Introduction
For decades, states in Africa, Asia, and Latin America have demanded the return of their cultural heritage that their former European colonisers had looted,Footnote 1 but European states denied or silenced them in the 1960s and 1970s.Footnote 2 While demands and debates on colonial heritage restitution are not new,Footnote 3 the past decade since roughly 2017 has seen renewed attention and new openness by European states to engage and consider colonial heritage restitution as an act of transitional justice. This new openness to righting past injustices includes changing museum practices, investment in provenance research, engagement with source communities, and actual returns of collection items. Over the past decade then, colonial heritage restitution has become an emerging international norm that manifests in increased appearance in domestic discourses and legal and institutional changes. But norm implementation levels differ: some states, including France, Germany, the Netherlands, and Belgium, have committed to restitution at the national level. Others, such as the United Kingdom (UK) and Portugal, have so far rejected broad national restitution policies. What helps to explain this divergent norm uptake?
The UK–Belgium comparison below is part of a larger project tracing and explaining European states’ implementation progress through analysing changed national discourse, institutions, and laws over the past decade. Both states are consolidated democracies and have been subject to similar renewed external and domestic pressures for restitution: some African states are demanding returns, neighbouring European states are advocating for restitution (e.g. France, Germany, the Netherlands), grassroots organisations and museums ask for the decolonisation of public spaces and museums. Both states also have extensive colonial-looted heritage in their national museums. They differ, however, in the extent to which they have committed to national restitution policies and whether museums can make their own return decisions. The comparison thus offers a most-similar case design where the cases differ on the outcome and an independent variable.Footnote 4 The analysis draws on scholarly sources, museum policies, government officials’ statements, policies, and newspaper articles. I also conducted 15 interviews with museum administrators, state representatives, and civil society organisations in Belgium and the UK in May 2024.Footnote 5
I show that implementation has been partial in both states. While various elements of the restitution norm are prevalent in domestic discourse as societal actors, particularly grassroots organisations and many museums, advocate for restitution, Belgium and the UK differ in their laws allowing returns. Belgium has passed a restitution law allowing returns of colonial collections once bilateral agreements with former colonies are concluded. No such national commitment has been made in the UK. In fact, the (now former, Conservative) British government has delayed legal changes to facilitate returns by national museums. Paradoxically, the picture is reversed regarding actual returns: while no large-scale returns have occurred (yet) in Belgium, British national museums have engaged in long-term loans, and non-national museums have returned items from their colonial collections.
The different domestic political structures (which actor has jurisdiction over museum collections) and government positions on restitution help to explain the implementation differences. In Belgium, museums with colonial collections are subject to (local, regional, national) government decisions, and the federal government handles bilateral restitution agreements with the Democratic Republic of the Congo (DRC), Rwanda, and Burundi. The federal government is the primary restitution actor for the national colonial collections, and since it has supported restitution (albeit decolonisation advocates would wish for more), a federal commitment to restitution has been enshrined in law. In the UK, museums are more in the driver’s seat: as arms-length bodies, most are not subject to the culture ministry’s decisions, but to their respective museum trustees. Non-national museums thus enjoy leeway to set their own rules and can thus return objects. Since the (previous) government has rejected returns by national museums, non-national museums have gone their own way by designing restitution policies and returning parts of their colonial collections.
The paper contributes to scholarship on heritage politics and norm implementation. First, I describe the components of the colonial heritage restitution norm and show that it can be usefully understood as an emerging international norm given its presence in international institutions and European states’ policy agendas despite lacking international legalisation. Second, the paper provides detailed case studies of norm implementation. While Belgium is applauded as a leader due to its restitution law, the analysis shows that the UK’s institutional structure has allowed non-national museums to return more colonial artefacts to source countries. Third, the paper highlights the hitherto-neglected role of museums as crucial implementation actors in International Relations.Footnote 6 Many museums have changed their practices, invested in provenance research, and are open to returning colonial objects. Fourth, I highlight the importance of institutional structure for restitution: depending on who is the main decision-maker on colonial collections, returns may proceed or not. Fifth, while often applauded as human rights leaders, the analysis also reveals that, on colonial history and restitution, European governments have long prevented restitution and instead fostered colonial amnesia.Footnote 7 The failure of the Belgian colonial truth commission and the UK’s stalled progress on legislation to allow returns show that European states have a long way to go to fully address their colonial pasts and engage with Global South states as equal partners.
The next section describes colonial heritage restitution as a new emerging norm, its content, and strength. I then present the findings of the Belgium and UK comparison, emphasising the role of museums, civil society, and the government as actors in the adaption of discourse, institutions, and law. I highlight common elements and crucial differences at the national level, specifically institutional context and government support for restitution. The last section summarises the findings and considers restitution’s future.
Colonial heritage restitution as an emerging norm and its implementation
The content of the colonial heritage restitution norm
Roughly since the 1990s, IR scholarship has seriously (re)analysed and discussed international norms, norm emergence, and their diffusion across states. A norm is ‘a standard of appropriate behavior for actors with a given identity’.Footnote 8 Rather than end products, norms are more usefully understood as ‘works in progress’, as contestation over their content, their status as norms, implementation, and compliance continue. This leaves many norms’ content vague, and this ‘enable[s] their content to be filled in many ways and thereby to be appropriated for a variety of different purposes’, benefiting their diffusion because they ‘may encompass different meanings, fit in with a variety of contexts, and be subject to framing by diverse actors’.Footnote 9 The vagueness is exacerbated for principle norms, such as the colonial heritage restitution norm. In contrast to treaty norms (binding legal obligations) and policy norms (rules within international organisations), principle norms are based on less formal principles and reflect ‘shared understanding that states either have not yet sought to codify or have chosen deliberately not to’.Footnote 10
The colonial heritage restitution shows this inherent vagueness. First, the term ‘restitution’ is defined differently in different states and world regions.Footnote 11 For some, including many academics and Western museum administrators, it is identical with returning heritage objects.Footnote 12 From an African perspective, restitution refers to a broader politics of justice and reconcilingFootnote 13: ‘Restitution is not only about the return of ancestral items, but it is about restoration or return of power, authority and voice.’Footnote 14 For instance, in the DRC ‘restitution’ refers to ‘a long process involving not only the reconstruction of history but also the reconstitution of knowledge, particularly among local Congolese communities’.Footnote 15
Second, no multilateral treaty obliges states to do this justice work or defines restitution actions. Thus, no model exists of what compliance with the restitution norm would entail. But the norm is embedded within a broader norm cluster of human rights and transitional justice norms,Footnote 16 i.e. mechanisms to overcome past legacies of human rights abuses. They have reached norm status since the mid-2000s.Footnote 17 Moreover, while some international documents such as the ICOM Code of Ethics of 2017, the UN Declaration on the Rights of Indigenous Peoples of 2007, and the 1998 Washington Principles on Nazi-Confiscated Art discuss cultural heritage restitution and the role of museums in fostering reconciliation, these are non-binding documents. And the binding agreements on stolen heritage mentioned below do not apply to colonial-looted heritage. Thus, if states are increasingly implementing the colonial heritage restitution norm, this would suggest that its normative pull is even stronger given the absence of an international legal obligation.
