Why doesn’t Abya Yala have its Accra?

Part two of “Cortés and Pizarro: conquest, thanatopolitics and imperial memory” and a continuation of “The Third Obligation: Accra and the Debt the West Refuses to Repay”

At the 2026 World Cup, two apparently minor scenes brought the colonial past back into the stands.

In Spain, during the celebrations for the championship, some young people once again invoked Hernán Cortés and Francisco Pizarro as signs of national greatness. Their names could circulate through chants, interviews and celebrations associated with conquest, victory and pride. That scene opened Cortés and Pizarro. The question was not why we still knew who they were, but what work of memory allowed two protagonists of sixteenth-century conquests to continue functioning, five centuries later, as positive emblems of a Spanish victory.

At matches involving the Democratic Republic of the Congo, another scene appeared. Michel Kuka Mboladinga, known as Lumumba Vea, remains standing and motionless for the full ninety minutes, dressed in the national colours with one arm raised. He has performed the gesture since 2013 and at the 2026 World Cup it became a recognisable image of the Congolese team. He is not representing a footballer. He reproduces the pose of a statue of Patrice Lumumba in Kinshasa and turns his own body into a public reminder of the history of Congolese independence (Cuesta, 2026; Paura, 2026).

It is worth pausing over who Lumumba was, because the gesture loses much of its meaning if his name is reduced to an icon. Patrice Emery Lumumba was one of the principal leaders of the independence of the Belgian Congo and the first prime minister of the independent Congo in 1960. He defended a united and sovereign Congolese state and publicly denounced the violence and humiliation of Belgian colonial rule. His government lasted only a few months. Amid the political crisis that followed independence, he was removed from power, detained and transferred to Katanga. On 17 January 1961 he was executed alongside Maurice Mpolo and Joseph Okito. Decades later, a Belgian parliamentary commission attributed moral responsibility to members of the Belgian Government for the circumstances that led to the assassination; in 2022, the Belgian prime minister officially reiterated that responsibility and apologised for the role played by his country’s authorities. US documents also show that, during the crisis, the CIA had developed a programme to remove Lumumba from power and had gone so far as to prepare plans to kill him, although his eventual execution took place in Katanga (Belgian Federal Government, 2022; U.S. Department of State, Office of the Historian, n.d.).

Lumumba therefore became much more than a former head of government. For broad sectors of Congolese and Pan-African memory, he came to represent an interrupted independence, the demand for political and economic sovereignty, and the violence with which colonial and Cold War interests cut across the construction of the new state. When Michel Kuka stands motionless in a stadium, he is not merely staging a person. He forces that history to accompany the present.

In one stand, the colonial past could enter the celebration. In the other, anti-colonial memory interrupted the celebration in order to become visible.

The contrast does not allow us to distribute “good” memory and “bad” memory comfortably, nor does it make Cortés, Pizarro and Lumumba symmetrical figures. The interesting difference lies in the work each symbol demands. Cortés and Pizarro can function as signs of victory if part of the experience of conquered populations remains outside the emblem. Lumumba works differently: to understand why a man remains motionless for ninety minutes, Belgian colonisation, independence, the Congolese crisis and the assassination must all be brought back into the image.

Two stands at the same World Cup thus show that the past does not return simply because it is remembered. It returns through different forms of political legibility.

That popular contrast found an institutional echo in 2026. On 25 March, the United Nations General Assembly adopted resolution A/RES/80/250 by 123 votes in favour, three against — Argentina, the United States and Israel — and 52 abstentions. The resolution declared the trafficking of enslaved Africans and the racialised chattel enslavement of Africans to be “the gravest crime against humanity”. The 27 member states of the European Union, including Spain, were among those that abstained. The resolution did not create a tribunal or establish binding compensation. It affirmed that claims for reparation represent a concrete step towards redressing historical wrongs and called on states to engage in good-faith dialogue on reparatory justice, including measures such as formal apologies, restitution, compensation, rehabilitation and guarantees of non-repetition (United Nations General Assembly, 2026).

Between 17 and 19 June, a sequence of technical, political and commemorative meetings in Accra sought to turn that language into an agenda. The process adopted the Accra Next Steps Commitments on Reparation Justice, a roadmap addressing restitution, compensation, debt, financial reform, climate justice, technological inclusion and permanent coordination mechanisms (Ministry of Foreign Affairs, Republic of Ghana, 2026). Accra did not inaugurate the reparations movement. It was one of those moments when decades of African, Caribbean and diasporic work acquired a political architecture capable of translating a known history into the international language of obligation.

The Third Obligation proposed reading that shift through Marcel Mauss. Giving, receiving and reciprocating did not describe some supposed reciprocity between enslavers and the enslaved. To transfer the scheme of the gift literally to a relationship founded on violence would be absurd. The question comes afterwards. What happens when a historical relationship produced accumulation for some parties and dispossession for others, when the harm eventually comes to be recognised, but those who inherited institutions, assets and advantages reject the idea that recognition should generate any obligation at all?

The first essay in this series began with young people who could once again turn Cortés and Pizarro into signs of victory. This one places Lumumba’s human statue beside that image and shifts the question.

Why doesn’t Abya Yala have its Accra?

The question does not call for an exact replica of Accra, nor does it seek to equate the conquest of the Americas, the encomienda, the mita, Indigenous enslavement and the racialised enslavement of Africans. Nor does it claim that Indigenous peoples lack rights or reparatory processes. It asks something more specific: what conditions allow a historical harm to become politically legible as a shared obligation.

The hypothesis of this essay is that one decisive difference lies in the political form of the archive. The Atlantic slave trade produced a harm which, in addition to being massive, was inscribed in legal and accounting devices comparable across territories: port registers, cargo manifests, contracts, insurance policies, censuses, plantation books, asientos, property deeds and litigation. The conquest of Abya Yala also produced immense archives, but the harm appears distributed across war, epidemic disease, encomienda, mita, tribute, evangelisation, enslavement, territorial loss and ecological reorganisation. It is not that one history is documented and the other is not. They are not documented in the same way, nor have they been politically assembled with the same effectiveness. That difference does not by itself explain the emergence of Accra, but it helps us understand why the passage from memory to obligation has followed different trajectories.

