions of the demos precedes rather than follows from any systematic theory, then forswearing a moral duty to specify the composition of the demos through democratic theory does not mean adopting an “anything goes” approach that is morally indifferent to any compositions that politics might generate.Footnote 23 Accordingly, we should not think that there is a binary choice between a democratic composition of the demos and an approach to compositional questions that is morally empty: avoiding the former does not mean adopting the latter.Footnote 24
Appeals to the systematic character of democratic theory, then, just like appeals to the completeness of democratic commitments, cannot justify a duty to specify the composition of the demos democratically. But it is also worthwhile to consider two more concrete ways to try to justify such a duty, both implicit in different BPR accounts. First, we can consider appeals to exclusion. Concerns about exclusion from the demos have been animating much of the literature on the democratic boundary problem. And it seems intuitive to think that determining the composition of the demos through democratic theory has a crucial advantage in mitigating the worry about exclusion: after all, the democratic “norm of inclusion”Footnote 25 generates a strong presumption in favor of a demos that, in Agné’s words quoted above, excludes “no one.”
Proponents of BPR, however, cannot invoke concerns about exclusion to explain why there is a moral duty to specify the demos democratically. For one thing, this appeal takes for granted the disputed notion that exclusion as such is inherently morally problematic.Footnote 26 But more importantly, the appeal to exclusion actually undercuts any independent force that BPR arguments might have in grounding global reform. The reason is this. If our opening assumption is that demoi ought not be composed in exclusionary ways, then we are already presupposing that some kind of a global demos—a global political community in which all of the world's individuals are members—is the moral default. But there must be an argument grounding this moral default, and this argument would generate a demand for radical global reform long before we turn to democratic theory for guidance as to how to compose the demos. Hence the democratic boundary problem would play no independent role in thinking about global reform. There would be some other argument that would be doing the real normative work in justifying reforms pertaining to membership in any demos.
A similar difficulty undermines attempts—equally familiar in the literature—to contrast a democratic approach to the composition of the demos with a historical approach. Consider, for instance, Sofia Näsström's claim that the main alternative to specifying the composition of the demos democratically has been to “run to the arms of history.”Footnote 27 This statement suggests that we ought to specify the composition of the demos democratically, because we ought not treat historically constituted demoi as our point of departure. But this argument cannot make the democratic boundary problem fundamental for thinking about global reform either. The reason is that here, too, the worry that a democratic composition of the demos is meant to allay turns out to be the key to thinking about global reform, in a way that undercuts any independent role for the boundary problem itself.
To see this, note that if historically constituted demoi ought not be our normative point of departure, this is presumably because of the various forms of historical violence that have been integral to their composition. Therefore, there has to be a theory of historical injustice underlying the contrast between a democratic composition of the demos and compositions that are rooted in sullied histories. But once we have such a theory, it is concrete claims about historical injustice, rather than abstract claims about the commitments of democratic theory, that take center stage in thinking about political reform. Suppose, for instance, that our theory of historical injustice leads us to think that all individuals ought to be able to become members of any demos they wish to join, given that virtually no existing demos can claim to have been originally constituted in a nonviolent manner. Or, along the same lines, suppose that our conception of historical injustice leads us to think that a sizeable portion of humanity ought to be allowed to join a wide range of political communities as a form of compensation for historical wrongdoing committed against their ancestors. Such arguments draw a direct link between historical injustice and present global reform, but there is no reason why this link has to go through the democratic boundary problem. The concrete demand not to condone past wrongs (or not to leave them unremedied) is entirely independent of abstract boundary problem questions regarding the commitments of democratic theory. So it turns out once more that the attempt to justify a duty to specify the demos democratically renders the democratic boundary problem moot for thinking about global reform.
2. Can the Boundary Problem Ground Global Reform? Two Prominent Arguments
I began with general doubts about the move from the democratic boundary problem to global reform. I now turn to criticize this move by examining the two specific BPR arguments mentioned at the outset, offered by Abizadeh and Goodin.
Other critics of the move from the boundary problem to global reform have also tended to focus on these two arguments. Sarah Song, for example, contends that Abizadeh's and Goodin's proposals ignore practical concerns regarding the size and stability of the demos.Footnote 28 Similarly, Louis Cabrera argues that Abizadeh and Goodin ignore the practical worry that global enfranchisement will only leave many vulnerable populations outvoted.Footnote 29 I do not wish to take a stance on these critiques here. Rather, I focus on showing (partly with the help of the previous section) that both Abizadeh's and Goodin's arguments suffer from an assortment of more fundamental problems. I want to suggest that, even before we get to the practical implications of Abizadeh's and Goodin's views, we can question the philosophical reasoning underlying these views.
