Severe poverty as a violation of negative duties: reply to the critics.

Pogge, Thomas

BASELINES FOR DETERMINING HARM

Mathias Risse discusses whether the global system of territorial sovereignty that emerged in the fifteenth century can be said to harm the poorer societies. This question is distinct from the question I raise in my book--namely, whether present citizens of the affluent countries, in collusion with the ruling elites of most poor countries, are harming the global poor. These questions are different, because present citizens of the affluent countries bear responsibility only for the recent design of the global institutional order. The effects of the states system as it was shaped before 1980, say, is thus of little relevance to the question I have raised. A further difference is that whereas Risse's discussion focuses on the well-being of societies, typically assessed by their GNP per capita, my discussion focuses on the well-being of individual human beings. This difference is significant because what enriches a poor country (in terms of GNP per capita) all too often impoverishes the vast majority of its inhabitants, as I discuss with the example of Nigeria's oil revenues (pp. 112-14). (1)

My focus is then on the present situation, on the radical inequality between the bottom half of humankind, suffering severe poverty, and those in the top seventh, whose per capita share of the global product is 180 times greater than theirs (at market exchange rates). This radical inequality and the continuous misery and death toll it engenders are foreseeably reproduced under the present global institutional order as we have shaped it. And most of it could be avoided, I hold, if this global order had been, or were to be, designed differently. The feasibility of a more poverty-avoiding alternative design of the global institutional order shows, I argue, that the present design is unjust and that, by imposing it, we are harming the global poor by foreseeably subjecting them to avoidable severe poverty.

The argument just summarized defines harm relative to a baseline that is different from the three baselines Risse considers: on my account, the global poor are being harmed by us insofar as they are worse off than anyone would be if the design of the global order were just. Now, standards of social justice are controversial to some extent. To make my argument widely acceptable, I invoke a minimal standard that merely requires that any institutional order imposed on human beings must be designed so that human rights are fulfilled under it insofar as this is reasonably possible. Nearly everyone believes that justice requires more, that an institutional order can be unjust even if it meets this minimal standard; and there is disagreement about what else justice requires. But I can bypass these issues so long as we can agree that an institutional order cannot be just if it fails to meet the minimal human rights standard. Because the present global institutional order falls short of even this minimal standard, and dramatically so, it can be shown to be unjust without invoking any more demanding and less widely acceptable standard.

Imagine for a moment a human world whose economic distribution resembles ours, but whose inhabitants have just sprung into existence. In this fictional world, the more powerful, constituting one-seventh of the population, impose on the rest an institutional order that reserves for themselves the vast majority of income and wealth, thereby leaving a nonconsenting half of humankind with insecure access to the most basic necessities. In regard to such a world, my argument and conclusion would be obvious and all but irresistible. In such a world, clearly, the global poor have a much stronger moral claim to the extra 1 percent of the global product they need for secure access to basic necessities than the powerful have to take 81 rather than 80 percent for themselves.

This thought experiment shows that if you, like most of this world's affluent, do not find my argument and conclusion obvious and irresistible, this is because the radical inequality of our world does have a history. You must be assuming that this history renders the moral claim the powerful have on the disputed a percent of the global product stronger, or renders the moral claim the global poor have on this disputed I percent weaker, than it would be without this history.

However widespread among the affluent, this assumption is wrong. To show this, I discuss actual and fictional histories, including the three additional (historical and counterfactual) baselines Risse considers. I can do this in a purely defensive way. To protect my argument, all I need to show is that considerations invoking such baselines cannot upset my argument.

Risse rejects as excessively speculative counterfactual statements to the effect that there is more severe poverty in the world today than there would be if either humankind had settled into some Lockean state of nature or if the continents had not been unified through European conquest and colonization. There are no knowable facts, he thinks, on the basis of which we could make such comparisons. This skepticism suits me well. If such comparisons are unsound, then they cannot be invoked to damage my argument. Then the moral claim the global poor have to the disputed a percent cannot be undermined by showing that severe poverty would have been at least equally bad without the European conquest or in a Lockean state of nature. And our moral claim to the disputed 1 percent cannot be bolstered by showing that we would have been no worse off without the European conquest or in a Lockean state of nature.

Putting these two counterfactual baselines aside, "the historical benchmark is the only benchmark among the three considered that we can make sense of" (p. 14 right-hand column, this journal), Risse writes, and judges that the last few centuries have brought fabulous improvements in human well-being. This is quite true--at least so long as we look at aggregates and averages. But if we look at individual lives lived near the bottom, the statistics are less rosy. According to the World Bank, the number of people living below its $2 per day international poverty line has increased from 2,478 million in 1987 to 2,735 million in 2001. (2) The number of chronically undernourished human beings continues to hover around 800 million) And the number of children under the age of five dying each year from poverty-related causes continues to exceed 10 million, such an early childhood death being the fate of 19 percent of all human beings born into our world these days. (4) How do Risse's statistics help him answer these individual human beings near the bottom when they ask us how we can justify imposing a global order designed so that it foreseeably produces a huge avoidable excess in misery such as theirs year after year?

Risse's glorious aggregate statistics--the increase in the global average income or in longevity--cannot silence these complaints. To the contrary, they show that the affluence of the nonpoor is increasing by leaps and bounds and that severe poverty is thus ever more easily avoidable. Such statistics can only exacerbate the scandal of severe poverty persisting on a massive scale.

Risse can say that, thanks to global population growth, the global poor constitute a shrinking percentage of humankind. (5) Or perhaps he can even say that, according to some statistical indicators, the world poverty problem is shrinking even in absolute terms. Such progress is better than no progress, to be sure. It means that severe poverty may one day be eradicated from this planet and that, over all of human history, fewer human beings will have suffered and died from severe poverty than would otherwise be the case. But all this cannot lessen the complaint of those who avoidably suffer and die against those who confine them to a life in grinding poverty.

To see this, consider a parallel case involving slavery. (6) Imagine once more a human world whose inhabitants have just sprung into existence. In this world, the more powerful whites impose an institutional order that facilitates and enforces the enslavement of blacks. This order and its imposition are unjust. Clearly, blacks have a strong moral claim to control their own bodies and labor power, and whites have no moral claim at all to treat black people as tradable commodities.

At this point, Risse's doppelganger enters the scene, arguing that this conclusion about the imaginary world without history cannot be simply transferred into the actual world of 1845, where the citizenry of the United States was imposing an institutional order that facilitated and enforced the enslavement of blacks. The actual world of 1845 was different, says the doppelganger, because it had a history, and a benign one at that: The proportion of slaves within the U.S. population (or even the absolute number of slaves) had been shrinking, the nutritional situation of slaves had steadily improved, and brutal treatment, such as rape, whipping, and splitting of families, had also been in decline. Let us stipulate, for the sake of the argument, that the doppelganger's historical assertions are entirely accurate. Do they weaken, in any way, the slaves' moral claim to legal freedom? Or do they support, in any way, a moral claim by the citizenry of the United States to perpetuate the institutional order that facilitated and enforced the enslavement of blacks?

Faced with this challenge, Risse has opted to answer these questions in the negative, thus dissociating himself from his doppelganger's argument. He recognizes that this saddles him with a new task. He must now explain why his invocation of an upward historical trajectory should have moral relevance against the complaint by the global poor when it has no moral relevance against the complaint by slaves in the United States of 1845. Risse begins to do this by highlighting three purported differences between the two scenarios: Blacks "were relegated to an inferior status. This evil can also be attributed to a group of perpetrators," and both groups were "participants in a single society sharing economic and political institutions" (p. 15 left-hand column this journal). Risse does not say which of these points render historical improvements relevant to present injustice, so let us consider all three.

The last two purported differences are easily denied: There is a group of perpetrators in both cases--namely, the citizens of the United States in 1845, and the politically influential global elite of the affluent in zoos. And just as there was a single society with shared social institutions in the United States of 1845, so there is now "one continuous global society based on territorial sovereignty" worldwide (p. 9 L, this journal).

The first difference is real: blacks were a rigidly designated group of persons with inferior legal status, while the global poor are not as such rigidly singled out and relegated to an inferior status by current legal instruments. But why should this difference make a decisive moral difference?

To see that it makes little moral difference, we need only imagine the U.S. system of slavery modified so that anyone can fall into hereditary slavery under universalistic rules, perhaps through failure to repay a debt on schedule. Let us couple this modified system with the previous stipulation that the proportion of slaves within the population (or even the absolute number of slaves) has been shrinking, that the nutritional situation of slaves has steadily improved, and that brutal treatment has also been in decline. Do the stipulated historical improvements, in this modified case, render justifiable the citizens' imposition of an institutional order that facilitates and enforces the enslavement of defaulting debtors and their progeny? If Risse answers in the negative, then he still owes us an explanation of why he thinks that a decline in the plight caused by severe poverty over the last few centuries renders justifiable our continued imposition of a global order that is designed so that it foreseeably reproduces avoidable severe poverty on a massive scale.

The Content of Cosmopolitanism

I respond only briefly to the second half of Risse's critique because I fully agree with him that we should reject what he calls "Pogge's claim that the sheer existence of states harms the poor" (p. 17 L, this journal). It is true that I consider myself a cosmopolitan. But if it is posited that "cosmopolitans take the existence of states, and a global order composed of them, to be wronging individuals by failing to respect their moral equality" (p. 15 L, this journal), then I must decline the label. I am not a cosmopolitan in this sense, because I do not believe Risse's empirical assertion that "the existence of states entails that life prospects differ vastly and are largely decided by birth" (p. 15 L, this journal). As my proposal for a Global Resources Dividend (pp. 196-215) makes clear, I think that radical inequality can be avoided and economic human rights securely maintained within a global system of states.

