Law and Philosophy Ó Springer Science+Business Media Dordrecht 2016 DOI 10.1007/s10982-016-9258-0
MICHAEL BLAKE
AGENCY, COERCION, AND GLOBAL JUSTICE: A REPLY TO MY CRITICS
(Accepted 8 April 2016) ABSTRACT. Mathias Risse, Andrea Sangiovanni, and Kok-Chor Tan have offered some subtle and powerful criticisms of the ideas given in my Justice and Foreign Policy. Three themes in particular recur in their critiques. The first is that the arguments I make in that book rest upon unjustified, arbitrary, or contradictory premises. The second is that the use of coercion in the analysis of distributive justice is a mistake. The third is that the global institutional set represents, contrary to my arguments, an independent first-order site of justice. I address these criticisms, and try to vindicate the methodology of Justice and Foreign Policy in the face of these objections.
The critical attention of very smart philosophers is a mixed blessing. On the one hand, that attention is the heart of the discipline of philosophy; being ignored is generally more painful than being criticized, and criticism contains at least the possibility of new ideas and new thoughts. We show respect for one another, on this understanding, through sustained and vigorous criticism. On the other hand, it is often a bit overwhelming to be shown the shortcomings in one’s work, especially by philosophers who are careful, thorough, and rigorous in their attentions. It is often a bit tempting to simply say: I wish I’d written a better book. Mathias Risse, Andrea Sangiovanni, and Kok-Chor Tan are, indeed, careful, thorough, and rigorous. My own thoughts on global and international justice could not have developed as they did without engagement with their writings; each of them has written work that is among the very best the field of political philosophy has produced. I am enormously grateful to them, both for this and for their willingness to engage critically with my own work. The criti-
MICHAEL BLAKE
cisms they have offered are powerful, and I think a full engagement with the subtleties of their arguments would require another book. What I hope to do in this paper, though, is comparatively more modest. I want to address three concerns that emerge across the critical essays written by these philosophers, and see what the ideas I use in Justice and Foreign Policy might have to say in response to these concerns. I do not think, of course, that what I say here should convince them. I do not, in fact, think that I want to offer a blanket defense of Justice and Foreign Policy as it stands; while I stand by the philosophical perspective that book explores, I hope that this perspective might be better developed and defended in later, better, books, whether or not those books are mine. I am, as before, grateful to Risse, Sangiovanni, and Tan, for the conversations from which those better books might emerge.1 I will therefore discuss three themes in what follows. The first deals with the status of what is assumed in my analysis of international justice and in the ethics of foreign policy. Several different versions of this criticism have been put forward by my critics: my arguments, these critics contend, rest on unmotivated assumptions about liberalism, inconsistent assumptions about what is changeable and what is not in public life, and irrelevant assumptions about the necessity of coercion. I deal with each of these in the first section of this paper. The second theme is the nature of coercion. Sangiovanni provides two powerful criticisms of my use of this concept. The first is that I cannot use a moralized baseline for the identification of coercion, without running into problems in the interpretation of state action. The second is that coercion cannot be meaningfully related to distributive justice on either an outweighing or compensating vision of how that justice makes coercion legitimate. I respond to these worries in the second section of the paper. The final section of this paper deals with the nature of global institutions, and with whether or not they might be sites of justice – first-order sites, to use my terminology. Tan argues that the global regime as a whole must be regarded as creating norms of property and resource rights, such that basic global norms of exclusion and resource holdings can only be understood with reference to the global system of rules; that system of rules, though, must be understood on its own terms, as an 1 I am also exceptionally grateful to Christopher Heath Wellman, both for his willingness to edit this exchange, and for more than a decade’s worth of conversation and friendship.
AGENCY, COERCION, AND GLOBAL JUSTICE
independent site of justice. Risse argues, similarly, that norms of justice must include both the distribution of resources and of people as distinct questions to which norms of justice apply. I disagree with Risse and Tan, and try to show how the tools I use in Justice and Foreign Policy can offer a reply to these objections. I. LIBERALISM AND GLOBAL JUSTICE: WHAT MAY WE ASSUME?
There are several points at which my critics charge that I have made unmotivated assumptions: about what can be taken as subject to collective political alteration, and what must be taken as fixed; about what can be inferred from the principle of autonomy, and about how precise this inference must be; and, finally, about whether or not coercion is worth taking seriously as the basis of a concern with distributive justice. These charges are, in one sense, well grounded; I do, indeed, make many assumptions, including assumptions about the possibility space over which my ideas might range. I am, though, less convinced that this is to the detriment of my theory. I think, instead, that it represents a difference in philosophical methodology, along with an associated difference in philosophical temperament. I am not making arguments that I intend to be useful at all times and in all contexts, as I explain at the end of my book; I am, instead, interested in what happens in the here and now, when we examine what can be done within certain defined limits to pursue and promote the autonomy of all persons. I am, in short, convinced that I can defend my work to some degree, even if my critics are quite right that my work rests upon particular (and not uncontroversial) assumptions. We can see this better, perhaps, by looking at some of the criticisms in greater detail. We can start with Risse’s argument that my defense of the universal validity of liberal democratic values is not adequately grounded in my principle of autonomy. I argue that we might derive some norms of democratic government from the thought that autonomy must be taken seriously both at the level of the individual and at the level of the collective. (‘‘We might, or we might not’’, Risse complains, with some mild irritation.) Risse then proceeds to argue that his own chosen ground principle – the idea that we seek to have all humans have lives worth living – does not entail anything so grandiose as the defense of liberal democracy.