2.1 Abizadeh's Argument
Let us start with Abizadeh's view. Abizadeh argues against the permissibility of separate demoi unilaterally controlling either their territorial borders or the composition of their membership—their “civic borders.” Abizadeh's argument, as he notes, is straightforward:
My thesis is that, according to democratic theory, the democratic justification for a regime of border control is ultimately owed to both members and nonmembers. The argument for this apparently radical thesis is surprisingly simple. The argument's first premise simply states the normative democratic principle of political legitimation; its second premise notices an obvious empirical fact. First, a democratic theory of popular sovereignty requires that the coercive exercise of political power be democratically justified to all those over whom it is exercised, that is, justification is owed to all those subject to state coercion. Second, the regime of border control of a bounded political community subjects both members and nonmembers to the state's coercive exercise of power. Therefore, the justification for a particular regime of border control is owed not just to those whom the boundary marks as members, but to nonmembers as well.Footnote 30
The boundary problem is, according to Abizadeh, a key reason why the argument's opening premise (“a democratic theory of popular sovereignty requires that the coercive exercise of political power be democratically justified to all those over whom it is exercised”) refers to all persons rather than merely all citizens of a polity—why the demos is “unbounded.”Footnote 31 In turn, Abizadeh understands the demand for justification at the heart of his thesis (“the justification for a particular regime of border control is owed not just to those whom the boundary marks as members, but to nonmembers as well”) as a demand for actual participation. Under the “democratic strategy” which Aabizadeh endorses, “saying that a justification is ‘owed to all those over whom power is exercised’ is to say… that all such persons must have the opportunity… actually to participate in the political processes that determine how power is exercised.”Footnote 32 This claim yields Abizadeh's version of global reform. Because (according to Abizadeh) all individuals in the world, “even those who never present themselves at the border or never seek citizenship,” are “really subject to border coercion,”Footnote 33 all individuals ought to be able to actually participate in decisions on the criteria governing all civic border regimes. In other words, all individuals ought to be able to participate in decisions defining the criteria for the compositions of all demoi.
The best way to begin our assessment of this argument is to consider what we may term a symmetrical world scenario. Imagine, then, a world in which the following conditions obtain. Separate states claim exclusive jurisdiction over separate territories. Each state automatically grants citizenship to all individuals born in its territory, and to their children. Each state allows all of its rational adult citizens to vote and run for office, and each state ensures that all citizens equally enjoy the familiar rights to which justice entitles them—bodily integrity, property, freedom of expression and association, rights to healthcare, education, a clean environment—and so on until the list of all possible familiar rights is exhausted. Imagine further that all states are identical in such a way as to essentially nullify the significance of being born in one rather than another. The safety level in each society is identical. The level of societal wealth and its distribution are identical. The size of each state's territory is identical and so are its natural resources. All states feature considerable levels of internal ethnic, religious, and linguistic diversity (with no ethnic, religious, or linguistic group enjoying clear dominance in any state).
Moreover, all states, in line with the “pretheoretical priors” above, respect obvious moral demands with regard to their criteria for citizenship. No state, for example, makes citizenship decisions on the basis of racial, ethnic, or religious discrimination. Each state grants citizenship to stateless individuals who arrive at its borders, and to individuals who have found themselves in the state's territory for a substantial time through no fault of their own (for example, individuals who have been brought to the territory as children and who grew up in the territory). However, outside of such cases, each state refuses to automatically grant permanent residence or citizenship to nonmembers. Additionally, each state refuses to allow nonmembers the right to participate in decisions over citizenship criteria. According to Abizadeh's argument, therefore, each state subjects a sizable portion of the world's individuals to illegitimate coercion through its border control regime.
More specifically, on Abizadeh's view, each state in the symmetrical world is violating the moral commitment to autonomy and equality that underlies democracy's value. “The democratic ideal of collective self-rule,” Abizadeh writes, “is grounded in the notion that securing the conditions of individuals’ autonomy and standing as equals intrinsically requires that they be the joint authors of the terms governing the political power to which they are subject.”Footnote 34 The symmetrical world conflicts with this notion: it features states that subject nonmembers to a coercive regime of civic border control, without allowing nonmembers to actually participate in decisions over the criteria governing this regime. So if we apply Abizadeh's argument, the symmetrical world would violate the fundamental moral commitment to respect individuals’ autonomy and equal standing.
