Migrants and potential migrants are often affected by decisions in which they have not participated. For that reason, and because of my previous work on immigration, the editors of this book asked me to think about whether the All-Affected Principle might provide a useful perspective on immigration. I have concluded that, with a few important exceptions, the All-Affected Principle does not help very much to reflect more deeply about immigration.
This does not mean that I reject the All-Affected Principle. On the contrary, as the other chapters in this book make clear, this principle can be a valuable theoretical tool for identifying and exploring a wide range of questions about democratic inclusion. But like all tools, it is best used only on certain objects. The All-Affected Principle does help us to think about who should be included in decision making, but, in my view, in the case of immigration, the important questions are not about who should participate in decisions but about what those decisions should be, or more precisely, the moral constraints on the acceptable range of decisions about immigration policies and immigration regimes. Moral principles, especially democratic principles other than the All-Affected Principle, greatly constrain morally permissible collective choices about immigration. There is much less room for discretionary democratic self-determination in this area than is commonly assumed. Thus, the question of who should participate in decision making is correspondingly less urgent.
So, think of this chapter as a cautionary tale. Exploring some of the limitations of the All-Affected Principle with respect to the topic of immigration can serve as a reminder of the ways in which the All-Affected Principle needs to be supplemented by other principles and other perspectives when we engage in normative reflection.
In what follows I will identify what I see as the most important normative questions about immigration. I will then sketch briefly my answers to those questions, drawing primarily – and occasionally explicitly, though without formal citations – upon a recent book.1 I hope that the importance of the questions I am asking will be self-evident, even to those who disagree with my answers. I will consider whether the All-Affected Principle helps to answer those questions or to identify important related issues that have been neglected.
Notes on Method
Every inquiry takes place against a background of presuppositions: normative, empirical, intellectual, linguistic, etc. There is no Cartesian starting point in political philosophy. We have to bracket some questions so that we can focus on others. There are often good reasons for adopting one set of presuppositions rather than another, given a particular intellectual goal, but we should never imagine that the intellectual goal we are pursuing is the only possible one. It is common to adopt a presupposition for one question that we subject to critical scrutiny for another. In what follows, I will try to be explicit about my key presuppositions, and I will change one important presupposition as I go along to illustrate both methodologically and substantively why it is so important to pay attention to presuppositions.
One of my key presuppositions is that it is possible to distinguish between the question of who has the right to make a decision and the question of whether that decision is morally acceptable. More specifically, I assume that it is possible for a political community to include everyone who ought to be included in a decision-making process and to include them in appropriate ways, and for the decision that the community makes to be morally wrong. To put it in a slightly different way, democracies sometimes act unjustly, not because there was a problem with who was included in the democratic process but because the outcome of that process was substantively unjust. I do not think of this as a particularly controversial presupposition, although, like every moral claim, it could be challenged. It is an important presupposition for this inquiry, however, because it draws attention to one of the limits of the All-Affected Principle.
Another key presupposition is that it is possible to assert that an agent, whether individual or collective, has the moral right to make a particular choice and yet to criticize the moral limitations of the options from which the agent is choosing. Everyone recognizes that the robber who says “Your money or your life” is not enhancing the victim’s autonomy by presenting the victim with this choice. But we sometimes do not notice the ways in which other choices are problematically constrained by entrenched institutions and norms so that who has a say in a particular decision may not get at the most important moral issues in a case.
Consider, for example, a famous passage from Robert Nozick that is often used, including by authors in this volume, to illustrate the limits of the All-Affected Principle. In this passage, Nozick says that if four men want to marry a particular woman, the fact that they (and others) will be greatly affected by her decision does not matter. Only the woman herself has a right to make that decision. She has a right to act autonomously.
Nozick is right, of course, that respect for autonomy precludes the men from having a claim to participate in the woman’s decision and so calls into question incautious, general formulations of the All-Affected Principle, but his construction of the example also obscures the important ways in which her choice is constrained and her autonomy restricted.
Nozick’s book appeared in the early 1970s. If Nozick had said then that two women and two men wanted to marry the woman in question and it was up to her to decide which if any of them to marry, the example would not have done the work he wanted in illustrating a widely accepted commitment to individual autonomy and limits on collective choice. At the time, same-sex marriage was not legally permitted. The democratic process had excluded that option. Moreover, same-sex marriage was not thought by most Americans to be morally permissible, whatever their general views on the importance of individual freedom. So, a woman’s autonomy with respect to her ability to make her own decision about which willing potential marriage partner to accept was constrained by democratic laws and by the public norms of a democratic society. Indeed, one might argue that because marriage itself was a patriarchal, heteronormative institution, both legally and socially, in the 1970s and because being unmarried in a patriarchal, heteronormative society entailed its own severe legal and social restrictions, to say that a woman facing such constraints was autonomous because she was free to decide whether or not to marry a man who wanted to marry her would be deeply misleading.2 In what follows, I will draw attention at times to the ways in which questions about immigration look different depending on what one assumes about the moral legitimacy of the background conditions.
Access to Citizenship
I begin with the issue of access to citizenship. The central question is this: On what terms should lawfully admitted immigrants and their children (and subsequent descendants) become citizens of the state in which the immigrants have settled?
Notice two of the presuppositions of this question. First, it simply assumes as a background the moral legitimacy of the division of the world into states, each of which has its own political process through which it exercises authority over immigration admissions and over access to its citizenship. The question here is about access to citizenship for immigrants whom the state has chosen to admit as permanent residents, not about access to citizenship for anyone who wants to live in a state. Second, it assumes implicitly that we know what citizenship means, and, in particular, what it entails as a legal status, at least within certain parameters. If we did not have some sense of that, how could we talk about who was or was not entitled to citizenship? For example, in democratic states in the modern world, citizenship is normally treated as a fundamental status that entitles one to equal rights with other citizens, with minor qualifications such as age restrictions on the right to vote. Moreover, under national and international law, citizens normally have some rights that noncitizens do not have, such as the right to vote in national elections, and the right to reside in a state whose citizenship one possesses and to enter that state if one is outside it. If citizenship did not have these familiar features, the question of who is entitled to be a citizen might appear quite different.
It would certainly be possible to challenge these presuppositions, either from the perspective of the All-Affected Principle or from some other perspective. Later I will explore one way of doing so. But doing that would make it impossible to explore the particular question that I want to ask, because that question arises precisely in the context of these two assumptions and others, including those previously identified. As I noted above, to explore one question, one must bracket others, at least temporarily.
Note that the question I have posed is not a question about whether immigrants themselves should be able to participate in a state’s decisions about the terms under which immigrants will be given access to citizenship. There are ways in which the All-Affected Principle could be used to raise that sort of question, as we will see at the end of this chapter. At this point, however, that question is simply precluded by my first assumption. My question is not “who should participate in deciding the terms under which immigrants gain access to citizenship?” but rather “what are the moral constraints upon that decision?” or, to put it another way, “what access to citizenship must a democratic state grant to immigrants and their descendants, if the state is not to violate basic democratic principles?”
How should we answer this question? Let’s start with access to citizenship for adults who have arrived as immigrants. My view is that settled immigrants should have relatively easy access to citizenship, in part because people have a right to participate in collective decisions that affect their lives on an ongoing basis and the right to participate fully is normally attached to the legal status of citizenship. Obviously, I am appealing here to a version of the All-Affected Principle. This is one of those cases in which I think the All-Affected Principle is highly relevant to a question about immigration, but it plays that role because citizenship is so deeply connected (in most states) to the right to participate in collective decision making.
Participation in collective decision making is not the only reason why immigrants should have easy access to citizenship, however. As I noted previously, citizenship status carries with it certain important rights under international law, a contingent fact but one important to the moral argument. Citizenship is also the way that we recognize people as full members of a political community in the modern world. In my view, simply living in a democratic society over time normally entitles one to that sort of recognition.
I do not think that the All-Affected Principle, understood as a principle focused on claims to participate in decision-making processes, encourages attention to this concern with the normative importance of social membership. I am not saying that the All-Affected Principle conflicts with the claim that social membership matters morally. The participation and social membership arguments for access to citizenship are complementary. The point is rather that the All-Affected Principle does not include all of the considerations that are morally relevant to this issue.
It is worth noting the indeterminacy of the phrase “easy access to citizenship.” There can be reasonable disagreements about what counts as easy access, and that sort of reasonable disagreement is the kind of thing best settled by democratic processes. So, different states might have somewhat different rules regulating access to citizenship, without violating the principle of easy access.
Turn now to the question of who should gain citizenship at birth. Birthright citizenship poses a puzzle. Why make infants into citizens? Infants are not agents. They are not capable of participating in collective decision making. So, the All-Affected Principle does not seem to apply to them, at least not directly, in the way it does to adults. Part of the answer as to why we make infants into citizens lies in the way the world is organized overall. It is divided into separate states and every human being is supposed to be attached to one state (at least) as a citizen. No one should be stateless. And, as I have noted, we assume that there are important rights that go along with this sort of attachment, even for infants, such as the right to enter and reside. But the requirement that everyone be attached to some state does not in and of itself provide any guidance as to what state anyone should be attached to. Moreover, I think that birthright citizenship has a deeper connection to our understanding of what a democratic political community is. We expect a child born to resident citizens to be an ongoing member of the political community and we grant citizenship at birth as a way of recognizing that belonging from the outset. But if that is indeed the rationale for birthright citizenship, then the same rationale applies to a child born to settled immigrants within the state where they have settled. So, the child of settled immigrants should also be recognized as a member of the community from the outset and should gain citizenship at birth. Again, what I am presenting is a certain kind of social membership argument.
It is probably possible to construct a complementary argument for birthright citizenship from the perspective of the All-Affected Principle, if we interpret the practice of birthright citizenship as a way to securing children’s right to participate in democratic decision making when they reach maturity. I don’t think that this way of defending birthright citizenship would be in conflict with my account, but I’m not sure about its adequacy, and, in any event, I don’t see what it really adds to the social membership argument.
Including Immigrants as Full Members
Full membership in a political community involves much more than having the legal status of citizenship. If immigrants or their descendants possess citizenship status but are excluded from the economic and educational opportunities that others enjoy, if they are expected to conceal things related to their immigrant origins in order to fit in, if they are viewed with suspicion and hostility by others, if their concerns are ignored and their voices not heard in political life, then they are not really included in the political community, even if they are citizens in a formal, legal sense. They are not likely to see themselves or to be seen by others as genuine members of the community. In many important ways, they will not belong. So, what is required for the full democratic inclusion of immigrants and their children besides granting them citizenship status?
