Returning the Bronze, Keeping the Key: Heritage as Power and the Comparativity of Restitution

Abstract. Restitution is now the preferred language through which Western museums narrate their own reform. Germany's 2022 joint declaration with Nigeria, the Smithsonian's return of twenty-nine Benin works in the same year, and the Sarr–Savoy report of 2018 are routinely read as a turning of the tide. This essay argues, through Comparativity Complex Theory (CCT), that the tide has turned inside a harbour the museum still owns. Restitution, as presently practised, transfers title while conserving the comparative categories (object, collection, art, universal value, loan, public access) that converted ancestral altars into property in 1897. The Nigerian presidential declaration of March 2023 vesting ownership in the Oba of Benin exposed those conditions by violating them. Read against the Osun-Osogbo Sacred Grove, the Akan logic of Sankofa, and the framework of Heritage Security Intelligence, the Benin case shows that heritage survives by morphing, and that the power to let it morph is precisely what has not been returned.

Keywords: heritage as power; restitution; Benin Bronzes; Osun-Osogbo; Sankofa; Comparativity Complex Theory; Heritage Security Intelligence

A Punitive Expedition Never Ends

In February 1897 a British naval force entered Benin City, burned much of it, sent the Oba Ovọnramwẹn Nọgbaisi into exile at Calabar, and carried away several thousand brass, ivory and wooden works from the royal palace and its ancestral shrines. The operation had a name that did its own ideological work: it was a "punitive" expedition, a correction administered by the civilized upon the disorderly, retaliation dressed as justice. The objects were auctioned, dispersed, sold to defray the costs of the campaign, and distributed across the museums and drawing rooms of Europe and North America, where they acquired a second life as "Benin Bronzes," a trade name for a violence.¹ More than a century later the same objects have become the emblem of a very different mood. Western museums now speak of provenance, shared stewardship, ethical returns and relational ethics. Ministers sign declarations; directors stand beside Nigerian officials at repatriation ceremonies; the language of restitution has migrated from the protest placard to the museum annual report.

This essay does not doubt the sincerity of that migration. It doubts its depth. Its thesis is contestable and deliberately so: the restitution consensus of the last decade has returned objects without returning power, because it has remained inside the comparativity of the museum itself. The categories through which the museum first rendered a Benin altar-head legible as "art," "specimen," "collection item" and "world heritage" survive the transfer of legal title intact. What travels back to Benin City is a museum object with a new owner. What stays behind in Berlin, Washington and London is the authority to define what an object is, how it must be kept, who may see it, and what counts as its safety. Heritage, in the vocabulary of Comparativity Complex Theory, morphs or it dies. The restitution consensus returns heritage on condition that it stop morphing.

The argument proceeds through three African cases and one framework. The Benin corpus, from 1897 to the Nigerian presidential declaration of March 2023, shows the museum's comparativity at work and then under stress. The Osun-Osogbo Sacred Grove in Osun State, inscribed on the World Heritage List in 2005, shows a heritage that survived by transforming, and shows the international heritage regime trying to fix it in place. The Akan principle of Sankofa supplies a logic of retrieval that is not a logic of preservation. And Heritage Security Intelligence (HSI), the framework this author has developed within the Comparative Heritage Project, reframes the whole debate as a question of security: who secures heritage, for whom, against what, and with what knowledge.

Heritage Is Made, and Made to Rule

The critical heritage literature has spent three decades dismantling the common-sense view that heritage is simply what the past leaves behind. Laurajane Smith's Uses of Heritage gave the field its most durable tool, the "authorized heritage discourse," a professional and state-sanctioned way of speaking that privileges monumentality, expert judgement, aesthetic value, time depth and the nation, and that assigns ordinary communities the role of grateful audience rather than author.² For Smith heritage is a cultural process, a practice performed in the present, rather than a thing awaiting discovery. David Lowenthal, working in a more ironic register, insisted that heritage is not history at all: history seeks to understand the past, whereas heritage exaggerates, omits and sanctifies in order to secure a group's claims in the present.³ Stuart Hall, speaking from the Caribbean-British margin of the national story, asked the question that turned heritage from a description into a demand, "whose heritage?", and showed that "The Heritage" of Britain had been built as an exclusionary settlement of who belongs.⁴ Barbara Kirshenblatt-Gimblett named heritage a mode of "metacultural production," culture produced about culture, and Rodney Harrison has since argued that heritage is an assembly of practices through which late modern societies manage their anxieties about loss.⁵ The older insight of Eric Hobsbawm and Terence Ranger, that many traditions claiming antiquity are recent and invented, and Ranger's demonstration that colonial administrators in Africa manufactured "customary" tradition to rule through it, sit beneath all of this work.⁶

