The Ancestors

On January 12, 2024, curators at some of the most prestigious museums in the United States began covering up the dead. At the American Museum of Natural History in New York, two entire halls were closed. At the Field Museum in Chicago and the Harvard Peabody in Cambridge, display cases were draped and shuttered, school field trips halted. The cause was not a discovery or a scandal. It was a deadline. New federal regulations had taken effect that morning — regulations implementing a law that had been on the books since 1990 — and for the first time in thirty-four years, the institutions could no longer display the human remains they had been showing the public for over a century without the consent of the descendants. The remarkable thing was not the scramble. The remarkable thing was that it had always been possible, and they had simply never been made to do it.

SECTION ONEThe Order

Every other installment of this series begins with a consent vacuum that no law was ever written to close. This one is the inversion, and the inversion is the point. The remains of Native American ancestors are the oldest bodies in this series — collected under a government order that predates the prison-autopsy contract, the foster system, the immigration agency, and the research university. And unlike every other population documented here, Congress eventually passed a law specifically to return them. The architecture absorbed that law and ran for another thirty-four years. Understanding how is the work of this installment.

It began with an order. In 1868, the United States Surgeon General issued Circular No. 2, directing Army medical officers stationed in what the government called "Indian country" to collect the skulls of Native Americans and ship them to the Army Medical Museum in Washington. The stated purpose was "to aid in the progress of anthropological science." The collection was to be large — George A. Otis, the Army surgeon who measured the skulls, emphasized quantity, because the racial science of the era required "sufficiently large series of adult crania of the principal Indian tribes" to furnish "accurate average measurements."

Read that sequence again, because it is the founding document of the harvest. The federal government, through its Surgeon General, mobilized its uniformed medical officers to collect the remains of a people the same government's army was at that moment in the field killing. The soldiers who fought the Indian wars and the surgeons who collected the skulls of the dead reported up the same chain of command. The remains were gathered, in the Army's own words, at "comparatively little cost to the museum," because the armed forces were already present, already producing the dead.

The intellectual engine driving the order was the work of Samuel Morton, the Philadelphia physician whose 1839 book Crania Americana classified human races by skull measurement and concluded that Native Americans were, in his ranking, inferior to Europeans. Morton's collaborators obtained crania however war and disease made available — one Army contact picked over the dead of the Seminole War; another forwarded the heads of Mexican soldiers killed at San Jacinto. Otis, applying Morton's method to the Army Medical Museum's growing collection, reached the conclusion the method was designed to reach: that "the American Indians must be assigned a lower position on the human scale than has been believed heretofore."

This is the architecture's deepest root. The Army Medical Museum collection — later transferred in large part to the Smithsonian — established the template that the universities and natural-history museums would inherit: Native remains as scientific specimens rather than as people with descendants, gathered without consent because the framing of "specimen" erased the existence of anyone who could consent. By the time anthropology departments were being built at Harvard, the American Museum of Natural History, the Field Museum, and the University of California — many of them seeded, directly, by the single collecting network of Harvard's Frederic Ward Putnam — the remains were already understood as institutional property. The question of return would not be seriously raised for another century.

SECTION TWOThe Scale

The numbers are knowable now only because of a federal reporting requirement that did not exist until 1990, and a journalistic effort — ProPublica and NBC News, building a public database from federal records — that compiled them into a picture the agencies themselves had never assembled. The figures that follow are drawn from that reporting and from federal NAGPRA data as of roughly 2023 through 2025; they are approximate and still moving, and a reader who wants the current count for any institution can look it up by name in ProPublica's Repatriation Database. The picture is this.

The holdings are concentrated in a roster of prestigious, well-resourced institutions. The University of California, Berkeley, through its Phoebe A. Hearst Museum of Anthropology, held about 9,000 Native American remains — more than any other university in the country. The Illinois State Museum reported the second-largest collection. The Ohio History Connection held more than 7,900 unrepatriated remains. Harvard University, through its Peabody Museum, reported more than 10,600 remains and, as of early 2025, still had not made roughly 5,400 of them available for return. Indiana University, the Field Museum, and the American Museum of Natural History each held the remains of many hundreds to thousands more.

And then the institution that holds more than any other, and answers to less. The Smithsonian Institution holds in storage the remains of roughly 10,000 people — more than any single museum in the country — but reports its repatriation under a separate and weaker statute, the National Museum of the American Indian Act of 1989, and does not publicly share what it has yet to return with the detail the 1990 law requires of everyone else. The largest holder of the dead operates under the lightest disclosure.