This conclusion, however, strains credulity. Why should we think that the symmetrical world represents any meaningful affront to the equality or autonomy of all individuals? Insofar as we would not think this, it follows that there is no reason to believe that the ability to participate in decisions on the criteria for the composition of all demoi has any profound—let alone intrinsic—moral value. But Abizadeh's argument hinges on the significance of this ability.
We can see the same point from another angle, as follows. Imagine that, convinced by Abizadeh's argument, certain individuals demanded an explanation of why they are not allowed to participate in decisions regarding the border control regimes of the states to which they do not belong. Confronted with this demand, government officials in each state might respond: “The fact that you are prohibited from participating in our border control decisions neither reflects nor generates any inequality between you and our own citizens. Our citizens, after all, are prohibited in exactly the same way from participating in your own state's border-control decisions. And these mutual prohibitions, in turn, do not generate or reflect any inequality in your resources, life prospects, or range of life choices, when compared with our own citizens. Therefore, the value of equality cannot ground any complaint on your part against the procedures related to our border control regime.”
Now, while I hope that many readers will be convinced by this reasoning, perhaps others will remain skeptical. Perhaps some readers will insist on Abizadeh's behalf that even symmetrical coercion is still coercion, and that it is not implausible to see the symmetrical world as violating the equal moral standing of all of the world's individuals, simply by denying them the right to decide on the criteria governing all states’ border control decisions.Footnote 35 I believe that the best response to such trenchant critics goes through the following thought. If, as Abizadeh holds, there is intrinsic moral value to being able to participate in decisions regarding the criteria governing border controls, then this value cannot be sensitive to substantive changes in these criteria. What matters on Abizadeh's view is not the content of these criteria, but who gets to decide on their content. This view, however, yields implausible results, which we can quickly see by adding some further details to the symmetrical world.
Suppose that, alongside all of the features of the symmetrical world given above, we further imagine that each state in the symmetrical world enacts a civic border regime that is highly inclusive: while each state still shies away from automatically granting residence or citizenship to nonmembers, the demands it imposes on the way to legal residence and eventual citizenship are truly minimal. To make this minimalism concrete, let us assume that each state in the symmetrical world announces to all outsiders that all they have to do in order to be eligible for legal residence and (eventually) citizenship is to prove the most basic level of competence in any single language out of the multiple languages that are widely used within the relevant state. We can make this requirement even more minimal through a host of further specifications.Footnote 36 But the point is that, within the symmetrical world, even border control criteria that would effectively allow virtually all of the world's individuals to (eventually) acquire citizenship in any state they choose would still, on Abizadeh's view, run afoul of individuals’ equal autonomy; after all, each state in the symmetrical world would still be imposing these border controls unilaterally, without allowing outsiders to participate in border control decisions. Yet here the charge that the symmetrical world represents an affront to the equal autonomy of each individual is clearly even less credible than it was before.
Given that this is the case, I think it fair to suspect that those readers who still find themselves drawn to Abizadeh's boundary problem argument, even when confronted with the permissive variant of the symmetrical world, are in fact being pulled by intuitions that have little to do with the boundary problem. Such readers might, for instance, be drawn by the thought that even the symmetrical world imposes limitations (albeit symmetrical limitations) on individuals’ freedom of movement. But freedom-of-movement arguments can be made without any recourse to the democratic boundary problem. Similarly, certain readers might be tempted by the thought that the artificially symmetrical world is inappropriate as a normative guide for the real world, where different states are extremely far from identical in their level of wealth, for example. While this point is obviously true, it only shows once more that the real normative work is done not by appealing to the boundary problem, but by appealing to much more familiar normative concerns—here, specifically, concerns about global economic inequality. And the same response would apply to any parallel complaint: any claim that will be made here as a way of motivating Abizadeh's BPR argument will in fact render it irrelevant to thinking about global reform.