Promoting the full inclusion of citizens of immigrant origin might include duties as well as rights for both the immigrants and the nonimmigrant population, and it may involve not only formal rules but also things like informal norms, incentives, practices of recognition, and conceptions of national identity. This question about inclusion requires us to explore issues relating to economic opportunity, multiculturalism, social interaction, and many other matters. For reasons of space, I will not spend any time spelling out the details of my analysis here. My main concern is to contrast this way of framing the problem of democratic inclusion with the one that arises from the All-Affected Principle, which is itself often presented as a (the?) principle of democratic inclusion, as the title of this volume illustrates. The All-Affected Principle is usually understood as a principle concerned with the question of who is entitled to participate in decision making, and as such, it does not invite attention to the kinds of concerns I am trying to explore in raising the question of full inclusion. Again, I am not saying that the All-Affected Principle contradicts the idea that these issues matter or precludes concern with them, but only that they are not the sorts of issues that the principle either naturally brings into view or offers much help in addressing once they have been raised.
Legal Rights of Legally Admitted Noncitizens
Some immigrants are not citizens, at least not yet. How should their legal rights resemble or differ from the legal rights of citizens? Again, this question is posed within the constraints of the presuppositions noted above, and again the question is not the procedural one of who should participate in deciding what these rights should be, but the substantive one of what those rights should be and why.
Let’s start with noncitizens who are permanent residents and with the actual practices of democratic states. The striking fact is that permanent residents now enjoy almost all of the legal rights that resident citizens enjoy except for some political rights (voting, running for office) and a few other relatively minor matters. We tend to think of citizen/noncitizen as the key dividing line when it comes to legal rights, but in reality, the key dividing line is resident/nonresident.3 The current practice of granting extensive legal rights to permanent residents is a major change from the practices in most democratic states in the late nineteenth and early twentieth centuries when there were sharp differences between the rights of resident citizens and resident noncitizens. In my view, this change was something that was morally required because moral claims to many legal rights, especially social and economic rights, derive from membership in society, which derives in turn simply from living in the society over time. (I do not claim that moral reasoning caused the change, however.)
Would the All-Affected Principle help us to see why residents ought to enjoy the same rights as citizens for the most part? Not if the primary focus is on participation in collective decision making. That’s the one area where resident noncitizens do not enjoy the same rights as citizens, which is why it is important for immigrants to have easy access to citizenship. But many permanent residents choose not to become citizens even when they can do so easily, in part because they enjoy almost all of the legal rights that citizens enjoy except the right to participate fully in the political process.
Immigrants are often admitted on a temporary basis at first and sometimes they are required to leave after a certain period of time. That is, they are not ever on a path to permanent residence. One could ask whether it is morally permissible to admit people with such restrictions on their ability to remain, but I want to bracket that question at this stage of the discussion, simply assuming (as most people do) that states are morally entitled to admit people on a temporary basis.
Assuming that temporary admissions are morally permissible does not mean that states are morally free to treat temporary residents any way they choose. Even with that assumption, we can and should ask, “What legal rights should temporary residents have?” In my view, temporary admissions must be truly limited in time or the state forfeits the (presupposed) right to require the immigrants to leave, and temporary immigrants are morally entitled to a wide range of legal rights, including many but not all of the legal rights that permanent residents enjoy.
I don’t pretend that my position is obviously correct or that there is no reasonable basis for disputing it. For example, some have argued that it would be morally preferable to grant temporary workers fewer legal rights because rich democratic states would then be willing to admit many more temporary workers.4 I do not have the space to explore the arguments for and against these different positions here. My main point is simply that the All-Affected Principle does not help us address this question.
Some might object that it could. The argument goes like this: Potential temporary workers have important interests both with regard to the terms of their admission and with regard to the numbers admitted. Therefore, the All-Affected Principle implies that these potential temporary workers should have a say in the tradeoff between these competing concerns.
From my perspective, this sort of argument illustrates the dangers more than the virtues of the All-Affected Principle when one applies it on an ad hoc basis. The All-Affected Principle works best, it seems to me, in contexts in which there are relatively clear parameters for the morally permissible options in some collective decision-making situation (or at least in which one assumes this to be the case for purposes of immediate analysis) and the question is who should be able to participate in identifying those options and choosing among them.
It is not enough to ask who should have a say. Some options should be off the table. If my argument that temporary workers are morally entitled to certain rights is correct, then depriving workers of those rights is not a morally permissible option for a democratic state, even if workers agree to their removal. The fact that some or even most potential temporary workers might consent to forego those rights in order to increase their chances of getting in does not, by itself, establish this as a morally permissible option. Desperate people will agree to almost anything.
Even most advocates of reducing rights for temporary migrants do not go so far as to argue that the bundle of rights possessed by temporary migrants should depend simply on what the receiving state and the migrants would agree upon. They criticize actual policies, like those of the Gulf States, which admit large numbers of migrants with only a temporary permit to stay, no matter how long they remain, and with very limited rights. The fact that no one forces migrants to go to these states does not make that package of policies morally permissible for any state, much less one committed to democratic principles. The All-Affected Principle is a principle to guide our thinking about democratic decision making in contexts of collective action, and it has a valuable role to play in that context. It should not be reinterpreted either as a version of utilitarianism or as a version of libertarianism.
Irregular Migrants
Turn now to the issue of irregular migrants (i.e. people who have entered and/or settled without the state’s permission). What legal rights should they have? Again, the background presuppositions of this question are crucial. I have been assuming that states are morally entitled to control immigration, at least for the most part, because that is the conventional view of the matter. So, it is useful to examine the claims of irregular migrants within the constraints of that view, even if one might want to challenge that presupposition in another context (as indeed I do).
Even if one accepts the conventional view, irregular migrants are morally entitled to a range of legal rights. I think that many of these legal rights should be protected by a firewall between those responsible for protecting these rights and those responsible for enforcing immigration rules. I also think that irregular migrants acquire a moral claim to legal status over time simply by living within a society.
My views on this issue have not gone uncontested, to put it mildly, but the important question for this chapter is not what position is correct but whether the All-Affected Principle helps us to think about this topic of irregular migration. So far as I can see, the answer to that question is “no.”
Neither my own arguments on this issue, nor the counterarguments that I have seen, appeal explicitly to the All-Affected Principle, and the arguments on both sides do not seem to me to flow from or to fit well with a concern for the question of who should participate in collective decisions on this issue. So, I do not see how the All-Affected Principle advances our thinking about irregular migration. Indeed, I worry that the discussion of the issue of irregular migration would be impoverished rather than enriched if one used the All-Affected Principle as a primary lens through which to view the issue.
Admissions
Turn now to questions about criteria of selection and exclusion of potential immigrants. The conventional view is that states are normally free, not only legally but also morally, to admit as many or as few immigrants as they choose and to decide what selection criteria to use with respect to admissions (e.g. education, skills, age, more distant family ties, etc.). But wide discretion is not absolutely unfettered choice. Even people who endorse the conventional view normally recognize that there are some immigration policies that would be morally wrong. So, what are the moral constraints on admissions policies within the conventional view? (Notice again how this question deliberately accepts the conventional view as a presupposition for certain analytical purposes.)
Set aside for the moment the issue of refugees, which will receive separate treatment below. One important negative constraint on admissions policies is nondiscrimination. At a minimum, most people who see themselves as committed to democratic principles think that democratic states ought not to exclude potential immigrants on the basis of race or religion. It is true that democratic states openly discriminated in the past and sometimes try to discriminate today without acknowledging that they are doing so (as with Donald Trump’s efforts, while he was President, to exclude Muslim immigrants). It is also true that this norm against discrimination is increasingly being subjected to overt challenges in ways that would have been unthinkable in the previous twenty or thirty years. Even so, the norm persists, and this constraint is reflected in various ways in national laws and international conventions, including the need to try to conceal the fact that one is discriminating when one does so.
On the positive side, most people recognize that the immediate family members of citizens and permanent residents have particularly strong moral claims to admission if they are not yet present. There are some qualifications to this duty and states do not always fulfill it, but, again, the principle is widely accepted, and it is widely reflected in laws and policies.
As always, this brief summary ignores many complications and complexities. But the question for this chapter is whether the All-Affected Principle helps in thinking about the normative limits on admissions policies within the constraints of the conventional view or whether it draws our attention to related but neglected questions. Again, I think the answer is no. The limited constraints on state discretion that I have identified (nondiscrimination and family reunification) flow not from the nature of the decision-making processes but from independent moral values that are supposed to constrain decision-making processes in this area. So, I don’t think the All-Affected Principle would affect the debate on the question I have posed.
Does the principle pose new and neglected questions that we ought to consider, perhaps about the ways in which current immigration policies fail to take into account the interests of citizens affected by immigration? I cannot rule out that possibility, but I have to say that I am skeptical that the All-Affected Principle will help to provide much moral guidance. The priority given to family reunification is clearly a response to affected interests, but it protects those interests by trying to remove them, at least to some extent, from the conventional calculations about interest that normally drive immigration policy. The same might be said of nondiscrimination rules.
Set aside these sorts of moral constraints. Some current citizens are undoubtedly more affected by immigration than others, but there are often disputes about what the effects of immigration are and about which effects are legitimately a subject of collective concern. The important general point is that in this respect – i.e. that a particular policy has a differential and contested impact on the interests of citizens – immigration policy is no different from most public policies. If we find a better way than the one provided by existing institutions to create more effective links between the nature and extent of a citizen’s ability to influence a public policy and the ways and extent to which that citizen’s interests are affected by the policy in question, and if we think that is desirable, as the All-Affected Principle might seem to prescribe, this is likely to require a wide transformation of the overall processes of democratic decision making within the state. In that sort of enterprise, the specific features of immigration policy are likely to seem relatively unimportant.
Perhaps someone will object that focusing only on the ways in which the interests of citizens are taken into account is a mistake. What the All-Affected Principle does is to draw our attention to the interests of those who are not citizens but who are affected by this policy. After all, those seeking to migrate clearly have an important interest in whether or not they will be admitted, and those who stay behind also have important interests at stake because they may be benefitted by the emigrants’ departure (e.g. through money sent home) or harmed by their departure (the brain drain). Wouldn’t the All-Affected Principle require that these people have a say in immigration policies as well?
This question just illustrates why it is so important to be clear about the presuppositions of one’s inquiry. Remember that the question I am asking here simply presupposes the moral validity of the conventional view in order to make it possible to see that the conventional view does contain certain (modest) moral limits on what states may do with respect to immigration. But the wide latitude provided to the state by the conventional view can exist only if a state has no moral duty to consider the interests of people outside its own population in constructing its immigration policy. The state is morally entitled to be self-interested (however that self-interest is defined) in what it does with regard to immigration so long as it respects the sorts of constraints I have identified. A state can choose to be generous if it wishes and take the interests of some external group into account, but it is under no moral obligation to do so. Of course, as I have noted before, we can refuse to adopt this presupposition and challenge the conventional view. The All-Affected Principle provides one way to do that, though there are other ways as well, as we shall see. But it is important not to introduce this sort of challenge in an ad hoc way, rejecting the conventional view for some purposes in a given argument but implicitly relying upon it for others in the same argument.