SECTION THREEThe Law That Was Supposed to Work

In 1990, Congress passed the Native American Graves Protection and Repatriation Act. Its requirements were not vague. Federally funded museums, universities, and agencies were to inventory the Native human remains in their holdings, determine cultural affiliation, notify the affiliated tribes, and repatriate the remains upon request. The mechanism existed. The mandate was clear. And the federal government did not leave the institutions to fund the work alone — over the years, roughly $60 million in federal funds was distributed to museums to carry out NAGPRA compliance.

This is what sets the Ancestors apart from every other population in this series. For the prisoners of Part I, the children of Part II, the unclaimed dead of Part III, the detained of Part IV, and the patients of Part V, the consent vacuum operated in the absence of any law requiring notice or return. Here, the law existed. The deadline was set. The money was appropriated. Every structural condition that the other parts identify as missing was, for the Ancestors, present.

And most of the bodies stayed on the shelves — not all, but the great majority, for a very long time.

SECTION FOURThe Word

The architecture closed the record on the Ancestors the same way it did on the unclaimed dead of Part III and the detained of Part IV: with a single administrative word, applied by the holding institution itself, subject to no independent review, that converted a mandate into permanent retention. In Part III the word was "unclaimed." In Part IV it was "abandoned." Here it was "culturally unidentifiable."

The mechanism was elegant. NAGPRA required repatriation of remains that could be culturally affiliated with a present-day tribe. So an institution that wished to retain its collection had only to determine — by its own assessment, applying its own evidentiary standard — that the remains could not be affiliated with any specific tribe. Labeled culturally unidentifiable, the remains could be held indefinitely while the institution remained, on paper, in compliance with the law. The determination required no tribe's agreement and no outside check. The institution that held the body decided whether the body could be claimed.

Here the honest account has to slow down, because the category was not pure invention. Affiliation is genuinely hard in some cases, and pretending otherwise would be its own kind of dishonesty. Many remains are fragmentary. Many were excavated generations ago with no usable record of where they came from. Some are ancient enough that the present-day tribal map cannot be confidently drawn back onto them, and some sit beneath overlapping claims produced by centuries of forced migration, displacement, and conquest — more than one nation, in good faith, identifying the same ancestors as its own. The "preponderance of the evidence" standard that governed affiliation existed for reasons that were not all cynical, and a determination that a particular set of fragmentary remains could not be affiliated was, in some real number of cases, a defensible scientific judgment rather than a dodge. Courts and good-faith curators sometimes found affiliation genuinely difficult, and any account that flattens every "culturally unidentifiable" determination into bad faith is not telling the truth.

But the difficulty is the alibi, not the explanation, and the data is what separates the two. The Ohio History Connection, holding thousands of remains, returned fewer than 20 ancestors to tribes in roughly the three decades after the law passed. The Illinois State Museum, as of 2022, had returned 2 percent of the 7,700 remains it had reported to the federal government — 156 individuals. A genuine evidentiary difficulty produces a scattered, case-by-case pattern of returns and retentions. It does not produce a 98-percent retention rate across a thirty-year span at the institutions holding the largest collections in the country. When the hard cases and the easy cases are retained at nearly the same rate, the difficulty has stopped being the reason and started being the cover. These were not obscure institutions struggling without resources. These were the largest holders in the country, recipients of federal compliance funds, and the aggregate rate at which they declined to return is the measure not of how hard affiliation is, but of how rarely they tried.

This is the finding that the Ancestors contribute to the architecture, and it is the most important one in the series for anyone who believes that passing a law is the same as solving a problem. The law was not the remedy. The law was a document the architecture learned to satisfy on paper while continuing to operate. The consent vacuum did not need the absence of a statute. It needed only an administrative word it could apply itself, and a penalty cheaper than compliance.

SECTION FIVEWhat Was Extracted While They Waited

It would be a mistake to imagine that the remains, during their century-plus on the shelves, simply sat. They were worked. The extraction did not stop when the collecting ended; it continued, in a quieter form, for as long as the institutions held the bodies.

The Army Medical Museum's crania produced the craniometric measurements that underwrote the racial science of the nineteenth century. When that science was discredited, the collections did not lose their institutional value — they were redeployed. The remains became the material basis for academic physical anthropology: the dissertations, the published studies, the careers, the departmental prestige of the universities whose anthropology programs had been built, in several cases, on the same collecting network. The Penn Museum's acquisition of the Morton cranial collection in the 1960s was described, in the institution's own telling, as opening "whole new vistas of research." The bodies were a research asset, and a research asset generates returns — in publications, in grants, in standing — for as long as it is retained and studied.