My basic charge against Abizadeh's BPR argument, then, is that it is ultimately unmotivated. The concern with coercion of nonmembers that is central to the argument cannot really be doing the normative work on which the argument relies: the force of the argument must come from somewhere else—most likely, from a much more familiar theory of individual rights or global distributive justice, which then turns the boundary problem issues into a distraction.Footnote 37
It might be useful, however, to deepen this general charge, through a more specific critique. Let us therefore consider in more detail a necessary component of Abizadeh's argument: the thought that there is a sufficient moral equivalence between the coercion that the state applies to outsiders and the coercion it applies to insiders. This thought is necessary, because if there are salient moral differences between the two contexts of coercion, then the mere fact that they share a certain family resemblance will not suffice to motivate the argument: the possibility would remain that the differences are decisive.Footnote 38
However, the idea that there is a sufficient moral equivalence between a government's coercion of its citizens and that of noncitizens creates other kinds of difficulties for Abizadeh's view. The reason is that, if no salient moral differences are identified between the coercion of citizens and noncitizens, and if noncitizens therefore ought to be able to participate in decisions over the border regime that coerces them, then there cannot be any salient moral difference between failing to fully enfranchise noncitizens on this matter and failing to enfranchise citizens on matters that have to do with their own coercion. But this result yields deeply counterintuitive implications for thinking about democratic political obligation.
This point obviously presupposes a certain conception of democratic political obligation, and I should therefore say something about the conception I favor. Following an increasingly dominant view in democratic theory, I assume that individuals have a general moral duty to obey the laws of their polity if and only if these laws embody a meaningful commitment to the moral equality of citizens at both the procedural and the substantive level.Footnote 39 At the procedural level, a meaningful commitment to moral equality requires that all adult citizens have an equal vote in free and fair elections that determine who will hold top political office. At the substantive level, all the major laws enacted by elected officials must be compatible with some defensible (even if suboptimal) interpretation of what it means to treat all citizens with equal respect (thus ruling out, for example, fundamental laws clearly based on racial, religious, or gender discrimination, even if these are supported by a majority of citizens). When these conditions obtain, equality itself requires that citizens obey even those laws that they think are mistaken or unjust.Footnote 40 However, when either the procedural or substantive condition is absent, citizens do not have a general moral duty to obey the law qua law.Footnote 41
This model of political obligation aligns with the strong intuition that fundamental laws which blatantly violate any reasonable interpretation of equality remove their victims’ general political obligation. But the model also aligns with the more radical thought that deeply flawed major laws undermine the general political obligation not only of their victims, but of other citizens as well. And I assume that many democrats, at least, will find this more radical thought congenial. Thus, for example, many democrats would want to say that African-Americans disenfranchised under official racial segregation in the United States had no moral duty to obey American law qua law; but that white Americans under official racial segregation had no general duty to obey the law qua law either.Footnote 42
Now, all of this matters here for the following reason. Equating, morally speaking, the kind of coercion that a government exercises over its citizens with the kind of coercion it exercises over all noncitizens through border control will commit democrats to saying that if the citizens of Australia, for instance, do not have a right to participate in decisions over American immigration policy, this is—in some key sense—morally analogous to withholding the right of political participation from black US citizens. After all, both of these cases feature individuals coerced by political decisions in which they do not get to participate. Therefore, if racial segregation and disenfranchisement dissolved Americans’ general political obligation in the 1950s, then the American government's current failure to allow Australians to participate in decisions over the composition of the American demos also dissolves Americans’ general political obligation at present. Yet this, it seems to me, is a decisive reductio. I see no reason why an American citizen could consider herself free of her general moral duty to obey the law qua law, simply because her government denies Australians the right to participate in decisions over the criteria governing the composition of the American people.Footnote 43
One potential objection to this reductio is that Abizadeh himself acknowledges gradations in the significance of border control for different individuals, endorsing “a greater participatory say to foreigners for whom entry actually represents a valuable option, an even greater say to those for whom the option of entry is necessary to have an adequate range of valuable options, and perhaps the greatest say to citizens themselves.”Footnote 44 However, the natural way to read this statement about gradations is to see it as presupposing that the key claim is accepted (“once we agree that all individuals ought to have a say over the composition of all demoi, we can discuss how strong of a say different individuals should have”). But then the appeal to gradations does too little to defeat the reductio, which challenges precisely the key claim. Alternatively, if one (less plausibly) interprets the appeal to gradations as an acknowledgment that there are fundamental differences between the ways in which a border regime subjects both citizens and noncitizens to state coercion, then this appeal does too much, because it destabilizes the argument's basic ground for the enfranchisement of outsiders—that “political boundaries… always subject both insiders and outsiders.”Footnote 45