The comparison itself requires limits. Article 28 of the United Nations Declaration on the Rights of Indigenous Peoples recognises the right to redress for lands, territories and resources confiscated, taken, occupied, used or damaged without free, prior and informed consent. Article 12 also recognises the right to the repatriation of human remains and to access and restitution of ceremonial objects (United Nations General Assembly, 2007). Nor is there an absolute regional vacuum. The first CELAC–Africa High-Level Forum, held in Bogotá in March 2026, dedicated a pillar to “Historical Reparation and Ethnic-Racial Justice”, and Colombia has had a National Intersectoral Commission for Historical Reparation since 2023. These are signs of a process under construction. The difference lies less in the existence of a reparatory vocabulary than in its degree of regional consolidation and in the capacity to gather historically heterogeneous harms into a common political claim.

Perhaps the decisive question is not yet how much should be paid, but an earlier one: what has to happen for a known harm to be formulated as a debt?

What can enter history

Michel-Rolph Trouillot offers an especially fertile route into this problem. In Silencing the Past, Haiti occupies a central place not as an absence of facts but as a crisis in the categories through which the West could understand them. The Haitian Revolution was “unthinkable” because people who had been legally transformed into property did not merely rebel. They formulated a general political project, built forces capable of defeating imperial armies and founded an independent state on the abolition of slavery (Trouillot, 1995).

Long before Trouillot, the Trinidadian intellectual C. L. R. James had carried out another decisive operation in The Black Jacobins. James narrated the Haitian Revolution through the political capacity of enslaved people and placed it at the centre, rather than the margins, of Atlantic revolutionary transformations (James, 1989). James and Trouillot do not perform the same intellectual task. The first reconstructs a revolution led by those whom the colonial order had turned into property; the second asks about the conditions that made that revolution difficult to think even while it was taking place. Read together, they move Haiti from the periphery of the narrative to the point where the narrative’s categories themselves enter into crisis.

That limit did not depend on a single racial theory. It formed part of an ontological order that ranked degrees of humanity and made the universality proclaimed by the Enlightenment compatible with colonial enslavement. The formulations that would later acquire the appearance of scientific racism reinforced that horizon. The Haitian Revolution was unthinkable because those who had been placed within such classifications as inferior beings appeared acting as autonomous subjects of a universal revolution.

For Trouillot, power does not act only by erasing documents. It intervenes when sources are produced, when archives are organised, when particular traces are recovered to build a narrative, and when some events, rather than others, retrospectively acquire historical significance. A silence can therefore be produced inside an enormous archive. Documentation may exist and yet be organised in such a way that certain relationships remain difficult to think politically.

That distinction is fundamental for Abya Yala.

The conquest has not been silenced because archives are lacking. It has been politically fragmented inside overwhelmingly abundant archives.

We know about wars, epidemics, expeditions, encomiendas, reducciones, displacement, tribute, mines, haciendas, religious conversion, territorial appropriation and multiple forms of resistance. Yet those processes can be organised as separate episodes in national, imperial or local histories without necessarily constituting a common historical harm capable of generating a transnational obligation. The power of history lies not only in hiding what happened. It also lies in determining the categories through which what happened can be related to the present.

The demographic dimension makes the problem particularly clear.

Koch et al. (2019) reviewed 119 regional estimates and calculated a population in the Americas before European contact of approximately 60.5 million people, with an interquartile range of 44.8 to 78.2 million. To estimate mortality, they used a standardised population of around six million by 1600. The detailed calculation in the article itself yields 54.5 million deaths, with an interquartile range of 39.0 to 72.4 million, equivalent to 90 per cent of the pre-contact population, with a range of 87 to 92 per cent (Koch et al., 2019, p. 22). In their conclusion, the authors round the estimate to 55 million. These are necessarily reconstructed figures and the subject of historiographical debate, not a retrospective census.

Nor does the figure mean that “Spain killed 54.5 million people”. Epidemics introduced from Eurasia were the principal immediate cause of mortality in many regions. But nor does it describe 54.5 million natural deaths that can be neatly separated from conquest. The study itself places war, enslavement and famine alongside epidemic disease. Displacement, the destruction of productive systems, coerced labour and the breakdown of community networks changed the conditions under which disease circulated and the possibilities for populations to recover (Koch et al., 2019).

That causal complexity is historically necessary. It can also have political consequences. Racialised Atlantic slavery constitutes a relatively identifiable legal institution, with documented markets, ships, contracts, insurers, ports, plantations and human beings explicitly transformed into property. The American catastrophe appears distributed across epidemic disease, invasion, war, enslavement, encomienda, mita, tribute, evangelisation, territorial loss and ecological reorganisation.

The fragmentation of causes can end up favouring a fragmentation of responsibility.

Quijano adds another layer. Coloniality of power does not mean that the colonial order remained static after independence. It points to the way some classifications of population, labour, authority and knowledge built during colonial expansion participated in the later formation of global capitalism and survived, transformed, beyond the formal end of empires (Quijano, 2000). Coloniality therefore helps explain why this fragmentation does not end with formal colonialism. The categories through which populations, forms of labour and forms of knowledge had been classified continued organising hierarchies after independence. That does not mechanically determine how each form of violence will be remembered, but it does condition the positions from which it can be narrated and claimed.

The question of Abya Yala therefore cannot be reduced to whether Europe “remembers” or “does not remember” conquest. It asks which categories allow known forms of violence to acquire the political capacity to generate responsibility.

The harm had an economy

History stops looking abstract when it is observed materially.

Eric Wolf argued that Europe and those converted into “people without history” had to be placed within the same history of connections. European expansion did not unfold in a world already divided between a dynamic Europe and static external societies. The relationships that would eventually form a world economy were built precisely through conquest, trade, circulation, coercion and the transformation of societies connected to one another (Wolf, 1982).