To be sure, I have advocated a vertical dispersal of political authority, which would expand the role and impact of supranational rules and organizations (pp. 168-95). But this view has become rather commonplace in the fifteen years since I wrote the essay on which that chapter is based. In fact, precisely such an expansion has been occurring and accelerating, paradigmatically in the ever more consequential rules and agencies of the WTO and the European Union. If I am to be characterized as a radical, then it should not be because I, too, advocate such an expansion, but because the design of supranational rules and organizations I envision differs substantially from the design that is being implemented by the world's affluent and politically influential.

Here I agree with Risse that our present "global order is ... imperfectly developed: it needs reform rather than a revolutionary overthrow" (p. 18 L, this journal). Minor redesigns of a few critical features would suffice to avoid most of the severe poverty we are witnessing today. In this sense, we are not far from a global institutional order that would satisfy the minimal human rights standard of justice. But I cannot agree with Risse that we should therefore refrain from calling the present global order unjust (ibid.). While the reforms needed for the sake of severe poverty avoidance are indeed small, the effects of our continued imposition of an unreformed global order are immense. It foreseeably causes millions of avoidable deaths from poverty-related causes each year. This is an imperfection. But it is also a massive crime against humanity.

A MINIMAL STANDARD OF SOCIAL JUSTICE

Alan Patten correctly identifies my central conclusion: that we, the citizens and governments of the affluent countries, in collusion with the ruling elites of many poor countries, are harming the global poor by imposing an unjust institutional order upon them. He rejects this conclusion on the ground that it runs into a fatal dilemma. This dilemma arises from two different interpretations of how I propose to understand the word "unjust." On the one hand, he might interpret me as endorsing a minimalist procedural conception of justice. This conception would indeed be widely accepted (at least as part of a plausible view of justice), and it would show the present global order to be unjust. But overcoming this injustice would do little to reduce the world poverty problem. On the other hand, he might interpret me as endorsing a maximalist substantive conception of justice. This conception would show the present global order to be gravely unjust. And overcoming this injustice would eradicate or substantially reduce the world poverty problem. But this conception is also much more demanding than advertised and far too radical to be widely acceptable. Thus, whichever way one interprets my argument, Patten concludes, it fails to achieve its purpose.

Patten's dilemma leaves my argument undented because I do not, and need not, accept either of his interpretations. Begin with the first horn of his dilemma. My critique of the asymmetrical requirements WTO rules impose on rich and poor countries suggests, Patten writes, "a baseline of justice that is quite minimal and intuitive" (p. 22 R, this journal). This is the procedural conception of justice that the first interpretation presents me as invoking--though I think the word "formal" might better capture what Patten has in mind: a global institutional order is unjust if it treats like participants unequally.

Patten correctly reports that I find the WTO treaty unjust on account of the special exemptions and privileges the rich countries managed to enshrine in it. Yet, when I point out a flaw in a novel, I am not suggesting that any novel without this flaw is flawless. Analogously, my critique of the WTO treaty does not suggest that any institutional order without such special exemptions and privileges would be just. I state often throughout the book, and also imply in its title, that the standard of social justice I invoke is a human rights standard. Evidently, an institutional order can be minimally fair, in the sense of treating its participants equally, and nonetheless foreseeably reproduce avoidable human rights deficits on a massive scale. That I would consider such an order unjust is clear beyond any reasonable doubt.

So let us consider Patten's second interpretation of my view and the objections he advances against it. I discuss these four objections in a different order from his, showing along the way how my view differs from Patten's second interpretation.

No Unduly Stretched Meaning of "Harm"

Patten's main objection is that my substantive standard of social justice is so strong

 that, in almost any situation in which an affluent
 person is connected with a badly off person,
 the affluent person will count as
 "harming" the badly off person, so long as
 there is some institutional scheme the affluent
 person could bring to bear that alleviates the
 suffering of the badly off. And this is tantamount
 to saying that by failing to help
 (through institutions) the affluent person
 would be harming. (pp. 26-27, this journal)

This characterization of my view leaves out the fact that my standard of social justice is sensitive only to human rights deficits. More importantly, I hold affluent persons morally responsible for a given human rights deficit only if four further conditions are all met: The affluent persons must cooperate in imposing an institutional order on those whose human rights are unfulfilled. This institutional order must be designed so that it foreseeably gives rise to substantial human rights deficits. These human rights deficits must be reasonably avoidable in the sense that an alternative design of the relevant institutional order would not produce comparable human rights deficits or other ills of comparable magnitude. And the availability of such an alternative design must also be foreseeable. These five conditions are surely not met by "almost any situation" in which someone can help a badly off person. In particular, my minimal standard of justice does not require us to create an institutional order with people whose human rights are unfulfilled, even when we can foresee that its creation would lead to the fulfillment of their human rights. (7)

Moreover, even when all five conditions are met, our obligation to compensate is limited to the amount of harm for which we become responsible by cooperating in the imposition of an unjust institutional order. My standard of social justice entails not, then, an open-ended duty to help the badly off, but a much narrower duty that is tightly limited in range (to persons subject to an institutional order you cooperate in imposing), in subject matter (to the avoidance of human rights deficits), and in demandingness (to compensation for your share of that part of the human rights deficit that is reasonably avoidable through an alternative institutional design).

So there is no abuse of language. Incorporating the five conditions, the standard of justice I employ really is minimal And I am not "skillfully reconceptualizing the concept of harm" (p. 26 L, this journal) so that any failure to help counts as a harming. On my view, you harm others insofar as you make an uncompensated contribution to imposing on them an institutional order that foreseeably produces avoidable human rights deficits.

Reaching Out to Libertarian Sympathizers With these matters set straight, we are better able to tackle Patten's second objection, which holds that few libertarians are likely to accept my standard of social justice. He illustrates this with my statement that a national institutional order is unjust when it gives rise to excessive domestic violence that could reasonably be avoided through an alternative institutional design. Patten objects that this view "rests here on a normative premise that libertarians do not accept: namely, that society should protect its most vulnerable members" (p. 27 R, this journal). It should now be clear that I invoke no such premise, but defend my statement quite differently: As citizens of this country, we are collectively imposing an institutional order on all persons residing within its borders. We hold ourselves entitled to promulgate rules and to enforce them on all, irrespective of their individual consent, and we hold them to be under an obligation to comply and liable to punishment if they do not. It would be unjust to do all this to those living within the jurisdiction we assert if the institutional order we impose on them were not minimally just--that is, designed so that human rights deficits are avoided insofar as this is reasonably possible.

Invoking not a positive duty to protect the vulnerable but a negative constraint on which institutional schemes it is permissible to impose, this line of argument may appeal to some libertarians. It is not unlike Nozick's famous (but flawed) argument to the effect that certain risky activities are permissible on condition of compensation--which leads him to conclude that dominant protection agencies may forcefully protect the rights of their clients only on condition that they also offer minimal protection to the rights of all those others living within range of their enforcement activities. (8) In any case, among this world's affluent there are few full-fledged libertarians, yet many who sympathize with their views. Most of these sympathizers reject a general justice requirement for tax-funded protection and aid of the vulnerable. But most also accept a justice requirement for tax-funded suppression of assaults, including domestic violence. This suggests that my argument is not in vain. Exactly contrary to Patten's diagnosis, it can and is meant to reach those who discard as phony or feeble all positive duties to aid and protect the vulnerable. My argument can justify to them stringent obligations toward the global poor that, like our obligations to suppress crime domestically, derive from a negative duty not to harm others by cooperating in imposing on them an institutional order that foreseeably produces avoidable human rights deficits. Each of us can avoid harming others in this way by making compensating protection and reform efforts for the victims of the injustice to which we are also contributing.

Justice Is Consistent with Fair Procedures

Patten's third objection is that my standard of social justice may run afoul of procedural fairness. The example he gives is that of a world in which severe poverty can be avoided only if poor countries are allowed to maintain protectionist barriers that are forbidden to rich countries (p. 25 R, this journal). Protectionist barriers may not, in fact, contribute to poverty avoidance, except perhaps for brief transition periods. Patten's example may then be unrealistic. But we can readily think of realistic examples: requiring higher tax payments from the more affluent, or permitting poor but not rich countries to impose currency controls against speculative inflows. My minimal standard of social justice would require such institutional asymmetries when they are needed to avoid human rights deficits. But I do not find them problematic--certainly not if they are couched in general terms. And I don't think that my endorsement of them is inconsistent with my critique of the exemptions and grandfather clauses through which the rich countries have served their own interests within the WTO scheme. (9) These latter asymmetries were imposed by the rich countries, without a sharable justification, through a naked exercise of their superior threat advantage. They violate the rich countries' own oft-proclaimed commitment to free and open markets, thus manifesting deceit and hypocrisy. Most importantly, these special privileges have foreseeably aggravated human rights deficits, adding uncounted millions to the ranks of the global poor. (10)

Justice Is Not Unfair in Its Demands

Patten puts most emphasis on his fourth objection: "The main problem I see with specifying the baseline in terms of the proposed substantive conception of justice is that, in focusing so much on the position of those who fall below the threshold, the proposal ignores possible claims of justice among people who are above the threshold" (p. 26 L, this journal). The example Patten gives is of poor countries with persistent severe poverty that could be eradicated either through domestic institutional reforms, imposing opportunity costs on the domestic ruling elites, or through global institutional reforms, imposing opportunity costs on the populations of the affluent countries. The poverty persists because neither side is initiating the requisite institutional reforms. The question Patten raises about this scenario is whether it makes sense to say that the affluent countries are harming the global poor. He rejects my affirmative answer as "unfair" to the affluent countries on the ground that they are counted as harming the global poor only because the ruling elites of the poor countries are not fulfilling their obligation of justice to reform their domestic institutional schemes.

