Consent Is Not a Substitute for Sovereignty: Tribal Data Governance and the Federal Genome

In 2004, members of the Havasupai Tribe learned that blood drawn for a diabetes study had been used for research they had never agreed to — ancestry studies, schizophrenia genetics, and other analyses conducted far from the reservation at the Grand Canyon's rim. The litigation that followed, settled in 2010, is usually taught as a consent case. It is better understood as a jurisdiction case. The question the Havasupai raised was not only whether a signature had been obtained, but who had the authority to authorize the use of a people's genetic material in the first place.

Two decades later, that question has moved from the courtroom into the machinery of federal research policy, and the answers being developed by tribal governments are among the most consequential contributions to American genomic governance now underway.

Two Grammars of Permission

Federal genomic data policy speaks, almost exclusively, the language of the individual. The Common Rule, the NIH Genomic Data Sharing Policy, and the standard data use agreement attached to controlled-access archives all presume a person who consents, a person who may withdraw, and a person whose interests are weighed against the public benefit of research.

Tribal frameworks introduce a second grammar. Under this view, a genome is not solely a private possession. It is simultaneously a family record, a community lineage, and — in the articulation of the CARE Principles for Indigenous Data Governance, published in 2019 — a resource subject to collective benefit, collective authority to control, responsibility, and ethics. The CARE Principles were drafted deliberately as a counterpart to FAIR data principles, which emphasize findability and accessibility. Where FAIR asks how data can be made more useful, CARE asks who gets to decide.

The practical expression of this second grammar is not a form. It is a government. Tribal research review boards, in several regions, function less as ethics committees than as regulators exercising a Nation's authority over what happens within its territory and to its citizens.

What Tribal Review Actually Looks Like

The Navajo Nation has maintained a moratorium on genetic research since 2002, and its Human Research Review Board requires that proposals be reviewed for community relevance, that results be returned, and that publications be reviewed before release. The Northwest Portland Area Indian Health Board operates an IRB serving more than forty tribes. The Alaska Area IRB has, for decades, reviewed research affecting Alaska Native communities with a structure that pairs institutional review with tribal approval.

The Native BioData Consortium, founded in 2018 and based in Eagle Butte, South Dakota, goes further: it is an Indigenous-led biological and data repository, positioning tribal nations as custodians of their own samples rather than as suppliers to distant institutions. Its founding premise is that storage, access, and benefit-sharing decisions should be made by the communities whose data are at stake.

These are not advisory bodies. When a tribal government declines to approve a study, the study does not proceed in that territory — a fact that federal program officers have had to learn, sometimes awkwardly, when grant timelines assume otherwise.

Where Federal Policy Meets the Friction

The tension surfaces most sharply in secondary use. An individual consent may authorize broad future research. But if a tribal nation has determined that certain categories of study are off limits, whose determination governs a sample already deposited in a federal archive?

Several structural features of the current system make this hard to resolve. Controlled-access archives are designed for portability and redistribution, not for ongoing governance by a third party. Data use agreements are contracts between institutions, and tribes are frequently not parties to them. Withdrawal provisions assume an individual actor and a single data custodian, not a government asserting a continuing interest across dozens of downstream analyses.

Consultation obligations add a further layer. Executive Order 13175, issued in 2000 and reaffirmed in subsequent administrations, directs agencies to consult with tribal governments before adopting policies with tribal implications. In practice, consultation often arrives late — after a funding opportunity has been drafted, after a consortium has been assembled — and is conducted with tribes treated as one interest group among many rather than as governments with regulatory authority of their own. The United Nations Declaration on the Rights of Indigenous Peoples, endorsed by the United States in 2010, recognizes in Article 31 the right of Indigenous peoples to maintain, control, protect, and develop their cultural heritage — language that explicitly reaches human and genetic resources.

Six Design Lessons for Agencies

For agencies now revising genomic data-sharing policies, the tribal experience suggests several concrete adjustments.

Recognize tribes as governments, not stakeholders. Consultation should be structured as government-to-government engagement, with standing mechanisms rather than ad hoc outreach tied to a single award.

Consult before protocol design. A tribe invited to comment on a finished study can only accept or refuse. A tribe engaged at the design stage can shape the question, the sampling, and the return of results.

Make withdrawal technically real. Federated archives need governance metadata that can carry a tribal restriction alongside an individual consent, and enforcement mechanisms that propagate it downstream.

Write benefit-sharing into agreements. Data use agreements that name tribal parties, specify permitted analyses, and include remedies are more durable than aspirational language in a consent form.

Fund Indigenous data infrastructure. Sovereignty requires capacity. Repository and IRB operations at tribal institutions remain chronically underfunded relative to the volume of research they are asked to govern.

Resist the template temptation. There are 574 federally recognized tribes, with divergent research policies, governance structures, and preferences. A single consultation protocol applied uniformly would replicate the error it is meant to correct.

The Limits of the Analogy

None of this resolves the underlying tension between collective authority and the open-science commitments that animate much federally funded genomics. Tribal restrictions can conflict with data-sharing mandates; researchers may find their work constrained; and the interests of individual tribal members who wish to participate in research are not automatically identical to those of their government.

These are genuine governance problems, not misunderstandings to be cleared up. But they are problems that individual-consent frameworks cannot even articulate, because those frameworks have no vocabulary for a collectivity that holds an interest in a genome.

What to Watch

The next round of federal data-sharing revisions will determine whether tribal authority is accommodated as an exception or incorporated as a structural feature. The difference matters. An exception can be waived by an official; a structure must be designed around. Tribal nations have spent two decades building the institutions and principles that make the second option possible. The question for federal agencies is whether they are prepared to treat those institutions as what they are — regulators, with standing.