Given these challenges, I understand the ‘colonial heritage restitution norm’ to refer to the idea that cultural objects that have been looted or acquired under duress during the colonial era should be subject to a process of acknowledgement and remedying. The key elements of this norm are provenance research, changing museum practices, engagement with source communities and countries as equal partners, and return of cultural objects. These elements are repeatedly mentioned when various actors discuss restitution – in interviews, museum documents, related treaties establishing a principle of return of stolen artefacts,Footnote 18and secondary sources – albeit many use restitution to refer to returns only. The issue is also deeply political: different actors’ expectations of what ‘restitution’ truly entails range between maximalist and minimalist understandings of the norm’s very elements and each element’s content.Footnote 19
First, governments should be increasingly funding and museums should be engaging in provenance research. This seeks to establish an artefact’s origin and history and reconstruct its journey into the museum.Footnote 20 Provenance research can establish whether an item was looted and thus subject to potential return. It also allows for proper contextualisation of museum collections. Second, museums should change their practices under the umbrella of ‘decolonising the museum’. At a minimum, museums should more openly engage with the history and provenance of their collections. This could mean changing artefact labels to more accurately explain a piece’s provenance to openly acknowledge colonial plunder and violence. While defining ‘decolonising museums’ is itself contested,Footnote 21 it can be usefully understood as
a long-term process that seeks to recognise the integral role of empire in museums – from their creation to the present day. Decolonisation requires a reappraisal of our institutions and their history and an effort to address colonial structures and approaches to all areas of museum work.Footnote 22
Museums may face resistance for their efforts to openly discuss and correct past practices.Footnote 23 Changing artefact labels and returning objects to ‘decolonise’ is much more modest than what many formerly colonised and Indigenous peoples demand. This ‘ideal type’ would mean rather ‘to reflect on the conglomeration of ideas or knowledges that make up “Western civilisation” and those ideas and knowledges that precede and continue during and after colonialization’.Footnote 24
Third, museums and governments in the Global North should be more openly addressing their colonial pasts and approach their counterparts from former colonies or origin states as equal partners in negotiations. Requests for item returns should be taken seriously; and these states and communities’ expertise should be sought out and included in exhibitions.Footnote 25 Fourth, the most discussed norm element is return, i.e. the material act of giving back heritage items.Footnote 26 Whereas the norm’s first elements involve primarily cultural institutions and museums, asking them to engage more critically with their role and history of collecting, governments are often the main actor regarding returns. Given that the return of colonial-looted heritage is not found in a binding treaty, international law has not offered avenues for returns. Instead, governments have engaged in isolated voluntary returns or concluded bilateral agreements.
Norm strength: Concordance and institutionalisation at the international level
The colonial heritage restitution norm remains weak: while it has achieved high concordance, it lacks institutionalisation at the international level. Concordance refers to ‘the degree to which international actors refer to and accept a principled idea as appropriate’Footnote 27 and can be observed in how often international organisations (IOs) and non-governmental organisations (NGOs) refer to this idea and encourage compliance, and how they treat non-compliance with the norm.Footnote 28
Concordance
Responding to calls from newly independent states that had suffered from looting and plunder, IOs have drawn attention to colonial heritage restitution since the 1960s. These new states form part of a pro-restitution group along with UNESCO and related agencies as well as archaeologists that would ultimately advocate for the 1970 UNESCO Convention and for continued attention to restitution.Footnote 29 Governments in market countries, museums, art dealers, and auction houses stifled these demands, instead advocating for retention and safeguarding cultural heritage as treasures of all humankind. While restitution advocates were successful in creating and adopting several international conventions, these do not apply retroactively to colonial heritage and are narrow in scope. Nevertheless, international organisations dealt with restitution matters as demanded by African advocates such as Ekpo Eyo and Mobuto Sese Seko. Zairian leader Mobuto demanded the return of cultural heritage stolen through ‘colonial pillage’ in a speech at the UN in 1973. In December 1973, the UN General Assembly (UNGA) affirmed restitution as ‘just reparation for damage suffered’ from looting, recognised ‘the special obligations in this connexion of those countries which has access to such valuable objects only as a result of colonial or foreign occupation’, and stressed cultural heritage’s integral role for the people’s future development.Footnote 30 Similar resolutions were adopted over decades, attesting to the topic’s continued relevance.
UNESCO played an integral role in promoting colonial heritage restitution under the leadership of Amadou-Mahtar M’Bow in the 1970s. In 1978, M’Bow issued his plea that stolen heritage should be returned to the ‘men and women who have been deprived of their cultural heritage’, and he threw UNESCO’s weight behind the ‘legitimate claims’ for return.Footnote 31 He called on member states to conclude bilateral agreements to facilitate returns and to promote long-term loans, sales, and donations. In 1978, UNESCO also created the Intergovernmental Committee (IGC) for Promoting the Return of Cultural Property to Its Countries of Origin, to which states, after attempts at bilateral negotiations, can direct restitution complaints. In 2008, the organisation promoted the conference ‘The Return of Cultural Property to its Countries of Origin’,Footnote 32 and it has adopted awareness-raising and training campaigns for museum and heritage staff.
