What does justice require from wealthy nations toward the global poor?

We must distinguish several questions that Thomas Pogge and Peter Singer are too quick to run together: humanitarian urgency, institutional fairness, and the proper subject of distributive principles. I do not deny the urgency. But the remedy is not simply to impose a global difference principle upon the society of peoples. If existing trade, finance, and patent rules prevent burdened societies from establishing just or decent political and social institutions, then those rules should indeed be revised. No reasonable law of peoples permits powerful societies to entrench arrangements that block others from achieving the conditions of self-respect, security, and effective political order. In that sense, I would reject keeping unjust rules and merely giving more as compensation. But I would also reject Pogge’s suggestion, if he means it, that justice requires continual cross-border equalization of wealth. The aim of the duty of assistance is a target: to help burdened societies become well-ordered. Once peoples can sustain just or decent institutions on their own, the duty has been met. Differences in wealth thereafter may reflect distinct political cultures, social choices, and conceptions of the good. Singer, for his part, presses a demanding moral ideal of beneficence, but justice among peoples cannot rest solely on the aggregation of urgent needs. It must specify fair terms of cooperation between free and equal peoples. So: abolish unjust rules where they undermine institutional development; reform the global framework accordingly; and provide assistance directed to enabling just or decent institutions—not endless transfers without a political objective.

If the rules are foreseeably and avoidably reproducing massive deprivations, then we must not “keep them and give more.” That proposal merely taxes the victims for the cost of their own subordination. Justice begins with negative duties: not to impose an institutional order that foreseeably produces life-threatening poverty. So yes: those features of trade, finance, resource, borrowing, and patent regimes that sustain radical inequality and premature deaths should be abolished or fundamentally redesigned. The question is not whether the affluent should be generous enough to share some gains. The question is whether we are entitled to keep gains extracted through a coercive global order that we shape to our advantage. We are not. Rawls is right that institutions matter. But he is wrong to frame the issue mainly as a duty to assist “burdened societies.” That formulation obscures our causal role. The global poor are not merely unfortunate; they are also harmed by rules that grant international borrowing privileges to dictators, resource privileges to any group controlling the state, protectionist asymmetries, and patent rules that price medicines beyond reach. When we uphold such arrangements, we do not merely fail to help. We collaborate in an unjust harm. Singer rightly stresses the urgency of saving lives, but charity-talk is morally too weak. What is required is not only more aid, though compensation may also be owed. What is required first is structural reform: redesign the rules so that severe poverty is no longer reproduced for our benefit. Only then can assistance cease to function as a moral cover for injustice.

If the rules really do save advantages for the affluent at the cost of millions of avoidable deaths and severe suffering, then of course those rules should be changed. To keep them and then hand back a fraction in aid is morally shabby. You do not push a child into a pond and then earn praise for throwing a lifebuoy. So on that point, Pogge is right against Rawls: if the global order is foreseeably producing deprivation, structural reform is required, not merely assistance aimed at getting societies to some threshold of decency. But I would add that even a juster institutional order will not discharge our responsibilities. People are dying now. If I can prevent something very bad from happening, without sacrificing anything comparably important, I ought to do it. Nations, no less than individuals, are in that position. Rawls’s target conception is too undemanding. Why should my obligations to a child threatened by malaria end once her society becomes “well-ordered”? If I can still save her life, at little cost to myself, justice — and ordinary morality too — continues to require it. Borders do not nullify the claim of suffering. So the answer is not one of the three options taken alone. Abolish or redesign unjust trade, finance, and patent rules where they inflict grave harm. Build decent institutions where they are lacking. And meanwhile, give much more — far more — through the most effective means available. The existence of structural injustice does not weaken the duty to rescue; it makes that duty even more urgent.

I should call refusal an injustice when, and because, the rules in question cannot be justified to other peoples as fair terms of cooperation. If affluent democracies maintain trade, tax, or patent arrangements that foreseeably keep burdened societies from achieving just or decent institutions, then refusal to surrender those advantages is not mere hard-heartedness; it is unjust. But I must resist both Pogge and Singer at the points where they overreach. Pogge is right to insist that institutional design matters. Yet he moves too quickly from the fact of harmful arrangements to a broadly redistributive conclusion. The Law of Peoples does not treat the world as one domestic basic structure to which a global difference principle applies. The object is not to maximize the position of the globally least advantaged indefinitely, but to secure a just or decent political order for all peoples. Singer, similarly, presses a principle of rescue so far that it threatens to dissolve the distinction between duties of justice and the wider claims of beneficence. That distinction matters. A reasonable political conception must specify what free and equal peoples owe one another as a matter of right, not only what compassion might counsel without limit. So my answer is this: yes, abolish or revise those rules whose benefits to the rich depend upon denying fair background conditions to poorer societies. No, justice does not require affluent democracies to surrender every advantage whenever doing so would further reduce inequality or suffering abroad. The duty is stringent, but it has an end: enabling peoples to stand on their own in a well-ordered society of peoples.

I would call such refusal an injustice, plainly so, when the advantages in question derive from an institutional order that foreseeably and avoidably reproduces severe poverty. Here Rawls still understates the matter. He asks whether rules can be justified as fair terms among peoples. I ask the prior question: are we, through these rules, actively harming human beings? If so, then refusal is not merely a failure to assist. It is participation in an ongoing wrong. This is why the contrast between “abolish the rules” and “give more” is morally clarifying. If patent rules price essential medicines beyond reach, if trade and tax rules facilitate extraction and capital flight, if resource and borrowing privileges empower corrupt rulers to sell a country’s wealth and bind its people to odious debts, then affluent democracies are not innocent bystanders. They are co-authors of the scheme. Keeping the gains and offering aid is, as I have often argued, morally perverse: it leaves intact the machinery of harm while advertising our generosity. Against Rawls, then, I deny that justice is satisfied once poorer societies reach some threshold of decency. If our institutional design continues to extract benefits by exposing many to avoidable hunger, disease, and early death, the wrong persists. Against Singer, I would say: yes, rescue matters urgently; but unless we center causal responsibility, beneficence talk remains too weak and too optional. The first demand of justice is negative: stop upholding rules that enrich us by foreseeably crushing the poor. Reform, and where harms have been imposed, compensation as well.