But is my answer really unfair? It is true that I would not count the affluent countries as causing severe poverty if the poor-country elites fulfilled their duties of justice so that severe poverty is avoided worldwide. But it is equally true that, in Patten's example, I would not count the poor-country elites as causing severe poverty if the affluent countries fulfilled their duties of justice so that severe poverty is avoided worldwide. The two groups are treated symmetrically: each counts as harming the poor and each ought to stop harming them through appropriate institutional reforms.

It follows from this symmetry that, if Patten is right to reject the charge of harming and the obligation to stop on behalf of the affluent countries, then the elites of the poor countries have equally good reason to reject this indictment and duty. They, too, can say that they should not be held responsible for severe poverty in their country when the domestic institutional order they impose would not give rise to such poverty but for flaws in the global institutional order for which they bear no responsibility. Patten's account absolves both parties and allows them, together, to produce great harms.

To see how implausible this is, consider two factories releasing effluent into one river. (11) Each factory's chemicals, by themselves, are harmless to the downstream population. But mixed together they are highly toxic and kill many. Given symmetrical placement of the fully informed factory owners, we must either hold both of them responsible or neither. It would be evidently absurd to contend that neither is harming the downstream population and that both may thus continue their releases. This contention would exemplify a "moral loophole," which can be displayed through fictional histories and puzzles of equivalence (pp. 85-88). We imagine that the two factories once belonged to the same owner, who came to realize that she was acting very wrongly by discharging pollutants that kill so many people downstream. Eager to avoid such wrongdoing, she sells one of the factories. Pollution and deaths continue as before, but now--if Patten were right--without wrongdoing. Clearly, this cannot be right. It is incredible that such a merely cosmetic change reverses the moral assessment of the chemical releases, thus enabling the people upstream to bypass the rights of the people downstream. And it is incredible that the merely cosmetic difference between the one-owner and two-owner scenarios has such a dramatic impact on what may be done to the people downstream.

To avoid these implications, Patten's absolution could be restricted to scenarios where some of the active parties are situated causally "upstream" from others and, in such scenarios, to the more upstream party. With this modification, Patten can say that the poor-country elites, whose decisions directly affect poverty, are harming the poor, whereas the affluent countries, whose decisions affect poverty only indirectly, should be absolved of responsibility.

Patten could afford to narrow his contention even further. He could say that the more upstream agent should be absolved of responsibility only if a further asymmetry also obtains: while the conduct of the midstream agent would cause some harm no matter how the upstream agent behaves, the conduct of the upstream agent would cause no harm if the midstream agent behaved in a harm-avoiding way. Patten could contend that the case of global poverty exemplifies this asymmetry as well: the global institutional order is such that severe poverty would be avoided completely if the poor countries had just institutional schemes and policies. But the converse is not true. As currently designed, their domestic institutional schemes and policies foreseeably give rise to some avoidable severe poverty, no matter how the global institutional order may be designed. These national institutional schemes are unjust in themselves, and it is only because of their injustice that our global institutional order is aggravating severe poverty. Under these circumstances, we should not (Patten might therefore conclude) hold the affluent countries responsible for the excess poverty that is avoidable through global institutional reform.

Let us build both modifications into our homely example. You own the upstream factory. You know that, no matter what you do, the midstream factory is releasing chemicals that kill some people downstream. You know that the chemicals you release, by themselves, would harm no one. And you continue to release your chemicals while knowing that your effluent, mixed with that from the midstream factory, greatly increases the harms that people downstream suffer from the pollution of their water. Is your conduct permissible, or is it unduly harming the people downstream?

Or consider another realistic case, closer to that of severe poverty. Under current global rules, arms manufacturers and arms merchants are free to sell most types of weapons to rulers in the poor countries. As a consequence, at least $167 billion worth of arms have been sold into the developing world in the 1996-2003 period. (12) Very predictably, these weapons are overwhelmingly used in immoral ways, mostly for internal repression, and people in the poor countries are also burdened with the cost of these weapons, which their oppressors buy without their consent and all too often deploy against them. This case satisfies both limitations I have added to Patten's account: The international arms trade is causally upstream from violent rule in the poor countries. And such violent rule would cause much harm even if there were no imported arms, while imported arms would cause no harm at all if the poor countries had governments that behave morally. By Patten's lights, this should then be a clear-cut case: the present arms export regime harms no one, and we may continue it even though we know that it greatly increases harms from wars, civil wars, and government repression in the poor parts of the world. This strikes me as a reductio ad absurdum of his position. When we have seen, time and again, how Indonesia's soldiers are using their weapons, we are morally required to bear the opportunity cost of foregoing arms sales to them. Yes, this requires self-restraint from us even while Indonesia's soldiers are not practicing the self-restraint morally required of them. But we owe this self-restraint nonetheless to all those innocent people whom the arms we deliver would kill, maim, and oppress.

In the last five paragraphs, I have assumed in Patten's favor that the cause of harm he wishes to exonerate does not influence the other causes, but merely aggravates the effects of these other causes. Thus, I have assumed that the upstream factory worsens not the chemical releases by the midstream factory but only the effects of these releases. I have assumed that the international arms trade does not worsen the quality of the rulers of the poor countries but only the effects of their bad rule. And I have assumed that the global institutional order does not worsen institutional schemes and policies in the poor countries but only aggravates their effects. In the last two real-world cases, however, this assumption is false. The international arms trade does worsen the quality of rulers in the poor countries, because superior weapons enable even widely hated rulers to maintain themselves in power. And the international resource and borrowing privileges, central features of the present global order, likewise strengthen greatly the staying power of repressive regimes as well as the incentives toward predatorial takeovers in the poor countries, especially those with large resource sectors (pp. 112-16, 153-66). We allow such putschists and autocrats to sell us natural resources and to incur repayment obligations to us in behalf of the people they oppress. We allow them to buy from us the high-tech weaponry they need to continue their rule. In these ways we ensure that many of the poor countries will be ruled by force of arms and ravaged by wars, civil wars, and coups. Yet Patten appeals to all this violence and repression, which we sustain, as his basis for arguing that we are harming no one and should not be expected to make any reforms!

This last thought provides a new perspective on Patten's protest that we in the affluent countries should not be asked to restrain ourselves for the sake of severe poverty avoidance so long as the ruling elites in the poor countries are continuing to enjoy the fruits of the domestic injustice they uphold. Some of the most plausible reforms we can institute at the global level are ones that will impose opportunity costs not merely on ourselves but also on those elites--by greatly reducing their ability to rule by force of arms, for example, and their opportunities to salt away their country's wealth in private bank accounts abroad.

THE CONTENT OF HUMAN RIGHTS

Rowan Cruft agrees, by and large, with what I write about our negative responsibility for severe poverty. He disputes what he takes to be my denial of positive responsibilities or positive duties to protect or assist. He bases his interpretation on my statement that "human rights entail only negative duties." The full sentence, however, reads: "The most remarkable feature of this institutional understanding is that it can go well beyond minimalist libertarianism without denying its central tenet: that human rights entail only negative duties" (p. 66; see also pp. 13, 70). By failing to deny this libertarian tenet, I am not asserting it. Rather, I am trying to build an argument that is widely acceptable by leaving open whether human rights entail any positive duties.

At the core of my book is the view that the human rights of others impose upon us a negative duty "not to cooperate in the imposition of a coercive institutional order that avoidably leaves human rights unfulfilled without making reasonable efforts to aid its victims and to promote institutional reform" (p. 170; see also pp. 70, 144). The human rights of others may impose further duties upon us, positive or negative ones, but my argument is meant to avoid any commitment, one way or the other, with regard to such duties.

It goes against this ecumenical spirit, and therefore was a mistake of mine, to have written: "In proposing this institutional understanding, I reject its interactional alternatives: I deny, for instance, that postulating that persons have a human right to X is tantamount to asserting that some or all individual and collective human agents have a moral duty--in addition to any legal duties they may have in their society--not to deny X to others or to deprive them of X" (p. 65). This mistake may have helped to mislead Cruft, and I apologize for it. I stand by my defense of an institutional understanding of human rights. But I do not want to deny (or assert) that human rights also impose positive or negative interactional duties. Taking a position on this matter is unnecessary for the book's argument and hence best avoided.

What Cruft asserts against me is as inconsistent with this ecumenical view that I do hold as it is with the view he incorrectly ascribes to me (that human rights entail only negative duties). He asserts that, whether I want to or not, I am committed to positive duties to protect or assist: "all rights--including negative rights--will, in appropriate circumstances, entail some duties that are positive in the act/refrain sense, and arguably some duties that are positive in the assistance/noninterference sense too" (p. 31 R, this journal). He provides three distinct arguments for this claim.

Before discussing these arguments (in a different order from his), let me clarify how the first two of them are distinct because they aim at subtly different conclusions. There are two ways in which a (human) right may entail a duty (see p. 65). It may do so directly, through correlativity. In this case, the right imposes the duty, and any violation of the duty is therefore ipso facto a violation of the right. Alternatively, a right may entail a duty indirectly, typically by virtue of the fact that any plausible justification of the right would also justify the duty. When a right entails some duty only indirectly, then this duty is not imposed by the right, and a violation of the duty is not a violation of the right.

Are There Positive Duties Correlative to Human Rights?

Cruft argues that there are. The access of human beings to food and drink may be threatened solely on account of climatic or environmental factors, or solely on account of severe mental or physical disabilities. And the fulfillment of their human right in such cases may well depend on positive acts of assistance. "If human rights have the content Pogge proposes, then they must entail some positive duties to assist those who are severely disabled" (p. 35 R, this journal).