MICHAEL BLAKE
I think Risse is right that nothing distinctively democratic can be derived from the idea of a life worth living; that, though, is part of the reason I do not rest my own theory on that idea. I rest it, instead, on the idea of autonomy, which does indeed have some content that tends to push against the legitimacy of unchosen forms of legal authority. The idea is that the autonomy of persons – which we can understand, linguistically, as involving the giving of a law to one’s self – is incompatible with certain forms of power relationship. Most centrally, it seems as if being subjected to a law over which one has no possibility of ownership – no possibility, that is, of meaningful power in the process of creation and interpretation of that law – is putatively a violation of autonomy. It seems to me still rightful that there is the possibility of some derivation of ideas traditionally associated with liberalism from this idea; most importantly, of the thought that one must have rights before the law, and rights to participate in the making of law, before that law is understood as anything other than naked and illegitimate power. Risse, though, is still within his rights to say: fine – show me what is implied, then, and let me see which ideas of democratic right are contained within that derivation, so that the theory you offer can be applied, criticized, or endorsed. This, though, is simply not my project. Risse’s own book – On Global Justice – is a magisterial work, and undertakes a thorough derivation of the project of global justice, from broad principle to relatively fine-grained application. That is all to the good; his book is an achievement, and ranks among the most important works the field contains. But I do not think that those who choose another path are therefore deficient. When a philosopher has a broad and abstract argument about how to understand the implications of his principles, but the proper derivation requires a great deal of factual information, he has one of two options: he can put forward his moral principles, and leave it to those more empirically-minded to figure out what ought to actually be done; or he can, instead, acquire that empirical information, and do the derivation himself. Risse chooses the latter path; I chose the former. What is interesting about my view, if anything, is that it offers the possibility of a continuing dialogue: a dialogue about what particular norms of liberal democracy are actually required for there to be respect for autonomy. I am, as I say in the book, convinced that
AGENCY, COERCION, AND GLOBAL JUSTICE
some forms of democratic practice, as well as some norms traditionally associated with liberalism, must be in place for there to be rightful rule. I am much less confident, though, that we have any solid sense of exactly what the specific norms are without which we fall into tyranny. Risse thinks a theory of justice must take a stand on this, as his book does. I want to resist this, and let my book represent as an invitation to further dialogue, rather than the end of that dialogue. Risse’s more damaging criticism, though, is that I am simply inconsistent in what I take as up for grabs for purposes of theorizing. I am, as I discuss in the book, an institutional conservative: what I mean by that is that I think we ought to see what the institutions we have now can do, to achieve justice, before we seek the creation of novel forms of institution. I think this perspective is defensible, if provisionally so, for two reasons: because the making of new institutions is frequently a more messy and difficult process than the alteration of policy by existing institutions, and because we learn philosophically interesting things from seeing what our institutions now can (and cannot) do. This conservatism is shared, in part, by Risse, but he charges me with simple inconsistency: I am not willing to argue directly for any transnational body with its own power and its own justification for that power, but I am willing to argue for the Alien’s Bill of Rights, a constitutional instrument that might limit state action when that action detrimentally affects the human rights of foreign citizens. The latter, though, is a rather fantastic thought, if taken seriously as a proposal for political action in the present. (Constitutional rights for foreign nationals are not likely to be part of anyone’s platform in the next American elections, to put it mildly.) Why, though, am I willing to argue for one vanishingly improbable dream, after a steady refusal to contemplate other improbabilities? Is this refusal to discuss new global institutions not, at this point, unjustified? I do not think Risse is entirely right on this point; there is, at the very least, a sense in which the two examples might be rightly treated in different ways. As I discuss above, my goal is to see whether or not we might, with current institutions, arrive at principles that enable us to do justice to all, and policies than enable us to live these principles. That means that there is a distinction, on my
MICHAEL BLAKE
view, between policies that are unlikely to be adopted, and institutions that are unlikely to be built. New institutions, to begin with, require novel sorts of information about the effects of new forms of human relationship; Risse has done more than anyone to demonstrate how little we know about what the effects of certain kinds of institutional novelty might actually be.2 More importantly, though, my question is whether or not we might do justice to all by insisting that our existing institutions alter what it is they do with their existing powers. I accept Risse’s intuition that the Alien’s Bill of Rights is unlikely to happen soon – although I think it might be closer than he thinks, given the general familiarity of liberal democratic countries with constitutional restrictions on democracy. I would insist, though, that we are right to treat that policy different from a theory of global justice that requires us to build novel transnational organizations from the group up. I want, at the very least, to see what is conceptually possible without such institutional innovation. If there is no possibility of doing justice with the institutions we have, that is itself an interesting (and not implausible) result. But I think it is worth restricting ourselves to the existent, if only to see what we can do, where we are, with what we have. This sort of consideration also affects the third, and final, point on which I am regarded by my critics as relying on an arbitrary or stipulative assumption. Sangiovanni’s criticism of my view begins with a thought experiment, on which the inhabitants of a given state cease to coercively enforce their laws; they are, instead, now reliant only on voluntary compliance, and perform their legal duties about as well without coercion as they did without that coercion. Sangiovanni argues that it would be implausible to think that the transition to this voluntaristic society would alter the duties of justice of those within that society; but if we accept the continuing relevance of duties of justice, we must therefore accept that those duties do not depend upon the presence of coercion. To this, my first response in Justice and Foreign Policy is to say: the people who are capable of behaving in this way are not subjects of justice as I understand them. This, though, is regarded by Sangiovanni as rather not enough, philosophically speaking; ‘‘I don’t see’’, says Sangio-
2
Mathias Risse, On Global Justice (Cambridge: Harvard University Press, 2012) pp. 304–324.