The International Council of Museums (ICOM) has been another important restitution advocate at the international level. Stanislas Adotevi of Benin discussed returns and criticised museums’ attachment to the European world at ICOM’s meeting in 1970, at a time when Global North states insisted on their museums’ role as guardians of world heritage and resisted returns.Footnote 33 Subsequently, two ICOM directors (Georges Henri Rivière and Hugues de Varine) advocated for rethinking museum practices. A 1979 ICOM report endorsed restitution and identified important considerations.Footnote 34 More recently, ICOM conferences include regular panels on restitution, ICOM created a restitution working group, and provenance groups can be found in different countries and museums. Many Global North museums have endorsed ICOM’s restitution recommendations or developed their own.
Institutionalisation
InstitutionalisationFootnote 35 means the idea’s codification in international law and can be assessed through international treaties, court decisions, and ‘international bodies that promote the adoption or implementation of a principled idea, monitor or evaluate compliance with it, and/or sanction violations’.Footnote 36 While the prohibition of plunder during war and the demand to return stolen heritage has been enshrined in international law,Footnote 37 the restitution of colonial collections has not been institutionalised, and not engaging in colonial heritage restitution does not result in sanctions.Footnote 38
That looted cultural heritage should be returned was first enshrined in the 1970 UNESCO Convention (Convention on the Means of Prohibiting and Preventing the Illicit Import, Export and Transfer of Ownership of Cultural Property). A preliminary draft of the convention included references to restitution of heritage removed before 1970 (which would include colonial looted heritage), but this focus was ultimately eliminated.Footnote 39 Most of the convention’s provisions guide states on monitoring and controlling heritage within their states, creating inventories, and advocating for cooperation between states.Footnote 40 Article 7 prohibits the ‘import of cultural property stolen from a museum or a religious or secular public monument or similar institution’ and requires the recovery and return such property when requested.Footnote 41 Moreover, Article 15 allows member states to conclude bilateral agreements to return cultural heritage ‘removed from its territory of origin before the convention entered into force’.Footnote 42 Former colonial powers and so-called market states like Switzerland and Japan long resisted this treaty,Footnote 43 and many only ratified in the 2000s. On the one hand, then, the UNESCO convention established the norm of returning stolen cultural property, but it was limited to items from museums or similar institutions, and it did not apply to colonial looted heritage.
Two more treaties dealt with the return of cultural heritage. In 1995, at UNESCO’s initiative, states adopted the UNIDROIT Convention on Stolen or Illegally Exported Cultural Objects, which requires the buyer of a cultural product to check its legitimacy and obliges the possessor of a stolen cultural object to return it. Noteworthy is Article 10(3), which states that the Convention ‘does not in any way legitimise any illegal transaction of whatever nature’ that occurred before the convention’s entry into force, nor does it limit a state’s right ‘to make a claim under remedies available outside the framework of this Convention for the restitution or return of a cultural object stolen or illegally exported before the entry into force’.Footnote 44 An additional step in strengthening the idea of return of stolen art and heritage occurred with the 1995 Washington Principles on Nazi-Confiscated Art, which led several European states to create commissions to identify original owners and return stolen artefacts. In addition to specific treaties, restitution can also be derived from the principle of cultural self-determination, which ‘affirms the right of communities and peoples to preserve, develop and manage their cultural heritage’.Footnote 45
While all three treaties do not apply to colonial looted artefacts, they do establish the norm that stolen heritage should be returned. Thus, while returning stolen heritage is an institutionalised norm, the return of looted colonial heritage is not institutionalised at the international level. Moreover, when states refuse to implement the colonial heritage restitution norm, they are not punished even though non-state actors may shame the state as discussed below.
In response to continued pressure from former colonised states but lacking international treaty obligations, European states and museums resorted to bilateral agreements, voluntary returns, and long-term loans. The earliest colonial restitution demands date back to 1872 Ethiopia and subsequent efforts toward the British and Italian governments in the 1920s and 1960s, resulting in some isolated returns.Footnote 46 Some states have engaged in bilateral negotiations, such as the 1975 agreement between the Netherlands and Indonesia which led to several returns.Footnote 47 Belgium returned one statue of King Bope Kena in 1976. In 2002, Nigeria and France reached an agreement over three sculptures which the French government had bought from a private dealer in 1998. Nigeria would retain ownership of the statutes but lend the sculptures to France for 25 years. Critics argue that the agreement ultimately legitimised the illicit trade because it had been illegally imported to France.Footnote 48 In 2008, following negotiations and court cases, Italy returned the Venus of Cyrene, stolen in 1915, to Libya. Some British museums such as the Horniman have voluntarily returned some colonial items from their collections.
Thus, isolated returns have occurred at the behest of newly independent countries. What is new now, I argue, and what has elevated colonial heritage restitution to a new norm, is that former European imperial states commit at the national government level to restitution through public statements, national policies, and bilateral agreements to return entire colonial collections in what I call the emerging colonial heritage restitution norm. In November 2017, French president Emmanuel Macron announced ‘temporary or permanent restitution’ as a priority because ‘African heritage can’t just be in European private collections and museums. African heritage must be highlighted in Paris, but also in Dakar, in Lagos, in Cotonou.’Footnote 49 Felwine Sarr and Bénédicte Savoy published their report on colonial looted heritage in French museums in 2018 (Sarr–Savoy report), recommending extensive investment in provenance research and large-scale returns. Other European countries soon followed: in 2019, Germany asked museums to provide collection inventories and promised to work with them on restitution. In 2021, the Dutch government endorsed recommendations to return colonial heritage. In 2022, Belgium passed a restitution law, allowing for returns to former Belgian colonies once bilateral agreements are concluded. Austria established an expert committee in 2022 to study its museums’ colonial collections. Thus, while newly independent states and international organisations such as UNESCO, ICOM, and the UNGA have long advocated for restitution, it is only now that European states have felt compelled to discuss and act on restitution at the national level. The fact that several ‘critical states’Footnote 50 of past European colonialism have committed publicly to the cause suggests that the norm has reached a tipping point. This norm emergence and cascade mirrors similar developments in settler colonial states where restitution and repatriation to Indigenous groups and the return of Nazi-looted art have become commonplace.Footnote 51
Norm implementation: Explanations and indicators
Scholars have developed various frameworks to better understand how norms are translated into domestic settings, ranging from norm salience,Footnote 52 localisation,Footnote 53 and policy convergenceFootnote 54 to implementation and compliance.Footnote 55 Implementation refers to ‘the steps necessary to introduce the new international norm’s precepts into formal legal and policy mechanisms within a state or organization in order to routinize compliance’.Footnote 56 Betts and Orchard theorise the process of norm implementation rather than as a domestic outcome in line with an international ideal type or treaty (as policy convergence and compliance imply). Understanding implementation as a process and considering the vagueness of principle norms also helps to explain why implementation across states can lead to significant norm contestationFootnote 57 and different domestic laws and institutions.Footnote 58 Much IR scholarship has shown that even when norms are anchored in treaties, ratification often results in partial state compliance or lack of domestic implementation into law.Footnote 59 Thus, implementation can take various forms and degrees that go beyond binary implementation and non-implementation or even follow a linear trajectory. While norm diffusion frameworks conceptualise differently how norms spread and become meaningful within states, they provide relevant insights into 1. explanations and actors and 2. Implementation indicators.