Comparativity Complex Theory accepts this inheritance and presses it further in two directions. First, CCT begins from an anthropological claim rather than a disciplinary one. Human beings, in the formulation this author offered in 2022, are "primarily a culture-producing machinery," and the theory stands or falls on the assumption that "heritage is deeply rooted and justifies most human engagements and performances of power."⁷ Culture-producing and domination-seeking are two faces of one capacity. The same hands that cast a commemorative plaque for an Oba's court also inscribe a hierarchy; the same museum that preserves a plaque also inscribes a hierarchy. Heritage is therefore neither innocent inheritance nor mere invention. It is power's favourite raw material, because it arrives already charged with the authority of the dead.

Second, CCT distinguishes comparison from comparativity. Comparison is an act: two things are placed side by side and judged. Comparativity is the prior field of relations, histories and institutions that makes things comparable in the first place, that decides in advance which axis of judgement will operate (art against artefact, written against oral, universal against local), and that distributes the authority to judge.⁸ A restitution dispute looks like a comparison between two claimants, museum and nation. CCT asks instead what comparativity produced "museum" and "nation" as the only recognisable claimants, and "object" as the only recognisable thing claimed. This is where the comparativist must dig, past the surface of the dispute, to the thresholds of discomfort at which the categories themselves begin to crack.

Out of these two moves comes the proposition this essay tests: heritage morphs or it dies. A living heritage is one whose custodians retain the authority to reinterpret, re-perform, re-make and even consume it. When that authority is removed, heritage may be preserved, but it is preserved the way a specimen is preserved, fixed at the moment of capture. Elsewhere this author has argued that African religious ethics is a lived and relational system rather than a code, sustained by practice rather than by text.⁹ The same holds for African heritage. Its mode of existence is use, and use is change.

Brass Into "Art": The Comparativity of the Collection

Before 1897 the works now called Benin Bronzes had no single name and no single function. The cast-brass plaques that once sheathed the pillars of the palace recorded court ceremony and precedence; the commemorative heads stood on ancestral altars, supporting carved ivory tusks, and received offerings; the casting guild worked under royal control, and the right to commission and hold certain forms was itself a mark of kingship.¹⁰ These were instruments of a political theology in which the Oba's authority was continuously re-enacted through ritual addressed to royal ancestors. They were not "art" in the European sense, and they were emphatically not "collection." They were heritage in the fullest CCT sense: the material through which a polity produced itself, generation after generation, and through which it exercised domination over its own subordinates and neighbours. There is no need to romanticize the pre-colonial court in order to recognize what happened to its objects.

What happened was a reclassification carried out by force and then completed by scholarship. Annie Coombes has shown how the looted works were absorbed into late Victorian and Edwardian England through the press, ethnographic museums and exhibitions, and how the same objects oscillated between two incompatible verdicts: evidence of a degraded "City of Blood" that justified conquest, and sophisticated castings that seemed too accomplished for Africans to have made without European influence.¹¹ Each verdict served the empire. The first justified the expedition; the second salvaged European superiority by assigning the skill elsewhere. Both assumed that the right to decide what the objects meant belonged in London. Tony Bennett's account of the nineteenth-century "exhibitionary complex" helps here: the public museum was an instrument for ordering citizens as much as objects, teaching visitors to see the world arranged in sequences of progress with Europe at the summit.¹² Once the Benin works entered those sequences, they served as African exhibits within a European narrative of human development.

Dan Hicks has pushed this argument to its sharpest point. In The Brutish Museums he contends that the anthropology museum did not merely receive the spoils of 1897 but extended the violence of 1897 across time, so that every year of display renews the original taking, and the museum becomes a technology for prolonging a military event.¹³ His critics have found the argument overheated. It is certainly polemical. Yet on the evidence of the British Museum Act 1963 it is difficult to call it exaggerated. Section 5(1) provides that the Trustees "may sell, exchange, give away or otherwise dispose of any object vested in them and comprised in their collections" only under narrow conditions, chiefly duplication or unfitness for retention.¹⁴ The law was not written with Benin in mind. It nonetheless converts a taking into a trust, and makes the museum's custody of looted altars a statutory duty. A comparativity is here written into statute: the British public is constituted as the beneficiary, the Edo court as a foreign claimant whose claim the law cannot hear.