AGENCY, COERCION, AND GLOBAL JUSTICE
vanni, ‘‘how the empirical likelihood that any of these cases might be actualized in the world as we know it is relevant.’’ I disagree with Sangiovanni on this latter point: the empirical facts are relevant to what duties there are, and the limits of what is empirically plausible are worth understanding as part of political philosophy. All political theorizing takes some things for granted, prior to beginning the task of theorizing: the fact of moderate scarcity, limited (but not non-existent) stocks of altruism, and so forth. We work within these conditions, not simply because we have to make some assumptions in order to get the task of theorizing started – although we do – but because these conditions represent morally relevant aspects of people as we know them, or the world as we understand it, or both. The idea that people must live in states, and that those states must of necessity involve the possibility of coercive enforcement of some norms, depends upon a particular vision of what kinds of people we have, and of what sorts of disputes they are likely to have. Imagine, instead, that we did political philosophy and decided that no one ever disagreed about who was entitled to what thing, and that all human disputes were solved through spontaneous agreement. The theory in question might be an interesting one, but I do not think it would provide much information about how humans as we know them might be capable of living together. The people it describes are simply not people as we know them to be. This is relevant, though, because Sangiovanni argues that the hypothetical voluntaristic society – regardless of how plausible it might be – is a society to which norms of distributive justice must surely apply, just as they applied when a coercive state were in place. I’m afraid I simply don’t see why. If we were able to arrive at a society in which there was never a need, or even a possibility, of coercive enforcement of particular norms or decisions, then it strikes me that those who live in that society have simply moved beyond justice. They exist, perhaps, as anarchic angels, freely associating for their own purposes, agreeing spontaneously with others (or agreeing spontaneously, at least, about how to solve what disagreements they have.) What norms of distributive justice might apply within that society, I think, is genuinely a new question – at the very least, it is a question that is unlike the one I have asked, which begins with
MICHAEL BLAKE
deeply imperfect people inhabiting a particular juridical society, and seeking what must be the case for the coercive institutions they face to be justified. I do not have any intuitions one way or the other about what these angels owe to one another. I am confident, though, that people as I know them are far from angelic indeed. Sangiovanni’s analysis of coercion, though, is worth examining in more detail; and it is to this task that I now turn. II. COERCION AND GLOBAL JUSTICE: (HOW) DOES THE STATE COERCE?
Over the course of several articles, Sangiovanni has developed a subtle and compelling analysis of global justice; this analysis begins with considerations similar to my own, on which there is an important relationship between the norms of distributive justice and the special sort of human institution understood as the state.3 Sangiovanni’s analysis, though, focuses on reciprocity, rather than on coercion; he has developed several arguments against the relevance of coercion to our theorizing about distributive justice. Space prevents me from dealing with all of his arguments here; I do want to address two of his arguments in this context though, both of which represent points on which I think the disagreement between us might be fruitful. What I mean by this is that I do not think I want to defend my view against his, so much as demonstrate that he and I simply have different visions of what it is that is special about the state.4 Sangiovanni focuses on the state as a place in which certain kinds of reciprocity occur; it is these sorts of political reciprocity that ought to be the focus of those who are examining distributive justice – and related forms of reciprocity might occur at both the global and domestic level. In contrast, I think the state is special precisely because of how it takes itself to be the final arbiter of disputes, and to determine who is right in a given contest between rival interpretations of justice. This coercive enforcement might be legitimated with reference to reciprocity – but the coercion itself begins our story, and we do well to keep it in mind. 3 Most importantly: ‘Global Justice, Reciprocity and the State’, 35 Philosophy and Public Affairs (2007) 3–39; ‘Global Justice and the Morality of Imposition, Coercion, and Framin’, in Miriam Ronzoni et al., eds., Social Justice, Global Dynamics (Routledge, 2011); and ‘The Irrelevance of Coercion, Imposition, and Framing to Distributive Justice’, 40 Philosophy and Public Affairs (2012) pp. 79–110. 4 This way of phrasing the question is borrowed from Helena De Bres, ‘What’s Special about the State?’ 23(2) Utilitas (2011) pp. 140–160.