And here a real objection has to be met head-on, because it is the strongest argument the institutions have, and it is not nothing. The same collections that began as trophies of conquest became, over the twentieth and twenty-first centuries, the material basis for genuine knowledge — the kind that did not exist when the skulls were gathered and could not have been imagined when NAGPRA passed. Bioarchaeology and paleopathology read these remains for the history of disease, diet, violence, and labor in the Americas before European contact. Newer methods — ancient-DNA sequencing, stable-isotope analysis — have extracted from old bones data about migration, kinship, and population history that no nineteenth-century craniometer could have retrieved, and that in some cases bears directly on questions descendant communities themselves care about, including the very affiliation claims NAGPRA turns on. Repatriation, when it ends in reburial, often forecloses that future study permanently. The osteologists who object that returning and reburying a collection destroys an irreplaceable archive of human history are not inventing the loss. It is real, and a serious account of this subject has to say so.

But the objection, taken at its full strength, does not rescue the architecture — it sharpens the actual question, which was never whether the remains hold scientific value. They do. The question is who decides what is done with the body of a person whose descendants are identifiable and are asking for him back. The scientific-heritage argument answers a question no one in this dispute is really asking ("is there knowledge in these bones?") in order to avoid the one that matters ("whose consent is required to keep them?"). Living research subjects with valuable tissue cannot be retained for study without consent, however useful the science; the value of the knowledge does not transfer the authority to extract it. The defensible version of the institutions' position — that some ancient, genuinely unaffiliated remains might be preserved for study under protocols developed with tribes rather than imposed on them — is precisely what consultation and tribal consent are designed to negotiate, and it is available to any institution willing to ask. And it is not hypothetical: tribes and researchers have built exactly these arrangements, from consent-based ancient-DNA studies designed in partnership with descendant communities to limited-study agreements that permit documentation or analysis before reburial on terms the tribe sets. Repatriation and future knowledge are not always incompatible; what makes them compatible is that the descendant community, rather than the holding institution, holds the authority to weigh them. What the architecture did instead was skip the asking. It treated the scientific value of the collection as if it were itself the authority to keep it, and held the bodies of identifiable ancestors on the strength of an objection that, even when sincere, answers the wrong question.

This last dimension is harder to quantify than the $900 torso of Part III or the $2,200 hospital day of Part V, and it should be stated with that honesty: there is no invoice for the prestige a skeletal collection confers on a department, and no ledger entry for the genuine discoveries the collections enabled. But the structural point does not require an invoice. The remains held value — reputational, academic, and increasingly scientific — for as long as they were held, and that value was the standing incentive against return. The architecture did not require anyone to act in bad faith or to disbelieve in the science. It needed only that the institution holding the asset also be the one to judge the claim against it — and that no one had the authority to make it ask first.

SECTION SIXWhat Three Decades Couldn't Do, Eighteen Months Did

The thirty-four years before 2024 were not uniform, frozen inaction, and the honest version of this story says so. Thousands of ancestors were repatriated over the decades. The roughly $60 million in federal compliance grants did real work at institutions that chose to use it. A 2010 regulatory change had already begun to push at how "culturally unidentifiable" remains were handled, creating a pathway some institutions used. A baseline existed — which is precisely what makes the 2024 contrast meaningful rather than miraculous. The story is not that nothing moved for three decades and everything moved in one year. The story is that the pace before 2024 was glacial and largely left to the institutions' own discretion, and that when the rules changed, the same institutions that had moved by inches began moving by thousands.

On December 6, 2023, the Department of the Interior announced a final rule revising the NAGPRA regulations. It took effect on January 12, 2024. And the institutions that had spent decades explaining why repatriation was slow, complex, and difficult began moving with a speed that recast every prior explanation as, at least in part, a choice.

The revised rule did several things at once. It set a five-year deadline for institutions to consult with tribes and complete updated inventories. It eliminated the "culturally unidentifiable" category — the word the architecture had used to close the record — and shifted the burden onto institutions to repatriate rather than onto tribes to prove their claims. It required institutions to obtain free, prior, and informed consent from tribes before displaying, allowing access to, or conducting research on human remains and cultural items. And it provided that a single line of evidence, including tribal oral history, could be sufficient to establish the cultural affiliation that institutions had spent three decades declaring impossible to determine.