Potosí allows us to recall that relationship without repeating everything developed in Cortés and Pizarro. Robins and Hagan (2012) estimate that around 18,000 metric tonnes of silver were produced there between 1574 and 1735 and that, between 1574 and 1810, some 39,000 tonnes of mercury were released as vapour during refining. The mita organised under Viceroy Francisco de Toledo also mobilised contingents of thousands of Indigenous workers; in 1578 the documented quota stood at 14,181 people (Gil Montero, 2011).

What matters here is not to reconstruct the entire history of mining again, but to retain its fundamental relationship. Value could travel from the Andes towards American, European and Asian markets while a substantial share of the health, ecological and human costs remained territorialised where extraction took place. If in the previous essay we followed Sánchez-Antonio (2020) in reading colonial thanatopolitics as a rationality capable of distributing exposure to death and harm differentially, it is enough here to retain one idea: coercion, too, could be transformed into economic infrastructure.

The plantation allows us to follow that question through other commodities and other labour regimes.

Sidney Mintz studied the plantation as a social form, not merely an agricultural enterprise. In his analysis of Puerto Rico he connects commercial monoculture, capital investment, concentrated control of production and different forms of enslaved, forced or coercive labour. The importance of his approach lies precisely in linking what happens in the tropical field to markets, consumption and accumulation far beyond the plantation (Mintz, 1974).

His Puerto Rican figures make that transformation visible. The recorded enslaved population rose from 17,536 people in 1812 to 51,216 in 1846. Meanwhile, sugar production grew from approximately 9,391 tonnes in 1827–1828 to 65,517 tonnes in 1860–1861. Between 1815 and 1876, the labour of enslaved people also coexisted with different forms of forced labour imposed on people who were legally free (Mintz, 1974).

Sugar also returns us to the beginning of this series. Gisela von Wobeser documents how Hernán Cortés promoted the cultivation of sugar cane in Cuernavaca at an early stage and the Tlaltenango mill, within a regional reorganisation of land, water and labour around sugar production (von Wobeser, 2004). But the map of the plantation remains incomplete if it begins and ends in Castilian possessions.

The slave-trading Atlantic: Portugal, Brazil and cotton

The Atlantic scale requires us to look towards Portugal and Brazil.

Portugal pioneered European slave-trading expansion along the Atlantic coast of Africa during the fifteenth century and, together with the circuits later developed from Brazil, ended up being by a wide margin the principal carrier in the entire transatlantic trade. SlaveVoyages estimates that between 1501 and 1866 5,848,266 people were embarked on ships classified as Portuguese or Brazilian, approximately 46.7 per cent of the estimated 12.52 million embarkations. British ships transported around 3.26 million; French ships, 1.38 million; those associated with Spain or Uruguay, 1.06 million; and Dutch ships, around 554,000 people (SlaveVoyages, n.d.; Eltis & Richardson, 2010).

The figures prevent this history from being narrated as a simple linear succession from Portugal to the Netherlands and then England. Portugal and Brazil retained enormous weight for centuries. The Netherlands did, however, play a strategic role during the seventeenth century. The Dutch West India Company occupied part of Brazil’s sugar-producing north-east from 1630, seized Elmina from Portugal in 1637 and Luanda in 1641. For several years it attempted to control both sugar-producing territories and crucial points in the supply of enslaved people. Britain later became the principal carrying power during much of the eighteenth century. These were not neat successions, but overlapping hegemonies within an increasingly extensive Atlantic infrastructure (Eltis & Richardson, 2010).

Brazil was the great American node in that system. SlaveVoyages estimates place approximately 4.86 million landings of enslaved Africans there, close to 45 per cent of those who survived the Atlantic crossing. South-east Brazil alone received around 2.26 million; Bahia, 1.55 million; and Pernambuco more than 850,000. The figure forces us to displace an image centred too heavily on the British Caribbean or the United States. For centuries, there was also a South Atlantic articulated directly between African coasts and Brazilian ports (SlaveVoyages, n.d.).

That scale left material sites of memory. Valongo Wharf, built in Rio de Janeiro from 1811 to concentrate disembarkations, received approximately 900,000 captive Africans. UNESCO regards it as the most important physical trace associated with the forced arrival of enslaved Africans on the American continent (UNESCO World Heritage Centre, n.d.).

The Spanish monarchy formed a substantial part of these circuits even though it did not lead them. Borucki, Eltis and Wheat (2015) estimate that 1,506,000 enslaved Africans arrived directly from Africa in Spanish America and another 566,000 through intra-American trafficking, from Portuguese, British, Dutch and other colonial territories. In total, just over 2.07 million people, with around two thirds of those arrivals occurring before 1810.

That growth should not, however, be narrated as a simple replacement of Indigenous people with Africans caused by the New Laws of 1542. Those laws attempted to restrict Indigenous enslavement and limit certain abuses of the encomienda, but they did not produce a clean break. Van Deusen (2023) has documented authorisations for Indigenous enslavement after 1542 in numerous parts of the empire. African slavery therefore expanded alongside a changing repertoire of encomienda, repartimiento, mita, personal servitude and legal and illegal forms of Indigenous enslavement; not simply after their disappearance.

Cotton in the southern United States shows another transformation. Continental North America directly received around 388,000 enslaved Africans during the Atlantic trade, a figure far below that of Brazil or the Caribbean (SlaveVoyages, n.d.). Yet the US census recorded 3,953,760 enslaved people in 1860. After the federal ban on transatlantic importation in 1808, the internal market in human beings expanded alongside the cotton economy. Historians estimate that approximately one million people were forcibly displaced from the Upper South to the Deep South in the process known as the Second Middle Passage (Baptist, 2014; Beckert, 2014).

The expansion was extraordinary. US cotton production rose from around 13,000 bales in 1792 to more than five million in 1860. By that year, the South had become the dominant producer on the world market and supplied approximately two thirds to three quarters of global cotton. Cotton also accounted for more than 60 per cent of the value of US exports on the eve of the Civil War (National Park Service, n.d.). Estimates of productivity also show that the amount picked per person per day increased roughly fourfold between 1800 and 1860. The interpretation of that increase is debated: Baptist (2014) emphasises quotas, surveillance and physical violence, while other studies also highlight seed selection and agronomic changes. We do not need to settle that debate here to recognise that productive innovation, industrial demand and the coercive discipline of labour combined within cotton expansion (Baptist, 2014; Beckert, 2014).