This first argument fails. Talk of the content of a human right is ambiguous. In a thin sense, a human right's content is its object: that which the right is a right to. In a fuller sense, its content is given by specifying which moral claims this right gives its bearer, with regard to its object, against which other agents. Content in the thin sense does not fix content in the full sense. We may agree that each human being has a right to life and yet disagree about whether one has a moral claim to a stranger's kidney, or to her funding one's medical treatment, when one's life depends on it. The human rights idiom affords such ample opportunities for empty rhetoric precisely because human rights have traditionally been formulated in terms of thin content only, leaving their full content unspecified. Everyone has a human right to X, they say in tones of moral indignation and confident of their deserved applause. And the applause comes, because all listeners can painlessly endorse the right so long as its fuller content is left unspecified: Yes, it ought to be the case that all human beings have access to X! And, yes, the human right to X gives each human being some moral claim with regard to X against some other agents. (Here each applauding audience member vaguely thinks of these moral claims as ones against a group of agents to which, as it happens, he or she does not belong.)

Going beyond such cheap and harm-entrenching rhetoric, I take steps to specify the moral claims that each human right gives its bearers. Here I do not endorse the "maximalist" view that a human right to X gives you a moral claim on everyone else that they each do whatever is in their power to ensure that you have X (p. 64). Instead, I defend an institutional understanding according to which a human right to X gives you a moral claim against all others that they not harm you by cooperating, without compensating protection and reform efforts, in imposing upon you an institutional order under which you lack secure access to X as part of a foreseeable and avoidable human rights deficit. I try to convince my readers that they are committed to at least this much. My argument asserts no more. In particular, it does not assert (or deny) that, in the absence of a shared institutional scheme, the human right to food gives the disabled or the draught-stricken a moral claim to aid.

Cruft finds my argument overly cautious because the denial of human-rights-based positive duties is "morally unattractive" (pp. 33 L, 35 L, this journal). (13) I disagree. When a duty is correlative to a human right, then any violation of that duty is a violation of this right. If I took the position Cruft urges me to take, I would be branding as human rights violators people who fail to aid and protect others whose access to the objects of their human rights is insecure. As human rights are generally understood, it is in principle permissible to use force to stop human rights violations. So I would be endorsing the further proposition that, even in the absence of a shared institutional scheme, such positive duties to aid and protect are in principle enforceable. I do not want to endorse this proposition because I do not find it obvious that the denial of such enforcement permissions is morally unattractive. This denial is compatible, after all, with stringent but unenforceable positive duties to aid and protect. Still, I am not urging or endorsing such a denial, but merely remain uncommitted on this question. The argument of my book does not in any way depend on asserting (or denying) any positive duties correlative to human rights.

Are There Positive Duties Indirectly Entailed by Human Rights?

Cruft's second line of argument takes the indirect route: "Pogge overlooks the fact that if human rights are individualistically justified then there will be prima facie justification for many assistance duties entailed by these human rights" (p. 37 L, this journal). His reasoning goes as follows (pp. 36-37, this journal): Human rights are individualistically justified by appeal to basic needs. If these basic needs are of sufficient moral importance to justify human rights not to be deprived of what is required to meet them, then they are also of sufficient weight to justify positive duties to aid and protect. I am committed, for example, to a human right not to have imposed upon one an institutional order under which one foreseeably suffers avoidable insecurity in one's access to physical integrity and minimally adequate nutrition. This commitment makes sense only on the assumption that it is of great moral importance that human beings have secure access to physical integrity and minimally adequate nutrition. But if this is of great moral importance, then it follows that we have positive duties to protect others from torture and to give them access to needed foodstuffs, when we can do so at low cost and risk.

Libertarians will disagree with this inference. They believe in a right not to be tortured, and they justify this right in individualistic terms. But they think they can justify the right in a way that does not compel them to acknowledge any positive duty to protect others from torture when one can do so at little cost and risk. Regarding this disagreement, I am on Cruft's side and share his belief that there are positive duties to aid and protect others whose human rights are under threat. But my book deliberately does not take sides in this disagreement, and its argument works equally well whether or not Cruft's inference can be made fully convincing.

Are There Positive Obligations Entailed by Human Rights?

Unlike Cruft, I draw a terminological distinction between duties and obligations (p. 172). (14) Duties are morally fundamental and apply to us always. Some are generative duties--that is, duties that, in conjunction with appropriate empirical circumstances, create more specific moral reasons for action: obligations. For example, a positive duty to assist persons in acute distress when one can do so at little cost and risk may, in conjunction with an acute emergency, generate a positive obligation to toss a life preserver to some particular drowning swimmer. A negative duty not to make and then break a promise may, in conjunction with a promise to repay some loan, generate a positive repayment obligation.

Cruft goes to some length to show that negative duties entail obligations to take positive action--or, as he prefers, "derivative positive duties" (pp. 30-32, this journal). This discussion is illuminating and I can endorse most of it, even within the argument of my book. I cannot, however, endorse the subtext that this discussion teaches me (and assorted libertarians) a needed lesson: "Pogge has perhaps made the natural assumption that any negative right will entail only negative duties" (p. 30 R, this journal). To the contrary, little would be left of my argument if I had assumed that negative duties cannot generate derivative positive duties ("positive obligations," in my language) of the sort Cruft adduces.

I hold that we have a negative duty not to harm others by cooperating, without compensating protection and reform efforts, in imposing on them an institutional order that foreseeably gives rise to avoidable human rights deficits. This is a generative duty that, in conjunction with our cooperation in imposing an institutional order that foreseeably gives rise to avoidable human rights deficits, generates obligations to make compensating protection and reform efforts for those whose human rights remain unfulfilled under this order. These are positive obligations. They require each of us to make up for our share of the harm we inflict together--by shielding its victims or by working for institutional reforms.

These positive obligations are generated by a negative duty that is correlative to human rights. Failure to fulfill such positive obligations therefore violates human rights. Privileged and competent adults who cooperate in imposing an institutional order that foreseeably gives rise to avoidable human rights deficits without making compensating protection and reform efforts for its victims are contributing to human rights violations.

Far from overlooking negative-duty-generated positive obligations, I have placed them at the center of my argument. One important reason for this is that stringent positive obligations are much more widely acceptable than stringent positive duties. For example, libertarians already endorse stringent positive obligations generated by the negative duty not to break a contract, and they already accept some of the remedial positive obligations Cruft mentions. By contrast, libertarians and their sympathizers tend to be quite skeptical of any and all stringent positive duties.

My argument does not invoke positive duties of any of the three kinds we have distinguished--positive duties correlative to human rights, positive duties indirectly entailed by human rights, and positive duties not entailed by human rights. It can therefore show no more than that we must be concerned for the human rights of those upon whom we help to impose an institutional order. Cruft regrets that I have not sought to show more than what can be derived from this narrow negative duty. But, in the world as it is, most human rights deficits would be avoided if we affluent took this one moral duty seriously--if we did not harm others by cooperating, without compensating protection and reform efforts, in imposing on them institutional rules that foreseeably give rise to avoidable human rights deficits.

PROFITING FROM INJUSTICE

Norbert Anwander appreciates that my main concern is with how the world's affluent contribute to social injustice that harms the global poor. He notes, however, that I occasionally use broader, disjunctive formulations that invoke a duty not to contribute to or to profit from injustice. Questioning whether profiting from injustice has any moral significance, he examines two ways in which it might: There could be a negative duty not to profit from injustice, over and above the negative duty not to contribute to injustice. Or profiting from injustice might modify our duties regarding contribution. Under this latter heading, Anwander considers two possibilities: the obligation of a contributor to injustice to make compensating protection and reform efforts for the victims of this injustice might be more stringent or more demanding when this contributor is also profiting from the injustice. Anwander asserts that profiting from injustice is not morally significant in any of these three ways. Closely following the structure of his essay, I discuss and refute these three assertions in this order.

But let me first point out that Anwander's discussion is slightly marred by a misunderstanding. As he reads me, I postulate negative duties not to contribute to and not to profit from injustice, and I then add that those who have violated either of these duties thereby acquire remedial duties to compensate for their wrongdoing. The duties I actually postulate require less than Anwander imagines. These duties do not make it wrong to contribute to, or to profit from, a collective injustice when one makes compensating protection and reform efforts for its victims. To explain why such conduct may not be wrong, I write about the first duty:

 It does not follow that one must stop contributing
 to the economy of an unjust society--though
 it may come to that in extreme
 cases. One can often continue to contribute
 and yet avoid collaborating in the undue
 harming of others by taking compensating
 action: by making as much of an effort, aimed
 at protecting the victims of injustice or at institutional
 reform, as would suffice to eradicate
 the harms, if others followed suit. (Endnote:
 Oskar Schindler, as depicted in Steven Spielberg's
 Schindler's List, compensated for his
 contributions to the economy of Nazi Germany
 by protecting some of its victims.) (pp.
 135-36)

Through his manufacturing activities and tax payments, Schindler cooperated in imposing the social institutions and policies of the Third Reich. But doing this allowed him to compensate (more than adequately) for his contributions to harm through protection efforts for its victims. His conduct complied with the negative duty I postulate--no less fully than if he had left Germany. In fact, Schindler did much better by the victims of the Third Reich than he would have done by emigrating.

In responding to Anwander, I take the liberty of modifying his criticisms so as to correct for this misunderstanding. His points are more interesting when brought to bear on the view actually articulated in my book.

The Duty Not to Profit from Injustice Regarding the first assertion, Anwander correctly rejects any general negative duty not to profit from injustice without compensating protection and reform efforts. Such a duty can be refuted by example. Illustrated by his Hiroshima case, one class of counterexamples features historical injustices that can no longer be mitigated and whose victims are now dead. These counterexamples are not fully convincing, because the compensation requirement might be interpreted as carrying an implicit insofar-as-possible rider. Another class of counterexamples may feature certain profitings that cannot be declined by their beneficiaries (see pp. 40-41, this journal). Whether they want to or not, all people everywhere profit from breathing air that is cleaner than it would be if large numbers of human beings were not unjustly kept in extreme poverty and thereby severely constrained in their polluting activities. Still, it is plausible that people not involved in sustaining this injustice owe no compensation to the global poor pursuant to a negative duty not to profit from injustice.