AGENCY, COERCION, AND GLOBAL JUSTICE
We can start with Sangiovanni’s first criticism of my account, on which the use of a moralized baseline leads to some difficulties. The baseline of expectations is that default set of entitlements against which coercion is understood; if we understand that set of entitlements as statistical or descriptive, then anything that proposes to take away a set of entitlements we happen to have is coercive. If, instead, we take a moralized baseline, then anything that proposes to take away what we ought to have is coercive. In my book, I defend a moralized baseline of expectations; what this means is that the one who proposes to leave an individual below a morally acceptable baseline – to take a central example – is engaging in coercion; the one who wanders by a pit and refuses to help until the victim signs over the deed to her house is engaging in coercion, not merely sharp dealing. Sangiovanni argues, though, that either the descriptive or the moralized baseline will lead to problems for my view. The descriptive view will take current entitlements as unduly important in moral analysis – which is, as Sangiovanni and I agree, a good reason not to accept it. The moralized view, though, seems to say that a state that enforces moral duties is not actually a coercive state at all. If, after all, it forcible prevents me from doing something I was not within my rights to do, has it threatened to take away anything to which I was entitled? It doesn’t seem as thought it has. This fact, though, is especially damaging for a view that takes taxation law to be coercive. I am committed, as Sangiovanni notes, to seeing taxation as coercive. But if the taxation is rightful, how can it be coercive? Does my view secretly depend upon a sort of ‘everyday libertarianism’, on which we are somehow morally entitled to whatever we happen to have gotten? I have been worried about these possibilities for a while; I tried to deal with them in the book by discussing malum in se and malum prohibitum, and noting that much of what the state did was not to insist upon moral duties but to define certain sorts of actions as legally wrongful. Sangiovanni finds this idea somewhat unhelpful, and I’m inclined to think he has a point. I’ll try again, though, with an idea that I think might do more work. Let’s imagine that there is a distinction between the moral truth, whatever that might be, and the power to insist upon a particular moral proposition’s being true – as
MICHAEL BLAKE
well as the power to interpret and specify that moral proposition’s implications. The gap between these is, I think, significant. On the one hand, we might think that there are true facts about the morality of abortion; if we are realists of a particular stripe, we might think that these facts exist, much as mathematical truths do, and are (sometimes) perceived by (some) human minds. On the other, we face a situation in which people of good will disagree wildly about the morality of abortion, while sharing a society; they must arrive at some particular conclusion about what principles shall rule the day, and then about what those principles actually mean for the inhabitants of that society. What is important about the state, on my reading, is that it decides which principles will win, and then how to interpret and execute these abstract principles in action. This, I think, is what I was imperfectly gesturing towards with my discussion of malum prohibitum: the thought that the state engages in coercion, even on a moralized baseline, simply in virtue of its ability to be final in its determination of what principles shall be used in the adjudication of disputes, not because it has special access to moral truth – it doesn’t – but because it is the state. The one who proposes to leave the victim in the pit proposes to do wrong, and so coerces that victim with reference to the rights she ought to have. The state that proposes to use force to insist upon particular norms of justice, and particular interpretations of those norms, is also engaging in coercion, although in a subtly different way: it insists upon its own final determination of what this broad, abstract, and controversial thing called justice shall mean. It is right, because it is final; it is not final because it is right. This means, though, that we might have some reason to think that the taking of property can be coercive, even if that property is ultimately the product of particular laws that are created by the government. To see this, take the case of David H. Lucas, who owned beachfront property in South Carolina, and saw the economic value of that land reduced through environmental regulation.5 The question was whether or not that reduction in value constituted a ‘taking’ of property, under the Constitution. The first court to hear his contention sided with him, and agreed that he had a property interest that had been taken from him; the Supreme Court 5 The Supreme Court case dealing with these facts is Lucas v. South Carolina Coastal Council, 505 U.S. 1003 (1992).
AGENCY, COERCION, AND GLOBAL JUSTICE
of South Carolina disagreed, and said that he never had that property interest at all; the Supreme Court of the United States disagreed still further, and said that he had, indeed, had that property interest all along. I am not especially interested in the substance of the legal dispute here, nor with the politics that undergird the concept of ‘regulatory takings’. I am interested, instead, in how to understand what was seemingly taken from Lucas, and then given back to him, and then taken, and then given back. Sangiovanni suggests that only an ‘everyday libertarianism’ could think that any thing was taken from Lucas, when the property interests affirmed by a lower court were determined by the Supreme Court of South Carolina to not exist. After all, property rights are ultimately the product of the legal system; if they do not pre-exist that legal system, we cannot use them to describe that legal system’s actions as coercive. This, though, simply seems wrong. Lucas genuinely did lose something between the lower court and the Supreme Court of South Carolina. It is not enough to say, as I think Sangiovanni wants to, that he never really had that thing, and so should not be regarded as subject to coercion on a moralized baseline. (We’re merely taking away, we might say, what you never had; you have no right to complain.) Instead, I think he did lose something, even if we do not assert the existence of pre-political property rights or any other libertarian vision of ownership. The Supreme Court of South Carolina issued its own definitive interpretation of what ‘takings’ really meant, and insisted that he must now regard his property rights as involving only what they thought it ought to mean. Their determination could not be overcome by a suitably annoyed Lucas; it could only be overcome by the United States Supreme Court. Lucas’s claim was not crazy – as is shown by the fact that it eventually convinced the Supreme Court. Yet, after the Supreme Court of South Carolina ruled, it asserted that its interpretation of property was superior to Lucas’s own – and, moreover, that the legal machinery of South Carolina was now potentially going to be deployed to defend that particular vision of property. At each stage in the proceedings, a court was exercising its legal right to be the final decision about how the abstract notion of ‘takings’ – and, indeed, the abstract notion of property – was to be understood. It is this finality – and the right to insist, through the threat of civil contempt and garnishment of
MICHAEL BLAKE