The strongest opposing position deserves a fair hearing. James Cuno has argued for the "encyclopedic museum" as a cosmopolitan institution that places the cultures of the world in conversation, resists the nationalist capture of antiquity, and holds objects for all humankind rather than for any modern state.¹⁵ Cuno's worry about nationalism is not frivolous; a Nigerian federal claim to Edo royal regalia is also a claim by a post-colonial state to a heritage it did not produce. But the encyclopedic argument quietly assumes what it needs to prove, namely that "humankind" is best represented by institutions in a handful of Western capitals, and that cosmopolitan access is realized by visitors who can obtain visas to see the collections. The universal, in this argument, has an address. CCT's objection is not that the universal museum compares cultures; comparison is inescapable. The objection is that it monopolizes comparativity. It decides the axis of judgement, grants itself the role of judge, and then calls the result universal.

The Restitution Consensus and Its Limits

The turn against that monopoly did not begin in 2018. Bénédicte Savoy's Africa's Struggle for Its Art reconstructs the restitution demands pressed by African states and intellectuals from the 1960s to the 1980s, and the coordinated institutional resistance that defeated them, so thoroughly that the defeat was then forgotten.¹⁶ The report she wrote with Felwine Sarr for President Emmanuel Macron in November 2018 therefore reopened a closed file rather than opening a new one. Its intellectual ambition remains the high-water mark of the restitution literature. Sarr and Savoy define restitution with legal precision: to restitute is to re-institute an item to its legitimate owner "for his legal use and enjoyment, as well as all the other prerogatives that the item confers (usus, fructus, and abusus)."¹⁷ They insist that what was taken from African societies was more than objects; it was "reserves of energy, creative resources, reservoirs of potentials," and they conclude that "Compensation here consists in offering to repair the relation."¹⁸ Following Simon Njami, they note that a returned object does not return as it was, that the point is "not about a return of the same, but of a 'different same'."¹⁹

This is a remarkable text, and CCT reads it as an ally. Notice, however, the third term in the Roman triad. Abusus is the right to transform, consume, alienate or destroy. It is the prerogative that separates an owner from a custodian. A restitution that returns usus and fructus while withholding abusus returns a lease. The question for the actual practice of return, as distinct from the theory, is therefore simple: has anyone returning Benin works returned the right to let them morph?

Consider the Joint Declaration signed in Berlin on 1 July 2022 by Annalena Baerbock and Claudia Roth for Germany and by Lai Mohammed and Zubairu Dada for Nigeria. Its preamble recognizes "the need to achieve not only the return of objects but also a new understanding of cultural cooperation," and its first article speaks of "returning Benin Bronzes unconditionally to Nigeria."²⁰ The document identifies 1,130 works in German public collections and commits the National Commission for Museums and Monuments and the holding museums to transfer agreements "including transfer of ownership." Then come the conditions that the word "unconditionally" was meant to exclude. German museums "will continue to display Benin Bronzes on loan as set out in the transfer agreements." Both sides undertake to support "the establishment of museum facilities in Benin City." Both declare their intention "to contribute to the universal role of the Benin Bronzes" by ensuring "their display is accessible to the general public and to research," and to facilitate their circulation through "international travelling and joint exhibitions."²¹

Read through CCT, the declaration is a precise map of what was and was not returned. Title returned. The museum form did not. The objects must be displayed, lent, circulated, researched, housed in museum facilities and assigned a "universal role." The Edo altar is returned as an exhibit, and the new owner inherits the old owner's obligations toward the global public that the museum invented. This is what it means for restitution to remain inside the museum's comparativity: the categories that governed the objects under German ownership (collection item, research resource, universal heritage, loanable asset) govern them still, now with a Nigerian signature on the deed.