Implementation factors
Domestic governments and civil society groups may act as implementation advocates or resisters.Footnote 60 These actors are usually powerful veto players that may, given the domestic opportunity structure, (re)shape norms to ‘suit their national interests and cultural contexts’ during the implementation process.Footnote 61 For instance, civil society organisations play pivotal roles in translating global health and human rights norms.Footnote 62 Moreover, national governments’ support for norm implementation matters whereas their resistance might impede norm uptake. For instance, the government change between Presidents Mbeki and Zuma significantly contributed to South Africa’s improvements in health governance.Footnote 63 Taken together, when powerful veto players reject a norm and domestic resonance is low, norm uptake is more likely to face resistance, include modified laws or reinterpretation of global frames.Footnote 64 The analysis below attests to the importance of government support: the Belgian government has supported colonial heritage restitution, and so new legislation was passed to facilitate returns.Footnote 65 In contrast, the British governments under Boris Johnson and Rishi Sunak resisted critical engagement with colonialism and the de-accessioning of museum collections.
In addition to national governments and civil society groups, museums staff have been part of norm implementation. By displaying objects from around the world, museums inherently speak to the unequal power relations that have always characterised international relations:Footnote 66 Global North museums have always been ‘the keepers of other people’s cultures, imposing their own classifications and interpretations onto objects from different peoples around the world’.Footnote 67 Thus, while much IR scholarship has analysed NGOs, international organisations, and national governments, museums need to be considered as distinct actors and sites of norm diffusion. I show that both Belgian and British museums have started changing museum practices and invested in provenance research. And when given policy leeway, as non-national museums in the UK have been, several have devised their own restitution policies and have returned collection items.
Lastly, the domestic ‘political system’ acts as a filter for norm implementation. As a catch-all phrase, it refers to the domestic political system’s characteristics that might empower or restrict the influence of pro-implementation actors. Betts and Orchard point to these as ‘institutional factors’ such as the division of labour between different ministries or between federal and state levels. Radaelli finds that European states’ varying bureaucracy types and policy process characteristics influence policy convergence.Footnote 68 I expand on this line of research by showing that jurisdiction over museums can explain divergent norm implementation. Depending on who is the main decision-maker over colonial heritage collections influences the extent to which returns have occurred. Several non-national British museums that own and can make decisions about their colonial collections have indeed returned items, whereas national museums have been restricted by national legislation prohibiting collections’ break-up. In Belgium, the national government owns the most extensive colonial collections, and so any returns depend on its future bilateral agreements.
Indicators for implementation
A norm’s occurrence in domestic discourse, on policy agendas, and in legal changes helps indicate implementation progress. For Zimmerman, localisation types can be assessed based on discourse and legal changes, as well as policy or institutional changes (what she refers to as implementation).Footnote 69 Similarly, Cortell and Davis measure a norm’s strength based on the idea’s appearance in domestic discourse or on policy agendas, changes in national laws and procedures, and changes in state policies.Footnote 70 Applying these insights to European states’ approaches to colonial heritage restitution shows that restitution has been embedded into societal and governmental discourses and agendas, museums have changed their own policies and practices and engaged in provenance research, and some museums have returned some colonial items. However, both states differ on legal and policy changes based on their domestic political context regarding who governs museum collections and the national government’s support for restitution.