wages, among other tools, that this final vision be accepted – that marks the state here as engaging in coercion. The state should, in short, be regarded as coercive whenever it defines a particular vision of what property should be, and then insists upon the right to use certain kinds of force in the defense of that vision. One need not endorse any libertarianism to think that this vision of coercion is plausible. I think more should be said on this front; I hope, though, that what I have said is at least enough to suggest that more could be said. Right now, though, I want to move on to what is an even more important part of Sangiovanni’s critique of coercion: how, exactly, does the imposition of norms of distributive justice relate to the existence of coercion? There are at least two possible answers, which Sangiovanni calls compensation and outweighing. On the former, the pro tanto wrong of coercion is compensated for through the creation of norms on which distributive justice holds among those coerced. On the latter, the interests involved are sufficient to justify the imposition of coercion; the disvalue of the coercion is, as it were, outweighed by the importance of the interests at stake that led to the coercion. In the case of distributive justice, we would think that the interests we have in not being coerced are, very simply, less weighty than those that might be at stake in the creation of a just world, complete with distributive equality. Sangiovanni, of course, argues that there are problems for my view on either interpretation. I agree with him that compensation is unlikely to work; I’ll focus, instead, on outweighing, which seems closer to what I take myself to be doing – although, as I’ll discuss, it’s not quite right either. The problem with outweighing, says Sangiovanni, is that it reduces questions of justification to the interests we have, and to a process of balancing those interests. If distributive justice is to be a partial response to coercion, it must be because we have some interest in living in a world governed by norms of distributive justice. And this means, in turn, that we do not find ourselves binding the scope of distributive justice with reference to the set of people who are actually facing coercion. We might, instead, simply see what it is that matters about distributive justice, and what interests it protects, and then figure out where we can best balance rights and duties so as to protect people’s interests. It is, in particular,
AGENCY, COERCION, AND GLOBAL JUSTICE
open to us to insist that the interests of foreign citizens of living in a world of distributive justice is so urgent that they ought to be brought within the scope of distributive justice, even though they are not within the set of people coerced by the domestic state. Once we start weighing interests, in short, we have to let everyone be counted; those who are coerced must be listened to, of course, but we have no reason to think only their voices matter. I said that outweighing was closer; I also said it isn’t quite right. I think the concept that I want to employ is best termed rectification. We can understand this idea, perhaps, by analogy to the justification of criminal punishment. On the one side, we might justify punishment with reference to the interests of all those affected. There are interests in not being punished, to be sure, and interests in not living in a society rife with crime. These are to be balanced, and the result is the ideal distribution of punishments to persons. The idea is fundamentally a hydraulic one: we are seeking the best overall outcome for certain specified interests, and we adjust our policies until the best overall level is obtained. Jeremy Bentham’s analysis of punishment, of course, takes this shape, and like Sangiovanni he refused to allow the scope of punishment to be determined by anything other than an overall consideration of interests. He was, most notoriously, willing to countenance lessened punishments for the socially beloved, given our interest in avoiding becoming deprived of their companionship.6 This might seem a bit unfair; Sangiovanni is a subtle thinker, and Bentham (despite his historical importance) is not. The reason I raise this, though, is because I think many of us react to the whole Benthamite account of punishment here in a way that is somewhat skeptical, and that skepticism tracks how I react to Sangiovanni’s analysis of outweighing. If we took a Kantian justification of punishment, though, we would start simply with the fact of someone’s having done wrong, and then take that as sufficient to justify the imposition of punishment. For Immanuel Kant, punishment is what is deserved.7 For Bentham, the set of people to be punished is guided by human interests, and this means that we should usually focus our 6 Bentham describes this as the relevance of the ‘particular merits or useful qualities of any particular offender, in case of a punishment which might deprive the community of the benefit of them’. Jeremy Bentham, The Principles of Morals and Legislation, Vol. II (London: W. Pickering, 1823) p. 30. 7 Kant’s theory of punishment changes over the course of his writings to some degree; the clearest analysis of his thinking is Jeffrie Murphy, ‘Does Kant Have a Theory of Punishment?’ 87(3) Columbia Law Review (April 1987) pp. 509–532.
MICHAEL BLAKE
punitive energy on the guilty. For Kant, though, it is a conceptual necessity that we ought to punish only the guilty, since only the guilty have done what gives rise to the need for punishment. Punishment may lead to the preservation of certain interests, but that is not why punishment is – for the Kantian – done. I think the justification of distributive justice as a rectification for the otherwise impermissible practice of coercion looks a bit more like this Kantian picture than Sangiovanni’s own analysis. It has, in the end, comparatively little to do with interests; I do not think we have an independent, standing interest in living in a society marked by distributive justice. Rather, I think my distributive story – like Kant’s criminal story – begins with respect for agency. The punitive state respects the agency of the criminal, even as it punishes him, by turning his maxim against him. The political state, in a similar way, respects the agency of the one coerced, by ensuring that the one who is coerced is placed in a situation as close as possible to that one his agency would have freely chosen. Sangiovanni is quite right, of course, that this is complex and rather poorly described; I think more needs to be said to explain how this respect for agency is to be understood. It begins, certainly, with the idea that there are some patterns of comparative advantage enforced by the law, and some forms of incapacity in the process of disputing and forming that law, that are incompatible with respect for agency. What I would insist upon here is not that I have the details right, but that the details should be worked out with respect to people understood as agents, and not as merely those who have interests. Once we have made this move, though, the binding of the scope of distributive justice comes rather easier; we look to those who are being made subject to the pro tanto wrong that is coercion. (When we punish, after all, we do not focus in the first instance on the criminal’s friends and their interests in his companionship; we focus on the one against whom the state acts.) Those who write on global justice are, on my view, right to continue to focus on coercion, rather than simply on interests. This might be rejected, by those whose philosophical proclivities disagree; pure retributivists are rare, after all, and this focus on agency may seem rather fetishistic. The view does, though, deserve at the very least the respect that is entailed by rejection on its merits, rather than with reference to the framework and concepts of a competitor view.