The Smithsonian's return of October 2022 shows the same structure at smaller scale. At a ceremony on 11 October the National Museum of African Art transferred ownership of twenty-nine works to Nigeria's National Commission for Museums and Monuments, the first official action under the Institution's new ethical returns policy; twenty were to travel to Nigeria, while nine would remain at the museum on loan.²² The Smithsonian's Secretary spoke of addressing "a historic injustice."²³ The Oba of Benin, Ewuare II, speaking through his brother Prince Aghatise Erediauwa, offered the sentence that the occasion deserved: "The truth is that no argument can change looted works into un-looted works or stolen works into un-stolen works."²⁴ The Oba's sentence is ontological. It denies that the museum's classification ever succeeded in turning the objects into anything other than what they were. The loan arrangement is procedural, and it keeps a third of the returned works in the building from which they were returned. Both statements were true on the same afternoon. The gap between them is the subject of this essay.

The Oba's Gazette and the Scandal of Sovereignty

That gap became a crisis in the spring of 2023. In March, weeks before he left office, President Muhammadu Buhari issued a notice published in the federal government's official gazette, titled "Notice of Presidential Declaration on the Recognition of Ownership and an Order Vesting Custody and Management of Repatriated Looted Benin Artefacts in the Oba of Benin."²⁵ The declaration recognized the Oba, rather than the National Commission for Museums and Monuments, as owner and custodian of looted Benin works, whether already returned or still abroad. Reporting in The Art Newspaper described European museums that had negotiated with the federal commission suddenly uncertain about who their partner was, and the commission itself apparently blindsided.²⁶ Plans elsewhere to house the works in a new museum in Benin City were thrown into doubt.

Much Western commentary treated the gazette as a problem of governance, even as a betrayal: museums had returned objects to a public institution for public display, and the objects might now vanish into a palace. L. Amede Obiora reads it otherwise, as the belated recognition that traditional institutions held these works under native law and custom before any state existed to claim them, and that cultural heritage law, national and international, has systematically excluded such institutions from standing.²⁷ This essay takes a position close to Obiora's, with one CCT qualification. The gazette was a scandal, but a revealing one. It was the first moment in the entire restitution process at which a Nigerian authority exercised the prerogative Sarr and Savoy called abusus: the right to decide, in an idiom not supplied by the museum, what the returned objects are and where they belong. The reaction showed at once that the restitution consensus had assumed the objects would remain museum objects after return. The expectation that returned Bronzes would be publicly displayed was a condition that had never been stated as a condition, and it surfaced only when it was breached.

The qualification is that heritage is power in Benin City as much as in Berlin. The contest among the federal commission, the Edo State government and the palace is a contest over who inherits the authority that 1897 destroyed and the Nigerian state partly absorbed. Jean-François Bayart's concept of extraversion is useful here: African political actors have long mobilized resources drawn from their relationship with external powers to consolidate authority at home.²⁸ Restitution is an unusually valuable external resource, and every Nigerian claimant has an interest in capturing it. CCT does not require the analyst to pick a virtuous claimant. It requires the analyst to see that the choice among claimants is itself structured by the museum's comparativity, which recognizes states and institutions and has only reluctantly learned to hear kings, priests and lineages. The palace's victory in 2023 does not end the question of power; it relocates it. What it does establish is that the question of power was always the real question, and that the language of stewardship had been used to postpone it.

A threshold of discomfort must be named here. Some readers, African and Western, would prefer that returned works be placed in a public museum where schoolchildren in Edo State can see them. That preference is legitimate. But it is a preference for one heritage regime over another, and it cannot be smuggled into restitution as a neutral condition of return. If the Oba's court chooses to return certain works to ritual use, to restrict their viewing, or to let some of them age in the conditions of a shrine rather than a climate-controlled vitrine, that choice is what restitution means. To demand otherwise is to insist that the object may go home only if it remains in the museum.

Ọ̀ṣun Does Not Live in a Vitrine

If Benin shows heritage frozen by capture, Osun-Osogbo shows heritage that stayed alive by changing. The sacred forest on the edge of Osogbo, through which the Ọ̀ṣun river runs, is the abode of the òrìṣà Ọ̀ṣun, and its annual festival re-enacts the covenant between the goddess, the founders of the town and the Àtáója, the town's king. UNESCO's description records a grove of some seventy-five hectares, first declared a National Monument in 1965 and extended in 1992, containing forty shrines, sculptures and artworks, "many created in the past forty years," along with palaces, sacred places and worship points "with designated priests and priestesses."²⁹ It was inscribed on the World Heritage List in July 2005 under three criteria. Criterion (iii) names it "the largest and perhaps the only remaining example of a once widespread phenomenon that used to characterise every Yoruba settlement." Criterion (vi) calls the annual festival "a living thriving and evolving response to Yoruba beliefs in the bond between people, their ruler and the Osun goddess." Criterion (ii) reads: "The development of the Movement of New Sacred Artists and the absorption of Suzanne Wenger, an Austrian artist, into the Yoruba community have proved to be a fertile exchange of ideas that revived the sacred Osun Grove."³⁰