The plantation did not remain identical. Commodities, territories, institutions, markets and the relationship between demographic reproduction and trafficking changed. That is precisely why continuity should not be confused with repetition. Brazilian and Caribbean sugar and US cotton belong to different historical configurations. What comparative political economy allows us to observe is how the legal appropriation of human labour could be integrated in different ways into commercial and financial circuits of enormous scale.

West and West-Central Africa: extraction and violence

Even this map begins too late.

Enslaved people did not simply appear on the African coast. They had to be brought there.

West Africa and, above all, West-Central Africa formed the other great space of this economy. SlaveVoyages identifies the principal regions of embarkation as Senegambia, Sierra Leone, the Windward Coast, the Gold Coast, the Bight of Benin, the Bight of Biafra and West-Central Africa. The latter — broadly, the territories of present-day Angola and the two Congos — was the largest regional point of departure for much of the history of the trade. Around 5.7 million people were embarked from West-Central Africa; close to two million from the Bight of Benin; about 1.6 million from the Bight of Biafra; and just over 1.2 million from the Gold Coast (Eltis & Richardson, 2010; SlaveVoyages, n.d.).

Atlantic demand did not encounter a human reserve waiting beside European ships. It connected with pre-existing African forms of enslavement, warfare and dependency, but altered their scale, incentives and destinations. In different places, captives obtained through warfare, raids, kidnapping, judicial conviction or commercial networks were moved from the interior towards points of embarkation. African states, elites, merchants and intermediaries participated in diverse ways; some political formations were temporarily strengthened by the system, while others were attacked, displaced or fragmented. Recognising African agency does not diminish responsibility for European and American Atlantic demand. It allows us to understand how an extraordinarily profitable external market reorganised relations of violence within the continent (Manning, 1990).

Dahomey makes that complexity visible. In its early phases, the monarchy attempted to centralise the trade and relied heavily on raiding; later it combined its own captures with the purchase of people from the interior and with private merchant networks (Law, 1989). European firearms formed an important part of those exchanges and could alter military capacities, but reducing three centuries of change to a simple “guns for slaves” cycle would be equally misleading. The relationship between Atlantic demand, warfare, state formation and commerce varied profoundly between regions.

Mortality before embarkation is far harder to quantify than mortality during the Middle Passage. People died during captures, armed conflicts, marches, transfers and periods of confinement. But there is no sufficiently robust continent-wide ratio that would allow us to claim that “one person died for every person embarked” as if this were a stable accounting relationship. Manning estimated millions of deaths directly related to processes of enslavement within Africa, but his model is not equivalent to an estimate of mortality in the Atlantic slave trade: it also includes people destined for internal African markets and other slave-trading circuits. What his work does show is that the intensity of exports reduced demographic growth and produced absolute population declines in many parts of West and West-Central Africa during periods of greatest extraction (Manning, 1990, 2010).

The thread of thanatopolitics can also be recovered here with caution: not because every African war can be attributed to Europe, nor because capture and extermination are synonymous, but because a transatlantic market made the systematic exposure of particular populations to the violence of capture, marches, confinement and the crossing economically tolerable. Death was not necessarily the objective, because the commodity had to arrive alive, but it could become an accepted cost in the production of captives.

Demographic composition also mattered. The Atlantic trade disproportionately demanded young people and, overall, more men than women. Removing people in central ages for labour and reproduction over generations was not simply a matter of subtracting individuals from a census. It altered households, agriculture and markets; it transformed gender relations, political capacities and possibilities of future growth. The impact varied between regions, but it was not marginal.

Walter Rodney had formulated this problem from another intellectual tradition. In How Europe Underdeveloped Africa, the Guyanese historian argued that European development and African underdevelopment should be thought relationally, not as separate economic histories. For Rodney (1972), the trade did not merely mean population loss: it also reoriented productive activities, political relations and development capacities towards an Atlantic economy whose centre of accumulation lay elsewhere. His argument is broader and politically stronger than a demographic reconstruction, and it should not be turned into a monocausal explanation of African history. Its importance here is different: it names a relationship that figures alone cannot describe.

Nathan Nunn (2008), using a very different methodology, identified a robust negative correlation between the historical intensity of the different slave trades and the later economic performance of the African regions affected. That relationship does not justify explaining contemporary African development through a single cause, nor does it allow four centuries of history to be reduced to an economic regression. It does allow us to reject the idea that the massive extraction of people ended without consequences when the last ships departed.

There were also concentrated episodes of political and military violence. In the Kingdom of Kongo, the Battle of Mbwila in 1665, in which King António I died and many members of the political elite were killed or captured, opened a prolonged crisis and civil war during which the capture and sale of people became still more important. Not all of that violence can be attributed to a single external cause, but neither can it be separated from the Atlantic market in captives that connected the region to Luanda and Brazil (Thornton, 1998).

And the conversion of human beings into accounting value did not end upon reaching the coast. The case of the Zong shows this with exceptional brutality. In 1781, during a voyage from Africa to Jamaica, the crew of the British ship threw around 130 enslaved Africans into the sea. The ship’s owners had insured the people being transported as cargo and subsequently claimed compensation for the loss. The case that reached London, Gregson v. Gilbert (1783), was a dispute between owners and insurers, not a criminal prosecution for murder. Granville Sharp tried to bring a murder charge, but did not succeed (London Museum, n.d.).

The Zong does not represent the entire trade, nor did the massacre take place in Africa. Its importance lies in showing how far the legal and financial logic of human property could go.

Killing could become an accounting operation.

None of this makes the mita, encomienda, Indigenous enslavement, the Brazilian plantation, Atlantic trafficking, the US domestic slave market or African wars of capture equivalent. They were different institutions, temporalities and relations of power. Historical political economy allows a different operation: observing how those differences could be connected within circuits that moved people, raw materials and value over long distances while distributing costs in profoundly unequal ways.