Refugees
Before turning to the challenge to the conventional view, I want to mention one other way in which the conventional view is constrained, even on its own terms. Most democrats think that refugees have a special moral claim to admission. Again, that is a view that has been under much sharper challenge in recent years than it was for most of the post–Second World War period, but, as with nondiscrimination, all democratic states have signed international conventions and passed domestic laws recognizing the special claims of refugees to some extent. My own view is that even within the constraints of the conventional view, democratic states have much stronger duties in this area than they have recognized. As always, I do not have the space to spell out those arguments here. Rather, I want to draw attention to one way in which I think the All-Affected Principle can be helpful in identifying an important and understudied question about refugees, namely, the question of who should decide where refugees are to settle when they need a new home.
The All-Affected Principle says that people (significantly) affected by a decision should have a say in that decision. Where one lives is something that has a major impact on most people’s lives. The existing refugee regime organizes decisions about where refugees will live in two, very different ways: asylum and resettlement.
Under the asylum regime, which all democratic states have accepted legally, where refugees will live depends primarily on the place where they first ask for protections, which in turn depends upon their ability to travel. So, it is a system that gives almost no voice (in principle) to the receiving state on the question of how many refugees will be admitted and which ones (though, of course, many states take various steps to prevent refugees from arriving on their territory). How much say the refugees themselves have in this matter depends very heavily on their economic and other resources.
In contrast to granting asylum, admitting refugees for resettlement is an entirely voluntary practice, in which only a few states participate, which involves admitting people who have been recognized elsewhere as refugees and providing them with a new home. This is a process in which the refugees themselves have almost no say about where they will go. Of course, they do have to agree to go to whatever state is offering resettlement, but when the alternative is remaining in a refugee camp, this is often not much of a choice.
A just refugee regime (if I may use that oxymoron) would clearly distribute the responsibility for refugees much more widely. One of the questions one would have to address in thinking about a just refugee regime is how much choice such a regime would provide to the refugees themselves in deciding where they would ultimately live and how much choice it would provide to states in deciding how many and which refugees to admit. Although the prospects for creating anything remotely resembling a just refugee regime seem quite remote at the moment, it can be helpful, nevertheless, to reflect upon these sorts of fundamental questions, if only as a way of providing clearer critiques of some of the efforts to defend existing arrangements. In pursuing such reflections, I think that the All-Affected Principle would provide a valuable reminder of the need to think about who ought to have a say about where refugees would go under a just refugee regime.
Open Borders
Consider now the possibility of challenging the conventional view in a more fundamental way. Suppose we stop treating the idea that states are generally entitled to control immigration as a presupposition and ask instead: “Is it true that states are morally entitled to wide discretion with respect to who enters and lives within their territory?” If one accepts some basic moral assumptions, such as the idea that all human beings are of equal moral worth and that social institutions must be compatible with that moral equality in order to be justifiable, one might well conclude that the current global order is not morally acceptable. The way the world is currently organized serves the interests of the few (i.e. those living in rich democratic states) much more than the interests of the many (i.e. most of those living elsewhere in the world), and giving states discretionary control over immigration is a crucial factor in maintaining this (unjust) order. From this perspective, asking questions (as I was doing earlier) about what morality requires with respect to immigration while simply assuming the contemporary world as the background context is like asking what autonomy for women requires while simply assuming a heteronormative, patriarchal society as the background context.
In my view, a just world would be one in which the economic and other differences between political jurisdictions would be greatly reduced. In such a world, people should and would also be largely free to move across jurisdictional boundaries and settle where they chose, and this sort of freedom would not generate huge problems.
This brief summary leaves out lots of nuances and qualifications and does not consider the many important objections to the position I have just outlined. Nevertheless, I hope that it is sufficient to highlight a few points for the purposes of this chapter. First, it illustrates the ways in which one can adopt presuppositions for certain analytical purposes and then move beyond them for other purposes. Second, it illustrates the point that even fundamental critiques of the status quo do not proceed without presuppositions. My open borders position presupposes the moral equality of humans and that social institutions serve certain purposes and require certain sorts of justifications. Third, and most importantly for this chapter, my claim proceeds without appealing to the All-Affected Principle. The argument for a just world with open borders that I have outlined is an argument about what substantive arrangements are compatible with justice, not an argument about who ought to participate in decisions.
Of course, if one probed further, one would quickly have to recognize that there would inevitably be many important areas of indeterminacy, even in a just world, and so questions would then arise as to how collective decision making should be organized to deal with such issues. And in that discussion, the All-Affected Principle would certainly have a place (though it might not be the only principle one would want to consider). For my immediate purposes, however, the important point is that one can construct a fundamental inquiry into, and even a fundamental challenge to, the idea of discretionary control over immigration without relying upon the All-Affected Principle.
There is, however, another way in which the All-Affected Principle can be brought to bear on the open borders debate which does focus on participation in a decision-making process. This is what Arash Abizadeh does in his well-known article on the unbounded demos, in which he argues that from the perspective of democratic theory, states are not morally entitled to decide unilaterally on immigration matters because border controls greatly affect the excluded, and so those entitled to participate in decisions about closure include, in principle, anyone who might want to move from one state to another.5 This then is another important exception to my claim that the All-Affected Principle does not help much in thinking about immigration.
Abizadeh actually constructs his argument on the basis of a more restricted principle of democratic theory, namely that all those subjected to coercion are entitled to participate in the decisions that coerce them, but he notes that the more expansive All-Affected Principle leads to the same conclusion.6 Abizadeh’s approach differs from my own, and I think that it adds something valuable to normative discussions of immigration through its strong link to claims to participation. As he makes clear, however, this participation-focused account leaves open to some extent the outcome of such a democratic process. He contends that democratic principles will almost certainly lead to porous borders but that they might not require open borders in my sense of the term.
As I read Abizadeh, he simply leaves open the question of whether there are reasons independent of democratic theory for requiring borders to be open. So, I don’t think that his democratic theory argument conflicts with the reasons I have offered for thinking that open borders are required as a matter of substantive justice. Our arguments are complementary rather than in conflict. On the other hand, from my perspective, the fact that an analysis that starts from a version of the All-Affected Principle leaves open the question of whether or not borders would be open illustrates again the potential dangers in relying only on procedural principles in normative discussions.
This is not a critique of Abizadeh’s analysis, which is very explicit about its goals and presuppositions and about the limits of his claims. As I noted at the outset, everyone has to bracket some questions in order to explore others. It is simply a cautionary note again about the importance of substantive claims about justice and the related limitations of the All-Affected Principle, even when that principle advances our understanding in some important respects.
Who should have a say in a given decision for it to count as democratic? This is the question with which the democratic boundary problem is concerned. Three main “solutions” have emerged in the literature: the All-Affected Principle (AAP), the All-Subjected Principle (ASP), and the Affinity Principle (AP).1 These principles respectively hold that, from a democratic point of view, a say should be given to all and only those affected by a decision, all and only those subjected to a decision, and all and only those who share national, social, or cultural affinities.2
As things stand, the AAP and ASP are the “front-runners” in the race for the best solution to the boundary problem, with the AP lagging somewhat behind. And as several scholars have observed, both the AAP and the ASP come with radically expansive implications as far as the scope of the franchise is concerned.3 Subscribing to either of them implies that democracy should go – to a greater or lesser degree – global.
This is, in a nutshell, the state of the debate on the boundary problem.4 My aim in this chapter is to question the presuppositions underpinning this debate. Scholars have proceeded by taking democracy for granted, treating it as an ultimate value. Consequently, the best solution to the boundary problem has been framed as the one that most loyally reflects the value of democracy. But it is not at all obvious that democracy is best conceptualized as an ultimate value. Arguably, democracy marks out a family of decision-making systems – characterized, at a minimum, by universal suffrage, free and fair elections, and broadly majoritarian voting procedures – that are themselves justified by appeal to how they reflect and promote important values in particular circumstances.5 The values in question range from equality and self-determination to peace, security, and respect for fundamental rights. In other words, what we call “democracy” is itself one of several possible solutions to the boundary problem (i.e. the problem of who should take part in a given decision) – a solution that is contingently justified by appeal to a variety of different values. Or so I shall argue.
The chapter proceeds as follows. First, I explore different moral dimensions of decision making – the “who,” “how,” and “what” dimensions – and locate the boundary problem in relation to them. Next, I explain that democracy is one possible and contingently justified answer to the “who” question. In the third part, I revisit the most popular purported solutions to the boundary problem: the AAP, ASP, and AP. I show that all of them fail to offer general principles for allocating decision-making entitlements. Instead, they often pick out features that matter to how decisions should be made. Overall, this leads me to conclude that the search for a “general democratic principle” to answer the boundary problem is misguided.6 Finally, I support this conclusion by looking at two real-world cases: the US presidential election and the Brexit referendum. I argue that neither the AAP nor the ASP nor the AP makes sense of the wrong involved in disenfranchising those who, intuitively, ought to have been given a say in these decisions. By contrast, taking into account a broader set of considerations that bear on the distribution of decision-making entitlements sheds light on this wrong.
Three Dimensions of Decision Making
Whenever a decision has to be made, there are at least three questions we may ask in relation to it: the “who,” “how,” and “what” questions.
Who should make the decision? Who should have the moral power to issue authoritative pronouncements about the matter at hand? (This is the question the boundary problem is concerned with.)
How should the decision be made? Through what process (e.g. reasoning, data gathering, consultation, etc.) should the decision makers make the decision? What considerations should they take into account?
What should the content of the decision be? What is the right answer to the question at hand?
To illustrate the difference between these dimensions of decision making, consider the following three issues, which we assume are up for decision.
Issue A: To which charity should Marc’s money go?
Issue B: Which questions should feature in the exam for the undergraduate module “Contemporary Political Theory” (CPT)?
Issue C: Who should receive an offer for an assistant professorship in politics at University College London (or any other university)?
In relation to issue A, if it is Marc’s money we are talking about, then presumably it is up to Marc to pick the relevant charity: this is our answer to the who question. Furthermore, in deciding which charity should receive his beneficence, Marc should consider the importance of the goals promoted by different charities, and the effectiveness and dedication with which each pursues them. This appears to be the most natural answer to the how question. Regarding the what question, probably several charities are worthy recipients of Marc’s generosity, and several are not, due to corruption or inefficiency.