Colonial heritage restitution after decades of colonial amnesia
Heritage restitution in the national discourse and museum policies: Museums and civil society
As did other European states, Belgium and the UK long emphasised the supposed positive aspects of imperialism, such as infrastructure and education provision, while denying or downplaying atrocities and colonial legacies.Footnote 71 After Belgian colonies gained independence, there was ‘post-colonial unease’, with attempts to normalise relations with the newly independent states as equal partners all the while ignoring the colonial past.Footnote 72 The atrocities committed in the Congo, especially under Leopold II’s reign, are not widely taught in Belgian schools, and countless roads and transit stops remain named after the former king. Likewise, in the UK, the publication of several works on British colonialism in the 1990s and 2000s failed to ignite a robust public debate about colonial massacres.Footnote 73 Instead, silence about the colonial past and some nostalgia characterized British engagement. Between the 1960s and the 2016 Brexit referendum, ‘the legacies of empire haunted political debates about Britain’s role in the world, yet did not coalesce into any memory culture, or a sustained effort to engage with imperial history or its meaning’.Footnote 74
This lack of engagement with their empires’ inherent inhumanity also coined the countries’ approaches toward their museums that displayed colonial heritage. Many newly independent states in Africa demanded the return of looted artefacts and found support in international forums in the 1970s and early 1980s, but these efforts were shut down by Western states and museums in what Savoy calls a ‘postcolonial defeat’.Footnote 75
But norms are changing. Previously accepted arguments about imperialism’s supposed positives are less accepted, and European governments are facing renewed pressure: externally, African states are demanding their heritage back, and other European states, such as France and the Netherlands, are leading the way in public commitments to restitution (albeit lacking large-scale returns). Focusing events such as Macron’s 2017 restitution speech and the Black Lives Matter (BLM) movement since 2020 have propelled the decolonisation of public spaces onto the political agenda, with politicians having to address these calls.Footnote 76 Shortly after Macron’s speech, several African nations asked French museums to return looted artefacts, including Senegal, Ethiopia, Chad, Benin, Mali, Madagascar, and the Ivory Coast.Footnote 77 Western governments cannot defeat these demands anymore with retentionist arguments: ‘Thus, former colonies play at least as important a role in this change as their European counterparts.’Footnote 78 Domestically, museums and civil society groups seek greater acknowledgement of colonial crimes and decolonisation of museums: ‘Decolonial activists … have amplified requests for the restitution of human remains and patrimonial objects, allying themselves with groups and states in the Global South.’Footnote 79 Coupled with this is the increasing acceptance that ‘good’ Western democracies should be open and upfront about their past human rights abuses,Footnote 80 and they seek to raise their international status by returning museum items.Footnote 81
United Kingdom
The UK has slowly started to discuss the legacy of colonialism thanks to the efforts of domestic civil society groups. The country experienced a reckoning over public spaces after the Rhodes Must Fall campaign at the University of Cape Town in spring 2015 when students toppled statues of British imperialists Cecil Rhodes and Edward Colston. This helped to end some British museums’ complacency in confronting their own colonial holdings.Footnote 82 According to Hicks, curator at the Pitt Rivers Museum in Oxford, it ‘was certainly African thinking that brought about an institutional watershed moment for the Pitt Rivers Museum’.Footnote 83 Diaspora groups have also advocated for the decolonising of museums. The group AFROMET was very active in the early 2000s to put returns of Ethiopian artefacts onto the political agenda. In 2014, a tabot from the Church of St John in Edinburgh was returned to Ethiopia.Footnote 84 Many civil society groups working on anti-racism and reparative justice participated in the October 2023 conference of parliament on reparations such as Birthmark of Africa, The Ubele Initiative, Operation Black Vote, the Society of Black Lawyers, and Stand Up to Racism. In 2020, a report by the organisation AFFORD showed that around 80 per cent of respondents favoured the return of looted African heritage and human remains.Footnote 85
Despite lacking a national government commitment to colonial heritage restitution, several non-national museums have started to decolonise and have created their own restitution policies. A 2020 report found that ‘all museum staff interviewed so far stated that they support the decolonisation agenda’.Footnote 86 Non-national museums are less restricted by existing laws that prevent collection de-accessioning. They have returned some items and included representation from source communities. Projects such as ‘Rethinking Relationships and Building Trust around African Collections’ between the Horniman and the Pitt Rivers Museums sought to develop best practices surrounding their Kenyan and Nigerian collections and included heritage professionals, community members, researchers, artists, and other stakeholders in Kenya, Nigeria, and the UK.Footnote 87 The Horniman Museum consulted with Nigerian groups in the UK to craft its own Restitution and Repatriation Policy. The policy lays out the process in which communities may approach the museum about a return and the return and appeals process. In summer 2022, the Horniman returned the Benin Bronzes in its collection to Nigeria: ‘This was considered a landmark return, because the Horniman is directly funded by the government through the Department for Digital, Culture, Media and Sport (DCMS), unlike other institutions which had announced returns of cultural objects.’Footnote 88 Similarly, the Pitt Rivers and Ashmolean Museums, and Cambridge’s Museum of Archaeology and Anthropology, stated their plans to return Benin Bronzes.
National museums, in contrast, cannot legally return given current laws, but they have promoted long-term loans and have found ways to return some collection pieces. Tristram Hunt, director of the Victoria & Albert Museum (V&A), urged the government to update the National Heritage Act to ‘ensure that museum trustees have the responsibility for deciding whether disputed items should be returned to their places of origin’.Footnote 89 The V&A started discussions with Ethiopia about looted items from the battle of Maqdala in 1868, which would be sent on a long-term loan. While initially well received, subsequent media reports suggested that the loan was seen as insufficient.Footnote 90 Similarly, in May 2024, a partnership between the Manhyia Palace Museum in Ghana, the V&A, and the British Museum led to the exhibition ‘Homecoming’, which allowed Asante gold regalia, part of the V&A’s collection since 1874, to be displayed in Ghana.Footnote 91 Thus, even national museums have cautiously acted in line with the restitution norm.
Many of these actions conform with the Arts Council England’s (ACE) museum guidelines in its 2022 ‘Restitution and Repatriation: A Practical Guide for Museums in England’. The guidelines seek to help museums prepare for restitution claims, assess them, and work together with claimants and, if found legitimate, to resolve the claim. Museums are asked to consider four aspects when weighing restitution claims: (1) the object’s significance to the claimant; (2) ‘how the object was removed from its place of origin or from a past owner; (3) how the museum has engaged with the object; (4) and who is raising the claim’.Footnote 92 But the guidelines have also been criticised, for instance, for its lack of references to the broader context of colonialism, racism, and looting.Footnote 93
Belgium
Belgian museums are investing time and resources to study their collections’ provenance and are very receptive to returns. In June 2021, Restitution Belgium – an expert group of museum staff and academics – published ‘Ethical Principles for the Management and Restitution of Colonial Collections in Belgium’. The authors note a ‘moral duty to return the colonial heritage [that] is emerging, inviting us to go beyond the limitations of the existing legal framework in order to make an ethical responsibility heard in law’.Footnote 94 The report recommended much of what is currently unfolding in Belgium, including a federal policy facilitating critical investigation of the colonial past, funding and execution of provenance research, and research cooperation with communities and countries of origin. Restitution is presented as an integral part of reconciliation for the injustice of colonialism and the lasting damage to local communities.Footnote 95 Museums are increasingly exhibiting previously stored collection pieces. In 2024, the second floor of the Museum aan de Stroom (MAS) in Antwerp was dedicated to ‘Visible Storage’, making visible what the museum usually holds in storage, including Congolese artefacts.