The harm had accounts, and wealth had destinations. From here, Accra acquires another dimension. It speaks not only from communities descended from those enslaved in the Americas. It also speaks from African societies from which millions of people were extracted and whose economic, demographic and political histories were shaped by that demand.

Haiti and inverted reparation

Haiti gathers these threads in a way that is difficult to improve upon with any abstract example.

On the eve of the Revolution, Saint-Domingue was one of the most productive plantation economies in the Atlantic world. In 1789 it had approximately 520,000 inhabitants. Around 452,000 were enslaved people, compared with about 40,000 white people and 28,000 free people of colour. Sugar represented approximately 48 per cent of the value of exports and coffee another 33.6 per cent (Ferrero, 2021).

This was not an unproductive margin of Atlantic capitalism but one of its major centres of value production.

The Haitian Revolution destroyed the ontological assumption on which that system rested. Those who appeared in the accounts as property emerged as political subjects. People whom the Atlantic order assigned a price defeated militarily those who defended that order and in 1804 proclaimed an independent state founded on abolition.

Trouillot helps us understand why that transformation was so difficult to conceive even while it was happening. The later history allows us to observe something even more paradoxical.

When France finally recognised Haiti’s sovereignty in 1825, it did so through a formidable inversion of reparatory obligation.

Charles X’s ordinance demanded 150 million francs as compensation for those who had lost colonial property. A French squadron carrying around 500 cannon accompanied the demand. Gaillard-Pourchet explicitly notes that the amount was equivalent to 15 per cent of the French state’s annual revenue at the time and, at minimum, to ten times Haiti’s annual revenue (Gaillard-Pourchet, 2023, section “Réparation aux ex-colons propriétaires”; Oosterlinck et al., 2022).

The mechanism becomes even more revealing when we look at what was being valued. The commission charged with distributing the compensation developed formulas that also allowed formerly enslaved people to be valued “per head”. Those who had lost the right to own human beings could thus appear as holders of a compensable loss (Gaillard-Pourchet, 2023).

Haiti, moreover, did not have 150 million francs. To meet the first instalment it contracted a nominal loan of 30 million francs on the French financial market at 6 per cent interest. After commissions it received only 24 million and had to use public resources to make up the 30 million France demanded for the first annual payment (Gaillard-Pourchet, 2023).

This is where the famous double debt begins. The new state had to pay the indemnity and, at the same time, borrow from French finance capital in order to pay it.

In 1838, after default and a long negotiation, the indemnity was reduced to a total of 90 million francs. But the loans and refinancing associated with it prolonged the transfer of resources for decades. The original indemnity was settled during the nineteenth century; the chain of debt connected to it continued far longer. The United Nations notes that in 1914 more than three quarters of Haiti’s national budget was still being used to pay French banks and places the final liquidation of financial obligations associated with this long sequence of debt in 1947 (United Nations in the Caribbean, 2025). Gaillard-Pourchet (2023) shows how the independence debt and related borrowing kept Haiti trapped for roughly a century in a financial relationship born directly from the recognition of its sovereignty. Oosterlinck et al. (2022) analyse it precisely as one of the major historical cases of odious debt.

French colonial sovereignty had ended and slavery had been abolished. Yet the power relationship could reappear through another technology. It was no longer legal ownership of people, but sovereign debt.

Debt therefore functioned as a financial technology of subordination at a moment when direct colonial domination was no longer sustainable. Haitian sovereignty existed legally, but part of the imperial relationship could be reorganised through payments, loans, interest and financial dependence. It is one of the places where Stoler’s notion of imperial debris takes on an almost accounting form.

Haiti was reparation in reverse.

A country born from a revolution against slavery began its international life compensating those who had lost property as a result of that revolution. Here Trouillot meets Mauss.

The third obligation and the limits of Accra

Mauss described the gift through obligations to give, receive and reciprocate. His scheme belongs to social contexts entirely different from conquest and slavery, and there is no reason to pretend reciprocity where there was appropriation and violence. But his question about obligation allows us to observe a later problem (Mauss, 2009).

Before one can discuss what should be returned, it must first be politically possible to admit that something was received. That is precisely one of the terrains on which imperial memory operates.

Where a historical relationship can be described through territorial appropriation, silver extraction, coerced labour, tribute and transfers of wealth, another narrative can reorganise it as encounter, evangelisation, mestizaje or the gift of civilisation. We are not dealing merely with different interpretations of the past. What is at stake is the grammar from which that past can generate obligations.

If what left America cannot be thought of as the receipt of land, labour, metals, food, resources, knowledge or human capacities, Mauss’s third obligation never even arises.

The dispute over memory therefore precedes the dispute over reparation. Haiti occupies the opposite extreme. There, a return did take place, but it was carried out by the previously dispossessed party. Those who had won their freedom compensated those who had lost the legal capacity to appropriate their labour and bodies.

Accra seeks to alter that grammar.

Resolution 80/250 neither calculates a global debt nor determines through a formula who should pay whom. Nor does it retrospectively convert every historical form of enslavement into present-day criminal liability. Its importance lies in the normative shift. It affirms that reparations claims constitute a concrete step in addressing historical wrongs and calls for international dialogue on restitution, compensation, rehabilitation, satisfaction and guarantees of non-repetition (United Nations General Assembly, 2026).

Nor did that moment appear from nowhere. Durban had already recognised in 2001 that slavery and the trafficking of enslaved people, particularly the transatlantic slave trade, constituted and should always have constituted a crime against humanity. What is new in 2026 is not the discovery of violence. It is the strengthening of the passage from historical recognition towards a reparatory architecture (United Nations, 2002).

The institutional difference is decisive here. CARICOM created its Reparations Commission in 2013 and adopted a ten-point programme in 2014. Its framework is not limited to African slavery. It also includes the genocide of Indigenous Caribbean peoples and extractive colonialism. In July 2026, CARICOM published a revised version of that programme and maintained those dimensions within its claim against former colonial powers (Caribbean Community, 2026a).