Turning to issue B, intuitively, whoever is in charge of CPT – i.e. whoever teaches it – ought to determine the structure and content of the relevant exam. Or, at least, this is how most universities operate. In setting the exam (the how question), the course convenor should consider the material covered and what students may be reasonably expected to know. Finally, with respect to the what question, the exam should be neither too easy nor too difficult, and allow markers to differentiate between well prepared and poorly prepared students. Here too several possible exam papers would presumably count as fair tests of the candidates’ knowledge.
While, with regard to issues A and B, answers to the who question are intuitively straightforward, issue C is a little more complicated. I would imagine that most would fall back on whatever departmental practice is adopted in the institution(s) where they currently work, or for which they have worked in the past. (Yes, I suspect my readers are academics.) Some would likely answer that the entire department should decide whom to hire, through a majority vote after deliberation. Others might insist that a committee of department members, appointed by the head of department, should make the decision. Others still might suggest that the committee should include members of the hiring department, members of other departments, as well as external experts.
This variation has some significance, and I will come back to it shortly. For now, it suffices to note that plausible answers to the who question will typically involve some combination of members of the university advertising the position and, possibly, third-party experts. Regarding the how question, decision makers should go about deciding whom to hire by carefully considering the candidates’ strengths and weaknesses as researchers, teachers, and prospective colleagues, with sensitivity to considerations of inclusion and diversity. Finally, the answer to the what question should be: “the best candidate.” Of course, there is likely to be reasonable disagreement about who that is. But even in those rare cases where a candidate is ranked top by every decision maker, they may turn out to be a disappointment. Giving right answers to the who and how questions does not guarantee the right decision outcome.
I have offered these three examples to illustrate the differences between the who, how, and what questions.7 Let us now zoom in on the who question in particular, and ask what reasons support our intuitive answers in each of the cases discussed. Regarding choosing a charity (issue A), the reason why Marc should be ultimately in charge is that it is his money at stake. It would be a violation of his autonomy if he were forced to donate to one charity or the other, contrary to his will. Furthermore, the very point of the practice of charitable donation – as opposed to, say, taxation – involves voluntary transfers from donors to recipients. To the extent that this voluntariness strikes us as valuable – say, because it expresses a sincere willingness to help others – that value can only be preserved by letting each individual decide where their donation should go.
Moving on to setting exam questions (issue B), here the justification for letting each course convenor be in charge probably appeals to values other than autonomy, e.g. education, fairness, etc. On the whole, assigning decision-making authority such that each course convenor sets the relevant exam papers is likely to ensure the best results – fairest, best-designed exams – overall. Since this distribution of decision-making authority typically promotes the values behind the practice of a university, it is, all things considered, justified.
Finally, as we have seen, when it comes to procedures for appointing a new assistant professor, a variety of possibilities appear intuitively plausible. Each of them has its advantages and disadvantages. The more-inclusive procedures give more people a sense of ownership over the new hire, but in sufficiently large and internally divided departments, they might exacerbate conflicts and result in stalemates. The less-inclusive ones are more agile and efficient, but may not be as accurate or as conducive to building a strong sense of joint ownership within a department. Which procedure is best, it seems to me, will depend on local factors, including the relevant departmental culture, composition of the department, and so forth. But once again, how decision-making power ought to be distributed rests on what best furthers the point and purpose of the practice of a university (within the constraints of fundamental rights): high-quality teaching and research.
In sum, depending on the particular practice and context at hand, different values and considerations will bear on our answer to the who question. With this general theoretical background in place, we can now turn to one particular answer to the who question, and its relevant context of application: democracy within political communities.
Democracy as a Rule of Regulation
The term “democracy” refers to a family of decision-making procedures, which we perceive to exhibit several moral virtues. Disagreement immediately arises, however, whenever we ask for a precise definition of democracy.8 This, I suggest, following Arrhenius, is because the term “democracy” is associated with both a particular set of institutions – involving free and fair elections, universal suffrage, and broadly majoritarian voting procedures – and the values that justify those institutions, such as procedural equality, substantively correct outcomes, stability, solidarity, security, inclusion, non-domination and so on.9
I suggest that, for clarity’s sake, we should use the term democracy to refer to a set of (contingently justified) institutions, rather than to the value or set of values those institutions are supposed to embody.10 To put the point in a language familiar from contemporary debates about justice, it may be best to think of democracy as a particular family of “rules of regulation,” the justification of which goes back to some fundamental values in conjunction with a variety of empirical facts.11 This terminological recommendation is well motivated. After all, if we ask ourselves why, within a given political community, decision-making power should lie with all citizens, via their elected representatives, our answer won’t itself appeal to “the value of democracy.” Such an answer would be uninformative. Instead, the answer will point to how allocating decision-making power in this way favours stability, respect for fundamental rights, expresses citizens’ equality in the circumstances at hand, and so on.
Furthermore, understanding democracy to designate rules of regulation is consistent with the widespread conviction that democracy cannot be the correct way of allocating decision-making power no matter which polity one is looking at. Polities characterized by considerable internal divisions, instability, and lack of mutual trust cannot be easily governed democratically. Instituting a democratic decision-making system when the conditions for its institution are not ripe can be inimical to the very values that justify democracy under the right circumstances.12
The foregoing observations, I should emphasize, are not dependent on a purely instrumental account of the value of democracy. They also hold for views that regard democracy as intrinsically valuable because it expresses a certain kind of respect for persons.13 After all, whether universal suffrage, free and fair elections, and broadly majoritarian decision procedures express that kind of respect depends on the background circumstances at hand. It is hard to see how a democratic institutional setup could be said to embody a commitment to equal respect if, when implemented in certain settings, it would foreseeably lead to chaos and instability.14 Similarly, the equal right to vote has now acquired a certain symbolic meaning, but this meaning is likely to be the result of a contingent historical process.
The unhelpfulness of treating democracy as an ultimate, context-independent value can be observed in debates about the justification of judicial review of legislation. By judicial review, I mean the practice – most typically exemplified by the US Supreme Court – of suspending the application of, or striking down, legislation approved by elected representatives. A good portion of the scholarly debate about judicial review concerns its democratic credentials (or lack thereof). Theorists such as Jeremy Waldron regard it as undemocratic, since it gives ultimate decision-making power to unelected judges, who are not representatives of the people.15 Theorists such as Ronald Dworkin, by contrast, argue that judicial review can be democratic, provided judicial decisions promote the egalitarian values at the justificatory heart of democracy.16 Others still, such as Corey Brettschneider, see judicial review as always “suboptimally democratic,” though itself contingently justified by appeal to democratic values.17
The use of the language of democracy, which dominates this debate, obscures rather than clarifies matters, precisely because “democracy” is sometimes used to refer to specific decision-making procedures and sometimes to the values that justify them. The debate on judicial review would be more fruitfully conducted if it were couched in terms of judicial review’s overall justification. This will likely appeal to the values that also justify democracy, including stability, procedural and substantive equality, respect for human rights, and much else. Whether judicial review is justified or not will depend on whether it facilitates or hinders expression and protection of these values, compared to feasible alternatives.18 These values, however, are not the same as democracy, even if several of them may contingently justify what we routinely describe as democratic decision procedures.
These considerations, about how best to understand the conceptual domain of democracy, support the hypothesis that there is something wrong-headed in framing the boundary problem as essentially democratic. This framing treats democracy as an ultimate value, and then proceeds to ask which distribution of decision-making power is most responsive to it. But, as I have suggested, democracy is best understood as designating a particular institutional distribution of decision-making power, the justification of which refers back to several values. The question we should be asking, then, is: Which distribution of decision-making power is justified in any given context, in light of the values that bear on the relevant practice and feasibility constraints? We should not be debating about which distribution of decision-making power is most democratic.
The Boundary Problem Reassessed: The No-General-Solution Thesis
The previous two sections have offered arguments in defence of the following theses:
a) the question “who should be given a say” ought to be answered in relation to each given practice – and, specifically, its underlying values – in light of the constraints imposed by the context in which the practice operates (first section);
b) “democracy” is a contingently justified answer to the who question, one that itself appeals to a variety of different values (second section).
Theses (a) and (b) should make us sceptical about the possibility of identifying a one-size-fits-all solution to the boundary problem: a general principle determining the scope of the franchise. And, as anticipated, they should also make us suspicious of the assumption that the relevant solution should always be democratic. Yet, the most prominent answers to the boundary problem – the AAP, ASP, and AP – have tried to do just that: articulate a general answer to the question of what makes a certain distribution of decision-making power democratic. To further support my theses, I critically discuss the AAP, ASP, and AP in turn. Doing so will reveal that each principle either captures considerations relevant to answering the how and what questions, or correlates with – but does not embody – considerations relevant to answering the who question.
Examining the AAP
The AAP cannot provide a general answer to the “who should decide” question. As Robert Nozick famously put it, the decision to marry someone is for each person to make. Potentially rejected suitors, albeit deeply affected, should have no say in it.19 Similarly, although candidates for an assistant professorship are highly affected by hiring decisions, it would be absurd to include them in the decision-making process.20
Proponents of the AAP might reply – as they typically do – that the scope of the AAP is restricted.21 The AAP is meant to operate within the constraints of individual rights (e.g. the right to choose a romantic partner) and of group rights (e.g. the right of a university to hire its employees). Even with this scope restriction, the success of the AAP as a principle addressing the boundary problem remains doubtful. Recall how, in the university hire example discussed above, even once we have ruled out candidates taking part in a decision, a number of different possibilities remain on the table as to who should be included. These range from all faculty in the department, to a committee of department members, to a committee of both department members and external assessors.22 While the first of these possibilities may be “rationalized” by appeal to the AAP, the other two cannot (or at least not straightforwardly).23 But it seems mistaken to suggest that only the first one ought to be adopted by any department or university.
As I mentioned earlier, each procedure has its virtues and vices. Which is best depends on the circumstances. And when circumstances take a certain shape, then some scope-restricted version of the AAP will correspond to the “right answer” to the boundary problem. Note, however, that the AAP won’t be what justifies the ascription of certain decision-making entitlements in that case. Instead, the AAP will just happen to match the ascription of decision-making entitlements recommended by underlying considerations about values and feasibility constraints.
Where the AAP seems to have greater – though still less than perfect – purchase is in relation to the how dimension of decision making. As other scholars have already pointed out, while it is doubtful that all those affected by a decision ought to participate in its making, it seems plausible that when a decision needs to be made, decision makers should take the interests of those affected into account.24 So, when deciding whom to marry, I should be sensitive to the feelings of the potential suitors I reject. Equally, when a small committee decides whom to hire, they should be sensitive to the interests of all department members.