I would call refusal an injustice, yes. If affluent democracies can prevent massive suffering only by giving up rules that enrich their own citizens, and they refuse, that is not merely a lack of generosity. It is a moral failure of a very serious kind. But I would put the basis a little differently from Pogge. I do not think we need to show, in every case, that we are causally implicated through an unjust global scheme before the obligation becomes stringent. Suppose I walk past a shallow pond and see a child drowning. If I can save the child at the cost of muddying my clothes, I must do it. That remains true whether or not I pushed the child in. Causal responsibility strengthens the case, certainly. It is not necessary to create it. So Rawls’s threshold view is too complacent. If trade, tax, or patent rules protect affluent interests while millions remain in extreme poverty, then asking whether poorer societies have yet become “well-ordered” misses the point. What matters first is the preventable suffering. If surrendering those advantages is the price of preventing it, and the sacrifice is not of comparable moral importance, then justice requires surrendering them. To Pogge I would say: you are right that keeping harmful rules and then giving aid is morally shabby. But even after reform, the duty to give does not disappear. We must change the rules, and we must do far more now, through the most effective means available. Justice is not exhausted by institutional tidiness while children still die for want of cheap remedies.

The primary ground, as I see it, is not sheer capacity to produce a better outcome, nor only causal involvement in this or that injury. It is the requirement that peoples relate on fair terms of cooperation under principles they can reasonably accept. From that basis follow stringent duties: first, to cease supporting institutions that block burdened societies from establishing just or decent regimes; second, to assist such societies until they can manage their own affairs under those conditions. Here I must press both Singer and Pogge. Singer’s view, if made the measure of justice, has no clear stopping point. It converts the political morality of a society of peoples into a limitless duty to maximize relief wherever suffering can be reduced. That, I think, fails to distinguish justice from beneficence, and it does not respect peoples as collective agents entitled to shape their own ends once fair background conditions obtain. Pogge is right to emphasize harmful institutional design. But he still lets the argument slide from rectifying injustice into a standing principle of global redistribution. I deny that. Once fair terms are secured and the duty of assistance discharged, continuing inequalities among peoples may reflect their different political cultures, choices, and social institutions. These are not, as such, injustices to be perpetually canceled. So what is concretely owed beyond institutional reform? Targeted assistance to burdened societies: support for basic public health, education, legal order, and the political conditions of self-government. But not the permanent surrender of prosperity and political independence whenever further transfer would improve aggregate well-being. Justice has a proper aim, and it is not endless leveling.

Rawls still lets the rich off the hook by building the debate around what affluent peoples can reasonably accept, rather than around what poor persons are in fact made to endure through an institutional order we impose. That is the wrong moral starting point. If the global rules under which we live foreseeably generate avoidable hunger, malnutrition, disease, illiteracy, and premature death on a vast scale, then our first duty is not generosity but to stop collaborating in that harm. So yes: if ending most extreme poverty required wealthy nations permanently to relinquish significant prosperity and political control, I would still call that justice whenever those advantages are bound up with unjust supranational rules. One cannot insist on the fruits of coercive arrangements and then describe their relinquishment as heroic self-sacrifice. It is more like returning what one has wrongfully taken, and ceasing to participate in a scheme that predictably crushes others. Singer is right to refuse complacency. But he still frames too much in terms of rescue by the capable. That rhetoric is morally weaker than the case warrants. The global poor are not merely people we might help if we are decent enough; they are people whom our governments and economic elites have long disadvantaged through the resource privilege, borrowing privilege, skewed trade rules, and intellectual-property regimes. What is owed beyond reform? Compensation for imposed harms, institutional redesign, and constraints on our own consumption of unjust gains. Not endless leveling for its own sake, Rawls is right about that. But as long as our prosperity depends significantly on avoidable grave deprivation abroad, surrendering that prosperity is not sainthood. It is the minimum demanded by justice.

If ending most extreme poverty really did require affluent nations to give up significant prosperity and political control, I would still call that justice, provided the sacrifice is not of comparable moral importance to the suffering prevented. That is the point Rawls continues to evade. He wants a stopping point built into justice so that the affluent can say: once poorer societies are “decent,” our serious obligations end. But a child dying from easily preventable disease is not helped by being told that the society of peoples has reached the right threshold. Rawls says my view collapses justice into beneficence. I do not think that is the decisive distinction here. If you can prevent something very bad from happening without sacrificing anything comparably significant, you ought to do it. Call that justice, call it morality, call it whatever you like—the labels do not rescue those who refuse. The practical demand remains. Pogge is right to insist that wealthy nations are often not mere bystanders but participants in harmful institutions. Where that is so, the case is stronger still. But I would resist making causal complicity the gatekeeper of obligation. Even if we could prove no unjust scheme at all, the sheer fact that we can save lives at relatively modest cost would still matter enormously. So why are the others letting the rich off the hook? Rawls does it by building in a moral exit clause. Pogge sometimes does it by tying too much to institutional causation. I would say: reform unjust institutions, certainly. But beyond that, keep giving, and give until further giving would require sacrificing something comparably important. Millions of lives are at stake; our comforts are not.