The history behind that sentence is well documented. Susanne Wenger arrived in Nigeria in 1950 with her husband Ulli Beier, became an initiate and priestess in the Ọ̀ṣun cult, and from the late 1950s worked with local artists and devotees, the group later known as the New Sacred Art movement, to rebuild ruined shrines and install monumental sculpture in cement and other materials throughout the grove, while resisting encroachment by farmers, loggers and land speculators.³¹ Beier's The Return of the Gods presented this work to an international public in 1975.³² Peter Probst's Osogbo and the Art of Heritage traces how the grove, the Osogbo school of artists, the Ọ̀ṣun festival, local chieftaincy politics, state tourism and UNESCO recognition became entangled, so that "heritage" at Osogbo is simultaneously devotion, artistic modernism, civic identity and revenue.³³

For CCT the grove is a nearly perfect demonstration that heritage morphs or dies. The shrines survived because they were rebuilt in new forms, new materials and an idiom that was neither "traditional" in the museum's sense nor European. The goddess was honoured with twentieth-century sculpture. The festival persisted by absorbing the town's modern politics, its diaspora pilgrims and its tourists. Had the grove been treated as an archaeological site, preserved as found in the 1950s, there would have been little left to inscribe. What UNESCO recognized in 2005 was the product of a half-century of transformation.

And yet the inscription reveals the other face of comparativity. The criterion that explains the grove's revival credits "the absorption of Suzanne Wenger, an Austrian artist." Whatever Wenger's genuine devotion, and it was genuine by every account, the authorized heritage discourse located the agency of revival in the European who joined the community, while the Yoruba devotees, priests, carvers and masons who did the same work appear collectively as a "Movement." This is the comparativity of the colonial exhibition in a gentler key: African heritage becomes legible as outstanding and universal at the point where a European hand touches it. The same UNESCO text worries that "the annual Osun Osogbo festival will need to be better managed so that the site will no longer suffer from adverse impacts of tourism during the festival."³⁴ The festival that criterion (vi) praises as living and evolving becomes, in the management section, a risk to be managed. A living heritage is welcomed for its vitality and then disciplined for the same quality. Michel Foucault's analysis of disciplinary power, in which institutions produce the subjects they claim only to observe, is apt: the World Heritage apparatus produces the grove as a site of "outstanding universal value" and then requires Osogbo to govern itself according to that production.³⁵

Lynn Meskell's research on UNESCO sharpens the point. In A Future in Ruins she traces the organization's founding dream of peace through the minds of men, and shows how the World Heritage system has been increasingly captured by the agendas of member states, so that inscription functions as diplomacy, prestige and economic leverage.³⁶ In an earlier assessment of the Convention at forty she documented the growing tendency of the World Heritage Committee to override the technical recommendations of its advisory bodies in favour of political bargaining among states.³⁷ Inscription, on this evidence, is a transaction within a state system. That is not a reason to reject it; the grove's inscription brought protection and resources. It is a reason to see that the authority conferred by inscription flows to the state party and the international regime, and only derivatively to the Àtáója, the priests and priestesses of Ọ̀ṣun, and the women who carry the calabash in procession.

Ferdinand de Jong's work on Senegal offers a way of holding both truths together. In Decolonizing Heritage he shows how Senegalese heritage institutions, shaped by Léopold Sédar Senghor's Négritude, treated museums and World Heritage sites as instruments of repair for colonial time and as archives for a future still to come; with Michael Rowlands he had earlier gathered West African cases in which heritage was reclaimed through alternative imaginaries of memory rather than through the official canon.³⁸ De Jong's point is that heritage in postcolonial Africa is oriented forward. That orientation is what the grove embodies and what the museum's comparativity finds hardest to accommodate. The grove is not a record of what Yoruba religion was. It is a place where Ọ̀ṣun is still being addressed, where ẹbọ is still offered, where àṣẹ, the power to make things happen, is still sought. A vitrine can display àṣẹ's residue. It cannot host its exercise.