With this issue resolved as he advocates, the controversy between us is then about whether there is a more specific negative duty not to profit from injustice without compensating protection and reform efforts. Anwander does not deny that it is sometimes wrong to act so that one benefits from injustice. His own case of the three planets shows this: When the Martians shower us Earthlings, in ways we cannot block, with the spoils of injustices they visit upon the Venusians, then we must restore these goods to the Venusians if we can. It would be wrong to keep the spoils and thereby to benefit from injustice. Anwander insists, however, that what makes such conduct wrong is that it violates some duty other than a duty not to profit from injustice without compensating protection and reform efforts. Thus, he invokes in the three-planets case our duty not to contribute to injustice: "In holding on to what rightfully belongs to someone else and thereby preventing the restoration of the goods to their owners we are perpetuating injustice. Alternatively, we can say that by holding on to what rightfully belongs to someone else we are preventing justice from being restored. This is a way of contributing to injustice" (p. 42 L, this journal).

One might win here rather too easy a victory for Anwander's conclusion by simply stipulating that each instance of wrongfully profiting from injustice without adequate compensation efforts ipso facto counts as a contribution to injustice. With this stipulation, the second duty I postulate would indeed become redundant--but only because it is now definitionally incorporated into the first duty. As I understand him, Anwander does not make this trivial move, but allows that profiting from injustice is defined independently from contributing to injustice. If this is right, then the conclusion he asserts would require him to demonstrate three points: that each instance of wrongfully profiting from injustice without adequate compensation efforts is, as a matter of fact, also an instance of contributing to injustice; that contributing to injustice figures in every such instance as a wrong-making feature; and that profiting from injustice does not figure in any such instance as an (additional) wrong-making feature.

I believe that none of these three points can be defended. But let me respond economically by recalling that, within the argument I address to adherents of broadly consequentialist conceptions of social justice, I formulate the first duty to cover only one specific contribution to injustice-namely, cooperation in imposing an institutional order that foreseeably gives rise to avoidable human rights deficits without making compensating protection and reform efforts for the victims of this injustice. Most U.S. citizens in 1845 were in violation of this duty through their uncompensated cooperation in imposing a national institutional order that facilitated and enforced slavery. They paid their taxes and complied with fugitive slave laws without working toward abolition or toward mitigating the horrific hardships inflicted on slaves. Did these U.S. citizens also profit from the injustice? Perhaps many, especially in the North, did not. But many, especially in the South, clearly did--most obviously by owning and using slaves, but also indirectly by purchasing cheap slave-produced commodities. I believe these latter citizens, insofar as they made no adequate protection and reform efforts, violated an additional negative duty not to profit from--or (as I write in subsequent work) not to take advantage of--a human-rights-violating institutional order without making adequate protection and reform efforts.

Anwander might say that by owning slaves (and by purchasing cheap slave-produced commodities without compensation?) one is doing one's slaves an injustice and thus contributing to injustice in some broad sense. But in the narrow sense that contributing to injustice has within my argument, it is quite unlikely that every instance of slave owning and every instance of purchasing cheap slave-produced commodities without compensation contributed to the imposition of the unjust institutional order. This shows that, in the context of my argument, the second duty is not redundant. U.S. citizens in 1845 had two distinct duties. They violated one negative duty insofar as they made uncompensated contributions to upholding an unjust national institutional order (by paying taxes, complying with fugitive slave laws, and so on), and/or another negative duty insofar as they, without compensation, took advantage of the unjust ownability of blacks.

The Duty Not to Profit from Injustice Applied to World Poverty

What does all this mean for today's great injustice of institutionally enforced severe poverty? Very little, thinks Anwander:

 There is a distinction to be made between
 profits that people actively seek out and benefits
 that simply accrue to them. But this may
 leave Pogge with a dilemma: while the descriptive
 claim that we are all benefiting from the
 global order owes its credence to the notion of
 passively being benefited by, the normative
 claim that by benefiting we are violating a negative
 duty is most plausible if this is understood
 as actively seeking to take advantage of.
 (p. 43 R, this journal)

Starkly put, Anwander maintains that I can either charge us affluent with merely passive profiting from injustice (such as breathing less-polluted air) or with active profiting. The former charge is ineffective because, though we are indeed passively profiting from injustice, doing so is not wrong. The latter charge is likewise ineffective because, though active profiting is indeed wrong, we are doing no such thing--at least not over and above our contributing to injustice.

I disagree. Like the slave owners of 1845, the world's affluent today are actively taking advantage of the injustice of the global institutional order all the time. Sex tourism and sex with women and children whom severe poverty has delivered into the hands of traffickers are common and dramatic examples. Most anything we buy is cheaper than it would be if severe poverty were avoided: If the bottom of the global wage scale were higher than it is today, products containing a poor-country labor component (coffee and textiles, for example) would be more expensive. If the poorer half of humankind were able to exert substantial market demand in competition with ours, or if their autocratic rulers were not recognized as entitled to sell us their countries' wealth, scarce natural resources (crude oil, metals) would also be more expensive.

Anwander may respond that we are merely passive beneficiaries of the avail ability of these things at current prices. This is true. But by buying them at these prices we are actively taking advantage of injustice--like the affluent foreigner employing an aboriginal driver at half the wage other drivers receive (p. 43 R, this journal). This does not mean that we do wrong to buy such things, of course--only that we should not pocket the gain. Like the affluent foreigner, we can supplement unjustly low prices for services we buy in poverty-stricken areas. Closer to home, we can buy fair-trade products or (perhaps more plausibly) make a compensating donation to an antipoverty organization such as Oxfam. In actively profiting without adequate compensation, we are violating a negative duty (or so I have argued). And this duty is distinct from our negative duty not to cooperate in imposing any human-rights-violating institutional order without making compensating protection and reform efforts for the victims of injustice.

Anwander speaks of a positive duty in this context, by the way, likening it to a positive duty of gratitude. As explained (p. 197), I see both as negative duties, albeit ones that may generate positive obligations: the duty of gratitude (supposing there is such a thing) requires no positive action unless one has accepted benefits. It is, therefore, a negative duty: we are not to accept certain benefits and then decline to reciprocate. Likewise with the duty here at issue: we are not to profit (actively) from an unjust institutional order and then decline to make compensating efforts for its victims. At least insofar as they suffer harm (because the compensating efforts made by the contributors to injustice are inadequate), these victims' moral claim to additional compensation is stronger than our moral claim to retain our profits from injustice.

Can Profiting Render More Stringent the Obligations of Contributors to Injustice?

Anwander's second assertion is that the obligation of a contributor to injustice to make compensating protection and reform efforts for the victims of this injustice is no more stringent when this contributor is also profiting from the injustice. He supports his assertion with an example: We have possessed for some time two stolen objects that we know to be of great and equal value to their respective rightful owners. One of these objects has been of great benefit to us, the other of no benefit at all. We now have an opportunity to compensate one, and only one, of these rightful owners. Ought we to compensate the one whose object has been of great benefit to us, in preference to the other?

Anwander believes that it makes no difference, morally, which owner we compensate. This may be plausible on his assumption that the debts are equal in magnitude. But if, as I have argued, we may owe more compensation to the owner from whose object we have benefited, then it may well be that, other things being equal, we should discharge the larger of our moral debts.

In any case, Anwander's example is far from my concerns in the book, which does not discuss where we have most reason to direct our compensation efforts (though this is very much worth discussing). It discusses the stringency of our obligations to make compensation efforts for the global poor. Here is a detailed passage on this issue:

 We can realistically end our involvement in
 their severe poverty not by extricating ourselves
 from this involvement, but only by ending
 such poverty through economic reform. If
 feasible reforms are blocked by others, then we
 may in the end be unable to do more than mitigate
 some of the harms we also help produce.
 But even then a difference would remain,
 because our effort would fulfill not a duty to
 help the needy, but a duty to protect victims of
 any injustice to which we contribute. The latter
 duty is, other things being equal, much
 more stringent than the former, especially
 when we can fulfill it out of the benefits we derive
 from this injustice. (p. 211, emphasis added)

What I had in mind here is that, because unjust features of the global institutional order advantage the affluent in many ways, we profit from injustice through most ordinary economic transactions: the rewards for our labor are higher, and many commodities we buy cheaper, than they would be under a global institutional order designed to avoid foreseeable human rights deficits. To be sure, it is impossible to quantify the extent to which we profit. But that we profit to some extent renders more stringent our obligation to compensate.

This can be illustrated with a small-scale example of the kind Anwander prefers. On your Belize vacation, you have been involved with four others in organizing a spectacularly successful beach party with fireworks. You are in charge of running the wet bar for your own account and, after all expenses are paid, are looking at a $600 surplus. There was a slight mishap at midnight, when the risky fireworks display you five had advertised and prepared misfired and destroyed a small fishing boat on which some very poor local families depend for their livelihood. Eager to attract continued tourism to the area, the local authorities are turning a blind eye. Nonetheless, each of you five organizers has a moral obligation to pay one-fifth of the $850 needed to replace the boat. But your obligation, I would think, is more stringent than that of the other four. It is wrong for them to fly home without paying, but more wrong for you to do so with your tidy party profit.

Can Profiting Render More Demanding the Obligations of Contributors to Injustice?

Anwander's third assertion is that, contrary to my view (p. 50), the obligation of a contributor to injustice to make compensating protection and reform efforts for the victims of this injustice is no more demanding when this contributor is also profiting from the injustice. The beach party example illustrates my response to this assertion as well. Suppose you pay your $170 share of the damage, but three of your friends do not. Here one may perhaps say of your remaining friend (who paid) that she has done all she was morally required to do. But one cannot say this of you, I think, if you fly off with your remaining $430 surplus, leaving the poor families with a $510 loss. If three of your friends refuse to pay, you should hand over your entire surplus to the poor families. You have a negative duty not to profit from your beach party with risky fireworks when doing so means that other, innocent parties are harmed by it.

