This essay is part of Boston Review’s special project Democracy’s Promise.

2001 casts a long shadow on contemporary global politics. The War on Terror, which was announced the day after the September 11, 2001, terrorist attacks, would be as Secretary Donald Rumsfeld warned, “a long war . . . that is not going to go away . . . is not going to be settled with a signing ceremony on the USS Missouri.” And indeed, with its defection from international constraints on unilateralism and preemptive war, it is among the United States’s longest wars.

The year associated with the War on Terror also gave rise to two alternative visions of global order—visions that were both couched in the idiom of “international responsibility.”

The War on Terror, however, did not spring fully formed only in response to the atrocity of 9/11. As Asli Bâli and Aziz Rana have argued, we should understand it instead as an escalation of U.S. global power since the end of the Cold War. The response to 9/11 intensified what had been slowly brewing: an “imperial arrogance,” as the Palestinian anti-colonial critic Edward Said has called it, “unschooled in worldliness, unfettered either by competence or experience, undeterred by history or human complexity, unrepentant in its violence and the cruelty of its technology.”

Eighteen years after the attacks on that fateful Tuesday—and still engaged in counter-terrorism efforts on six continents and in eighty countries—the United States has not veered from this course. Indeed, with the election of President Donald Trump and his unsanctioned bombing of Syria, withdrawal from multilateral agreements such as the Iran nuclear deal, sanctioning of Israeli expansion, and recognition of Juan Guaidó as the president of Venezuela, the country has only hastened its departure from international institutions and more clearly projected its unmoored and unfettered imperial arrogance.

This moment of “imperial unraveling” has occasioned a much-needed debate about the crisis of U.S. power, and it presents an important opportunity to remember that the year associated with the War on Terror also gave rise to two alternative visions of global order—visions that were both couched in the idiom of “international responsibility” and that are worth revisiting today. First, at the United Nations World Conference Against Racism, held in Durban from August 31 to September 8, 2001, Caribbean states and non-governmental organizations from the global south called for reparations for slavery and colonialism. The Caribbean reparations movement demanded that the international community take responsibility for its part creating some of the Caribbean’s most intractable problems. At about the same time, in December 2001, the International Commission of Intervention and State Sovereignty (ICISS), which was organized by the Canadian government with United Nations representation, articulated a different model of international responsibility: the principle of “responsibility to protect” (commonly referred to as R2P), which essentially held that wealthy nations have a responsibility to intervene in times of crisis and protect vulnerable populations when states failed to do so. Central to both was an effort to create an international order where states are held accountable for injustice. Indeed, as the reckless and unprincipled War on Terror got off the ground, the architects of “international responsibility” worked in its shadow, seeking to provide models for how we might instead imagine international order.

From climate change to migration, the scale and frequency of the crises that confront us today require a far more robust model of international responsibility than R2P can offer.

R2P is, by far, the better known and more frequently discussed model—especially as Samantha Power’s new memoir has brought its terminology back into popular discussion. Calls to return to its “rule-based,” liberal international order are particularly strong now. But while a return to any sense of “responsibility” is tempting given our current president, it is crucial to understand that these two models view the character of injustice and the scope of states’ responsibility in radically different ways. They draw on divergent intellectual and political genealogies and, in fact, represent competing world views. R2P, for instance, begins with a view of domestic and international politics as bifurcated. As a result, it can only imagine international responsibility as duties of assistance and intervention in times of crisis.

But from climate change to migration, the scale and frequency of the crises that confront us today require a far more robust model of international responsibility. The Caribbean movement for reparations is thus worthy of our attention because it offers an alternative starting point—one that situates responsibility in the historically entangled relationships between the domestic and international. It recovers a tradition of anticolonial internationalism that recognizes all states as unequally positioned within a globalized world that emerged from the legacies of imperialism. On this view, international responsibility is not an extraordinary moment of intervention, but an on-going commitment to build a more egalitarian and just world order.

To fully appreciate the potential of a more robust model of international responsibility, we must first understand the origins and limitations of the conventional model.

