The Invisible 90 Percent
Walk through any major art or natural history museum and you are seeing, at best, a fraction of what the institution actually owns. The Smithsonian Institution — the world’s largest museum complex, comprising 21 museums, 21 libraries, the National Zoo, and a network of research centers — holds more than 157 million objects, specimens, and artifacts in its care. On any given day, less than one percent of those items are on public display. The rest exist in climate-controlled storage rooms, in rolling metal cabinets stacked floor to ceiling, in acid-free tissue paper and archival boxes that most visitors will never see.
This invisible majority raises a question that museum professionals have grappled with for decades, and that the broader public is now demanding answers to with increasing urgency: What gives an institution the right to keep everything it has accumulated? And when something should go — whether sold, donated, or returned to communities from which it was taken — who gets to decide that, and how?
The answers involve arcane professional ethics codes, federal legislation, colonial history, financial pressure, and a fundamental reckoning with what museums are actually for. They are also, increasingly, political.
What Deaccessioning Actually Means — and Why It’s So Controversial
Deaccessioning is the formal process by which a museum removes an object from its permanent collection. It sounds administrative. It is anything but.
The term covers a range of outcomes: an object might be sold at auction, transferred to another institution, donated to an educational facility, destroyed if it has deteriorated beyond utility, or — in the most fraught cases — returned to a community or nation that argues it was wrongfully taken in the first place. Each pathway carries its own ethical weight and professional consequences.
The American Alliance of Museums (AAM) and the Association of Art Museum Directors (AAMD) have long maintained strict ethical guidelines around deaccessioning, rooted in a core principle: proceeds from the sale of collection objects should only be used to acquire new objects or, in extraordinary circumstances, to care for existing ones. The moment an institution sells a Rembrandt to pay the electric bill, the argument goes, it has broken a fundamental covenant with the public — the promise that these objects are held in perpetual trust, not as liquid assets.
That covenant was tested severely during the COVID-19 pandemic. In 2020, the AAMD temporarily relaxed its guidelines, allowing museums to use deaccession proceeds for direct care of collections — a broader definition that some institutions interpreted as permission to monetize their holdings for operational survival. The Brooklyn Museum, the Baltimore Museum of Art, and others announced high-profile sales that drew criticism from peers and applause from fiscal pragmatists.
The Baltimore Museum of Art planned in 2020 to sell three works — by Brice Marden, Clyfford Still, and Andy Warhol — for an estimated $65 million, with proceeds earmarked for funding staff salaries, equity initiatives, and collection care. Critics, including some prominent curators and scholars, argued the sales violated sacred professional norms, and the outcry forced the museum to call the sale off. Supporters countered that those norms had always served the interests of wealthy donors and institutional prestige more than the public good. The debate remains unresolved.
The Science of Deciding: Collections Management Policies
Before an object can be deaccessioned, it must first be formally evaluated — a process governed by what museums call a Collections Management Policy, or CMP. Every accredited institution is required to maintain one. These documents, sometimes hundreds of pages long, define what a museum collects, how it cares for collections, and under what circumstances it may remove objects.
The decision to deaccession typically begins with a curatorial review. Staff assess whether an object fits the institution’s current mission (which may have evolved since the object was acquired), whether it is in good enough condition to be useful, whether it duplicates other holdings, and whether another institution might steward it more appropriately.
Provenance research is a critical and increasingly resource-intensive component of this process. Before any object can be sold or transferred, institutions must investigate its ownership history — particularly to determine whether it may have been looted, stolen, or acquired under conditions of duress. The 1970 UNESCO Convention on the Means of Prohibiting the Illicit Import, Export and Transfer of Ownership of Cultural Property established that date as a benchmark: objects documented in institutional collections before 1970 are generally considered to have cleaner title, while anything acquired after requires more rigorous scrutiny.
That standard, however, is widely criticized as both arbitrary and inadequate. Many objects that entered collections before 1970 were nonetheless acquired through colonial extraction, grave robbing, or outright theft. The 1970 date was a political compromise, not an ethical one.
The Metropolitan Museum of Art, the Getty, and other major institutions have established full-time provenance research departments — a development driven partly by ethical commitment and partly by the spectacular embarrassments of the 2000s, when several American museums were forced to return looted antiquities to Italy and Greece amid criminal investigations and international diplomatic pressure.
NAGPRA, the Smithsonian, and the Long Road to Repatriation
Perhaps nowhere is the complexity of collections decisions more viscerally apparent than in the domain of Native American human remains and sacred objects.
The Native American Graves Protection and Repatriation Act, passed in 1990, requires federally funded institutions to inventory their holdings of Native American human remains, funerary objects, sacred items, and objects of cultural patrimony, and to repatriate them to lineal descendants or culturally affiliated tribes upon request. It was a landmark piece of legislation — the first time the United States government formally acknowledged that museums had, in many cases, built their collections through the desecration of Indigenous graves and the theft of sacred materials.
Thirty-plus years later, the results are mixed at best. According to a ProPublica analysis of federal data, as of early 2025 institutions had made roughly 58 percent of the approximately 210,000 Native American remains they reported available for return — leaving more than 90,000 still held. ProPublica’s investigation, launched in 2023, found that a handful of institutions, including the University of California, Berkeley, still held thousands of Native American remains that had not been repatriated despite NAGPRA’s requirements, and that many institutions had made little meaningful progress in decades.