That architecture also has a Caribbean intellectual genealogy. In Britain’s Black Debt, Hilary McD. Beckles had articulated a history of colonial enrichment, Indigenous genocide, slavery and reparation written from the Caribbean and explicitly oriented towards building a political claim (Beckles, 2013). As chair of the CARICOM Reparations Commission, Beckles summarised the shift of phase in July 2026 with a phrase particularly significant for this essay: “The case has been made. The question now is implementation and demand” (Caribbean Community, 2026b). The historical documentation did not appear that year. What changed was the institutional capacity to turn it into a claim.

The African Union has in turn built specific mechanisms, including the African Union Committee of Experts on Reparations and the African Union Reference Group of Legal Experts on Reparations. On 17 June 2026, these bodies and the CARICOM Reparations Commission held their first joint meeting to operationalise an African-Caribbean mechanism for reparatory justice, in preparation for the high-level segment in Accra on 18 and 19 June. The African Union itself places this process within more than two decades of institutional work following Durban (African Union Commission, 2026).

This is why “why doesn’t Abya Yala have its Accra?” does not describe a total absence. The CELAC–Africa Forum in Bogotá shows that a regional space for reparation that expressly includes Indigenous experiences is also beginning to be built in Latin America. The difference is one of political and institutional consolidation.

There is also a specific historical difficulty. Many of the Latin American states that would today have to participate in a continental conversation on reparation did not emerge as Indigenous states liberated from metropolitan rule, but as republics that inherited and transformed previously colonial structures of property, racial classification and territorial control. Creole elites were not simply external victims of empire. In many places they later participated in new dispossessions of Indigenous and Afro-descendant peoples.

A Latin American tradition of critical thought named part of this problem internal colonialism. González Casanova (1965) and Stavenhagen (1967) used the concept to challenge the idea that political independence and national integration automatically eliminated colonial relations within the new republics. The term does not turn contemporary Latin American states into copies of European empires, but it helps explain why, for many Indigenous peoples, independence did not necessarily coincide with decolonisation.

That makes it much more difficult to identify a simple division between a party that must repair and a party that must receive.

Spain and Portugal?

The present-day American states?

Religious institutions?

Companies and private fortunes?

Financial institutions?

Those who inherited land obtained under colonial regimes?

Who should be the subjects of reparation when affected peoples cross borders subsequently created by the states themselves?

Even before a reparation can be calculated, Abya Yala forces us to debate who inherits responsibility and who has authority to make a claim.

The European explanation for abstaining on A/RES/80/250 shows another boundary. The member states of the European Union jointly justified their position. Their statement unequivocally condemned the transatlantic trade and slavery, recognised their persistent consequences and supported action against structural racism. But it objected to the term gravest, expressly invoked the principle of non-retroactivity in international law and argued that references to reparations claims “lack a sound legal basis” (European External Action Service, 2026).

The distinction is particularly revealing for our argument: memory can be accepted while obligation is rejected. From a Maussian reading, that distance is precisely what matters. Recognising that something happened is not yet the same as admitting that it generates an obligation to return.

Imperial remains

Ann Laura Stoler proposes that empires should not be thought of only through their ruins, as dead remains of a finished era, but through processes of ruination. Attention shifts towards the ways past imperial relationships leave enduring traces on territories, bodies, institutions and possibilities of life in the present (Stoler, 2013). The idea allows us to avoid two opposite errors: claiming that nothing has changed since the sixteenth century, or imagining that independence and abolition automatically closed every relationship built over centuries.

Haiti offers an especially clear version of the problem. French sovereignty formally ended, but part of the power relationship could be reorganised through debt. Quijano provides another way of naming this mechanism without confusing continuity with immobility. Coloniality does not mean that present inequalities are intact fossils of colonialism, but that some ways of classifying population, knowledge, authority and labour acquired an organising power that outlasted the legal duration of empires (Quijano, 2000).

From AIthropology Lab, one final contemporary question remains, but its point of contact should not be a genealogy of extraction. Artificial intelligence does not continue the Atlantic slave trade, the plantation or Potosí. These are historically distinct institutions, technologies and human positions. The useful connection lies elsewhere: in the politics of what an infrastructure manages to make legible.

In April 2026, UNESCO described African languages as one of artificial intelligence’s major blind spots. Systems trained on content dominated by a small number of languages provide poorer coverage for much of the continent’s linguistic diversity, precisely because many languages have few digitised corpora or because much of their knowledge circulates orally (UNESCO, 2026). A language can be socially rich, historically deep and fully alive while remaining poorly legible to a model.

Nor does the answer have to be to turn more experience into data as quickly as possible. Masakhane, a Pan-African natural language processing community, frames the issue in terms of participation and sovereignty: strengthening technologies for African languages “for Africa, by Africans”, and ensuring that those who produce these languages and forms of knowledge participate in research as protagonists rather than merely as sources of data (Masakhane, n.d.).

This is where the comparison ends. A training corpus is not a slave-ship manifest, and the digital absence of a language is not equivalent to its colonial persecution. But Trouillot’s question acquires a limited contemporary echo: which infrastructures decide what can be made visible, classifiable and usable, and who has power to intervene in that decision. Accra now includes technological inclusion within the reparatory agenda. That does not make AI a debt inherited from slavery. More modestly, it places the capacity to participate in contemporary knowledge infrastructures within the discussion of what repair can mean.

What we still do not know how to return

Accra was not possible because someone discovered slavery in 2026. Enslaved people knew the violence of the system while they endured it, resisted it, escaped from it and rebelled against it. C. L. R. James could reconstruct the Haitian Revolution decades before Trouillot asked about its condition as “unthinkable history”. Historiography, archives and descendant communities had long known about ships, plantations, contracts, markets, resistance and profits.

But knowing is not the same as obligating.

The Atlantic trade was legally legible as a relationship of property long before it became politically legible as debt. That is precisely why the Zong is so revealing. The system could identify people as insured cargo, register their loss and debate before a court who should bear that loss economically. For centuries, the archive could count property before it could formulate reparation.

The transformation leading to Durban, CARICOM, the African Union and Accra consists in partially turning that archive against the relationship for which it had been built. Ships, policies, contracts, account books, censuses and laws did not produce justice by themselves. Political movements, research, institutions and communities were what managed to assemble those traces within a common language of responsibility.