This sort of case is relevant to world poverty, which persists on a massive scale because so few affluent people are adequately compensating for their contribution to the harm we together do. In such a situation, I believe, we must not merely compensate for our share of the harm, but also for any profit from injustice that remains after we have done so. Those who suffer from an injustice we contribute to have a stronger moral claim to these remaining profits than we do. To be sure, it is next to impossible to quantify the compensation efforts we owe for contributing to and (especially) profiting from the injustice of the global institutional order. I will briefly address this issue at the end of my response to Satz.

SOME COMMON MISUNDERSTANDINGS

Debra Satz criticizes various views that are distinct from those I have defended. Because these misunderstandings may be common and because I may be partly to blame for them, I am eager to clear them up. (15)

First, my approach is meant to be "ecumenical" in that I seek to show adherents of different moral conceptions that the world's affluent, in violation of their negative duties of justice, are massively harming the global poor. (16) This is my central thesis, which I seek to demonstrate to diverse audiences by appeal to diverse arguments and baselines. I do not defend the view "that we have a general obligation to aid other human beings in severe need" (p. 47 L, this journal). Nor am I super-ecumenical by asserting merely the disjunction that the affluent are failing to fulfill some positive or negative duties toward the global poor--or, even more blandly, that the status quo should be condemned (p. 53 R, this journal).

Second, in supporting my central thesis, I do not "adopt the libertarian premise that only the failure to perform negative duties counts as harm" (p. 53 R, this journal). For me, the coincidence of duties not to harm and negative duties holds by definition (pp. 13, 66-67, 130-36,144-45, 172, 210). This definitional connection does not prejudge any substantive questions about the stringency of duties.

Third, by focusing on negative duties, I do not "attempt to derive all of our obligations to the global poor from the need to refrain from harming others" (p. 47 R, this journal). Rather, I simply leave positive duties aside. (17) Showing how the affluent are violating their negative duties, I leave aside their failure to fulfill their positive duties toward the poor.

Fourth, Satz argues at length that eating another's candy (in violation of a negative duty) is not worse than letting a child drown (in violation of a positive duty), thereby refuting the claim that each negative duty is more stringent than every positive duty (pp. 51-52, this journal). But I never made or suggested this claim. I do say that negative duties are widely thought to be more stringent than positive duties and that I share this judgment. I go to great length to stress that this unequal stringency is thought to obtain when "what is at stake for all concerned is held constant." (18) Satz's refutation is irrelevant to the much weaker claim I endorse.

Fifth, I am not committed to the view "that we have strong duties not to harm but only weak duties to benefit people we have not harmed" (p. 47 R, this journal). My argument does not invoke positive duties, and I leave open how much or how little they differ in stringency from their negative counterparts.

Sixth, I am not committed to the view "that we can be obligated only by those harms for which we are causally responsible" (p. 51 R, this journal). This view entails that there are no positive duties to protect others from harms inflicted on them by third parties. It is true that I do not invoke such positive duties. But I am not denying them either.

Seventh, I do not endorse "the idea that all our obligations of justice are negative" (p. 51 R, this journal). I do argue, against Rawls, that two important duties of justice related to social institutions are negative duties: We must not participate in just social institutions without a willingness to comply with them; and we must not cooperate in upholding unjust social institutions without making compensating protection and reform efforts for their victims (pp. 134ff.). By arguing for these two negative duties, I do not imply that justice allows social institutions to be designed so that the needs of children and the disabled remain unmet (pp. 52-53, this journal). In fact, these two negative duties of justice imply nothing about the content of social justice, about what justice requires of social institutions. (19) If justice requires that any institutional order must be designed so that all its participants receive fresh flowers every morning free of charge, then we are harming our compatriots (in violation of a negative duty) by cooperating in imposing on them an institutional order under which they do not receive such daily flowers.

Satz seems to recognize that I tie the notion of harm to justice in this way when she objects that my view "erodes the distinction between harming and failing to remedy" (p. 54 L, this journal). It would indeed be absurd to make the accusation that, by helping to impose an institutional order under which most compatriots receive no free daily flowers, we are harming (rather than, at most, failing to benefit) these people. But the absurdity of this accusation does not refute my contention that the imposition of an unjust institutional order harms some of those on whom it is imposed. Rather, the absurdity of the accusation refutes the conjunction of my contention with an eccentric conception of social justice. One of them must go; and, I submit, it should be the latter. The absurdity of the accusation does not illustrate that the imposition of unjust social institutions sometimes does no harm. Rather, it illustrates that the nondelivery of free flowers does not constitute a social injustice.

Satz's worry about erosion brings out an important point: readers will accept that we are harming the global poor by cooperating in imposing unjust social institutions upon them only if I can convince them that these social institutions really are unjust. To make it easier to convince readers on this score, my argument invokes not the full account of what I believe social justice requires, but a much weaker conception of justice that requires merely that any institutional order must be designed so that, insofar as reasonably possible, the human rights of those on whom it is imposed are fulfilled. This minimal requirement implies that those who impose an institutional order that, foreseeably and avoidably, exposes women to domestic violence, blacks to enslavement, serfs to starvation, or the poor to severe deprivation are thereby harming the victims of these avoidable human rights deficits--even if they have no personal contacts with brutalized women, slaves, serfs, or the severely impoverished.

This minimal conception is proposed as merely a necessary condition of social justice. I do not hold that human rights exhaust what justice requires. I merely hold that justice requires at least that any institutional order we impose must fulfill the human rights of those on whom it is imposed insofar as this is reasonably possible. I work solely with this minimal requirement of social justice because it is widely acceptable and suffices to reach my conclusions. That some institutional order foreseeably reproduces an avoidable human rights deficit is thus on my view merely a sufficient, not a necessary, condition for this order being unjust and for its imposition being a violation of a negative duty of justice and hence a harming.

I hope to make this minimal requirement acceptable to nearly all who think of social justice in broadly consequentialist terms-that is, those who assess institutional schemes by their foreseeable effects on human lives. But this requirement may not be acceptable to those who assess institutional schemes in light of their actual or fictional histories. For such readers, different arguments are needed to establish harm, as sketched in the introduction to this symposium.

The preceding four clarifications should lay to rest the idea, floated by Patten, Cruft, and Satz, that I am some kind of libertarian. It is true that I try to avoid commitments to antilibertarian views that libertarians would find offensive. Thus I make my argument without any appeal to positive duties. But it is also true that I try to avoid commitments to libertarian views that antilibertarians would find offensive. Thus, I never deny that there are very stringent positive duties to aid and protect. To be sure, I do deny that positive duties are as stringent as their negative counterparts (holding constant what is at stake for all concerned). I deny, for instance, that an affluent person who, in order to save $80, declines to sponsor a child in Mali with the predictable result that this child dies is acting as wrongly as another affluent person who kills such a child for an $80 benefit. Libertarians would endorse this denial. But so would nearly everyone else.

Eighth, Satz suggests that, in analogy to the explanatory nationalism I reject, I am an adherent of explanatory globalism. This is incorrect. (20) I do not seek "to explain all local failure in terms of failures of the global order" (p. 49 R, this journal). No global institutional order, no matter how well designed, could possibly forestall all local failures. I do hold that most of the severe poverty today would be avoided if the design of the global order were just. I concede that most of today's severe poverty would also be avoided if the poor countries had just social institutions and policies. These two beliefs are consistent because causal contributions are often not additive. (21) The case of the two polluting factories (p. 63, this journal) illustrates this: either one of the factory owners could single-handedly stop most of the harm they together cause downstream.

Satz is right that there is considerable empirical uncertainty about why exactly severe poverty persists at such a high global rate. She is right that "various countries are unlikely to agree as to how much harm is caused by global as opposed to local institutions" (p. 50 L, this journal). But I need not achieve agreement on this. I must make plausible that most severe poverty today is avoidable through reforms in the design of the global institutional order and that it is possible to design specific reforms that would work. I need not show of any such reform that it "would necessarily generate better alternatives across the board" (p. 50 L, this journal). A reform can be justified even if it will not reduce severe poverty everywhere (for example, in North Korea), and even if the substantial harm reduction we expect of it is not necessary but merely highly likely in light of the empirical facts and correlations as we understand them.

Consider, for example, the present global rules for incentivizing pharmaceutical research. These rules reward the inventors of new drugs by allowing them to charge monopoly prices for twenty years. (22) During this period, even lifesaving drugs must not be produced and sold cheaply by generic producers and are therefore priced far out of the reach of the global poor. Moreover, these rules skew medical research toward the affluent: medical conditions accounting for 90 percent of the global disease burden receive only 10 percent of all medical research worldwide. Of the 1,393 new drugs approved between 1975 and 1999, only 13 were specifically indicated for tropical diseases, (23) and 5 out of these 13 actually emerged from veterinary research. (24) I have sketched a detailed reform proposal envisioning instead that pharmaceutical research is rewarded according to its impact on the global disease burden. (25) Such incentives could be funded, for instance, through a global Polluter Pays regime that raises funds from countries in proportion to their citizens' and corporations' contributions to transnational environmental pollution.

This reform proposal needs to be more fully worked out--something I hope to do in the next few years. Even then we will not be able to predict with precision what the difference in deaths and suffering would be between the two alternative regimes for incentivizing pharmaceutical research. This is so because we cannot foresee what remedies against currently neglected diseases pharmaceutical companies would invent if attempting to do so were worth their while. Still, it is undeniable that the present rules foreseeably lead to an avoidable lack of affordable drugs for the most destructive communicable diseases, including pneumonia, HIV/AIDS, diarrhea, tuberculosis, malaria, measles, meningitis, hepatitis, dengue fever, leprosy, sleeping sickness, Chagas disease, river blindness, leishmaniasis, Buruli ulcer, lymphatic filariasis, and schistosomiasis (bilharzia). These diseases account for millions of deaths and unimaginable suffering each year. (26) This is justification enough for undertaking the reform effort, even if we cannot know in advance by what percentage the reform would reduce this burden of communicable diseases.