The Smithsonian operates under a separate but related law — the National Museum of the American Indian Act of 1989 — which established the National Museum of the American Indian and created its own repatriation framework. The Smithsonian’s National Museum of Natural History maintains a dedicated repatriation office and has returned tens of thousands of objects and remains to Native communities. In 2022, the Smithsonian adopted a sweeping new ethical returns policy allowing its museums to return objects collected under circumstances considered unethical by contemporary standards — a more expansive stance than the law technically requires. In 2024, a Smithsonian task force went further, recommending that the institution not collect, display, or use human remains for research without the consent of the deceased or their descendants.
The Smithsonian’s commitment is not without complications. Some tribes lack the infrastructure to receive, store, and care for large volumes of returned materials. Some objects have become so physically integrated into museum storage systems that identifying and separating them is genuinely difficult. And some scholars — a diminishing but vocal minority — continue to argue that keeping certain remains available for scientific study serves a legitimate public interest that should factor into repatriation decisions.
That argument carries increasingly little weight with tribal nations, many of which point out that their ancestors’ bones were collected not through neutral scientific inquiry but through the systematic exploitation of communities with no power to refuse.
The International Dimension: When Nations Want Their Heritage Back
The repatriation debate extends far beyond Native American contexts into a roiling international arena where governments, museums, and advocacy organizations clash over some of the most iconic artifacts in the world.
The Elgin Marbles — the sculptural program removed from the Parthenon in Athens by British diplomat Lord Elgin in the early nineteenth century and now held by the British Museum — remain the most famous and politically charged example of a contested collection. Greece has demanded their return for decades. The British Museum has consistently refused, arguing that it serves as a “universal museum” for all of humanity and that the marbles are better preserved and more accessible in London than they would be in Athens. Critics note that this argument — that a wealthy northern institution knows better how to care for another culture’s heritage — encapsulates exactly the colonial logic that the repatriation movement seeks to dismantle.
Recent years have seen significant movement on other fronts. In 2022, the German government formally transferred ownership of Benin Bronzes — extraordinary cast-metal works created by the Edo Kingdom and looted by British forces in 1897 — to Nigeria. Several major German museums then handed over physical custody of their holdings. The Smithsonian followed, transferring 29 Benin objects to Nigeria in the same year, making it among the first American institutions to do so.
France has been particularly active under President Emmanuel Macron, who in 2017 pledged to return African cultural heritage held in French collections — a commitment that has yielded high-profile restitutions to Benin (formerly Dahomey) and Senegal, though critics argue the pace and scale remain insufficient given the scope of French colonial extraction.
These international repatriations operate differently from NAGPRA cases — there is no binding legal framework compelling American museums to return objects to foreign nations, so decisions rest on institutional ethics, political pressure, donor relations, and board discretion. This gives individual institutions enormous latitude, which produces wide variation in outcomes.
The Financial Reality No One Wants to Discuss
Behind every collection management decision runs an uncomfortable financial current. Museums in the United States are largely nonprofit institutions dependent on a combination of ticket revenue, government support, and private philanthropy — a funding model that has been under strain for years and was severely disrupted by the pandemic.
Maintaining a large collection costs money — significant money. Climate control, pest management, conservation, cataloging, provenance research, and insurance all accumulate into budgets that smaller and mid-sized institutions struggle to sustain. Many regional museums hold tens of thousands of objects with no realistic prospect of ever displaying or studying them, acquired through decades of well-meaning but strategically undisciplined collecting.
This reality has prompted a growing conversation about “right-sizing” collections — a term that sounds corporate but reflects genuine professional reflection about whether hoarding objects serves any legitimate purpose. An institution that holds 200,000 natural history specimens it lacks the staff to study or the space to display is arguably not serving the public interest.
The financial temptation created by valuable deaccessions is real and must be acknowledged. A museum sitting on a Monet worth $80 million while its heating system fails and its education staff gets cut faces a genuinely difficult ethical calculation. The AAM and AAMD guidelines are meant to prevent institutions from taking the easy way out — liquidating cultural assets for operational comfort — but critics argue those guidelines were designed in an era of museum abundance that no longer exists for most institutions.
Most museums in America, after all, are not large encyclopedic institutions with substantial endowments but smaller regional ones with chronic funding problems and collections that haven’t been strategically managed in decades.
Where the Calculus Is Heading
The pressures converging on collections management — financial strain, Indigenous rights movements, international repatriation claims, evolving professional ethics, and a public increasingly skeptical of institutional authority — suggest that the field is at a genuine inflection point.
Several trends appear likely to shape the next decade. Digital technology is transforming what “access” means: high-resolution 3D scanning and online databases can make an object globally accessible without requiring physical custody, potentially weakening one of the traditional arguments against repatriation. The Smithsonian has invested heavily in digitizing its collections, making millions of records freely available online — a model others are following.
Collaborative stewardship models are also gaining traction, in which a museum and a source community share custody, decision-making authority, and interpretive control over objects. Some museums and tribal nations have developed such partnerships around specific holdings — arrangements that recognize Indigenous authority without requiring the complete dismantling of institutional collections.
The legal landscape has already shifted. Revised NAGPRA regulations that took effect in January 2024 eliminated the “culturally unidentifiable” category that had allowed institutions to delay action on thousands of remains, and require museums to defer to the traditional knowledge of tribes and lineal descendants rather than demanding scientific proof of cultural affiliation.
What seems clear is that the default posture of acquisition and retention — the assumption that collecting is inherently good and returning is inherently a loss — is losing its grip on the profession. A new generation of museum professionals, many of them from communities historically excluded from or harmed by these institutions, is asking harder questions about legitimacy, consent, and purpose.
The museum of the future, in this view, isn’t the one with the most stuff. It’s the one that can honestly account for how it got what it has, and make principled decisions about what it’s right to hold and what it’s right to let go.
That accounting — rigorous, uncomfortable, and long overdue — is finally, haltingly, underway.