However, museum renovations, new exhibitions, and government efforts do not go far enough for many activists. Civil society groups have been advocating that the government do more to address discrimination against Black Belgians, and they raised concerns about colonial heritage in museum exhibitions.Footnote 96 Diaspora groups are especially influential in the capital Brussels, where their larger population share prompts local politicians to take up their concerns. For Nyanchama Okemwa, chair of the European Network against Racism, Belgium’s professed progress on restitution amounts to ‘lip service’ because those in power are not prepared to have their privilege called into question and to acknowledge guilt and that the country’s well-being is ‘greatly anchored upon the extraction that they got and are still getting from their colonized countries’.Footnote 97 She criticises the state-to-state focus of bilateral agreements, which sidelines communities of origin. Geneviève Kaninda, Advocacy and Policy Offer at African Futures Lab, echoes these concerns. Even the widely applauded Belgian restitution law might be considered neocolonial given that the Belgians demand a restitution discussion when in fact neither Rwanda or Burundi have officially demanded the return of looted artefacts.Footnote 98 Thus, it is Western states that drive the discussions on their terms at the expense of source countries and civil society actors, threatening to perpetuate colonial power relations.
The central museum in Belgian restitution debates is the Royal Museum for Central Africa in Tervuren, now called AfricaMuseum.Footnote 99 The federal museum, opened in 1897 to display colonial heritage, holds around 85,000 objects from the Congo of which 58 per cent were ostensibly obtained legally, ‘and the remaining 40-plus percent of the collection required further research’.Footnote 100 After extensive renovations that involved the consultation of African diasporas through RMCA-African Associations Advisory Committee (COMRAF), the museum reopened in 2018 with a new approach to its colonial holdings, providing more contextualisation and seeking to present a multifaceted African past and present. Museum Director Bart Ouvry does not oppose returns: ‘I open cupboards with 40 of the same masks. What’s the problem in sharing?’Footnote 101 Today, the museum hosts a provenance exhibition to educate visitors about the collection’s origins and invests in provenance research through the Proche programme. But the museum’s decolonisation is far from perfect as the collaboration with the COMRAF during the planning and renovation process was ‘far from egalitarian’.Footnote 102
Table 1 presents the two states’ main characteristics. The previous sections have shown that restitution of colonial collections has become a topic of domestic discourse and that where possible, museums have adjusted their policies and practices and invested in provenance research. Demands from African states, civil society groups and museums as well as focusing events have created more societal awareness about imperial legacies.
Table 1. Norm elements and implementation in Belgium and the UK.
Based on the different jurisdictions over colonial collections and government support for restitution, as I explain in more detail below, Belgium and the UK differ on existing restitution legislation. This combination of factors has led to the paradoxical outcome that non-national museums in the UK have engaged in more returns than Belgian museums because they have more leeway to break up their collections despite the UK’s national government’s long opposition to restitution and the Belgian government’s support for it.
Legal changes: Institutional contexts and government support for restitution
Institutional context: Who owns the colonial collections and decides on returns?
Who owns museum collections and can make return decisions differs in both states and partly explains differences in implementation. In Belgium, national, regional, and local politicians oversee museum collections, whereas in the UK, museum trustees can make decisions. While both states’ federal governments oversee national museums, colonial looted artefacts in Belgium are more concentrated in the national museums, thus placing the government in the driver seat on restitution. In the UK, colonial heritage is more dispersed, and non-national museums own larger colonial collections. Thus, non-national museums in the UK have decided to return more colonial heritage despite the lack of national commitment.
In Belgium, the federal government oversees the collections of 11 national institutions. Given the country’s 2022 restitution law, returns from federal collections can be finalised once bilateral agreements with former colonies have been ratified. Federal museums then, such as the AfricaMuseum, are not responding to restitution demands themselves, but this is settled through official state channels. Other museums, like the MAS in Antwerp, are not federal collections and thus subject to one of the three federated communities (Flemish Region, Brussels Capital Region, Walloon Region) and local government’s decisions. While some isolated returns have occurred – such as the tooth of assassinated Congolese prime minister Patrice Lumumba and a kakuungu mask as a long-term loan in 2022 – large-scale returns depend on future agreements. The draft agreement between the DRC and Belgium is currently awaiting the DRC’s response. The AfricaMuseum presented a list of 84,000 artefacts in its collection to the DRC so that both states can work on provenance research.Footnote 103 Meanwhile, Rwanda and Belgium have collaborated on the restitution of colonial archives which Belgium has been digitising.Footnote 104
In the UK, national museums are subject to the British Museum Act (BMA), the National Heritage Act (NHA), and the Museums and Galleries Act 1992 (MGA). In response to the first restitution demands from newly independent African states, the 1963 BMA ensured that national collections could not be broken up or de-accessioned.Footnote 105 Museum trustees make decisions about collections, not the UK government itself, so they enjoy some flexibility and independence. British Museum trustees considered returning collection items as ex gratia payments based on a moral obligation, as suggested in the 2011 Charities Act, but the High Court clarified in 2005 that ‘the Trustees could not use the ex gratia principle to circumvent the restriction on disposals set out in the British Museum Act 1963’.Footnote 106 While national museums have long refused to restitute based on existing laws preventing deaccession and safeguarding world treasure, regional and university museums have returned some colonial heritage items. However, parliament has been discussing changes to the Charities Act as described below.