Similar considerations apply to the other modifications of the global institutional order I discuss: reduced protectionism that would open rich-country markets to exports from the poorest countries (pp. 17-18), a stake in seabed resources for the global poor (pp. 125-26), a global resources dividend (chapter 8), and an opportunity for poor populations to disallow future resource sales and debts incurred in their name by autocratic rulers (chapter 6). Even if I have not spelled out any of these reforms in full detail, it is clear that each of them, if well designed, would have a huge impact on the incidence of severe poverty. The two last-mentioned reforms would partly have this impact indirectly, by reducing the staying power of putschists and autocrats and by reducing the incentives toward the undemocratic acquisition and exercise of political power (pp. 163-64, 206-207). These reforms would not end autocratic rule and severe poverty everywhere. But they would substantially reduce their incidence and thereby promote the national and local factors Satz adduces to dispute my belief in the great causal importance of global institutional arrangements: "sound financial institutions, laws that enforce contracts and promote competition, and independent judiciaries ... decent policies for women and children, educational opportunities, and a free press" (p. 50 L, this journal). These factors are not exogenous but strongly dependent on the extent to which global institutional arrangements encourage or discourage good governance.

THE RESPONSIBILITY OF INDIVIDUAL CITIZENS

Satz's most fundamental challenge regarding the allocation of responsibility among the world's affluent is her suggestion that most ordinary citizens of wealthy countries bear no responsibility for world poverty:

 Can we hold citizens of all of the world's developed
 nations responsible for the policies of the
 IMF and the World Bank? Here, the agency relationships
 are more indirect. The IMF is
 accountable to finance ministers and central
 bank governors, and its officers are not elected
 but rather appointed by agreement of governments.
 Further, the voting arrangements in the
 IMF ensure the disproportionate influence of
 only a few developed countries, in particular
 the United States. Because IMF policies are
 most often debated in secret, most people are
 unaware of the policies they debate. There is little
 accountability for international institutions
 and even less information about their policies
 than about domestic ones. What exactly is our
 responsibility here? To what extent do the
 unfair (indeed sometimes ruinous) policies of
 the IMF render us responsible participants in an
 unjust global order? (pp. 50-51, this journal)

Satz is surely right that responsibility for decisions that foreseeably result in millions of avoidable deaths rests in the first instance with the politicians and negotiators who make them. Such "honorable" and otherwise unremarkable people have knowingly committed some of the largest human rights violations the world has ever seen. But does their guilt absolve ordinary citizens of responsibility?

Our politicians and negotiators wield powers we delegate. Their decisions and agreements would be of little consequence if they were not so empowered by us. We may choose to pay no attention to, and we may even allow them to conceal, what they are doing with the powers we lend them. Can we thereby disconnect ourselves from responsibility for how our collective power is wielded in our name? An affirmative answer would once more manifest a moral loophole, indicated by the vested interest we would have in perpetuating such convenient intransparency: So long as political decisions are made in obscurity, we get the benefits of unjust policies and of an unjustly structured world economy plus the clean conscience of bearing no responsibility for the vast human rights deficits such injustice produces. If our international financial organizations were to inject transparency and accountability into their proceedings, we would lose this sweet spot (afforded by Satz's morality) and would be forced to choose between our gains from injustice and a clean conscience.

Satz is quite wrong to believe that the obscurity of political decision-making disconnects us from responsibility. We cannot disown responsibility for how our politicians and negotiators wield our collective power by appeal to our own failure to insist on transparency and accountability. The fact that we choose to remain ignorant, choose to allow important structural features of the world economy to be shaped by unknown bureaucrats in secret negotiations, cannot negate our responsibility for the harms that our governments inflict upon the innocent.

Satz may be right that some affluent countries are too small to affect the global rules of the game by themselves. But several together can certainly influence the outcome of the negotiations in small ways. And small differences in global institutional design can make a large difference to human rights fulfillment. Thus, even the citizens of Luxembourg and New Zealand have a responsibility to instruct their governments to oppose and to help reform global institutional arrangements that foreseeably cause avoidable human rights deficits on a massive scale.

Satz believes that, in order to hold any particular affluent person responsible, I would need to perform two impossible tasks. I must show what the present design of the global institutional order contributes to the persistence of world poverty. And I must show who bears how much responsibility for this present design.

I have not shown these things--certainly not "exactly." It would indeed be good if this could be shown with some precision. One would then have a way of estimating how much each of us affluent individuals ought to contribute toward shielding the global poor from the harms we also cooperate in producing. This would provide an answer, within my negative-duty account, to the question often posed to positive-duty theorists, such as Peter Singer: How much is enough? And it would reinforce my above response to Patten, showing that the negative duty I postulate is indeed "tightly limited" (p. 61 L, this journal).

One could surely do better than I have done. One can estimate that the typical affluent person, by the time of his or her death, bears responsibility for roughly one poverty-related death, for about 200 human life-years spent in severe poverty, and for about 20,000 hours of children suffering intense pain from hunger or diarrhea. But this sort of rough calculation underestimates our responsibility. Most of us belong to some subset of the world's affluent that could single-handedly bring about the needed global institutional reforms. If a substantial number of U.S. citizens mobilized in favor of a poverty-avoiding design of the global institutional order, the U.S. government would promote such a design and would carry the other affluent countries along. The same is true of the European Union, and perhaps of Japan. And the same may be true even for much smaller groups, such as EU academics, U.S. journalists, or employees of the international financial organizations (WTO, IMF, World Bank). A substantial proportion of the members of each of these groups could, by itself, initiate a reform process that would eradicate most severe poverty worldwide within a few years. Insofar as the typical affluent person belongs to such smaller groups, his or her responsibility is correspondingly greater. In this way, the sum total of what we are individually responsible for may greatly exceed the total. There is nothing odd about this. If three persons each make a necessary contribution to a homicide, each of them is fully responsible for the resulting death. As it is, many groups of affluent persons make necessary contributions to the massive persistence of severe poverty.

The calculation of personal responsibility becomes even messier when we take account of individual circumstances. I agree with what Satz suggests in her example of the laid-off steelworker (27)--that citizens who were born into an affluent family, have enjoyed an excellent education, and have a good job, wealth, and influence bear more responsibility for their country's policies than citizens with the opposite characteristics. Yet, even with all the care and information in the world, we cannot model such judgments algorithmically to calculate, say, Warren Buffet's exact share of responsibility, or the ratio of his share to Oprah Winfrey's.

How important is it that a precise allocation of responsibility for complex collective harms is impossible? This impossibility certainly does not show that individuals bear no responsibility. (It would be absurd to hold that members of the Nazi Party bore no responsibility for the concentration camps because the precise share of each cannot be calculated.) Nor does this impossibility invalidate the judgment that nearly all affluent persons today are evidently failing to compensate for their share of the collective harm, on even the most miserly estimate. (28)

Moreover, for those who are genuinely sensitive to moral reasons, a precise allocation of their share of the collective responsibility would not fully settle the matter. Recall the Belize beach party (p. 73 R, this journal), and now assume that none of you five organizers has profited from it in any way. You pay to the poor families your one-fifth share of the replacement cost of the boat that your risky fireworks destroyed. But your friends leave without making any compensation. Would you be leaving Belize with a clear conscience of having harmed no one through your contribution to the fireworks?

Averting one's fair share of the harm we together cause may also seem insufficiently political. We are citizens of countries whose governments are producing great harms in our names. This calls on us to try to put the issue on the political agenda. Here, too, one might envision a sharp limit, so that we can discharge our negative duties "by making as much of an effort, aimed at protecting the victims of injustice or at institutional reform, as would suffice to eradicate the harms, if others followed suit" (p. 136). But again, this limit will leave a moral person uneasy: Having made the required effort on the political front, would you no longer feel any responsibility for the harm your country inflicts?

These thoughts are related to the forward-looking, practical intent of my book. My concern is not to pass judgment, in the polite company of academics, about ordinary citizens and their shortcomings. I mean to talk to my fellow citizens in the affluent countries about how we can, together, fulfill our responsibilities of citizenship (see p. 166). Included in the conversation, the laid-off steelworker may be more insulted than relieved by being put, with children and the severely mentally disabled, among those unfit for the responsibilities of citizenship. Exempting poor and marginalized citizens from responsibility for their countries' policies is an affront to those laid-off steelworkers, janitors, and single mothers in the affluent countries who take responsibility and act accordingly.

We all have a vote, a voice, friends, influence. Even a few thousand of us can change the world forever, as Margaret Mead is said to have emphasized. (29) The people of Manchester proved this in 1787 when they joined the uphill battle against slavery with a petition signed by 11,000 of them. The campaign against slavery endangered their livelihoods, because much cotton from slave-labor plantations was processed in Manchester. And most of them were already much poorer than we can imagine. Those who had no money to give supported the cause in whatever way they could. Women especially, though greatly constrained by law and convention, supported the movement, contributed needlework with antislavery images and inscriptions ("Am I Not a Woman and a Sister?"), and refused to buy sugar (slave-grown in the West Indies). These ordinary people did not blame it all on the African slave hunters or on Liverpool merchants or British politicians. They did not stop to ask what the degree of their country's complicity was, exactly, or their own exact share of responsibility, nor how much could reasonably be expected from people like themselves. They did not plead poverty, powerlessness, or ignorance. Nor were they deterred by the low odds of success. These working-class men and women of eighteenth-century Manchester understood better their shared responsibility for the misery inflicted half a world away than do today's sophisticated journalists and political philosophers. If they could recognize and stop their country's crime, then so can citizens of a depressed steel town in today's United States.