Abya Yala poses a different difficulty. The conquest does not lack an archive either. We know about demographic catastrophe, encomienda, mita, Indigenous enslavement, Potosí, territorial appropriations, military campaigns and the reorganisation of economies and ecologies. The problem is that these traces belong to different causal, legal and political regimes and were subsequently distributed across national, imperial and local narratives. Building a common reparatory claim cannot consist in pretending that everything was the same crime. It requires precisely assembling what has been fragmented without erasing its differences.

Haiti shows why this operation is political before it is archival. In 1825 the international order found the patrimonial loss of former French owners perfectly legible and managed to turn it into an obligation enforceable against the new Haitian state. What remained much harder to formulate for far longer was the reverse operation: that the appropriation of lives, labour and territories could obligate those who had received its benefits.

This is where Trouillot and Mauss converge. Trouillot helps us ask which relationships manage to enter history with the capacity to produce consequences. Mauss allows us to ask what happens when a recognised relationship ceases to be merely past and begins to oblige.

The day Abya Yala has something we can call “its Accra” will not be because a secret archive has appeared or because every form of colonial violence can be reduced to a single category. It will be because a political framework has been built that can turn dispersed archives and different harms into a shared question of responsibility without falsifying their history.

That past will no longer be merely something that happened. It will also be a relationship whose effects impose obligations.

And, at last, the third obligation will have to be confronted.

What to return. To whom. And who is obliged to do so.

Primary and documentary sources

African Union Commission. (2026, 17 June). African Union and CARICOM reparations mechanisms hold first joint meeting operationalizing a joint mechanism on reparative justice. African Union. https://au.int/fr/node/46539

United Nations General Assembly. (2007). United Nations Declaration on the Rights of Indigenous Peoples (A/RES/61/295). United Nations. https://digitallibrary.un.org/record/606782/?ln=en

United Nations General Assembly. (2026). Declaration on the qualification of the trafficking of enslaved Africans and the racialized chattel enslavement of Africans as the gravest crime against humanity (A/RES/80/250). United Nations. https://digitallibrary.un.org/record/4106660?ln=en

Caribbean Community. (2026a, 11 July). The CARICOM Ten Point Plan for Reparations: A Manifesto for the Coming Enlightenment. CARICOM Secretariat. https://caricom.org/documents/the-caricom-caribbean-community-ten-point-plan-for-reparations-a-manifesto-for-the-coming-enlightenment/

Caribbean Community. (2026b, 10 July). CARICOM endorses revised Ten-Point Reparations Manifesto. https://caricom.org/caricom-endorses-revised-ten-point-reparations-manifesto/

Cuesta, S. (2026, 23 June). Lumumba Vea, el hincha de la RD del Congo que ve los partidos como una estatua. elDiario.es. https://www.eldiario.es/mundial-2026/lumumba-vea-hincha-rd-congo-ve-partidos-estatua_1_13325747.html

European External Action Service. (2026, 25 March). EU explanation of vote – UN General Assembly: Action on A/80/L.48 – Declaration of the Trafficking of Enslaved Africans and Racialized Chattel Enslavement of Africans as the Gravest Crime Against Humanity. European Union. https://www.eeas.europa.eu/delegations/un-new-york/eu-explanation-vote-%E2%80%93-un-general-assembly-action-a80l48-declaration-trafficking-enslaved-africans_en

Belgian Federal Government. (2022, 20 June). Discours du Premier ministre lors de la cérémonie de restitution de la dépouille de Patrice Emery Lumumba. News.belgium. https://news.belgium.be/fr/discours-du-premier-ministre-lors-de-la-ceremonie-de-restitution-de-la-depouille-de-patrice-emery

Masakhane. (n.d.). Masakhane: A grassroots NLP community for Africa, by Africans. https://www.masakhane.io/

Ministry of Foreign Affairs, Republic of Ghana. (2026). The Accra Next Steps Commitments on Reparation Justice. High-Level Consultative Conference on Reparations. https://reparations.mfa.gov.gh/

United Nations. (2002). Report of the World Conference against Racism, Racial Discrimination, Xenophobia and Related Intolerance, Durban, 31 August–8 September 2001 (A/CONF.189/12). United Nations. https://digitallibrary.un.org/record/456677?ln=en

National Park Service. (n.d.). The Eve of War. U.S. Department of the Interior. https://www.nps.gov/features/waso/cw150th/reflections/eve-of-war/page3.html

Paura, C. (2026). El hombre que revive a Patrice Lumumba en el Mundial 2026: un guiño a la historia del Congo. Historia National Geographic. https://historia.nationalgeographic.com.es/a/hombre-que-revive-patrice-lumumba-mundial-2026-guino-historia-congo_26376

U.S. Department of State, Office of the Historian. (n.d.). The Congo, Decolonization, and the Cold War, 1960–1965. https://history.state.gov/milestones/1961-1968/congo-decolonization

UNESCO. (2026, 2 April). African languages, the blind spot of AI. https://www.unesco.org/en/articles/african-languages-blind-spot-ai

United Nations in the Caribbean. (2025, 19 April). How Haiti paid for its freedom – twice over. United Nations. https://caribbean.un.org/en/293392-how-haiti-paid-its-freedom-%E2%80%93-twice-over

Vice-Presidency of the Republic of Colombia. (2026a). Comunicado de los copresidentes del Primer Foro de Alto Nivel CELAC–África. Government of Colombia. https://www.vicepresidencia.gov.co/prensa/Paginas/Comunicado-de-los-copresidentes-primer-foro-de-alto-nivel-Celac-Africa.aspx

Vice-Presidency of the Republic of Colombia. (2026b). Foro de Alto Nivel CELAC–África Colombia 2026: Agenda. Government of Colombia. https://www.vicepresidencia.gov.co/CELAC/assets/files/ES--AGENDACELAC-AFRICA-vff-3-Compress.pdf

Bibliography

Baptist, E. E. (2014). The half has never been told: Slavery and the making of American capitalism. Basic Books.