Even with respect to the how dimension of decision making, though, the AAP fails to qualify as a fully general principle. Consider, again, the simple case of an academic hire. Even though those most affected are, arguably, the candidates themselves, taking their interests into account – beyond treating them professionally and fairly – is not what selectors ought to do. Doing so would be contrary to the purpose of the practice of hiring.
In sum, the AAP appears to be neither a general principle for addressing the boundary problem, nor a general principle articulating how decisions should be made. Although it may, on occasion, match our considered judgements about who should participate in a given decision, it does not offer a plausible explanation of those judgements, which are ultimately traceable to our assessment of (i) the values at stake and (ii) which allocation of decision-making entitlements best honours and realizes those values in the circumstances at hand.
Examining the ASP
The ASP states that all and only those who are subjected to a decision ought to have a say in it. This formulation is ambiguous between at least two interpretations of subjection. The first sees subjects as the targets of coercion. A, in this case, is subjected to B, if and only if B coerces A (i.e. if B forces A to perform certain actions by threatening sanctions). The second focuses on someone’s being the addressee of certain de facto authoritative commands. A, in this case, is subjected to B, if and only if B issues authoritative commands directed at A. Everyone falling within a state’s jurisdiction counts as subjected to the state’s commands in both senses.25
The ASP too, in both formulations, is not a plausible general principle for the allocation of decision-making entitlements. Consider, for instance, the coercion interpretation of the principle. If you insist on entering my apartment, I (or the police) may well permissibly coerce you to prevent you from getting in. It would seem absurd to suggest that such coercion could only be justified if you had participated in my decision to keep you out. While, to be sure, the use of coercion is subject to strict normative standards, those standards are typically not participatory. Coercion may be legitimately employed to protect people’s rights, and what people’s rights are is, at least to some extent, independent of the outcomes of collective decision-making procedures. Instead, those outcomes can be valid only if they respect the relevant rights.
It might be objected that although some rights are independent of collective decision making, others are not. The latter are rights on which there is reasonable disagreement, but whose content needs to be settled. This means that respect for everyone who will be subjected to coercion in the name of those rights requires them to contribute to deciding what the boundaries of those rights are.26
The difficulty with this argument is twofold. First, it is unclear why only coercion in the name of certain (potentially controversial) rights is disrespectful towards the prospective coercees, if they have not themselves taken part in the decision-making process determining the contours of those rights. What is so special about coercion as compared to, for example, affectedness more generally? After all, on the face of it, both can be inimical to individual autonomy, though in different ways: coercion through will bending, affectedness through reducing one’s options, sometimes quite dramatically.27 But if the rationale for focusing on coercion collapses into affectedness, then we already know, as stated earlier, why it fails to offer a general account of how entitlements to participation should be allocated.
Second, even if we just focus on coercion narrowly construed (i.e. as involving the threat of sanctions), it is unclear why participation is necessary to make such coercion in the name of reasonably contested rights justified. The argument here often refers back to how participation preserves individuals’ autonomy. But this argument is known to be mostly metaphorical. It is unclear how an outvoted minority coerced in the name of what it regards as a mistaken understanding of rights preserves its autonomy through having participated. More likely, participation in such cases contributes to conveying, symbolically, the equal status of those coerced and to fostering a sense of ownership over the relevant decision and broader political institutions. These are, of course, important values, but they are, as we have seen, not the only ones that matter when it comes to allocating decision-making entitlements. Stability, respect for fundamental rights, and much else matter too – at least in contexts we regard as paradigmatically “political.” So, while there may be contingent reasons in favour of some form of participation in decisions that coerce one, those reasons don’t add up to vindicating a general principle.
Let us now turn to the second meaning of subjection, concerning one’s claim to authority over others. Here, too, reflection makes it apparent that the ASP could not serve as a general basis for distributing decision-making entitlements. Consider a teacher and her pupils. She clearly claims authority over them, in certain domains. For instance, she claims authority over the content of the exams she will set, she claims authority to release students from the obligation to remain seated during her lecture, and so forth. It would appear absurd to suggest that her claim to authority is invalid unless students also contribute to making the relevant decisions. Such a principle would be contrary to the purpose of educating children.
Even once we move to broader contexts – including entire political communities – the validity of claims to authority does not seem systematically dependent on participatory entitlements being granted to those who are subjected to the relevant authority. To be sure, participation may (again) contribute to generating a sense of ownership over a given decision, and to conveying a special type of respect towards the subjects of authority. But it is unlikely to be sufficient (or necessary) for a claim to authority to be justified.
This can be easily seen by considering a scenario inspired by similar ones famously proposed by Robert Nozick and John Simmons.28 Assume that I, and some others in my neighbourhood, want to set up a street-beautifying scheme. You’re also a resident in the neighbourhood, but are not particularly interested in making our streets prettier or cleaner. The neighbourhood holds a referendum on whether to institute this scheme. You, qua resident, are given a right to vote. However, you don’t exercise that right. The majority of voters select a fairly expensive street-beautifying scheme. This results in a decision to implement the scheme that is presumptively binding on you. Does the fact that you were given a say validate the claim to authority made by the neighbourhood and place you under an obligation to contribute? The intuitive answer appears to be “no,” which suggests that entitlements to participate in a decision – even assuming its content is fully morally acceptable – are insufficient for claims to authority to be justified. In sum, the ASP too – in its different interpretations – does not offer a general answer to the boundary problem.
Examining the AP
Finally, I very briefly turn to the Affinity Principle (AP). This holds that, when it comes to political decisions, those who belong to the same nation or people should be given a say. I won’t dwell much on the AP, since it has already been successfully challenged by others. I just limit myself to noting that one’s cultural, social, or national affinities appear irrelevant to the issue of whether one should be given a say in certain decisions as a matter of principle. What the AP, however, seems to point to are the empirical conditions under which institutions we would most readily define as democratic can function reasonably well.29 These are conditions characterized by sufficient mutual trust, commonality of interests, and mutual understanding. If so, the AP may be a proxy for one or more of the de facto conditions under which a democratic distribution of decision-making power, within a certain context, is justified.30
We have seen that the AAP, ASP, and AP do not offer general principles justifying the allocation of decision-making power. This further corroborates my hypothesis, namely that there is no single solution to the boundary problem: different assignments of decision-making power are justified in different circumstances.
Implications
To complete my discussion, I now wish to focus on two recent political events that raise the “who should decide” question quite prominently: the 2016 referendum on the UK’s membership of the European Union (EU), and the 2016 US presidential election. Let me give you a bit of context for my choice of focus. I am an Italian citizen, and until December 2020 I was a resident in the UK, where I had lived, with short interruptions, for several years. Like any other non-British European citizen, I had no say in the decision about whether the UK should remain in the EU. And like any other non-US citizen, I had no say over whether Hillary Clinton or Donald Trump should be the next US president. I have found both forms of disenfranchisement somewhat troubling, but the former considerably more than the latter. What could explain these feelings?
I know I am not alone in having felt like this – at least within the relevant demographic – so I hope that taking my intuitions as provisional “data points” to be explained is not too idiosyncratic. I will attempt to show that, while neither the AAP, nor the ASP, nor the AP can make sense of how I have felt in relation to the Brexit vote and US presidential election, looking at the plurality of factors that bear on the who question can.
Consider the AAP first. It is far from clear that my stakes in the Brexit decision were greater than my stakes in the US presidential election. Arguably, the reverse was the case. To be sure, I had lived in the UK for many years, I identified with the UK a lot more than I did with the United States (for that very reason), so my immediate moral intuitions may have been clouded by a sense of disappointment, rejection, and alienation. But the truth is, there is only so much damage that Brexit could do to someone in my position. If things took a turn for the worse, I told myself, moving back to the European continent, for instance, would be a feasible alternative – as it turns out, an alternative I eventually took, as I am now residing in Germany. And even a significant downturn in the British economy wouldn’t have had deleterious consequences for someone in a fairly privileged position, like myself.
Who becomes US president is, bluntly put, a much bigger deal. I can (and did) escape Brexit, but I doubt anyone could escape the Trump presidency, no matter their location in the world. Just consider Trump’s statements about climate change, his lack of diplomatic skills, his “America first” mantra, and his war-mongering tendencies. I was and still am upset about Brexit, but I was a lot more anxious about Trump (and still am, although he is no longer president). This alone suggests that when it comes to stakes, at least in my perception, the much more consequential decision was taken in the United States. If the AAP were the correct principle for defining the demos, then I should feel a lot more aggrieved for not having been given a say in the Clinton-vs-Trump race than for my lack of participatory entitlements in the Brexit referendum. Or, at the very least, I should feel equally aggrieved in the two cases. Yet my intuitions go in the exact opposite direction.
Now consider the ASP. Might it be that, because my degree of subjection to UK law – whether in terms of coercion or in terms of being the addressee of obligations – was much greater than my subjection to US law, this explains my intuitions? I doubt it. People like me are obviously subjected to several pieces of US law – e.g. US immigration law as well as law for which the United States claims extra-territorial jurisdiction – both in the sense of being coerced by it and in the sense of being addressees of presumptively binding commands.31
Still, one could insist that the degree to which I was subjected to UK rule while still a resident in the UK was much greater than the degree to which I was and still am subjected to US rule. But is this really so? For what does subjection amount to? In the eyes of proponents of the subjection argument, it involves the issuing of commands (backed by the threat of sanctions). But consider the following commands: “Don’t steal on US territory, otherwise you shall be punished in accordance with US law” and “Don’t steal on UK territory, otherwise you shall be punished in accordance with UK law.” Someone like me is subjected to both of them, simultaneously, all the time. It just so happens that, since until December 2020 I was physically present in the UK a lot more than in the United States, the latter command was arguably more likely to have an impact on me than the former. But in both cases, the commands take the form: “Don’t perform action X, otherwise you’ll face consequence Y.” It seems that how likely I am to perform action X is irrelevant to my degree of subjection to the corresponding commands. After all, I was and am supremely unlikely to commit theft in the UK (or anywhere else), yet, as a UK resident, I was most definitely subjected to UK criminal law – both in the sense of being coerced by it and in the sense of being presumptively bound by it.32
So, even as far as subjection is concerned, it is not clear that there was much of a difference between my relation to the United States and to the UK, even though I was a resident of the latter. And if there was one, this appears to reduce to a matter of affectedness: UK subjection was more likely to affect me than US subjection, for the reasons just mentioned. But as we have seen, overall, the Brexit vote arguably affected my interests less than the US presidential vote.
