Europe’s recent restitution legislation raises an important question about African archaeological material: can a legal mechanism designed to address historical appropriations also respond to objects removed through the modern illicit market?
Nok terracottas provide a difficult test. Their importance lies not only in the objects themselves, but also in their archaeological context. When archaeological material is excavated and removed without adequate recording, the loss is not merely proprietary. Information about provenance, use, chronology and cultural setting may be lost irretrievably.
A more careful account of the Paris question
Published accounts describe a dispute involving Nok and Sokoto sculptures acquired for the French museum collection that later became associated with the musée du quai Branly. They also describe a 2002 Nigeria–France arrangement under which Nigeria’s ownership was recognised and the objects remained in France on loan. Those specific propositions should be supported, before publication, with the text of the bilateral instrument and contemporaneous museum or government records. I have not treated the detail of that arrangement as independently verified here.
That evidential caution matters. The legal consequences of any prior agreement depend on its precise terms: whether it recognised title, created a loan, included a release or final-settlement language, and identified the particular objects concerned.
What the French legislation demonstrably changes
France enacted Law No. 2026-351 of 9 May 2026, concerning the restitution of cultural property that has been subject to illicit appropriation. The law amends the Code du patrimoine, including by inserting a statutory framework for cultural objects subject to illicit appropriation. It also requires government reporting to Parliament on restitution requests, decisions removing objects from the public domain, completed restitutions and unsuccessful requests. Law No. 2026-351 of 9 May 2026
This is a material institutional development. It creates a general statutory route in an area where restitution from French public collections has often depended on legislation directed to particular objects or claims. But the statute should not be used to make unqualified claims about any particular Nok object without first checking the current text of the relevant Code du patrimoine provisions and the full documentary history of that object.
The legal question is therefore not simply whether an object is culturally Nigerian. It is also whether the statutory criteria apply, what evidence establishes illicit appropriation, and whether an earlier state-to-state arrangement has consequences for a renewed request.
The international-law gap
The 1970 UNESCO Convention and the 1995 UNIDROIT Convention are often discussed together, but they do different work. Any published treatment should distinguish carefully between them and should cite the authentic treaty texts and the current depositary status of the states concerned.
In particular, it is unsafe to say that an unprovenanced archaeological object was “never stolen” under the 1970 Convention, or that one convention necessarily provides a remedy in a particular national court. The availability of a remedy turns on the relevant treaty provision, the states’ treaty status, implementing legislation, limitation periods, and the forum’s private-law and procedural rules.
Why South Africa is relevant
South African law provides concrete border-control and enforcement mechanisms, although their operation depends on the statutory conditions being met. Under section 33(1) of the National Heritage Resources Act 25 of 1999, foreign cultural property may be imported only through a customs port of entry, and the export permit or other permission issued by the country of origin must be produced to a customs officer. Section 33(2) prohibits the import of foreign cultural property illegally exported from a reciprocating state after a cultural-property agreement between South Africa and that state has entered into force. A customs officer who reasonably suspects a contravention may withhold the object while provenance is investigated. National Heritage Resources Act 25 of 1999, s 33
The Act also provides that every member of the South African Police Service and every customs and excise officer is deemed to be a heritage inspector. National Heritage Resources Act 25 of 1999, s 50(2)
These provisions do not mean that every object said to be unprovenanced or foreign is automatically seizable or returnable. The factual and legal predicates must be established, including the object’s identity, origin, applicable export law, export history, documentation, and—where section 33(2) is invoked—the relevant cultural-property agreement.
The practical question for the market
For an archaeological object offered for sale, provenance diligence should be capable of answering more than the broad question, “Is there paperwork?” It should address:
- the object’s documented chain of custody and ownership;
- the country and date of excavation or discovery, where known;
- the export permission or other authority specifically connected to the object;
- whether the object falls within a published risk category; and
- any claim, restriction, seizure history or unresolved provenance gap.
A missing answer is itself material. It may not resolve title or criminality, but it should affect the level of risk a buyer, dealer, museum or insurer is prepared to accept.
Conclusion
Nok terracottas show why restitution cannot be reduced to a simple opposition between colonial-era taking and legitimate modern collecting. The modern market may preserve objects, but it can also reward the destruction of archaeological evidence. The most defensible response is rigorous documentation before acquisition, transparent publication of provenance, and legal analysis tied to the actual object and the actual instruments—not assumptions about either.
Verification note before publication
The following claims in the earlier draft require primary-source citations before they should be reinstated: the dates and terms of the alleged 2002 Nigeria–France arrangement; the 1998–1999 acquisition history; the 1995 Patrick Darling report and its quantitative claims; the stated ICOM Red List date and scope; particular customs returns; and the treaty signature, accession and entry-into-force dates for France, Nigeria and South Africa. The revised text deliberately omits or qualifies them pending that verification.