The response was immediate, and it is the proof of the thesis. Within days, the American Museum of Natural History closed two halls and covered display cases. The Field Museum shuttered exhibits. The Harvard Peabody began closing displays on January 12, the very day the rule took effect. These were not the actions of institutions that needed years to figure out how to comply. They were the actions of institutions that could move in forty-eight hours when the alternative was losing federal funding — and that had chosen, for thirty-four years, not to.

The repatriations followed the same pattern. In 2024, after the rule took effect, UC Berkeley — which had held more remains than any university in the country, and which the Muwekma Ohlone tribe said had "always shut the door" on its decades of requests — filed a notice committing to return 4,440 ancestral remains and nearly 25,000 items to Bay Area tribes. The Illinois State Museum, which had returned 2 percent of its holdings in the prior three decades, made 1,325 ancestors available in a single February 2024 Federal Register notice. The Ohio History Connection, which had returned fewer than 20 ancestors in three decades, made more than 150 available.

"Berkeley was very unique because they always shut the door on our people." — Charlene Nijmeh, Chairwoman, Muwekma Ohlone Tribe

Read those figures in sequence and the architecture's central explanation buckles. The work that was described for decades as too complex, too under-resourced, and too evidentiarily uncertain to complete was substantially advanced, at several of the largest institutions, within a single year — once the regulations made retention harder than return, and once the burden of proof was moved off the tribes and onto the holders. This is not to claim that every prior delay was bad faith, or that the difficulties named earlier in this installment were fictional. It is to say that whatever share of the delay was genuine difficulty and whatever share was institutional choice, the 2024 acceleration revealed that the choice component was large — larger than three decades of institutional explanation had implied. The institutions did not suddenly acquire a capacity they had lacked. Many of them acquired a reason.

The progress is real, and it is also incomplete, and the frictions ahead are real on every side. Most of the more than 100,000 remains have still not been returned. The five-year clock is running, not finished. Tribes face their own capacity limits in consulting on enormous collections; conflicting claims between nations still have to be adjudicated; the sheer logistical volume is daunting; and the lowered evidentiary bar that finally unlocked returns carries its own tensions about standards that thoughtful people on both sides are still working through. None of that is trivial, and the next five years will test whether the federal side — the National NAGPRA Program and its Review Committee — is resourced well enough to make the deadline mean anything. But the core lesson is already written, and it is the lesson a legislator most needs: the largest single variable separating the glacial decades from the rapid year was not money, not complexity, and not evidence. It was enforcement design. The bodies moved fastest when the rules were rewritten to make not moving them more expensive than moving them.

SECTION SEVENThe Architecture, Applied

The Ancestors complete the pattern this series has traced across six populations, and they extend it in one direction the others could not. The prisoners, the children, the unclaimed, the detained, and the patients all show the architecture operating where the law was silent. The Ancestors show it operating where the law spoke — clearly, with funding, with a deadline — and was simply absorbed. The consent vacuum is not merely the absence of a rule. It is a structure robust enough to digest a rule and keep running, until the rule is rewritten with teeth sharp enough to cost more than compliance.

SECTION EIGHTThe Legal Throughline

The constitutional throughline shifts here, and the shift is itself a finding. The first five installments leaned on the Fourteenth Amendment — the next-of-kin property interest in a decedent's body recognized in Newman v. Sathyavaglswaran, the due-process and liberty protections of O'Connor and Addington. The Ancestors rest on a different foundation: tribal sovereignty, the federal trust responsibility toward Native nations, and treaty rights — and, uniquely in this series, a federal statute that already existed and was, for most of its life, badly under-enforced.

Beneath the legal frame sits a genuine philosophical conflict that this installment should name plainly rather than pretend away, because it is the conflict on which thoughtful people actually divide. On one side is the claim of descendant communities to govern the disposition of their own ancestors — a claim grounded in cultural and religious sovereignty, in the specific trust obligations the United States owes to tribal nations, and in the plain moral intuition that the dead belong, first, to those who remember them. On the other side is the universalist claim, inherited from the Enlightenment, that certain collections constitute the scientific and historical heritage of all humanity — that a skeletal archive bearing on the migration, health, and history of the human species is not the property of any single community, present or past, and that destroying or sequestering such an archive impoverishes everyone, including future members of the very communities seeking return. This is not a contest between a real interest and a pretextual one. Both are real. An osteologist who argues that these collections enable medical training, comparative study, and questions not yet formulated — questions better methods will let the bones answer a generation from now — is making an argument in good faith, and the knowledge at stake is not imaginary.