Finally, regarding affinity, I think it is fair to say that I had (and possibly still have) greater “affinity” to the UK, rather than to the United States, for the simple fact that I had lived there for many years. My sense of identification with the British polity, both cultural and political, was certainly greater than my sense of identification with the United States. But while this affinity might explain why being disenfranchised from the Brexit vote was somewhat upsetting to me, it does not explain why I ought to have been included in it. To begin with, as I have already argued, it is not clear why identification and affinity are in principle relevant considerations when it comes to assigning participatory entitlements. Moreover, I am not aggrieved about the fact that I am not a British citizen and that I was not allowed to vote in UK national elections, despite residing in Britain. Yet exclusion from the Brexit vote had a very different effect on me. Mere affinity cannot explain this disanalogy. I thus conjecture that something to do with the nature of the question being decided on, in the Brexit case, may do part of the explanatory work.
What explains my intuitions here – i.e. that EU citizens’ exclusion from the Brexit referendum was problematic in a way that their exclusion from the US presidential election was not – is a combination of factors, which cannot be captured by a single, overarching principle. First, while I think “affected interests” should be taken into account in political decision making pretty much across the board, how they should be taken into account – e.g. whether via enfranchisement or some other mechanism – is a contingent matter. A rule of regulation that required all those affected by the US election to have a say in it would be utterly impracticable in the world in which we live, incompatible with the state system as we know it. Trying to implement it would be disastrous. By contrast, allowing long-term EU residents to vote in the Brexit referendum would have been entirely feasible. The different feasibility of including someone like me in the relevant demoi for these two decisions has certainly influenced my intuitive judgements. If ought implies can, as a matter of principle, the United States cannot have “wronged me” by not including me. The same is not true of the UK.
Second, I had moved to the UK on the bona fide assumption that this came with the guarantees of EU citizenship and, quite suddenly, those guarantees may be unilaterally revoked. It is, therefore, as if an implicit “social contract” had been breached. By analogy, when the terms of a contract are changed, either all parties to the contract get to participate in the change in terms and conditions, or at least changes in the contract should not alter the rights and duties of the parties excluded from the amendment process. Such exclusion, then, appears particularly problematic in the Brexit case, for reasons that do not equally apply to the United States.
It might be tempting to add to these two reasons – one pragmatic (feasibility), the other principled (unilateral change in terms and conditions) – that EU residents in the UK had acquired a sort of “moral citizenship,” by virtue of their long-standing cooperation within British society. Such cooperation, in turn, would entitle them to becoming co-authors of the relevant terms of cooperation, including in the case of Brexit.
These considerations about moral membership, however, would do little to explain the sense of unfairness felt by EU residents in the UK for being excluded from the Brexit vote. Typically, long-term residents would have qualified for citizenship. Had they taken up citizenship – which, admittedly, involves a number of hurdles – they would have been automatically included in the franchise. The option of becoming a UK citizen was available. To that extent, long-term EU residents who chose to exclude themselves from the franchise have little to complain about, or so it could be argued.33
This response has some merit, but it ignores the fact that, for many, the choice not to take up citizenship was made under the assumption that their rights would be guaranteed by virtue of the UK’s membership of the EU. Those were the terms and conditions under which many EU citizens moved to the UK in the first place. This is why the sense of unfairness for being excluded from the Brexit demos is best explained by appeal to a perceived “unilateral change in terms and conditions,” rather than by a failure to acknowledge long-term residents’ moral entitlement to citizenship.
An objector might counter that, in fact, “terms and conditions” were not changed. After all, at least as a matter of principle, each EU country retains a right to leave the Union. This raises interesting questions about how to ascertain the terms of an implicit contract, such as the contract that arguably exists between a legal immigrant and their host state. Depending on our answer, the “unilateral change in terms and conditions” argument for the unfairness of excluding permanent EU residents from the Brexit vote will either stand or fall. The point to be emphasized in this context, though, is that the peculiar sense of unfairness felt by EU residents in the UK excluded from the Brexit vote cannot be explained solely by reference to the AAP, the ASP, or AP. Rather, it likely also hinges on feasibility considerations as well as on the impression – the validity of which, I have suggested, is arguable – that an implicit “deal” has been unilaterally broken by one of the parties.
Conclusion
In this chapter, I have tried to cast doubt on the framing of the boundary problem. Instead of asking who should be given a say for a decision to count as democratic, we should focus on the question of what distribution of decision-making power is justified in any given circumstance. Democracy is itself a contingently justified answer to this broader question. To call the boundary problem “democratic” from the start is to put the cart before the horse, to prejudge our answer to the “who should decide” question. In any given circumstance, how decision-making power should be distributed depends on the values underpinning the practice within which the decision has to be made, broader applicable moral principles (e.g. concerning legitimate expectations, implicit contract making, and so forth) in conjunction with feasibility constraints. This means that our answer will sometimes look like democracy and sometimes not, and that the search for a general one-size-fits-all principle for answering the boundary problem is, ultimately, misguided.34
A final objection is worth considering. This is that, even though decisions shouldn’t always be made democratically, there is always some value in the democratic pedigree of a given decision. So it may still be useful to ask “who should be given a say for a decision to instantiate democratic value?” If my argument is correct, however, there is no such thing as “democratic value” per se. Instead, there are several values that contingently justify our attachment to democracy. Of course, one might concentrate on one such value in particular – say, autonomy, or political equality, or non-domination – and call it “democratic.” But it is unclear what the advantage is, other than rhetorical, of using language in this way. It would be better to instead say that a commitment to autonomy, or political equality, or some other value gives us a pro tanto reason to confer decision-making power on some people rather than others. This, however, falls short of a conclusive answer to the “who should decide” question. And to the extent that reference to democracy is often seen to point to a particular set of concrete decision-making mechanisms, labelling particular values “democratic” obscures the pro tanto nature of such determinations. Moving away from a democratic framing of the boundary problem, then, still allows us to capture all that democrats find valuable in substance, without becoming blind to the full range of morally relevant considerations that bear on the “who should decide” question.
What procedures should be used to determine the boundaries of political communities? What principles should govern the structure of these procedures? This chapter sketches an answer to these questions rooted in democratic thought and based on the idea of the All-Affected Principle (AAP) – the idea that people who are affected by a decision should have a voice in it.
I start with a commitment to democracy as a foundational normative theory, in the sense that “no prior or more basic institutional commitment rightly commands our allegiance.”1 The plausibility of this position hinges in part on democracy’s ability to address fundamental questions of political morality. The challenge of delineating the boundaries of political communities is one such question. The AAP, though not without its difficulties as a principle to guide our thinking about boundaries, is a promising starting point for this exploration.
Yet the AAP on its own is ambiguous. What does it mean to be affected? And what does it mean to have influence over decisions that affect oneself? Though often it is lauded for its intuitive appeal, there is much disagreement surrounding the appropriate interpretation of the AAP. The AAP offers a simple and powerful idea, which seems to track core democratic intuitions. And yet, this simple version of the principle has attracted serious and valid criticisms that require addressing. In my account, I aim to salvage the core idea of the principle. To do so, this chapter provides a pluralistic interpretation of the AAP, joining other scholars who move away from identifying one simple principle to determine boundaries of membership.2
Three Dimensions of Decision Making
What are the boundaries we seek to determine? Two kinds of boundaries are often mentioned in the literature: boundaries of membership, delineating a set of people who comprise the community, and scope of jurisdiction, delineating the affairs a community has authority over. Typically, considering the question of inclusion (or the problem of the demos, as Robert Dahl calls it) turns our attention to criteria for membership and away from questions of jurisdiction.3 While the abstract question is sometimes phrased as “Who should be included in the demos?”, the discussion often concerns a specific (though not typically specified) context of assumed institutional jurisdictions, such as “Who should get to vote in the elections of representatives who rule state institutions that have ultimate authority and de-facto control in a given territory?”
Compounding the confusion between membership and jurisdiction, political and ethical disagreements on questions of immigration focus on power over territorial borders – the right of a state to exercise control over a clearly defined territory, including movement of people (and sometimes goods) across territorial lines. These discussions, however, make assumptions on, or else implicate, a host of issues regarding the rights associated with membership as well as the jurisdiction of the states involved. Whether a state should be permitted to bar nonmembers from entering a given territory depends on the plausibility of assumptions regarding the state’s right to grant or deny membership status to these particular individuals, as well as its impact on the circumstances that led them to want to move.4
The point is that the plausibility of an ethical judgment passed over a policy which concerns boundaries, for example in the discussion of naturalization laws, hinges on background assumptions regarding jurisdiction and border control. Therefore, I argue that questions of membership, jurisdictions, and borders should be examined, and answered, together. How we should do this will, I hope, become clear as we discuss the AAP, but we can already see what we are aiming at: a judgment regarding the relationship between membership, territory, and jurisdiction.
Before proceeding, I should note that the three dimensions of democratic decision making I am concerned with – membership (who can take part in decisions), jurisdiction (what the decisions should be about), and territory (where do they apply) – are substantively connected. In other words, it is not unusual for people to make assumptions that connect these topics, because judgments about one ordinarily come with the other. Thus, arguments about movement often assume that members cannot be barred from movement or that people who inhabit a territory have a right to membership in whatever political association governs that territory.5
I should also note at the outset that in discussing political communities, I am thinking of a wide array of social institutions and not just about states, especially not the misleading conceptions of the state as either a Hobbesian ultimate (as in unlimited and undivided) authority or the Weberian centralized monopoly on legitimate power in a territory.6 Not only do these notions of statehood measure poorly when examined in light of the rich variety of political orders that exist in the world today, they also restrict democratic theory without good reason. If we examine the theoretical foundations of democratic theory, we find that the reasons for inclusion and membership in the state apply to other political structures, including local authorities and global governance bodies. The boundary problem of democratic theory is much more prevalent than has been appreciated, and it is interesting to note that the philosophical debates about democratic boundaries more often engage the case of citizenship than disagreements regarding the proper jurisdiction of political power, of the kind that is typical in the literature on (say) federalism.7 Yet the disputes that arise around boundaries of local communities are of the same kind as the controversies surrounding borders of states. A principled democratic response is needed in all these contexts, at least for those of us committed to democracy as a foundational normative commitment.
To see the force of the last point, I briefly present my approach to democratic theory and sketch the considerations that lead us to be concerned about boundaries. The discussion lays the foundation for a presentation of the AAP in the following section.
