The Wrong First Question

ARepatriation is one word for several different transactions, and the confusion begins there. It can mean the return of an object carried off as loot in a documented act of war; the return of one bought in a colony from a seller who had little room to refuse; the return of human remains gathered for anatomical study; or, increasingly, the transfer of legal title to an object that stays exactly where it is. In the decade to 2023 the Kingsbridge Museum, a large civic collection in northern England, logged two hundred and fourteen formal requests for return. Thirty-one ended in a transfer of ownership, forty-six were refused, and the remainder are unresolved. Newspaper coverage tends to compress all of this into a single quarrel between museums that wish to keep things and nations that wish to have them back. The compression is convenient, and it has made the sharpest disagreements in the field almost invisible to the public.

BThe legal picture explains why so few disputes are ever tried. The international convention agreed in 1970 to curb the illicit movement of cultural property binds only the states that have ratified it, and it does not reach backwards: an object exported in 1897 is untouched by a rule adopted seventy-three years later. Domestic law is the second obstacle. Several national collections are governed by statutes drafted in the middle of the twentieth century which forbid trustees to part with anything in the permanent collection unless the item is a duplicate, is unfit to be kept, or has been damaged beyond repair. Trustees who sympathise with a claim may therefore be unable to act on it without new legislation. The consequence is that claims are settled, when they are settled at all, by negotiation rather than by judgment; fewer than one request in twenty has ever reached a courtroom.

CAdaeze Nwosu, a legal historian who has advised three claimant governments, argues that the field has been asking the wrong evidential question. Museums, she says, ask whether an object was stolen, and treat a receipt or a bill of sale as the end of the matter. Her own test is consent: could the people who handed the object over have refused? A purchase concluded in a town under military occupation, on her account, is not automatically a sale. Surveying eleven hundred West African objects in four European collections, she found that only twelve per cent were accompanied by documentation naming the person who transferred them, which she takes to show how rarely the question was thought worth recording. She is careful to separate ownership from custody. Transferring title, she points out, does not require the object to travel, yet the two decisions are routinely bundled together by institutions that would rather discuss shipping crates than law.

DBeatrix Sommer, who has curated ethnographic displays for thirty years, is read as Nwosu's opponent, though she does not defend nineteenth-century collecting and has called parts of it indefensible. Her objection is to what she names the single-origin fallacy. A carved figure may have been made in one polity, taken as tribute by a second, traded to a third and sold to a European buyer by a fourth, all before 1880; to speak of returning it to its place of origin is to assume a history the object itself often contradicts. The modern state that lodges the claim, she adds, is not always the heir of the community that made the thing, and in one case she has documented, two neighbouring national museums claimed the same group of objects on incompatible grounds. Restitution, in her formulation, is frequently the right outcome but the wrong first question; the first question is who is entitled to decide.

EPractice has moved faster than either argument. The Wexford agreement of 2021 transferred ownership of twenty-seven of the forty-one objects under claim to a national museum overseas, while leaving them physically in Britain on a five-year loan from their new owners — an arrangement that reversed the direction of the lending without emptying a gallery. Digitisation has been offered as a further compromise, and high-resolution scanning now yields a copy accurate to a fraction of a millimetre. Nwosu's response is that a scan answers a research need and leaves the ownership question precisely where it was. An independent audit published in 2022 followed three hundred objects returned since 2010 and found that seventy-one per cent were on public display within two years of arriving, a figure that surprised commentators who had predicted that returned material would vanish into storage.

FWhat has changed, then, is less the balance of the argument than its subject. Twenty years ago the contested question was whether a European museum could ever be obliged to give anything back; the practical question now is who decides, on what evidence, and how quickly. Nwosu and Sommer disagree about that far more sharply than they disagree about the wrongs of the nineteenth century, and it is on procedure — who sits on the panel, what counts as proof, whether a refusal can be appealed — that the next decade of claims will turn. The universal museum, an institution that presented itself as holding the world in trust for everyone, was always an argument as much as a building. Its defenders are now making a narrower and more interesting case: not that the collections should stay where they are, but that nobody has yet designed a body competent to decide where they should go.