R2P’s roots go back to the aftermath of the Cold War, but its support really started to grow with the perceived failure of the international community to act in Rwanda in 1994. At the time, the central question of humanitarian intervention was: How do you square morally legitimate international action with the dictates of international law? This question was brought into stark relief with the 1999 NATO intervention in Kosovo, which was largely perceived as morally legitimate but found to be unauthorized under international law because the UN Charter’s principles of sovereign equality and nonintervention protected states from aggressive wars. For the then-Secretary-General of the UN, Kofi Annan, and members of ICISS, the answer lay in rethinking sovereignty itself. As Annan argued, “The world cannot stand aside when gross and systematic violations of human rights are taking place, [but] if it is to enjoy the sustained support of the world’s peoples, intervention must be based on legitimate and universal principles.”

R2P promised to put the genie of U.S. imperialism back in the bottle of international law.

Responding to this charge, the 2001 ICISS report redefined sovereignty in keeping with the demands of international morality. State sovereignty, it concluded, should entail the legal and political responsibility to protect populations; the failure or inability to fulfill these responsibilities should trigger international action. By 2005—the second year of the illegal war in Iraq—the United Nations World Summit unanimously endorsed a circumscribed principle of “responsibility to protect,” which authorized international action in the specific instances of genocide, war crimes, ethnic cleansing, and crimes against humanity. With humanitarian commitments newly embedded in the definition of sovereignty, intervention was framed not as a violation of sovereignty, but rather a mechanism for realizing its underlying principles. At the same time, in the case of the four humanitarian crimes, international military intervention was seen as a duty and responsibility, a defense of the existing world order.

In the context of America’s War on Terror, this institutionalized understanding of humanitarian intervention promised an alternative to unilateralism and preemptive military action. The idea that intervention might be legally and institutionally circumscribed was a rebuke of the U.S. war, and R2P offered the possibility of reasserting international law and the UN system after years of their marginalization. It promised to put the genie of U.S. imperialism back in the bottle of international law.

But while R2P aims to delimit the scope of intervention and reassert humanitarian law, it stems from an impoverished account of international responsibility. The idea that states hold the primary responsibility for preventing humanitarian crisis within their borders relies on a concept of the international order in which domestic politics can be disentangled from their international context. On this model, we conceive of states as products of a domestic social contract and frame international politics as a second-order social contract, into which states enter after their domestic constitutions are settled. Understood in this way, the international community can be positioned as a wholly external actor that only intervenes once a crisis ensues.

To call this concept a fiction would be polite. From transnational financial and trade regimes to global governance institutions, the international and domestic spheres are impossible to disaggregate. Indeed, the basic order of operations is wrong: the international order has always preceded the nation-state, which took its shape within an already imperialist and hierarchical system.

The entangled character of national and international politics make it impossible to view humanitarian crises as solely domestic. International institutions constitute the conditions in which humanitarian crisis emerge. They are present prior to and during the forms of violence that then justify intervention. For example, in the period before the Kosovo intervention, Yugoslavia and its successor states were subject to decades of IMF structural adjustment that exacerbated economic instability, diminished constitutional minority protections, and undermined social cohesion. Similarly, in Rwanda—the exemplary case of international inaction—international development programs funded by the IMF and World Bank as well as bilateral foreign aid helped to create a “model developing state” prior to the genocide. In effect beginning in the late 1980s, the policies of international actors contributed to inequality in the country and deepened its political divisions. At the same time, they strengthened the administrative capacity of the state, contributing to the scale and speed of the genocide.

R2P relies on the fiction that domestic politics can be disentangled from their international context.

The point here is not to identify international institutions as the primary causal actors in instances of ethnic cleansing and genocide. We need not absolve local actors of responsibility. But we cannot understand the origins of their actions if we treat the nation-state as an enclosed cell. The political and economic conditions of the world’s “failed states” are embedded in and a product of the wider international order. To argue in this context that the blame for violations of human rights falls first and foremost with states burdens them with responsibilities for political, economic, and humanitarian conditions that are not of their own making. It assumes that all states are sovereign in the same way, saddling weak and poor states with responsibility for structural problems that they have limited power to transform.

At the same time, under R2P’s framework, international responsibility amounts to a limited duty of assistance or intervention. It both relieves powerful states of burdens and enhances their authority. Responsibility for the origin of crises in underlying political and economic structures is, admittedly, much harder to pinpoint, but the R2P calculus does not even attempt it. Viewing the international community as external to humanitarian crisis also endows international institutions—in particular the unrepresentative Security Council—with the power to decide when and how to discharge that responsibility. As the Bolivia representative to the UN declared in 2009, “the application of [R2P] is discretionary and not universal” because of the “geopolitical interests at work within the Council.”