AGENCY, COERCION, AND GLOBAL JUSTICE
III. GLOBAL INSTITUTIONS: WHAT DOES THE WORLD DO?
The last topic I want to consider today is, perhaps, the most important one: what, exactly, is the nature of global society? This question is important, because how we understand the global institutions that now exist must of necessity affect how the norms of justice ought to be applied at the global level. I have suggested, in Justice and Foreign Policy, that the institutions we find today at the global level are best understood as second-order sites of justice: that is, as sites to which considerations of justice apply in virtue of their ability to help (or hinder) justice at the domestic level. I say this, in part, because I believe we are frequently misled by the appearance of transnational institutions. That which is genuinely an independent body, with its own sources of both power and legitimacy, is frequently confused with that which is simply a cover for naked power politics undertaken by wealthy Western states. I am less convinced than my critics, that is, that we ought to take the global institutional set as itself a first-order site of justice, to which norms akin to those found within the state might apply. I recognize, though, that this is ultimately a question that depends to some degree upon empirical truth – and that I am ill-equipped to arrive at that truth myself. The world might, indeed, simply be other than I imagine it to be. In the present context, I want to focus on the moral criticisms that have been brought against my view, rather than the view’s empirical plausibility. I will look at two versions of these worries, given respectively by Tan and by Risse. Both of these thinkers provide new variants of my hypothetical pair of states (Borduria and Syldavia), and seek to show that some first-order considerations of justice must be thought to hold sway in any plausible account of justice between states. We can start with Tan’s argument. Tan’s question is deceptively simple: what, if anything, justifies the territorial state in keeping its resources in – and in keeping unwanted outsiders out? The response to this question, suggests Tan, must entail the recognition that some rights held by states are best understood as having been produced by the international institutional set itself. To demonstrate this, Tan develops a hypothetical extension of Borduria and Syldavia. The Syldavians are somewhat less well-off than the Bordurians, and find that their lands are simply less well-resourced than the land of the
MICHAEL BLAKE
Bordurians. There are, moreover, chunks of land in Borduria that are simply unoccupied, and might be annexed by the Syldavians without any negative effect upon the lives of the Bordurians. What possible reason might the Bordurians cite, in order to justify their continuing claim to territorial integrity? The easiest way, suggests Tan, is with reference to some sort of Lockean propery right, on which the Bordurian state has mixed its labor with this particular parcel of land, and thereby has acquired the right to maintain that land and exclude others from it. This Lockean methodology, though, sits poorly with my own institutional vision, as Tan notes. The alternative, though, is for me to give up my insistence that the global institutional set is not a first-order site of justice. It might, instead, be a first-order site, albeit a different sort of first-order site; it is comprised of that set of global institutions that define, and defend, particular entitlements to land and to resources. Tan argues that I cannot, without running into absurdity, avoid the conclusion that the global institutional set is itself part of the institutional story about global distributive justice. This is a powerful argument; it derives from, but is not reducible to, Tan’s own institutional analysis of global justice.8 My response, though, is to push back against Tan’s analysis of the state’s right to its resources. Tan argues that the right of a state to its territory is either justified with Lockean principles, or with reference to the institutions of global society. I this isn’t quite right: I think the right of a state to its resources emerges from considerations of sovereignty, rather than considerations of property, and that these considerations of sovereignty are independent of – and logically prior to – the emergence of any particular global institutional set. Sovereignty, for me, is the right to rule; and, while global society is obligated to respect the legitimate state’s right to rule, it is a mistake to think that this latter right is a creature of the institutions of global society. We can see this, I think, by looking at a bit of John Rawls’s theory of justice that is sometimes overlooked: distributive justice applies to things that are created or extracted through human activities organized through the basic structure. We are right, on this account, to think that there ought to be political principles governing the distribution of money to persons; we are not right, though, to think that bodily organs ought to be redistributed. We did not, collec8 Tan’s own view is an exceptionally important one, laid out in his Justice, Institutions, and Luck (Oxford: Oxford University Press, 2012).
AGENCY, COERCION, AND GLOBAL JUSTICE
tively, create those organs, and we have no right to think that they are proper subjects for political reallocation. (This does not mean, of course, that we have no duties of justice in healthcare to help those who are missing kidneys; we have such duties, but they must be derived from the just distribution of health care resources, and not from the just distribution of kidneys.) We are, in short, only entitled to regard a given thing as a viable subject of distributive justice, if those things are social primary goods, rather than natural primary goods.9 We can redistribute widgets, since we made them together, and ore, since we have dug it out together; we cannot redistribute parts of bodies. With this in mind: how should we understand the right of a state to its own resources? Are they more like widgets, or more like body parts? At first glance, the question seems almost silly; the borders of states are arbitrary, and any given state in the world right now came into the world through processes that ranged from the random to the downright evil. Surely, then, no state can have title to its resources, as a person does to her kidneys! This easy answer, though, isn’t quite right, once we start examining the idea that what the state has over those resources is not the right of property, but the right of sovereignty – the right, that is, to be the final arbiter in disputes about how those resources shall be owned and transferred among agents within that jurisdiction. The reason this matters is that we have a reason to resist the thought that the state can only acquire rights over resources in the way an agent can acquire property rights over land – namely, by improving it, and then making it owned. On my analysis, the state is not best viewed as the owner of the land in question, but as holding rights to rule over that land; these rights of sovereignty, though, are not produced by the global institutional set, nor do they depend upon that institutional set’s consent. Tan, in this, is quite right to note the Kantian flavor of my analysis of sovereignty. I accept, with Kant, the idea that the first duty of all humans is to leave the state of nature, and to set up a political society within which their disputes might be rightly (and finally) adjudicated. This first duty, though, comes with other duties – namely, the duty to respect those societies that have been created by others, in their own quests to create political communities. The 9
See John Rawls, A Theory of Justice, original edition (Cambridge: Belknap Press, 1971) p. 62.