Another potential objection to my reductio is that anarchists of various sorts will not accept the conception of political obligation that I am presupposing here.Footnote 46 But, for one thing, if Abizadeh's argument can be appealing only to anarchists, this already diminishes its force considerably. Furthermore, it is not clear that anarchists are a relevant audience here, insofar as most democrats reject anarchist assumptions from the outset. At least the vast majority of democrats consider it obvious, pace anarchists and philosophical anarchists, that the existence of the state and its laws is morally necessary: that the state and its laws make social life possible, solving crucial coordination problems, allowing individuals to develop stable expectations regarding others’ behavior, and specifying the content of individual rights (such as property rights) that anarchists and philosophical anarchists mistakenly believe to be determinate prepolitically. From a mainstream democratic viewpoint, at least, a decision to disobey the law ought to be driven by a hope for a different law, obedience to which will be morally necessary. There is rarely (if ever) a democratic acceptance of, or a hope for, no law at all.Footnote 47
2.2 Goodin's Argument
Let us now turn to Goodin's version of BPR, which argues from the principle of affected interests to the conclusion of global reform. In Goodin's view, decision-making procedures gain a key part of their democratic credentials by considering the interests of those they affect. At the same time, Goodin believes that affected interests should also shape the composition of the group that makes democratic decisions: the group ought to include all individuals whose interests will be affected by its decisions. As he puts it:
Protecting people's interests is… the most plausible candidate principle for bringing the “who” and the “how” of democratic politics into alignment. That principle dictates who should constitute the decision-making group (“all affected interests should have a say”). It also dictates how that group should be governed (“making decisions democratically,” which well-established results tell us is the best way to protect and promote people's interests).Footnote 48
Such a position amounts to the claim that democratic theory necessarily leads to a global demos. One's interests, after all, are clearly affected not only by any policy that is actually pursued, but also by the various other options that were available: one must take into account the hypothetical consequences of alternative decisions that were never taken. Furthermore, many decisions are path-dependent (can only be possible if certain prior decisions are taken). Thus any tenable account of “affected interests” must cover a huge range of decisions across space and time, affecting a number of individuals that is impossible to calculate. Combine these points with the fact that numerous decisions states make in the twenty-first century clearly affect interests beyond their borders, and the conclusion easily follows that in order to include “all affected interests” in the demos, democrats must pursue a global demos, or at least approximate it as much as possible.
There are several reasons, however, to think that this conclusion follows too easily. This is true even if we put aside the enormous difficulties involved in defining “affected” interests, in assessing when interests are strong enough to be included, and in weighting interests of varying strength.Footnote 49 Even independently of such issues, multiple objections can be raised against Goodin's move, through the affected interests argument, from the democratic boundary problem to global reform. The best way to see these objections is to ask whether Goodin's argument can be supported either on an instrumentalist or on a noninstrumentalist approach to the justification of democracy. I wish to show that in both cases the answer is negative.
Let us start with the instrumentalist camp. Instrumentalists deny that democratic procedures for making political decisions can have inherent as opposed to merely instrumental moral value. This claim can be based on different grounds: for example, it can be rooted in a belief that there is an inevitable conflict between the idea of equality as the bedrock of political morality on the one hand, and on the other hand the real-world exercise of political power which necessarily involves, even in a democratic setting, unequal relations between rulers and ruled.Footnote 50 The instrumentalist denial of democracy's inherent moral value can also be grounded in a more straightforward insistence that democratic procedures cannot have moral value that is independent of their outcomes. However, I wish to argue that no instrumentalist approach to democracy—no matter what its particular grounds are—can support Goodin's move from the boundary problem to global reform.
The reason is fairly simple. Goodin's argument takes the democratic boundary problem as its starting point, but this is something that no instrumentalist account of democracy can do. If democracy is not a foundational value—if it is only an instrument—then the argument's starting point must be whatever it is which democracy is an instrument for. But if that is the case, then it becomes unclear why the boundary problem has any real normative role.