Consider the one and only instance of R2P’s invocation in military action: the NATO bombing of Libya. Although touted at the time for its multilateralism and use of the UN Security Council process, it contributed to civil war and further destabilized the region. Should it come as any surprise that this outcome paralleled the reckless U.S. war in Iraq?

To secure a more just world order, we must instead wrestle with a more expansive account of international responsibility. Since the 2001 Durban conference, the demand for reparations has moved from the margins of political debate to the center. After victories in the cases of sterilization in North Carolina and police torture in Chicago, plus British compensation for torture during the anticolonial Mau Mau rebellion in Kenya, advocates of reparations have a renewed sense of political possibility.

In 2013 Hilary Beckles, the Vice Chancellor of the University of West Indies, published Britain’s Black Debt, which systematically outlined Great Britain’s “criminal enrichment” through slavery and colonialism. Beckles’ book has served as the basis for the demands outlined by the Caribbean Reparations Commission, which represents fifteen island-states and is seeking reparations from Britain, France, and the Netherlands. The Reparations Commission has outlined an ambitious 10-point program for redress. It includes political and economic demands associated with the Global South, including debt cancellation for the region, technology transfer, and investments in an indigenous people’s development program.

Central to reparations claims is the argument that building a Caribbean future depends on redressing the past.

Despite growing support, however, these demands face an uphill battle. When the Caribbean Reparations Commission was formed, the British Treasury was still repaying an 1835 debt it took out to bailout slave owners after slavery was abolished in the British Empire. To compensate slave owners for the loss of property after emancipation, the British Treasury paid them £20 million—roughly 40 percent of the British government’s yearly income at the time and what would be equivalent to more than £300 billion today. Half of that money went to slave-owning families in the Caribbean and Africa. Ignoring this history during his September 2015 visit to Jamaica, former Prime Minister David Cameron urged West Indians to “move on from this painful legacy and continue to build for the future.”

Central to reparations claims, however, is the argument that building a Caribbean future depends on redressing the past. The islands are what they are because of slavery, colonialism, and the very uneven distribution of resources left in their wakes. The book is not closed on these relationships. Since independence, island states have had to borrow to try to lift social standards from the abject levels produced by colonialism; their modern debt is a direct consequence of their efforts to overcome the colonial legacies of poverty and underdevelopment. This debt, the Commission argues, “properly belongs to the imperial governments who have made no sustained attempt to deal with debilitating colonial legacies.” Caribbean debt is, as Beckles argues, transformed into Europe’s “black debt”—a debt accrued through “criminal enrichment.” European wealth and Caribbean poverty are tethered by a long history of exploitation and domination.

But the colonial past is not only present in the structural relations between the European and Caribbean states. It is inscribed on Caribbean bodies. According to the Commission, the Caribbean is home to the “highest incidence in the world of . . . hypertension and type two diabetes.” These are the “direct result of the nutritional experience, physical and emotional brutality, and overall stress profiles associated with slavery, genocide, and apartheid.” The “painful legacy” of slavery cannot simply be relegated to the past as Cameron suggested. It is inherited and transmitted, carried on the bones of the descendants of the enslaved and colonized.

The example of the Caribbean movement for reparations is focused on the specific relationship between the Caribbean and former European empires, but it provides resources to rethink international responsibility more broadly. It draws our attention to the persistent entanglements between the domestic and international realms of politics and urges us to consider what responsibility means in the face of everyday political, economic, and social crises. From this perspective, the international community cannot be pictured as an external actor with a supervisory responsibility to protect. Nor can meaningful international responsibility be limited to assistance and intervention in instances of acute humanitarian crises. Instead, international responsibility involves undoing the mechanisms that produce hierarchy, poverty, and violence.

The liberal internationalism of R2P demands little of us. It conceives of responsibility primarily in the form of a bomb.

From here, we can build a broader left alternative to liberal internationalism. Within the U.S. empire, such a political project would reject the position of the world’s policeman and embrace the role of global partner in transforming the political and economic structures that contribute to contemporary crises.