MICHAEL BLAKE
political community, though, is at least in part a geographic venture: it is a commitment that disputes within a particular chunk of the world’s surface shall be settled with reference to the norms and decisions of a particular state. The duty of all humanity is to respect that political community, when that political community is actually doing the job. This means that – for me – the right to resources is best understood with reference to the right to continue existing as a political community, which in turn is best understood with reference to political justice generally. A state which is doing the job of justice, to put it pithily, has the right to continue doing that job – and that entails the right to continue being the final arbiter about resources, within this particular bit of the world.10 With this in mind, we can now return to Tan’s imagined Syldavian critique. Why, say the Syldavians, can we not simply absorb this bit of your resources? You don’t need it, and we do; we do not take from you anything you require, in order to have good lives. The Bordurian response, though, would be to question the very terms of the Syldavian request. We consider these resources not with reference to the human interests they make possible – although, of course, we would have to use them to help you, were you to fall below a threshold of functioning. We do, instead, consider the Bordurian state to have the continuing right to determine which Bordurians shall have the right to trade in which resources, and how, and what limits there shall be on transfers, and so on. We are, in short, sovereign over these resources, and this sovereignty derives not from the international institutions in which the Bordurian state participates, but from the most central duty of all: the duty to create justified political societies. The Bordurian state, having done this, has the right to continue doing so, even if our abandoning of that claim might help Syldavia out considerably. This response is, I think, persuasive – at least as applied to the request for a redistribution of fungible natural resources, or to the right to carry those resources out of Borduria and into Syldavia. I am less confident that this methodology can reject Tan’s argument about land. Tan imagines the Syldavians requesting the right to take land that is currently uninhabited by Bordurians, and to absorb that land into their own society. Everything I have said above, I should 10 I argue more explicitly for this in ‘Civil Disobedience, Dirty Hands, and Secession’, in Fernando R. Téson, ed., The Theory of Self-Determination (Cambridge: Cambridge University Press, forthcoming).
AGENCY, COERCION, AND GLOBAL JUSTICE
make clear, involves the right to continue serving as the final arbiter of disputes between people within a particular geography. It has comparatively little to say, though, about chunks of land in which there aren’t any people to be found. If the Bordurians want to take an uninhabited chunk of Syldavia, and populate it, the claim of the Syldavians to resist this might have to be made in some other way. They might contest the very notion of ‘unpopulated’, to begin with. Some land without permanent inhabitants nonetheless is used by humans for particular purposes; Central Park has no permanent human inhabitants, but it would be a mistake to think of it as therefore terra nullius. Perhaps the Bordurians could resist the appeal of the Syldavians in this way; perhaps not. I am, at least, not confident that all possible cases can be dealt with in this manner. This means, though, that Tan’s imagined Syldavian appeal might sometimes have to succeed. This result, though, does not entail describing the global context as a first-order site of justice. The respective rights of the Bordurians and Syldavians are not derived from the interactions of states, but from those more simple moral materials on which states are justified to begin with. Risse, finally, has some criticisms of my view that begin with ideas similar to those of Tan’s. Risse’s version of the criticism, though, is somewhat different from Tan’s own, and the cases merit independent discussion. Risse imagines several hypotheticals, on which Borduria makes a claim against Syldavian society; Risse argues that these cases demonstrate that the global institutional set is, itself, an independent site of justice. One of these – Borduria and Syldavia, Part Seventeen – deals with resource rights; Syldavia contains, through no merit of its own, the world’s supply of Castafiorium, a necessary mineral for certain industrial processes, and the relatively less well-off Bordurians want to enter so as to take advantage of that Castafiorium. Another – Borduria and Syldavia, Part Eighteen – deals with population pressure. All Syldavians, except for fifty, die out, leaving a vastly underpopulated society. Bordurian citizens apply to enter into that society, and are rejected by the (surviving) Syldavians. In both of these cases, Risse argues that the Syldavians are behaving very badly indeed; though they are not actually unjust, the Syldavians are engaging in actions that the Bordurians can resist, and Risse
MICHAEL BLAKE
concludes that – at a minimum – the Bordurians have no obligation to obey the Syldavian law against entry into Syldavia. I have a similar reaction to these cases; I am, in fact, concerned that, on these points at least, Risse and I are hiding our agreement beneath slightly different terminology. I accept that the Syldavians are acting quite badly; I would potentially invoke the language of callousness, or the persistent absence of beneficence, to explain the failures of the Syldavians here.11 I also share his contention that the Syldavians here do not engage in injustice. Since I think the stories we would tell here are slightly different, it might be worth examining mine in more detail. The discussion of resources can, perhaps, be had relatively quickly. The Bordurian claim here seems to be dispelled by a view of resources that focuses on sovereignty, instead of property. The Syldavians have sovereignty over the land on which Castafiorium is found; so long as other polities are able to justly administer their own political processes, there is no wrong involved in Syldavia continuing to administer the distribution of Castafiorium in accordance with its own political conception of justice. The idea of population pressure, though, is more difficult to dispel. What, exactly, can be said by the underpopulated state to the relatively more populated one? Does population pressure not represent an independent site of justice? This theory is a complicated one, and I had some hand in its development; Risse and I worked together to develop an early version of these ideas, which Risse has now developed into a significantly more sophisticated and powerful ground of global justice. (I am, therefore, now arguing against a previous version of myself, as well as against Risse.) I wanted to avoid focusing on land as a resource in my own book, though, because I have become convinced that the primary function of land is political; it is, to put it most simply, a place on which politics happens, on a suitably broad notion 11 I would say something similar about Risse’s Borduria and Syldavia, Part Nineteen, on free trade. On this hypothetical, Syldavia takes advantage of a vulnerable Borduria, and ensures through biased trading rules that the vulnerability continues. I think Risse is right to be worried about this case; I think he is wrong to think that I cannot be. I would understand this case with reference to the moral importance of democratic self-government; the Syldavian actions may make democratic self-government impossible, when they identify only particular, and particularly bad, pathways for development. I would also argue that the notion of the callous and the uncharitable might be invoked, for less obviously egregious cases of this sort. Even if the Syldavians are not unjust, it does not follow that they cannot be subject to moral criticism.