The federal government’s role in restitution
Museum and civil society actors in Belgium have had the federal government’s support, which has helped create political change on restitution. Younger Belgian politicians, like Zuhal Demir and Thomas Dermine, advocated for more critical engagement with the country’s colonial past, propelled forward by grassroots organisations, museum professionals, and academics. This was aided by changing political majorities in the early 2000s, when the Christian Democrats were forced into opposition.Footnote 107
Several events show a national commitment to openly discuss the country’s colonial legacy since the late 2010s, albeit sidelining accountability and reparations. In 2019, Belgian prime minister Charles Michel apologised for the kidnapping of ‘Métis’ children born to Belgian settlers and local women in Belgian colonies, and parliament officially recognised that the children were subject to ‘forced removals’. A lawsuit by five Métis women led a Belgian court to decide in 2024 that this amounted to crimes against humanity and to order compensation in ‘a landmark win for the reparations movement’.Footnote 108 Two restitution conferences in 2020 and 2021 included discussions with high government involvement.Footnote 109 In 2020, the Belgian parliament created a Special ‘Truth and Reconciliation’ Commission on Congo, Rwanda, and Burundi to better understand and address the legacies of colonialism (commonly referred to as the Congo Commission). The commission faced criticism for its initial composition without diaspora actors and organisations, its broad mandate, and short timeline.Footnote 110 The commission heard testimony from nearly 300 people (mainly experts in history, law, and socio-politics, but also diaspora representatives) and visited the DRC, Rwanda, and Burundi. In late December 2022, when it came time to vote on the commission’s recommendations, members from the liberal and Christian Democrats resisted a proposed apology, Remembrance Day, or the creation of memory institutions.Footnote 111 The commission’s mandate was not renewed and a second vote failed in late 2023. According to committee chair Wouter De Vriendt, ‘minds were not ready’.Footnote 112 Facing a general election in June 2024, disagreements persisted over an official apology for colonialism.Footnote 113
In June 2021, the Belgian government announced an agreement with the DRC to collaboratively study the provenance of colonial artefacts in Belgian museums and the possibility of returns. Thomas Dermine, state secretary for science policy, said ‘they don’t belong to us’.Footnote 114 In 2022, the parliament passed the ‘Bill Recognizing the Alienability of Goods Linked to the Belgian State’s Colonial Past and Determining a Legal Framework for Their Restitution and Return’ (restitution law). This legislation makes Belgium ‘the first country in the world to adopt legislation allowing for large-scale restitutions of colonial collections’.Footnote 115 The bill allows the ‘breaking up’ of federal collections in the 11 institutions over which the federal government has jurisdiction and promotes bilateral agreements to study collections’ provenance and allow eventual returns. The bill passed within two months without much controversy while the Congo Commission was ongoing. The public’s attention on the Commission’s broader topics may have contributed to the restitution law’s swift and uncontroversial passage.Footnote 116
While the restitution law represents a binding commitment, academics and decolonisation activists have criticised its limited scope and the marginal involvement of source communities.Footnote 117 It only applies to movable property in federal institutions and owned by the Belgian state, thus not including archives, human remains, artefacts owned by private individuals, or those owned by regions and communities.Footnote 118 Hence, the Congolese collection in the MAS, owned by the city of Antwerp, is not subject to any future restitution treaty. Moreover, the power to restitute ultimately rests with the Belgian government and not with countries of origin, leading some to argue that ‘the effort seems to be more about future relations with former colonized States than about justice for what happened in the past’.Footnote 119 Thus, rather than equal engagement with Global South states, Western European states remain in the driver’s seat and, throughout this long diplomatic process, Northern museums continue to exhibit and financially profit from plundered heritage.
The UK is also showing greater engagement with its colonial past. While some members of parliament have actively advocated for restitution and legislative change, the past administrations under Boris Johnson and Rishi Sunak resisted these efforts by delaying the entry into force of a revised Charities Act and by decrying decolonisation efforts as ‘woke’ politics. The All-Party Parliamentary Group for Afrikan Reparations (APPG-AR), chaired by MP Bell Ribeiro-Addy, has campaigned for a holistic view of reparative justice including addressing inequalities in education and health and ‘the theft of cultural artefacts’.Footnote 120 The existence of the APPG, composed of members from various parties, shows the presence of colonial restitution advocacy among British political elites.
The restitution debate has recently focused on revisions to the Charities Act of 2011.Footnote 121 Following recommendations from the Law Commission, parliament passed a revised Charities Act in 2022. The law allows museum trustees, including those of national institutions, (1) to make ex gratia transfers, meaning returns, from their collections upon obtaining the authorisation of the court, Attorney-General, or the Charity Commission, where there is a moral obligation to do so, and (2) to make ex gratia transfers ‘of low-valued property as an ex gratia transfer without needing the approval of the Charity Commission, Attorney General or the court’.Footnote 122 Most members of parliament expressed support for the moral duty to return but brought up complexities surrounding who should be returned to (e.g. national governments or local communities). However, the Johnson and Sunak governments delayed the legislation’s entry into force until it can ‘fully understand the implications for national museums and other charities’.Footnote 123 According to Herman, the government did not foresee the law’s implications for colonial collections at national museums.
Whereas a broad political consensus supported the Belgian restitution law, the issue has become entangled in a culture war over ‘wokeness’ in the UK.Footnote 124 The Johnson administration adopted a policy of ‘retain and explain’ on ‘contested’ heritage such as including public statues and museum collections. The argument was that these items should be contextualised and explained rather than removed or returned. A reversal of the Heritage Act 1983 was ruled out. In 2020, when BLM and calls for the decolonisation of public spaces were at their apex, culture secretary Oliver Dowden wrote to several national museums and rejected returns.Footnote 125 He also ‘complained about a “noisy woke brigade” rewriting British history [and] told national museums in 2020 that he was opposed to the removal of objects with “difficult and contentious” histories’.Footnote 126 This has made restitution a more controversial political topic in the UK.
Add to this the museums’ reliance on government funding. Dowden advised museums against becoming involved in ‘activism or politics’ and reminded them of their financial dependence on the government which ‘many curators interpreted … as a thinly-veiled threat’.Footnote 127 Moreover, museums are guided by the specific priorities of their leadership. Most of the British Museum’s 25 trustees are appointed by the prime minister and skew more conservative politically.Footnote 128 The museum has been almost stubbornly steadfast in its rejection of Greek restitution efforts for the Parthenon Marbles, ‘confirming its reputation in many parts of the world as the curmudgeon in the room whenever the topic of restitution arises’.Footnote 129
The long-standing British government’s resistance to restitution is also backed by a public that has expressed more support for empire than any of the other former European colonial powers: 32 per cent of respondents said the empire was something to be proud of and 33 per cent responded that colonised countries were better off overall for being colonised.Footnote 130 A broader reckoning with the past has thus been made difficult in a country where ‘colonial history is still a touchy subject’.Footnote 131
Both states built on experience with repatriation and restitution of human remains and Nazi-looted art. Belgium lags behind the UK on the repatriation of human remains. After the Belgian Bioethics panel stated in 2023 that there is no justification for keeping and displaying human remains taken from former colonies,Footnote 132 the Belgian government introduced a draft bill to allow for the repatriation of human remains still in Belgian institutions, but a vote has yet to be scheduled. The law would allow repatriation after a formal request from a foreign government for funerary purposes. The DRC government had not been consulted or informed of the draft law.Footnote 133 Regarding Nazi-looted art, Belgium created the Office of Economic Recovery in 1946 to deal with economic damage from the war, including restitution of cultural property.Footnote 134 While some original owners or beneficiaries recovered their art, many remained unaccounted for.Footnote 135 In 1997, the government created the Study Commission on Jewish Assets and the ‘Cell for the recovery of assets looted during the Second World war in Belgium’ to allow researchers and interested parties to consult archives on looted art. However, Belgium did not create a formal panel to consider restitution claims as done in the UK and other European states.