To make this point more concrete, consider, for instance, Richard Arneson's instrumentalist account of democracy. Arneson argues that “the right to a democratic say” (“an equal vote in a democratic political procedure”) necessarily involves power over others, meaning that it cannot be considered fundamental: “rights that involve power over others are among the nonfundamental rights and their existence, strength, and shape should be fixed according to what is maximally productive for fulfillment of fundamental rights.”Footnote 51 If one holds such a derivative view of a democratic say, one would not start one's normative inquiry from the democratic boundary problem. Rather, one would start by spelling out the list of fundamental rights in question. In turn, there would be no analytical necessity that these rights will require democratic enfranchisement, let alone that they will require any kind of global reform. But even if they did—even if it turned out that some form of global enfranchisement is “maximally productive for fulfillment of fundamental rights”—the democratic boundary problem would have no role to play in the argument. The argument would turn solely on whether one can defend one's preferred list of fundamental rights, and show that the rights in question indeed mandate some form of global enfranchisement. But this kind of argument would have little resemblance to a boundary problem argument.
This point obviously goes beyond Arneson's specific view. It applies to any attempt to motivate Goodin's argument from an instrumentalist perspective on democracy. If democracy is ultimately an instrument for some X, then it must be that X from which the argument starts, and in that case, there is no reason why the argument has to move through the democratic boundary problem, nor is it clear what the argument will gain from such a move. This complaint would apply to Goodin's own instrumentalist account of democracy, for instance, just as it applies to Arneson's: just as in Arneson's case the task would be to unpack the list of fundamental rights and examine their practical implications, in Goodin's case the task would be to unpack the list of interests which democracy is an instrument for “protecting and promoting,” and to defend the claim that the best way to protect and promote these interests is to move towards a global demos. Here, too, it is unclear how delving into the boundary problem would advance the task.Footnote 52
When we move to noninstrumental approaches to democracy, however, Goodin's argument faces problems that are just as significant. Like instrumentalists, noninstrumentalists come in different forms: some argue that democratic procedures are intrinsically valuable as a way of expressing equal respect toward all citizens;Footnote 53 others see the inherent value of democracy in its embodiment of mutual justification of the terms of public life (the “public reason” tradition);Footnote 54 others still emphasize an analogy between individual and collective autonomy as explaining democracy's inherent value.Footnote 55 What all noninstrumentalists share, however, is a rejection of paternalism that conflicts with Goodin's “affected interests” reasoning.
One familiar source of conflict here derives from the noninstrumentalist worry that “affected interest” reasoning might be compatible with support for benign autocracy. If the ultimate aim of public policy is—as Goodin holds—to “protect people's interests,” then we have no principled grounds with which to object to a benign autocrat doing away with democratic procedures and protecting people's interests himself. Goodin's suggestion that “well-established empirical results” show that democratic procedures best “protect and promote” popular interests falls short of the kind of principled reasons noninstrumentalists offer for the moral significance of popular self-government. For this camp, the core democratic concern is that the people, the right agents, govern themselves, prior to and independently of the question of whether their self-government would best advance their interests. Noninstrumentalists emphasize that the demos can legitimately make decisions that set back even important popular interests.Footnote 56 The principle of affected interests thus sits uncomfortably with those who believe that democracy is first and foremost about agency rather than interests.Footnote 57
Another source of conflict between Goodin's affected-interest reasoning and noninstrumentalist approaches is very much related. Noninstrumentalist approaches to democracy hold that rational adults ought to decide themselves in what their interests consist, and that neither benign autocrats nor “enlightened utilitarian” elites have any moral permission to act as “guardians”Footnote 58 protecting their interests. And while some, to be sure, have attempted to accommodate such worries,Footnote 59 it is worth noting that (whether or not these attempts succeed) Goodin's efforts have gone in quite the opposite direction. Employing empirical evidence, Goodin has insisted that people constantly fail to act on their interests, among other things owing to “weakness of will,” or because “people are notoriously bad at judging interests that hang in any important way upon probability calculations,” as well as because “people manifest a wholly irrational bias in favor of present over future interests.”Footnote 60 Building upon these findings, Goodin has argued that “these are far from uncommon elements, but whenever they are present, there is a case to be made for discounting a person's own judgment of his interests. … Just look down the list of circumstances that make individuals misperceive their interests and notice how largely immune from those forces would be public officials judging their interests on their behalf.”Footnote 61 Whether or not one can ultimately justify this kind of paternalistic reasoning, those who are noninstrumentalist about democracy, at least, will be reluctant to endorse it.Footnote 62
With these remarks in hand, I wish to note two final objections to Goodin's move from the democratic boundary problem to global reform. Both build on my doubts in the opening section as to whether there is a moral duty to specify the composition of the demos democratically. The more seriously one takes these doubts, the more willing one should be to take separate demoi, constituted through sources external to democracy, as a starting point. But if one adopts this starting point, the affected-interests principle which Goodin invokes becomes even less convincing. First, whatever appeal this principle may or may not have as a criterion for how to constitute the demos as a collective agent whose composition is undecided, it has even less appeal for how to guide the behavior of an agent whose composition is already decided. This is because, at least according to commonsense morality, agents have no general moral duty to continuously advance the interests of others. Hardly anyone will deny that you can permissibly choose to refrain from purchasing property that I wish to sell, for example, even if I am extremely keen for you to make the purchase, to such a degree that your choice significantly affects my interests: your choice may set back my interests, even profoundly, but this by itself does not mean that your choice wrongs me. As far as I can see, the only line of defense against this criticism in the context of our discussion is to argue that in the case of the demos, the very composition of the agent bearing duties is itself unclear, and that it is in order to decide this composition democratically that we turn to something like the principle of affected interests.Footnote 63 However, if, as I have argued above, there is no duty to decide on the composition of the demos democratically, then the composition of the duty-bearing agent is not in fact at stake.