Social Power and Political Communities
Political Justice and Social Power
On my account, democracy is a demand of justice. Democratic theory is a theory of political justice, as distinguished from a distributive conception of justice.8 This view is contrasted with much of the literature that sees democracy as a standard of legitimacy, often leaving ambiguous its relationship to justice. Nonetheless, several democratic theorists have advanced views that locate democracy’s value in its contribution to a just society. My view aligns with those that are also tied to a “relational” view of justice or equality.9
Political justice concerns the structure of society and the way people relate to one another. In particular, political justice is concerned with the regulation of social power.10 This position resonates with central concerns of deliberative democracy, even if much of the theory aims to provide a standard of legitimacy because “the point of deliberative democracy is to subject the exercise of power to reason’s discipline, to what Habermas famously described as ‘the force of the better argument.’”11 Given the centrality of this idea to deliberative democracy, we can see why it makes sense to see it as a theory of political justice.
What is social power? Humans form relations that give rise to social facts that are “intersubjective” – they are true in virtue of the fact that people believe that they are true, yet they are not a matter of personal preference or opinion. Some such social facts concern the ability of some people to make others do things they would not do otherwise. In other words, some social facts are about power. Shared beliefs bestow power on some people and it is that power that I call, following Miranda Fricker, social power.12 Social power is the ability to make others do as they would not otherwise in virtue of the shared beliefs that people have. This analysis points us towards formal institutions and decisions that take place within them, but the focus is not exclusive. Democratic theory focuses on regulating the exercise of social power, and so follows social power where it appears. The proliferation of sites of power is reason to look beyond state institutions.13
Social power is rooted in shared perceptions that form among groups of people. Relationships change over time, but shared perceptions typically arise in the context of relatively stable relationships, which together comprise social orders. The hallmarks and delineators of social orders are norms – social rules that dictate behavior in particular situations, due to a formal role or specific relationship; they are “a standard of appropriate behavior for actors with a given identity.”14 Norms are enforced and circulated through a variety of social mechanisms such as social pressures and sanctions, ostracism, imitation, and internalization.
Norms are often contrasted with laws, because norms are weakly enforced by social pressures while laws are backed by coercion or the threat of it. The distinction is important, as coercion plays a fundamental role in various normative theories and is often said to require a special justification.15 Yet the fundamental concern of democratic theory, on my account, is with social power and not coercion, and therefore the distinction obfuscates the important similarity between norms and laws as exercises of social power that require justification.
I should add that my expansive operationalization of social power aims to include subtle or invisible forms of social power like what Stephen Lukes calls the “third dimension” of power,16 which manifests in latent, rather than actual, conflicts. Lukes maintains that the exercise of power need not be conscious or intentional but rather can be the product of “socially structured and culturally patterned behaviour of groups, and practices of institutions, which may indeed be manifested by individuals’ inaction.”17 The first two dimensions, Lukes argues, cannot account for the way in which power is exercised to shape individuals’ preferences such that they become unaware of their real interests. Absence of conflict, or even consensus, may not reflect an authentic agreement but be the result of manipulation or thoughtless acceptance of cultural norms.18 This aspect of power illuminates how interests of certain people are ignored or undermined by prevalent beliefs, cultural norms, and the status quo without any observable conflict. And yet, a focus on state institutions or legal enforcement completely misses this form of social power.
Exercises of power along any of these dimensions can take various forms. Democracy, as a theory of political justice, concerns itself with relatively stable patterns of interactions that I will refer to as social orders.19 A focus on stable patterns is implicit in the analysis of both the first and second dimensions of power, where the discussion referred not to ephemeral observable conflict but to the social structures where such conflict tends to arise, and these are commonly accepted as the subjects of democratic principles, being the central venues of decision making.20
However, I argue that democratic theory also aims to regulate the third dimension of power. Cultural and social norms are not the product of decision making in a straightforward sense, but they are “decisions by accretion” since “widespread societal conclusion[s] … [are] reasonably described as … collective decision[s].”21 In other words, cultural norms allocate social power. To say that they are on par with decisions of the kind made in more formal institutions is to treat them as exercises of power. When I accept and uphold a norm that also serves my interests, I am exercising power over someone even though I may not intend it. This interpretation of norms dovetails with Lukes’ analysis of power, as he notes that the third dimension of power reflects the fact that “the bias of the system can be mobilized, recreated and reinforced in ways that are neither consciously chosen nor the intended result of particular individuals’ choices.”22 The power of norms is real; as John Stuart Mill noticed, “social tyranny” can be “more formidable than many kinds of political tyranny” and the use of social sanctions “leaves fewer means of escape, penetrating much more deeply into the details of life, and enslaving the soul itself.”23
To recap briefly: political justice concerns the regulation of social power. To locate social power, we examine stable patterns of interactions, social orders, which form informal arrangements as well as formal institutions and organizations. The delineators of social worlds are norms, which I consider “decisions by accretion” for the purposes of democratization. These social worlds are the contexts within which we should allocate participation and voice.24
Political Communities
The picture we have is one of multiple overlapping social worlds that are diverse and slowly evolving, as “norm entrepreneurs” challenge entrenched norms, others push back, and various other factors (demography, technology, etc.) interact to shape the social world.25 Social norms are powerful but not irresistible. Prevalent norms are followed by most members of a social world because people accept that “this is how they should behave” (prevalent and uncontested norms may be accepted somewhat thoughtlessly). In addition, violations lead to social pressures, including shaming and ostracism, that are often enough to deter people from breaking these norms.
Yet there are always people who, for various reasons, do break norms. The stigma rightfully attached to murder, with the accompanied threat of social shaming and ostracism, act as important deterrence to prevent murder, but as a society we do not accept that as a sufficient means for ensuring certain important norms are adhered to.26 Societies have therefore developed all sorts of social technologies to strengthen the enforcement of certain important rules: codification in written documents, threat of punishment, processes of reputation, and so forth. Legal systems are a particularly relevant social technology; they typically aim at enforcement of norms as well as adjudication of conflicting claims regarding norm breaking. What makes legal systems complicated in this regard is that they both enforce norms but also are upheld, to an extent, by norms. Following the law becomes a norm on its own and in turn the legal system also functions to enforce it, typically by using the threat of punishment.
These various social technologies used to enforce norms are what we call political institutions, and the people who find themselves within the same political structure form a political community. I use the term political here because political institutions reflect the attempt to create a community and manage the way it operates. Sometimes political institutions are conscious and intentional, other times they are implicit in the practice of social norms. Political institutions are those social institutions that exercise the social power embedded in the social world; if there are norms, there are consequences to violating them, and the political institutions are those rules and norms that govern those consequences.
In other words, political institutions are the subset of social institutions that attempt to institute a common good. We often focus on the fact that political institutions establish a “good” and argue about the permissible limits of such a practice, but political institutions also establish the community for which that good is common. Political institutions create boundaries. Sometimes, political institutions trace existing patterns of norms, defining the boundaries of the target population as close as possible to existing social patterns. In other words, sometimes the institutions that police enforcement of rules are applied mostly to people who already accept, perhaps reluctantly or thoughtlessly, the set of norms that the political institutions enforce. Yet the opposite is also true, as social patterns form around existing institutions. Being subjected to a system of social rules may give rise to a shared identity and the creation of a social world.27
In discussing political communities, we tend to think mostly about states, but that tendency is misleading – first, because states are not the only political communities. Cities, for example, also “express and prioritize different social and political values” and are “sites of collective self-determination” whose boundaries matter for the same reasons.28 Political communities overlap and reside within each other as people, members of various social worlds, create, reform, and inhabit institutions. But more importantly, the term state is not particularly useful because it lumps together political orders that are relevantly distinct. The most problematic assumption is that states are unitary actors that enjoy ultimate authority over a territory, associated with the idea of sovereignty. In fact, many states do not enjoy ultimate authority as various internal and external actors restrict their effective power or their perceived legitimacy, yet these instances are dismissed as exceptions.29 Furthermore, even assuming states as unified ultimate authority over a territory, the institutions we call states are different in relevant ways. Some states are federative systems with robust substate structures that exercise a certain measure of autonomy. Some are part of regional organizations that have a great deal of influence over their internal affairs, while others depend on market forces, donations, or other states in order to provide basic functions from security to economics. Some share power with a variety of non-state organizations in and outside of their territory. Some states control disputed areas whose inhabitants have strong affinities to other states, who in turn contest and challenge that control.
Lastly, many states are not actually unitary agents. While it makes sense to look at them as such for the analysis of certain situations, they are in fact collections of agents and institutions that are intermingled together in a variety of complicated relations. For example, it is typical to consider states as sole actors that can deploy coercive force to back their demands. Yet if we look closer at the state, there are various agencies and organizations with capacity to use violence in various situations. The executive branch may need approval from the legislative branch to deploy military force, but actually does so on its own in an unauthorized way, sometimes; or it may be that state leaders have a legal authority to use military force, but they are limited because they believe the military will refuse to obey certain commands; or there is a civilian authority that technically rules over the military commander but in fact is restricted by it. In addition, there are often various domestic actors with varying degrees of coercive power or coercive threat: courts, police forces of different levels, intelligence organizations, bureaucrats, civil society, illegal organizations, and so forth.
Therefore, in thinking of political communities we shouldn’t restrict ourselves to thinking about states. Instead of assuming the power of coercive threat is concentrated in one entity, we should inquire into who can authorize coercion and under what circumstances. It is an open question whether any political institution should have any given set of powers. The idealized version of a system of states is not our point of departure.30 In absence of such point of departure, what should guide us as we consider the boundaries of various political communities? In the next section, I defend the AAP as a better normative starting point for thinking through these issues.
The All-Affected Principle
The All-Affected Principle is the simple and intuitively appealing idea that those who are affected by a decision should have a voice in it. Scholars who support the AAP treat it as a freestanding principle central to democratic theory. Archon Fung describes the All-Affected Principle as a “starting point” of democratic theory, “perhaps the most basic of democratic intuitions,” while Mark Warren calls it a “defining norm of democracy.”31 I start by examining a simple and clearly stated idea of the AAP:
AAP: The people that are relevantly affected by a decision ought to have, in some sense, influence over it.32
Despite its commonsensical appeal, the AAP is quite revolutionary; existing decision-making structures rarely follow power rather than membership, especially when considering the boundaries of political communities.33 The most obvious context where our institutions do not even attempt to track power is the system of states in the global (or “international”) realm, where the legal norm is to exclude noncitizens and completely ignore the voices of nonresidents. However, the AAP also offers a critical perch in many contexts of asymmetrical power including racial and socioeconomic inequalities within states, where those arise out historical or circumstantial factors that never did, and still do not, give voice to the people on the weaker side of the asymmetry.