This installment comes down on the side of descendant consent, and it should say why rather than assume it. The universalist claim is strongest as an argument about what may be studied and weakest as an argument about who decides. It establishes that the knowledge has value; it does not establish that the institution holding the bones is the party entitled to weigh that value against the wishes of identifiable descendants. In every other domain of research on human remains and human tissue, that weighing belongs to the person or the next of kin, and the importance of the science does not override the requirement of consent — it is precisely the importance of the science that makes consent matter. The universalist position, properly held, is therefore not an argument against repatriation but an argument for negotiation: for protocols, developed with tribes and adopted with their agreement, under which genuinely unaffiliated or ancient remains might be preserved and studied. Some tribes have entered exactly such arrangements; others have chosen reburial; the point of consent is that the choice is theirs to make.

The universalist can press one genuinely hard case, and it deserves a direct answer rather than a glide past it: the truly ancient remains — thousands of years old, predating any tribal polity that could be drawn on a present-day map — for which no living community can establish the lineal descent that grounds a consent claim at all. Here the "whose consent?" question does not resolve cleanly, because the honest answer in a small number of cases may be that no specific descendant exists. This is the universalist argument's real ground, and it should be conceded as such. But two things follow, and neither rescues the architecture. First, that hard case is a narrow exception, not the rule the institutions actually relied on; the collections at issue are overwhelmingly of remains recent enough that affiliation was possible and was simply not pursued, and a genuine doctrine for the unaffiliable ancient dead cannot be retrofitted onto a century of retaining the affiliable recent dead. Second, even for the genuinely ancient, the choice between preservation and reburial is a question of public and Indigenous policy — properly resolved through tribal consultation, federal trust obligation, and a transparent process — not a private entitlement of whichever university happened to acquire the bones. The hard case argues for a deliberate, consent-centered process to decide the residual category. It does not argue for the institution keeping the residual category by default, which is what it did.

What cannot be defended, in other words, is the arrangement NAGPRA was passed to end and the architecture preserved for another generation: the institution as the sole judge of both the science and the claim, weighing humanity's interest in the collection against the descendants' interest in their dead, and ruling, with remarkable consistency, for itself.

That difference carries the sharpest legislative lesson in the entire series. For every other population, the reform is to build a protection that does not yet exist. For the Ancestors, the protection existed for thirty-four years and accomplished comparatively little, because a right enforced by a $2,955 penalty is a right the holder can buy out of. The 1990 law was not wrong. It was under-enforced. And the 2024 regulations demonstrate, in real time, what the difference between an under-enforced right and an enforceable one looks like when measured in bodies returned: deadlines, the elimination of the self-applied loophole, consent requirements with funding consequences, and the burden of proof placed on the institution rather than the dispossessed.

The principle generalizes to every part of this series. A law that names a harm is necessary. A law that makes the harm more expensive than its remedy is what actually moves the bodies. The Ancestors are the proof, because they are the one population for whom we can watch both the under-enforced version and the enforceable version operate on the same institutions, a single regulation apart.

SECTION NINEWhat Can Be Done

The Ancestors offer legislators something rare: a reform whose effectiveness has already been demonstrated. The 2024 regulations are not a hypothesis. They are a working model of what enforcement design accomplishes. The remaining work is to extend, fund, and harden it.

CLOSINGWhat Remains

The Ancestors are the oldest bodies in this series and the proof of its hardest claim: that the architecture is strong enough to survive its own cure. Congress named the harm in 1990, funded the fix, set the deadline, and freed the dead on paper — and more than 100,000 ancestors were still on the shelves thirty-four years later. They began coming home in 2024 not because anyone's conscience changed, but because the rules were finally rewritten to make retention more expensive than return.

That is the lesson the other six parts can borrow, and it is the reason this installment matters beyond its own subject. In every part of this series, the architecture closes the record through whatever administrative word it is handed — "unclaimed," "abandoned," "next of kin unreachable," "culturally unidentifiable." The word is never the truth. It is the device. And the device holds until someone with the authority to do so forces the word open — not by passing a law that names the harm, but by building one that costs the institution more to ignore than to obey.

The bodies in this installment waited a century and a half. They are coming home now, slowly, because the price of keeping them finally exceeded the price of returning them. Every other population in this series is still waiting for the same arithmetic to be done on their behalf.

Part VII — The Market — will follow the money the whole series has been circling: the demand side, where bodies and parts are bought and sold, and where the question that has haunted every installment is finally answered — not how the bodies become available, but what they are worth, and to whom.