Sankofa Is Not Nostalgia

The Akan concept of Sankofa supplies a logic that restitution discourse badly needs. Rendered in the Adinkra repertoire as a bird whose head turns back toward its tail, and glossed in Akan as the injunction to go back and fetch what has been left behind, Sankofa has become one of the most widely circulated African concepts of the last half-century.³⁹ Christel Temple's history of Sankofa practice shows how the concept was taken up in the United States as a philosophy of cultural recovery across education, the arts, community organizations and scholarship, to the point where it now functions as a general name for the retrieval of African memory.⁴⁰ That wide circulation has costs; the symbol is easily reduced to an emblem of heritage pride. What deserves recovery is its grammar.

The grammar of Sankofa is a return undertaken for the sake of the future. One goes back and fetches in order to move forward with what was missing. The past is not revisited for contemplation; it is raided for use. This is the opposite of the authorized heritage discourse, which looks backward in order to conserve, and it is the opposite of the museum's temporal regime, in which the object is permanently held at the moment of its acquisition. Sankofa and CCT's axiom converge: the only heritage worth fetching is heritage that can be put back to work, which means heritage that will change in the working. Sarr and Savoy's "different same" is Sankofa translated into the idiom of a French ministerial report.

Applied to Benin, a Sankofa logic would measure restitution by what the returned works are able to do, not by where they are kept. Can they re-enter the ritual life of the court, if the court so decides? Can new works be commissioned from the casting guild to stand alongside them, as the Joint Declaration's article on contemporary production faintly anticipates, without being treated as mere craft reproductions of a closed classical canon? Can the objects be allowed to acquire new meanings in Edo, including meanings that Western curators would find distasteful? A restitution that answers yes to these questions has returned power. A restitution that answers yes only on the condition of public display and international loan has returned property and retained the key.

Sankofa also disciplines the restitution activist. Fetching is not the same as freezing. A demand that returned works be treated as sacred and untouchable in perpetuity, that no new meaning be permitted, reproduces in African dress the very museum logic it opposes. Heritage that is restored only to be sealed has been rescued from one vitrine in order to be placed in another. CCT's axiom cuts both ways: it condemns the museum that will not let heritage morph, and it condemns the nationalist or traditionalist who will not let it morph either.

Heritage Security Intelligence: Who Secures What, Against Whom

The language of the Benin debate has always been a language of security, though rarely acknowledged as such. The expedition of 1897 was presented as a punitive and security operation, a response to the killing of a British party in January of that year and a pacification of a hostile trading frontier.⁴¹ The retention of the works through the twentieth century was defended in a security register: the objects were safe in London, protected by climate control, guards, conservators and law, whereas their return would expose them to theft, decay, trafficking or neglect. In the 2023 controversy the security register reappeared at once, as commentators asked whether works entrusted to a palace would be secure. Restitution, in other words, has been argued throughout as a dispute about safety, and in every round the institution that held the objects claimed the expertise to judge what safety meant.

Heritage Security Intelligence (HSI), as this essay develops it, begins by inverting that claim. HSI treats heritage as a security question in the full sense: a field in which communities are rendered secure or insecure, in which knowledge about heritage is also knowledge about power, and in which the decisive intelligence is knowledge of how heritage is being weaponized. On this view the principal insecurity of the Benin works was never humidity. It was the removal of the Edo court's authority over its own heritage, an insecurity inflicted in 1897 and prolonged by every legal, curatorial and diplomatic arrangement that kept that authority elsewhere. A vitrine can make an object physically secure while keeping its community insecure. The security of heritage, in HSI's terms, is measured by the security of the community's capacity to produce, reproduce and transform it.

Three propositions follow. The first concerns sovereignty of interpretation. A heritage is secure only when those who inherit it decide what it means; any arrangement in which meaning is supplied by an external authority, however benevolent, is a standing vulnerability. The second concerns custody of use. Heritage is secured by being used, and use includes forms of handling that museum conservation forbids. The Ọ̀ṣun festival is safer as a festival that sometimes strains its management plan than as a performance staged for inspectors. The third concerns the capacity to morph. A heritage that cannot change is already dying; securing it means securing the conditions under which it can change on its own terms, including the economic conditions, the legal standing of its custodians, and their freedom from the obligation to perform authenticity for outsiders.