That said, even a simple version of the AAP suffers from conceptual challenges that threaten its plausibility. The rest of this section outlines these challenges and offers revisions to the principle that aim to salvage its core appeal. First, critics argue (rightly) that the AAP’s focus on decisions is misguided and conjures an image of a single set of choices among a clear set of options, one after another. To reform the AAP, we therefore need to start by turning away from thinking about decisions and towards thinking about decision-making structures. My earlier concern with social power led us to look at stable patterns of interactions, social institutions, through which social power is typically exercised. Decision-making structures are a subset of these social institutions that explicitly involve transforming a group of people into a collective agent by instituting procedures that create a unified collective perspective to guide action. Decision-making structures create collective agency.34
The more important conceptual difficulty that plagues the literature on the AAP is an ambiguity regarding the conditions of affectedness that trigger the principle and the content of the influence that it grants. Many criticisms of the principle stem from the implausible mismatch between a weak affectedness condition, such as being “merely” affected, and a strong conception of influence, such as being entitled to full voting rights.35
The first step to dispel this ambiguity is to accept a pluralistic account of the AAP, according to which people who are affected in different ways are granted different forms of influence over decisions. The main benefit of a pluralistic account is that it abandons binary conceptualization of affectedness and influence. Once we clear up the ambiguity, we can see that a binary version of affectedness and influence is the main grounds for criticizing the AAP. For example, Dahl criticizes the AAP based on the observation that “the logic of the Principle of Affected Interests is that for every different set of persons affected there [will] be a different association or decision-making unit.”36 This kind of criticism hinges on a conception of influence that is conceived only in terms of direct participation in decision making.
For Dahl, it is other democratic requirements, the criteria of economy and competence, that “argue strongly against this degree of direct participation; they argue instead in favor of indirect participation.”37 Yet the AAP is not committed to direct participation as the unique mode of influence. Advocates have no reason to ignore the fact, pointed by Dahl earlier in the same passage, that “people affected by a decision are by no means affected equally.”38 Sometimes affected people have the right to participate directly, but the principle allows, indeed requires, that some of the people affected by decisions will only be granted indirect influence. Thus, the AAP does not require a different association for every different set of people affected, but rather that every association provides a variety of avenues for influence, both direct and indirect.39
The preceding discussion suggests that the AAP is actually a scalar or proportional principle, granting more influence to individuals who are more affected. The proportional logic fits with the intuitive appeal of the AAP. If the reason people deserve a voice is that they are affected, shouldn’t they be granted more voice if they are affected more?
Yes and no. In fact, I think the temptation to interpret the AAP in a strictly proportional manner should be resisted. The main reason is related to the way we conceptualize affectedness. The relevant effect that triggers demand for voice, on my account, is being subjected to social power. Yet social power operates in a variety of different ways, from soft pressures to conform, to punitive sanctions, to internalizing self-depreciating stereotypes – and none of these easily fall along a scalar continuum. Proportionality may play a role in interpreting the AAP as we consider clusters of cases as having a greater or more serious effect on people’s lives, but the pluralistic framework allows for more differentiation in form rather than scale.
In the same manner, there are many possible avenues of voice that do not fit easily into categories of voice. An important avenue of voice is voting rights. Binary accounts of the AAP often take voting rights as synonymous with being granted a voice. Proponents of the AAP typically think that anyone that is affected (in whatever way that is conceptualized) deserves to be included in the sense that they deserve full voting rights. The problem with this view is that it both over- and underemphasizes the role of voting in granting voice to people. On the one hand, voting is only one of the ways in which political institutions allow individuals to participate in, and shape, the way norms are enforced. A democratic society is one where members are active and participate regularly, beyond voting, in public discussion, protests, public meetings, hearings, and so forth. On the other hand, voting rights are rightly tied to an ongoing relationship with a particular political community. Granting voting rights to everyone in the world, as in Goodin’s account, or to other people whose relationship to the community is only temporary, risks making the act of voting even less influential than it is today.40 A pluralistic perspective avoids these problems by locating voting rights within the context of other avenues of voice.
Thus, the pluralistic account of the AAP presents a picture where decision-making structures allow multiple channels of influence, each open to a group of individuals affected in a manner that fits that kind of influence, and individuals find themselves in many such groups. In short, the AAP requires that, in creating decision-making structures, we make sure to match scope and domain – that is, match the people participating in the decision with the array of matters that they have power over. It requires that decision-making structures bring together power with constituency. This is what makes the AAP a foundational democratic principle – it is concerned with empowering the people who are subjected to social control.
My account departs from the AAP as it is often understood. To clarify these differences, I offer this reformulation of the principle:
Decision-making structures rightly wield social power as long as the people that are subjected to it have sufficient influence over it.
This formulation is still abstract, as it does not specify what influence is sufficient for any particular case. But it contributes to addressing the question of boundaries by providing a criterion for evaluating the justifications of political structures. The premise behind the concern for political justice is that social power requires justification. The AAP, as a democratic principle, addresses this concern by specifying the form of justification required, namely that decision-making structures must demonstrate that they provide avenues of influence to the people subjected to their social power, and explain why such influence is adequate. This may seem like a weak conclusion to be drawn from the principle, especially given the promise that it would resolve the question of what makes the boundaries between two or more communities legitimate. However weak this conclusion may be, I contend that it has sufficient implications to merit our attention. To conclude, the next section explores one such implication: the need for democratic decision making surrounding the establishment of boundaries.
Procedures for Determining Boundaries of Political Communities
What does adopting the AAP as I have defined it entail for the institutionalization of boundaries? What are the implications of embracing my pluralistic interpretation for the resolution of boundary-disputes? The first answer is that we ought to have democratic procedures for determining the boundaries of political communities. Since the question of inclusion, of boundaries, of the demos, is a fundamental determinant of the character of the power wielded by a social structure, there is no way in which people have sufficient influence over the systems that wield power around them if they do not, among other things, have a voice in the question of boundaries. Even if they have a voice in the substantive decisions made by the structure, they ought to have a voice in the procedural question of who gets to have a voice. Most importantly, the AAP suggests that the group of people entitled to be involved in boundary-drawing procedures may be different than the groups that we normally think of as entitled. It cannot be the case, for instance, that “current voters” have the power to decide whether nonvoters have the right to vote because the latter group is obviously implicated by that decision. Thus, the pluralistic account of the AAP points to the need for having a separate and special procedure for determining boundaries of political communities.
As trivial as it sounds, this is a controversial position. The commonplace assumption, in political life as well as political theory, is that the way political communities have formed is less important than the way they are managed. The atrocities of the past may give rise to claims of historical justice, the view goes, but they do not undermine the legitimacy (or justice) of existing political structures so long as those are governed according to democratic principles such as majority rule and rule by representatives.41 We have come to think of political communities as if they are born in one constituent moment, out of thin air, and from then onward there exists a political community where before there wasn’t one. Democracy, we’ve been led to believe, is what happens in between such “constituent moments” where ordinary politics is suspended, and “the people” comes into being momentarily as people step forward to change the existing definition.42 Each such change is thought to be permanent as the boundaries are presumed to be, at least in principle, stable.
This position is untenable. Political structures create boundaries by organizing the way social power enforces norms and rules. This practice must be justified not only in principle but also in practice. This idea is central to deliberative democracy, which centers around a “reason-giving requirement.”43 The reason-giving requirement restricts the kinds of reasons that can justify policies but also requires establishing procedures that facilitate participation and provide opportunity for political influence. Reason is not alone required; for a system to be democratic there must be actual reason-giving. Likewise, the AAP requires that political communities provide an account of the ways in which they empower the people subjected to their power. The only way to do that is to institute procedures whereby such arguments can be articulated, contested, defended, and ultimately decided in some manner that reflects a collective judgment. In other words, it requires a democratic procedure. Despite the violent and dark history surrounding the formation and delineation of many political communities, the practice of democracy is ahead of the theory when it comes to procedures for setting boundaries.
Consider referendums. In recent years, we have seen referendums used to determine the boundaries of political communities. Referendums raise many complicated questions and there is no doubt that the practice of referendums is far from ideal. Yet at least in some cases they are unique as examples of boundary setting procedures that are intentionally designed to reflect democratic values.44 For example, two recent UK referendums focused on questions of boundaries: the 2014 vote on Scottish independence, and the United Kingdom European Union Membership Referendum (a.k.a. the Brexit vote).45 Both were touted for their democratic credentials even though they also raised serious concerns from a democratic perspective. Even if these referendums failed to live up to the democratic ideal, they represent the potential for democratic procedures to determine borders on an ongoing, and forward-looking basis. Theory can learn from these practices and offer revisions that address the challenges they bring up. One of the challenges raised by the Scottish independence and Brexit votes was the arbitrariness of the timing of such votes, and the way voting eligibility was manipulated to try to achieve particular results. Many of the people who voted against Scottish independence argued, reasonably, that they would have voted differently if the results of the Brexit referendum were taken into consideration, especially its economic implications for the UK as a whole. The problem is that many intended their vote to be a protest that initiated a further process of reason-giving. Instead, these referenda were presented as “once in a lifetime” opportunities to determine boundaries “once and for all,” reflecting the same position that democratic systems need to assume boundaries in the background rather than subject them to continual examination.46
The concern that a constant reexamination of boundaries may interrupt effective operation of the political community is understandable. Brexit discussions dominated British politics for a long while and many other policies and issues had to be put on hold until the question of membership in the EU was settled. Yet the conclusion cannot be that questions of boundaries need to be determined “once and for all” without any additional opportunity for revision and reexamination, but rather that they cannot be too frequent. Just like ordinary election of representative cannot occur too frequently, or constitutional amendments require supermajorities or other exacting constraints, referenda on boundaries cannot be frequent or commonplace. Yet there must be in place a principled procedure that can trigger them. That is, I argue, an implication of the AAP.
For an illustration of how this balancing act might be accomplished, Stuart White suggests a system he calls PAR, which stands for petition, assembly, referendum.47 According to this scheme, when enough citizens sign a petition in favor of a bill (for White, this is limited to constitutional amendments), the communities affected are required to set up a citizens’ assembly to look at the proposal and decide what, if anything, should be put out to a referendum. Thus, there is an institutionalized deliberative process, which relies on popular participation, that shapes the agenda of any such referendum. I cannot here provide a full account of such procedure for boundaries referenda, but the proposal serves as a fruitful starting point for developing democratic procedures for setting the boundaries between political communities in a way that is, in principle, revisable and open-ended.
Conclusion
In this chapter, I have argued that the AAP urges us to think about instituting processes that open the question of the demos for periodic contestation and allow revision of the boundaries of the demos. Democratic rule is temporary and elected representatives must periodically submit themselves to the judgment of the people. In the same vein, the people itself must be periodically reexamined and constituted. This periodic rebirth need not only be the result of a revolution or moments of rapture, it can also be incorporated into the structure of democratic institutions. If these kinds of procedures were integrated into our political structures, we would have a more expansive view of what kinds of decisions ought to be made democratically – not just how we rule, but also who.
Open access