AGENCY, COERCION, AND GLOBAL JUSTICE
of politics, and the value of land can be understood with reference to this political function. This means, though, that any simple metric of how many people can be found within a particular jurisdiction tells us very little of any great importance. What is important, of course, is that the society in question is capable of maintaining a form of life for its inhabitants that respects their rights to autonomy; that provides an adequate political society, for instance, as well as an adequate civil society within which they can develop a plan of life. But the number of people per square mile tells us very little indeed about this. In this, an undernoticed figure in the history of political life is Elisha Graves Otis, who developed the safety elevator in 1854.12 Prior to Otis, humans were limited in practice to buildings of four stories or less; artificial ‘land’ on the tenth floor is of comparatively little use, if one must walk up ten flights each day to get there. After Otis, though – and under pressure from industrialization – urbanization has taken root; more people now live in cities than in rural areas, for the first time in human history. The importance of this, of course, is that we now should be very careful about using ‘people per square mile’ as a number that provides us with the right sort of information for purposes of global justice. On the average weekday, the population density of Midtown Manhattan is 980,000 people per square mile. Most people work in high-rises in Midtown.13 Most ‘land’ in Midtown, moreover, is artificial – the surfaces on which people walk were built by human labor, and not by God. The population density of Mongolia, in contrast, is 1.87 people per square mile.14 What, though, does this tell us? Perhaps if Manhattan were worse off than Mongolia, Mongolia would be callous to refuse entry to outsiders. But I cannot see why Mongolia’s relatively underpopulated countryside creates anything more normatively significant than that. Risse’s metric is not so simplistic that these facts are sufficient to demand an obligation, on Mongolia’s part, to open its borders to entry from the Manhattanites. But I think it is hard to see why these numbers should be part of our theorizing about global 12 Otis, strictly speaking, designed the safety system, which enabled the widespread adoption of previously available elevator systems. 13 See Mitchell L. Moss and Carson Qing, ‘The Dynamic Populaton of Manhattan’, Rudin Center for Transportation Policy and Management, Wagner School of Public Service, New York University, 2012. Available at https://wagner.nyu.edu/files/rudincenter/dynamic_pop_manhattan.pdf. 14 World Bank Figures, as of 2014. Available at www.tradingeconomics.com/mongolia/populationdensity-people-per-sq-km-wb-data.html.
MICHAEL BLAKE
justice at all – unless one has already decided that land itself, rather than the forms of life it makes possible, is a good to which norms of justice ought to apply. I confess I do not understood why we should do so. This focus on land, I think, is slightly fetishistic; while land is needed for humans, our focus should be on the latter, and not on the former. You need land only because you need a life in which your autonomy is respected – and you don’t need much land for that, and it doesn’t actually have to be land, if there are enough buildings around. This means, in short, that there is nothing about the Syldavian maintenance of its borders that is made unjust because of its low population density. Syldavia cannot keep out the truly desperate, whose lives would be saved through entry into Syldavia. I believe Risse agrees with me on that. Nor can Syldavia escape some forms of moral criticism, when it uses its rights in a manner that demonstrates a lack of empathy or care for the lives of outsiders. I think Risse agrees with me on that, too. The only point of disagreement between us, then, is whether or not sharing a single globe represents an independent site of justice, so that population pressure represents an independent basis for justice claims against states. I am convinced, despite Risse’s eloquence, that it does not. We should care about the rights of people to live autonomous lives; land is valuable in the struggle to defend these rights, but we have no reason to think it ought to be anything more than that. IV. CONCLUSIONS
I have been living with the ideas in Justice and Foreign Policy for several years now; as I say in the book itself, it is a short book with a long history. I do not think that I have necessarily succeeded in defending the ideas in the book. I am well aware that – in global justice – we are still in the infancy of the debate; much of what I say in the book will appear wrong, or worse, in years to come. The justification of books like mine, then, must be found somewhere other than in a lasting influence. If there is a justification, I think it is twofold. In the first place, I think it is a valuable exercise to figure out what we might do – here and now – with the institutions we have been forced to inherit at this point in history. This is not the only way in which political philosophy might be done; indeed, it is
AGENCY, COERCION, AND GLOBAL JUSTICE
not the only way in which it should be done, and we might genuinely welcome a more diverse and pluralistic set of methodologies by which global justice might be understood. The second justification, though, is through the conversations that begin from even imperfect books. This, I think, is where the justification for Justice and Foreign Policy might really be found; even if the arguments of that book are not widely accepted, they are – at the very least – likely to be rejected in interesting ways. For these conversations, as for so much else, I am enormously grateful to my critics. I thank them for their specific arguments, as well as for the ongoing arguments we have shared about global justice. I am grateful for the chance to be a part of these arguments, and hope they continue in the future. University of Washington, Box 353350 Seattle, WA, 98195, USA E-mail:
[email protected]