Presumably Satz would think that the working-class people of 1787 Manchester shared no responsibility for the enslavement that happened half a world away, on that horrific "middle passage" and in the Americas. She would not fault them for mobilizing against slavery, to be sure. But she would judge that they had no obligation to do this. They really knew too little, they were not educated enough, they were too poor and too challenged by their personal poverty, their political clout was insignificant, and so on.

It is noteworthy that the people of Manchester did not want to be counted out in this way and did not see their effort as supererogatory. "We know about slavery," they would say to Satz, "we know that the cotton we touch every day has been planted and gathered by slaves. We know that ships flying our country's flag are carrying slaves across the Atlantic. We know we are part of this injustice and we know we must try to put a stop to it--for the sake of our black brothers and sisters and also for the sake of our country. We understand that success is unlikely and may be impossible. But we know that we must make the attempt."

Perhaps Satz would accept their assessment. But then they would ask her on what grounds she exempts many of us from responsibility for the global institutional order our affluent countries impose: "Are your compatriots poorer or less educated than we are? Do they have less spare time or lesser opportunities to inform themselves? Are they less able to make their voices heard and to influence your society's political process?" In all these dimensions, disadvantaged citizens of today's affluent countries score much better. They do and can know much more about the world and the horrific poverty it contains on such a massive scale. They are much better educated and vastly more secure economically and in their civil rights. They have much more free time and much greater political opportunities in today's affluent democracies. So what will Satz answer the working-class men and women of Manchester?

I do not profess to know what citizens in Manchester thought and felt 218 years ago. But what I know about them suggests that they saw their mobilization not as an onerous task regrettably required by religious or moral duty but as a necessary component of a life worth living and as an urgent service to their country. In any case, this is the inspiration I want to convey to the citizens of the affluent countries today. I am not writing against you, to make you feel guilty or to present you with an itemized bill for wrongful damages done. I am writing for you, to suggest that we can lead much better, happier lives in a much better country if we are willing to do without that bit of extra affluence now purchased for us with rivers of blood, sweat, and tears of the global poor.

It is not hard to see that the opportunity cost of eradicating severe poverty is small: though 44 percent of the people in the world are counted as poor by the World Bank's $2 per day threshold, their collective shortfall from this international poverty line amounts to barely 1 percent of the global product. The U.S. share of the rich countries' opportunity cost of avoiding severe poverty worldwide would be around $100 billion annually, not much more than the $70 billion annual cost of the current "humanitarian" occupation of Iraq. (30)

It is harder to see that a determined effort at reform is likely to succeed. Knowing the enormous magnitude of death and destruction caused by world poverty, one is inclined to assume that many things must go right for such a huge problem to be solved. But, to the contrary, many things must go wrong for it to persist on such a massive scale year after year. In particular, many features of our global institutional order must all be designed and adjusted without giving weight to the imperative of poverty avoidance. And the overwhelming majority of the affluent must manage to avoid facing up to what we are doing. Despite its fearsome magnitude and destructiveness, the world poverty problem is fragile. An intelligent effort by even just 11,000 people could trigger its defeat, as the Manchester mobilization of 1787 triggered the defeat of slavery.

(1) Thomas W. Pogge, World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms (Cambridge: Polity Press, 2002). All in-text citation references are to this book, unless otherwise noted.

(2) Shaohua Chen and Martin Ravallion, "How Have the World's Poorest Fared since the Early 1980s?" World Bank Research Observer 19 (2004), p. 153.

(3) This figure is reported each year by the United Nations Development Programme. The latest figure is 831 million. See UNDP, Human Development Report 2004 (New York: UNDP, 2004), pp. 129-30.

(4) UNICEF, The State of the World's Children 2005 (New York: UNICEF, 2005), inside front cover.

(5) For example, those living below $2 a day in 1987 constituted 49.3 percent of the global population then, whereas those living below $2 a day in 2001 constituted only 44.5 percent; see www.census.gov/ipc/www/worldpop.html.

(6) See Thomas Pogge, "Real World Justice," Journal of Ethics 9 (2005), pp. 38-39. This essay is a defense of my book, first presented at the American Philosophical Association Eastern Division Meeting, Washington, D.C., December 30, 2003, where Risse, Patten, and Satz were the featured critics of an author-meets-critics session.

(7) See the example of human rights deficits on Venus in Pogge, World Poverty and Human Rights, p. 198. To be sure, there may well be a duty to create a common institutional scheme when this is needed to fulfill human rights. But this question falls outside the scope of my book, which focuses on the moral claims persons have, by virtue of their human rights, on any institutional order imposed upon them and hence against those who are imposing this order.

(8) Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), ch. 4. See also Thomas W. Pogge, Realizing Rawls (Ithaca: Cornell University Press, 1989), p. 19, n. 6.

(9) Examples abound and have now come to be publicly deplored even by pillars of the establishment. A classic example of grandfathering is the so-called Peace Clause, Article a3 in the WTO Agricultural Agreement, which protected the agricultural subsidies of the affluent countries; see www.tradeobservatory.org/head lines.cfm?RefID=18901. Some of the abuses are summarized in a recent speech,"Cutting Agricultural Subsidies" by World Bank chief economist Nicholas Stern; available at globalenvision.org/library/6/309. See also the book by Stern's predecessor, Joseph Stiglitz: Globalization and Its Discontents (New York: W. W. Norton, 2003).

(10) Citing research by the IMF, Nicholas Stern estimated that rich-country protectionism in textiles alone entails 27 million lost jobs in developing countries. "Every textile job in an industrialized country saved by these barriers costs about 35 jobs in these industries in low-income countries" (Stern, "Cutting Agricultural Subsidies"). By depressing wage levels, such unemployment aggravates severe poverty far beyond the ranks of the unemployed and their extended families.

(11) See Pogge, "Real World Justice," p. 48.

(12) See Conventional Arms Transfers to Developing Nations, 1996-2003 (Washington, D.C.: Congressional Research Service, 2004), p. 43; available at www.fas.org/man/crs/RL32547.pdf.

(13) I suspect that Cruft's sentiment here is partly explained by his occasionally losing sight of the fact that there are two other kinds of positive duties: those indirectly entailed by human rights and those not entailed by human rights at all. Thus, he writes, for example: "If human rights are negative rights that entail no other-directed precautionary duties, then you cannot be subject to such a duty. But this overlooks the ties of community and fraternity that should bind us all" (p. 33 R, this journal). Clearly, there are important positive (and negative) duties that are not entailed by human rights--duties to ensure that others not inflict cruelty on animals, for instance (and duties not to lie).

(14) The distinction is most fully developed in Thomas Pogge, "O'Neill on Rights and Duties," Grazer Philosophische Studien 43 (1992), pp. 133-47.

(15) I leave her important challenge regarding individual responsibility to the following section.

(16) See Pogge, "Real World Justice" pp. 36ff.

(17) Ibid., pp. 34-36.

(18) Ibid., p. 34. See Pogge, World Poverty and Human Rights, pp. 132, 240 n. 207, 241 n. 216.

(19) See Pogge, "Real World Justice," pp. 45-46.

(20) Ibid., pp. 44-53.

(21) Ibid., pp. 47-48.

(22) This regime was created through the Agreement on Trade-Related Aspects of Intellectual Property Rights, concluded in 1995.

(23) Medecins Sans Frontieres, Fatal Imbalance: The Crisis in Research and Development for Drugs for Neglected Diseases (Geneva: Medecins Sans Frontieres, 200l), pp. 10-11; available at www.msf.org/source/access/2001/fatal/fatal.pdf.

(24) Patrice Trouiller, Els Torreele, Piero Olliaro, Nick White, Susan Foster, Dyann Wirth, and Bernard Pecoul, "Drugs for Neglected Diseases: A Failure of the Market and a Public Health Failure?" Tropical Medicine and International Health 6, no. n (200l), pp. 945-51; available at www.neglecteddiseases.org/tmih.pdf.

(25) See Thomas W. Pogge, "Human Rights and Global Health: A Research Program," in Christian Barry and Thomas Pogge, eds., Global Institutions and Responsibilities, special issue of Metaphilosophy 36, nos. 1-2 (2005), pp. 182-209.

(26) See, e.g., WHO, Macroeconomics and Health: Investing in Health for Economic Development (Geneva: WHO Publications, 2001); available at www.cmhealth. org/www.cid.harvard.edu/cidcmh/CMHRepor t.pdf. The commission that produced this report, chaired by Jeffrey Sachs, concluded that some 8 million deaths could be prevented each year in the poor countries through real access to medical care at a cost of about $60 billion annually.

(27) She asks: "Is a laid-off American steelworker ... really more responsible for global poverty than a rich citizen of a poor country?" (p. 51 L, this journal). The answer is: certainly not. The present global institutional order is designed by, and for the benefit of, the political and economic elites of both rich and poor countries. The rich citizen of a poor country thus typically shares responsibility for global institutional arrangements. In addition, he or she also shares responsibility for the national institutional order of his/her country, which typically promotes domestic corruption and severe poverty.

(28) Of course, we ought not to employ such a most miserly estimate of our compensation obligations because, seeing how very badly off the global poor are relative to us, underestimates (to their detriment) are vastly more consequential than overestimates. See Christian Barry, "Applying the Contribution Principle," in Barry and Pogge, eds.: Global Institutions and Responsibilities, pp. 210--27.

(29) The statement, "Never doubt that a small group of thoughtful, committed citizens can change the world. Indeed, it is the only thing that ever has," has been attributed to her.

(30) See United Press International, "Iraq War Topping $5.8 Billion A Month," November 17, 2004.

Thomas Pogge, Many thanks to the editors of Ethics & International Affairs and my fellow symposiasts for making this exchange possible, and to David Alvarez Garcia, Nicole Hassoun, Keith Horton, Rekha Nath, and Ling Tong for their critical comments.