The second objection here is closely related. When we consider how decisions by existing demoi affect outsiders who do not take part in making these decisions, it is not the general possibility of adversely affecting outsiders’ interests that is of concern to us. Rather, the concern is much more specifically that decisions by existing demoi wrong outsiders by violating their rights. But, for one thing, we have seen above that there is no reason to believe in an independent right to decide on the composition of all demoi. This means that the boundary problem has little bearing on which rights outsiders should be seen as possessing. To settle this question we need to return to familiar global justice debates on individual rights—precisely the debates that the BPR arguments were supposed, through their minimalism, to avoid. Furthermore, even if the theory that arises from the relevant debates establishes that individuals have certain rights that are endangered by the current decisions of at least some demoi, the most natural response is not to demand that these demoi enfranchise those whose rights they are violating. Rather, the natural demand is that these demoi stop violating the rights in question. As Ben Saunders puts it in response to Goodin, “rights are better respected by limiting the power of groups to infringe them, rather than by requiring them to include or enfranchise the rights-holders.”Footnote 64 If the mafia, for example, violates my rights, my response would not be to demand a voice in the mafia's decisions, but to demand that the mafia stop (or be stopped).
Conclusion: Not Taking Seriously the Need for Global Reform?
We can now take stock. My main aim in this essay has been to show that a focus on the democratic boundary problem, rather than advancing normative discussions of global reform, in fact sets these discussions back, by distracting us from the real site of the normative action. I have sought to advance this thesis first by presenting general doubts as to whether the democratic boundary problem can motivate arguments for global reform, and then by elaborating specific doubts as they apply to the particular arguments presented by Abizadeh and Goodin.
I would like to reiterate, however, that my aim here was not to defend the global status quo in any way. In particular, nothing I have said here should be taken as ignoring weighty moral concerns about global economic inequality and the way in which global institutional arrangements perpetuate historical injustice. Global inequalities and historical injustices ought to command our moral and practical attention. But I hope to have shown that our duty to give attention to these issues only provides a further reason to put the boundary problem aside.
As I have argued repeatedly, there is no necessary analytical connection between the boundary problem and either the historical injustices that underlie present borders or the global inequalities that they demarcate. The boundary problem is neither necessary nor sufficient to motivate a focus on historical injustice. The concrete demand not to ignore the historical violence that has produced existing demoi is distinct from the abstract question of who should be included in the demos. Similarly, the fact that individuals born on two sides of the same border often face radically unequal life prospects is morally disturbing entirely independently of any questions about whether and how democracy should determine the composition of a people. We may have very strong reasons to think that global economic inequality mandates a variety of global reforms. But, for one thing, there is no necessary reason why these reforms will involve changing compositions of demoi. Furthermore, even if we did think that in order to curb global economic inequality we ought to soften civic borders, such that citizens of poorer countries may become citizens of wealthier nations, we have to see that what is doing the real normative work in the argument is not any question about the internal requirements of democratic theory as to the composition of the citizenry. What is doing the real normative work is a familiar theory of global distributive justice. That is a further reason why the focus of global political philosophy ought to remain with these kinds of theories, rather than with the boundary problem.
