The Lottery You Cannot Leave

NATIONS do have moral obligations beyond their borders. But they are not obligations of charity. They are obligations of justice, and they arise because every country takes part in a single global order – a web of trade rules, financial systems and treaties – that decides, before anybody has done anything to deserve it, how good a chance at a decent life a newborn will get.

That argument is worth making against the most careful defence of the opposite view. John Rawls, probably the most influential political philosopher of the twentieth century, was right to reject limitless cosmopolitan redistribution, and right to insist that a theory of global justice describe something states could actually do. But his case rests on a single assumption – that countries are independent enough to be held responsible for their own fate – and under modern economic dependency that assumption collapses.

Once we recognise that the global order is itself a “basic structure” in Rawls’ own sense, his logic runs straight past the borders he drew around it, and it produces a ladder of duties: first to stop causing harm, then to correct the structures that produce it, and last to help anyone still suffering.

Rawls’ World of Peoples

Rawls provides the most sophisticated defence of limited global obligation there is – and he does it by constructing a world that no longer exists, one containing only liberal societies. In The Law of Peoples, published in 1999, he extends the social contract – the idea that a society’s rules are whatever free and equal people would agree to – outwards to the world. He imagines a second “original position”: a bargaining table whose representatives sit behind a veil of ignorance, choosing the rules of international life without knowing which country they will turn out to represent, so that nobody can rig the terms in their own favour.

Crucially, the parties at that table are not individuals. They are “peoples”. For Rawls, peoples rather than states or single human beings are the moral actors of international justice, because a people can sincerely commit itself to an obligation in a way that a state – a legal machine, with no moral intentions of its own – cannot. That choice does a great deal of quiet work. If justice operates between societies rather than between individuals, then distributive justice – the business of deciding how the good things in life are shared out – stays inside borders.

What crosses the border instead is what Rawls calls the Duty of Assistance. Reasonably well-ordered societies owe targeted help to “burdened” ones: societies whose conditions are severe enough that they cannot build decent basic institutions on their own. He attaches two guiding points: a society does not need to be rich in order to be well ordered, and the duty switches off the moment a country can sustain those institutions by itself, however poor or unequal it remains afterwards.

The Premise That No Longer Holds

Rawls justifies his framework through a collection of overlapping political and moral commitments. He places great weight on self-determination: societies must be free to determine their own political and economic arrangements without external influence, which leads him to reject paternalism – the habit of overruling people for their own supposed good – even where the resulting outcomes are grossly unequal.

This is reinforced by his claim that principles of distributive justice apply only within a shared basic structure, meaning the framework of major institutions – constitution, courts, markets, property law – that shapes people’s life prospects from birth onwards. In the absence of a global basic structure, Rawls argues, there is no basis for extending egalitarian principles beyond borders, which rules out any global version of his difference principle – his domestic rule that inequalities are acceptable only if they leave the worst-off better than they would otherwise have been. What remains is a Westphalian logic, named after the 1648 settlement that made the sovereign state the basic unit of European politics: morally distinct countries, owing one another nothing beyond the harm they immediately cause.

But this premise is precisely what globalisation has dissolved. Allen Buchanan makes the case bluntly: Rawls’ Law of Peoples is a set of rules for a vanished Westphalian world, one in which states genuinely determined their own economic fate. That is no longer true.

Trade agreements set the terms on which a country can sell what it produces. Financial architecture decides whose currency and whose central bank effectively set the cost of borrowing for everyone else. Intellectual property regimes decide who can manufacture a medicine and who must pay a licence fee for it. None of these were chosen from behind anything resembling a veil of ignorance; they were negotiated by unequal parties, and they now function as a genuine “basic structure” – Rawls’ own term – that operates above and across the level of the individual state.

If that is right, then Rawls’ own logic, applied consistently, no longer stops at the border he drew. He argued that principles of justice apply wherever a basic structure exists, because a basic structure shapes life prospects that individuals never chose and cannot exit. A global trade and financial architecture does exactly that. The Duty of Assistance, designed for a world of separate self-determining societies, cannot be the whole answer once the structure producing a “burdened” society’s condition is partly external to it.

A Ladder of Duties

None of this collapses into limitless cosmopolitan redistribution, which is where thinkers like Peter Singer go further than the argument here needs to. Singer’s case – that a comfortable person is obligated to give until the marginal cost to them equals the marginal benefit to someone in extreme poverty – asks too much and specifies too little about who exactly owes what to whom. The better answer is a ladder, not a flat obligation.

The first rung is a negative duty: stop causing harm. A rich country’s trade rules, tax policy or resource extraction that actively worsens conditions elsewhere is not a case for charity; it is a case for stopping. Thomas Pogge’s argument that current global institutions actively contribute to the persistence of severe poverty belongs here – if a wealthy state benefits from an arrangement it helped design and that arrangement predictably harms others, its first obligation is simply to stop benefiting from the harm.

The second rung is a duty to correct: reform the structures – trade rules, debt architecture, intellectual property regimes – that a state helped build and continues to profit from, even where the harm was not deliberately intended. The third and weakest rung, arising only once the first two are satisfied, is a genuine duty to assist those still suffering, closer to Rawls’ own Duty of Assistance but now grounded in participation in a shared global structure rather than in bare compassion.

This is, in a sense, an old idea recovered rather than a new one invented. Luke Glanville traces a “law of nations” tradition running from Vitoria through Grotius to Kant, in which even a limited, sceptical view of intervention still held that a state should “sacrifice a little to help others a lot” – a phrase he attributes specifically to Emer de Vattel. The three-order ladder above is a modern version of exactly that instinct, made precise enough to apply.

Conclusion

Rawls built a careful, defensible theory of global justice for a world of independent peoples, and its caution is not a flaw. But that world’s central premise – that a country’s fate is substantially its own to determine – no longer describes the one we live in.

A newborn’s chance at a decent life is shaped before birth by a global basic structure that nobody outside the wealthiest states had any real hand in designing. Recognising that does not require abandoning Rawls for Singer’s more demanding cosmopolitanism. It requires taking Rawls at his own word: that justice follows wherever a basic structure reaches, borders included.