The UK’s reluctance to address colonial heritage restitution at the national level marks a contrast to its past position on repatriation and restitution of human remains and Nazi-looted art. The UK has been a leader on human remains since the early 2000s when it started returning Australian Aboriginal remains in UK collections. The government under Tony Blair and the DCMS set up a working group in 2001 under the leadership of Norman Palmer. This group then recommended guidelines on human remains in museum collections, which were later incorporated into the 2004 Human Tissue Act.Footnote 136 This law allowed nine national institutions, including the British Museum and the V&A, to return human remains if they are not older than 1,000 years. Since then, many UK museums, including the British Museum, have returned ancestral remains:Footnote 137 ‘Despite its reputation for being arch-conservative in matters of restitution, the UK is in fact far ahead of the European norm when it comes to human remains.’Footnote 138 In early 2025, a new report by the APPG-AR has demanded that also human remains older than 1,000 years should be returned.
The UK created the Spoliation Advisory Panel (SAP) in 2000 to deal with restitution requests for Nazi-looted art in British collections.Footnote 139 The SAP reviews restitution requests and recommends to the culture secretary whether to return an artefact or order monetary compensation.Footnote 140 In 2009, the government then passed the Holocaust (Return of Cultural Objects) Act, which ‘allowed the disposal of objects from national collections, but only in cases where the SAP had recommended the return and the Secretary of State had approved it’.Footnote 141 But even this approach to restitution of Nazi-looted art, on which there was consensus and moral clarity ‘that museums here should not be the beneficiaries of Nazi looting’, took around 10 years to enact.Footnote 142 One reason for this different approach could be that Nazi-looted art was about horrific acts perpetrated by Germans. Taking seriously the return of colonial heritage confronts Britain with its own imperial atrocities. It has a clearer destabilising potential on their national stories, whereas the return of Nazi-looted art more squarely fit into the national story of defeating Nazi Germany.
Conclusion
Colonial heritage restitution is an emerging norm. Former European imperial powers increasingly see it as the appropriate thing to do. European governments and museums change their practices, invest in and conduct provenance research, and ponder returns of colonial heritage. In contrast to the 1960s and 1970s when newly independent states first demanded their heritage back, European states can no longer deny, ignore, and delegitimise these demands.
But this emerging norm has been implemented only partially. While restitution has emerged as a topic of discourse in both the UK and Belgium, promoted by museums and civil society groups, national laws differ. Belgium passed a restitution law to facilitate returns, but large-scale returns have not (yet) occurred. In the UK, non-national museums are actively returning items while national museums, legally prevented from de-accessioning, engage in long-term loans. Thus, we see norm implementation despite government rejection of the norm.
The cases have also shown that power relations remain skewed in favour of Global North states and their museums. The norm element of equal engagement is largely lacking. First, British national museums’ focus on long-term loans means that Northern institutions retain property rights as long as governments resist restitution. The creation of a museum in Benin City where objects would be displayed permanently but only as loans
has been met with unsurprising dismay amongst much of the West African Press … There is therefore something uniquely humiliating in the relationship between the West and Africa, and that humiliation is played out again and again by tone-deaf museum professionals in their concern for objects.Footnote 143
Second, Northern states such as Belgium write restitution treaties without any or much input from the Global South governments and communities from which the cultural objects were stolen, and they ultimately would have to approve any returns. Thus, while all interviewees praised states’ recent moves toward restitution, even those interviewees for whom restitution would ideally mean much more, the sheer fact that so much colonial heritage remains in Northern museums necessitates more critical assessments: ‘For the present, this work is more words, advice and policy documents than deeds.’Footnote 144
The comparison has shown the importance of institutional context and government support for restitution. Government turnover can thus quickly turn the tides. The Labour Party’s 2024 election victory in the UK may change the country’s path as many prominent politicians who advocate for restitution are from Labour, such as Bell Ribeiro-Addy and Ed (Lord) Vaizey. New culture minister Lisa Nandy has already announced openness to restitution and promised a break from the past government’s ‘era of culture wars’.Footnote 145
Governments can learn lessons from past large-scale returns of Nazi-looted art. In line with the 1998 Washington Principles, five European states created their own restitution committees. In France and the UK, this involved changing existing legislation that had prevented de-accessioning of national collections. Governments supported restitution, then legislation was changed, extensive provenance research undertaken, and returns completed. Austria’s approach stands out for its ‘automatic, ex officio, and proactive’ restitution.Footnote 146 Austrian museums must undertake provenance research, and if items found their way into collections through theft or duress, the commission will find the owners or their heirs. Nothing but political will prevents governments from amending this approach to colonial heritage. Yes, provenance research is harder as multiple generations have passed, and items often belonged to entire communities. But it is conceivable that a proactive and automatic policy of return could be adopted that relieves source communities of acting as claimants and rather puts the onus on those who perpetrated the theft and have been profiting from it ever since.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0260210525101113.
Acknowledgements
I would like to thank the editors and anonymous reviewers of RIS for their invaluable advice and help in bringing this article into its present form. I would also like to thank the special issue editors Elif Kalaycioglu and Jelena Subotić for their suggestions and help along the way.
Funding statements
Fieldwork for this project was funded by the Research Enhancement Program, Texas State University.
Open access