Beckert, S. (2014). Empire of cotton: A global history. Alfred A. Knopf.

Beckles, H. McD. (2013). Britain’s black debt: Reparations for Caribbean slavery and native genocide. University of the West Indies Press.

Borucki, A., Eltis, D., & Wheat, D. (2015). Atlantic history and the slave trade to Spanish America. The American Historical Review, 120(2), 433–461. https://doi.org/10.1093/ahr/120.2.433

Eltis, D., & Richardson, D. (2010). Atlas of the transatlantic slave trade. Yale University Press.

Ferrero, M. (2021). Accidental socialism: A natural experiment in Haiti 1796–1820. Journal of Institutional Economics, 17(3), 393–409. https://doi.org/10.1017/S1744137420000491

Gaillard-Pourchet, G.-K. (2023). La dette de l’indépendance d’Haïti. L’esclave comme unité de compte (1794–1922). Bibliothèque nationale de France, Patrimoines Partagés. https://heritage.bnf.fr/france-ameriques/dette-lindependance-dhaiti-lesclave-comme-unite-compte-1794-1922

Gil Montero, R. (2011). Free and unfree labour in the colonial Andes in the sixteenth and seventeenth centuries. International Review of Social History, 56(S19), 297–318. https://doi.org/10.1017/S0020859011000472

González Casanova, P. (1965). Internal colonialism and national development. Studies in Comparative International Development, 1(4), 27–37. https://doi.org/10.1007/BF02800542

González Senosiain, M. (2026a, 20 August). Cortés and Pizarro: conquest, thanatopolitics and imperial memory. From Abya Yala to contemporary Spain. AIthropology Lab. https://aithropologylab.org/en/essays/cortes-pizarro-thanatopolitics-imperial-memory/

González Senosiain, M. (2026b, 26 June). The Third Obligation: Accra and the Debt the West Refuses to Repay. AIthropology Lab. https://aithropologylab.org/en/essays/the-third-obligation-accra/

James, C. L. R. (1989). The Black Jacobins: Toussaint L’Ouverture and the San Domingo Revolution. Vintage Books. (Original work published 1938).

Koch, A., Brierley, C., Maslin, M. M., & Lewis, S. L. (2019). Earth system impacts of the European arrival and Great Dying in the Americas after 1492. Quaternary Science Reviews, 207, 13–36. https://doi.org/10.1016/j.quascirev.2018.12.004

Law, R. (1989). Slave-raiders and middlemen, monopolists and free-traders: The supply of slaves for the Atlantic trade in Dahomey c. 1715–1850. The Journal of African History, 30(1), 45–68. https://doi.org/10.1017/S0021853700030875

London Museum. (n.d.). The Zong Massacre Trial. https://www.londonmuseum.org.uk/collections/london-stories/zong-massacre-trial/

Manning, P. (1990). Slavery and African life: Occidental, Oriental, and African slave trades. Cambridge University Press.

Manning, P. (2010). African population: Projections, 1850–1960. In K. Ittmann, D. D. Cordell, & G. H. Maddox (Eds.), The demographics of empire: The colonial order and the creation of knowledge (pp. 245–275). Ohio University Press.

Mauss, M. (2009). Ensayo sobre el don: Forma y función del intercambio en las sociedades arcaicas (J. Bucci, Trans.; F. Giobellina Brumana, Ed.). Katz Editores. (Original work published 1925).

Mintz, S. W. (1974). The history of a Puerto Rican plantation. In Caribbean transformations (pp. 95–130). Aldine.

Nunn, N. (2008). The long-term effects of Africa’s slave trades. The Quarterly Journal of Economics, 123(1), 139–176. https://doi.org/10.1162/qjec.2008.123.1.139

Oosterlinck, K., Panizza, U., Weidemaier, M. C., & Gulati, M. (2022). The odious Haitian independence debt. Journal of Globalization and Development, 13(2), 339–378. https://doi.org/10.1515/jgd-2021-0057

Quijano, A. (2000). Coloniality of power and Eurocentrism in Latin America. International Sociology, 15(2), 215–232. https://doi.org/10.1177/0268580900015002005

Robins, N. A., & Hagan, N. A. (2012). Mercury production and use in colonial Andean silver production: Emissions and health implications. Environmental Health Perspectives, 120(5), 627–631. https://doi.org/10.1289/ehp.1104192

Rodney, W. (1972). How Europe underdeveloped Africa. Bogle-L’Ouverture Publications.

SlaveVoyages. (n.d.). Estimates: Trans-Atlantic Slave Trade. Retrieved 24 August 2026, from https://www.slavevoyages.org/assessment/estimates

Stavenhagen, R. (1967). Seven erroneous theses about Latin America. New University Thought, 4(4), 25–37.

Stoler, A. L. (Ed.). (2013). Imperial debris: On ruins and ruination. Duke University Press. https://doi.org/10.1215/9780822395850

Sánchez-Antonio, J. C. (2020). Tanato-política, esclavitud, capitalismo colonial y racismo epistémico en la invasión genocida de América. Tabula Rasa, 35, 157–180. https://doi.org/10.25058/20112742.n35.07

Thornton, J. K. (1998). Africa and Africans in the making of the Atlantic world, 1400–1800 (2nd ed.). Cambridge University Press.

Trouillot, M.-R. (1995). Silencing the past: Power and the production of history. Beacon Press.

UNESCO World Heritage Centre. (n.d.). Valongo Wharf Archaeological Site. https://whc.unesco.org/en/list/1548/

Van Deusen, N. E. (2023). Why Indigenous slavery continued in Spanish America after the New Laws of 1542. The Americas, 80(3), 395–432. https://doi.org/10.1017/tam.2023.33

von Wobeser, G. (2004). La hacienda azucarera en la época colonial (2nd ed.). Universidad Nacional Autónoma de México, Instituto de Investigaciones Históricas.

Wolf, E. R. (1982). Europe and the people without history. University of California Press.

Image credit

Patrice Lumumba, Brussels, 26 January 1960. Harry Pot / Anefo / Nationaal Archief. CC0 1.0, via Wikimedia Commons.