Consider the unevenly distributed effects of climate change. From Hurricane Maria in 2017, which devastated Puerto Rico, to Cyclone Idai, which tore through Mozambique, Zimbabwe, and Malawi in March, poor and weak states are first in the path of climate change. In Puerto Rico, the PROMESA bill, signed by Congress in 2016, created an unelected board that enforced steep austerity measures as part of the restructuring of the island’s debt. Budget cuts hollowed out state institutions such as hospitals and schools, while the bankruptcy of the public utility company ensured that services were not maintained even prior to the hurricane. This deterioration of public infrastructure exacerbated the unprecedented storm’s effect and slowed recovery efforts; residents were left without electricity and clean water for months. In Mozambique, the cyclone’s path highlighted the injustices of today’s global economic order more plainly: the hardest hit regions lie not far from ExxonMobil’s recent efforts to develop gas deposits.

In these contexts, international responsibility requires holding private corporations and international financial institutions accountable as well as supporting public and democratic institutions. Like the Caribbean demands for reparations, recent calls for ExxonMobil to pay its “climate debt” insist that the burdens of creating post-carbon economies be borne by those who have benefited from the current economic order. Reckoning with this debt and ensuring a just and egalitarian global transition requires supporting democratic institutions at the local, national, and global level that empower those most affected by climate change.

We are, after all, all embedded in a global web of economic relations, and acknowledging that basic fact also reveals how migration patterns are products of global structures of violence and inequality. The foreign and economic policies of western states have contributed enormously to the increasing numbers of refugees and migrants. From U.S. support of authoritarian and military regimes in Central America to the North American Free Trade Agreement that has transformed the conditions of Mexican agriculture to the wars in the Middle East, the connection between U.S. policy and displacement is direct. Seen from this perspective, the closure of borders and violent policing of migrants in the United States and Europe is not only inhumane and immoral, it is also a dereliction of responsibility. Migration policies that redress this past are indispensable in the broader effort to transform economic and political relations between states. To abandon the human products of imperial irresponsibility is to contribute to the further destabilization of the world.

If we are ever to escape the now eighteen-year War on Terror, we need to take genuine responsibility, which requires both historical self-knowledge and the international solidarity that it implies.

Emphasizing these broader structural responsibilities is not meant to shun the specific international responsibility in instances of humanitarian crime to which R2P is directed. Critics of humanitarian intervention are always maligned for refusing to “do something” in atrocities, but more often than not that “something” is equated with military intervention. Instead, responses to humanitarian crises and civil wars such as the kind that prompted NATO’s airstrikes in Libya should prioritize multilateral and diplomatic processes that emphasize the role of regional partners, aim at the de-escalation of conflicts, and include all stakeholders in political transition. It should be remembered that precisely this option was foreclosed in the case of Libya when African Union officials were prevented from flying to Tripoli due to the Security Council’s declaration of a no-fly zone. To be sure, there may be cases where diplomatic options are exhausted or unavailable and intervention is considered a last resort. In those contexts, a responsible approach to intervention would follow what Aziz Rana describes as a “do no harm principle,” which involves a careful consideration of the possible consequences of coercive action.

But ultimately our responses to specific cases of humanitarian crisis must be nested in a broader campaign of demilitarization. The United States spends more on its military than any other country in the world and remains the largest exporter of arms. Local conflicts are fueled by this global trade in arms. Both directly through its own wars and indirectly through military aid and the glut of the global arms market, the United States remains, as Martin Luther King put it, “the greatest purveyor of violence in the world today.” A dramatic scaling down of the U.S. military and a transnational effort at collective disarmament must be part of a wider response to violent conflicts around the world.

The liberal internationalism of R2P demands little of us. It conceives of responsibility primarily in the form of a bomb. If we are to ever escape our entrapment in the now eighteen-year War on Terror we need to take genuine responsibility, which requires both historical self-knowledge and the international solidarity that it implies. These are taxing undertakings, but we have much to gain from embracing this perspective. It is not a question of altruism or charity. Redressing histories of domination and creating new global institutions will do more than alleviate the conditions of others; they are necessary parts of remaking democracy at home—a task that will remain impossible so long as we cling to a view of ourselves as benign sovereigns, white saviors, the cavalry who arrive just in time.