HSI is also an intelligence discipline in the plain sense: it asks what the parties to a heritage dispute know, conceal and misrepresent. The Benin case yields an instructive example. The restitution consensus presented its conditions (display, loan, universal access) as neutral professional standards rather than as the continuation of a particular regime of power. Uncovering that concealment is intelligence work, and it is the work CCT calls digging to the thresholds of discomfort. Ciraj Rassool's account of the return and reburial of the remains of Klaas and Trooi Pienaar from Vienna to South Africa in 2012 shows the same method applied to human remains: return became meaningful only when it was joined to a process of "rehumanisation" that undid the racial-scientific categories through which museums had turned persons into specimens.⁴² Rassool's insistence that the museum was one of empire's primary sites, an institution of extractive and hierarchical knowledge, is an HSI claim avant la lettre. Returning the body without undoing the category returns a specimen.

This is also where HSI meets the long history of violence against Black bodies and memories that this author has examined in the American context, where the display of the dead and the management of their memory served as instruments of terror long after the events themselves.⁴³ The museum's custody of African ancestors, whether in brass or in bone, belongs to the same family of practices. Security, for the communities concerned, has never meant only protection from theft. It has meant protection from the power of others to decide what their dead signify.

Against the Comfort of Return

A reasonable objection must be met. Is it not ungracious to criticize institutions for doing, belatedly and imperfectly, what critics demanded for decades? Germany did transfer title to more than a thousand works. The Smithsonian did adopt a returns policy and act on it. The Sarr–Savoy report did change the vocabulary of European cultural ministries. Surely the right response is to consolidate these gains rather than to question their depth.

The answer is that consolidation on the present terms would lock the gains into a form that forecloses their completion. Aníbal Quijano's account of the coloniality of power, in which the social classifications produced under colonial rule persist long after colonial administration ends, describes exactly this risk: a formally decolonized relation that keeps the colonial classification intact.⁴⁴ V. Y. Mudimbe showed how the very idea of Africa was produced within a Western order of knowledge that decided in advance what African things could mean.⁴⁵ Edward Said showed how the authority to represent the Orient was itself a form of possession.⁴⁶ The restitution consensus has transferred possession of certain things. It has not transferred the authority to represent them, classify them, or determine the conditions of their care. Until it does, returns will remain gifts from the holder of the key, revocable in spirit if not in law, and the 2023 reaction to the Oba's gazette showed how quickly the spirit of a gift can harden into disappointment when the recipient behaves as an owner.

Nor does this critique require that museums empty their galleries. It requires that restitution be redefined so that its success is measured by the recovered authority of the source community rather than by the elegance of the transfer ceremony. On that measure the Nigerian presidential declaration of 2023, whatever its internal politics, did more to test the meaning of restitution than any European policy document, because it was the first act that treated the returned works as belonging to a living institution with its own rules. Savoy's history of postcolonial defeat is a warning that restitution can be defeated by forgetting as well as by refusal; the present danger is a subtler defeat, in which restitution succeeds so smoothly that no one notices the categories have stayed behind.⁴⁷

The Key Stays in the Lock

Restitution debates have been conducted, for the most part, as arguments about where objects should be. This essay has argued that the more consequential question is what objects are allowed to become. The Benin works were made to be used by a living court, reclassified by force into "art" and "ethnography," held by statute as a public trust in foreign capitals, and are now being returned under conditions that preserve that reclassification. The museum's comparativity, its authority to define object, value, care and access, has survived the transfer of title. That is why restitution in its present form returns objects without returning power.

The Osun-Osogbo grove shows what an alternative looks like and what threatens it. The grove lives because its custodians, with a European initiate among them, allowed it to change. Its inscription as World Heritage honours that change while importing a regime of management that treats the festival's vitality as a risk and credits a foreigner with the revival. Sankofa supplies the principle that should govern return: go back and fetch, so that what is fetched can be used. Heritage Security Intelligence supplies the measure: a heritage is secure when its community holds the authority to interpret, to use and to transform it. By that measure the Benin works are not yet secure, whatever the quality of their new cases.

Heritage morphs or it dies. The morphing has already begun, in Benin City's palace politics, in the quarrels among Nigerian claimants, in new castings and in the arguments of Nigerian lawyers and scholars about what native law and custom require. These are not signs that restitution has failed. They are the first signs that it might succeed, because they are the sound of heritage being argued over by the people it belongs to. The institutions that returned the works can help by doing the one thing the restitution consensus has so far avoided. They can hand back the key, and then stop asking what will be done with the door.