The FAIR principles have a default direction of travel: toward openness. Make data Findable, Accessible, Interoperable, and Reusable, and the implicit aspiration is that data should flow as freely as good metadata allows. For a great deal of research that aspiration is exactly right. But for data about Indigenous peoples, lands, and knowledge, “open by default” can reproduce the very extraction that Indigenous communities have spent generations resisting. The CARE Principles exist to address this, and the key to understanding them is that they sit alongside FAIR, not against it. This article sets out how, drawing on the Indigenous data care domain.
FAIR is about data; CARE is about people
The cleanest way to grasp the relationship is this: FAIR is concerned with the technical properties of data, while CARE is concerned with the rights and interests of the people the data are about. FAIR asks whether data can be found and reused; CARE asks whether they should be, by whom, under what authority, and to whose benefit. The two are complementary axes, and the considered position of the Indigenous data sovereignty movement is not to reject FAIR but to insist that FAIR is necessary and insufficient on its own. A fuller treatment of the relationship is at FAIR vs CARE.
CARE stands for four principles: Collective benefit (data ecosystems should be designed so Indigenous peoples benefit), Authority to control (their rights to govern data about them must be recognised), Responsibility (those working with the data are accountable for supporting self-determination), and Ethics (the rights and wellbeing of Indigenous peoples are the primary concern). They were articulated by the Global Indigenous Data Alliance (GIDA) as a complement to FAIR, and are now widely cited in research-data policy.
The foundational concepts
CARE rests on a small set of concepts that need to be precise, because the whole framework turns on them.
- Indigenous data is data about Indigenous peoples, their lands, resources, or knowledge — a broad category that includes administrative data, environmental data on Indigenous territories, and records of Traditional Knowledge.
- Indigenous data sovereignty is the right of Indigenous peoples to govern the collection, ownership, and application of data created by, about, and for their communities. It is the principle from which the rest follows.
- Free, prior and informed consent (FPIC) is consent given freely, before activities begin, and with full information — the foundational consent standard for research engaging Indigenous communities, and distinct from the individual informed consent familiar from biomedical ethics in that it often operates at the level of the community.
A recurring feature is that consent and benefit are frequently collective rather than individual. A community may hold rights in knowledge or data as a collective, which is why collective consent and a benefit-sharing agreement are central rather than peripheral.
The TK Labels: governance made machine-readable
Principles need mechanisms, and the most influential mechanism for expressing Indigenous governance in digital systems is the Traditional Knowledge (TK) Label, developed by Local Contexts. A TK Label is a digital label attached to cultural material or data that communicates the community’s protocols for its use — who may use it, in what contexts, with what attributions and restrictions. The companion Biocultural (BC) Label serves the same function for biocultural materials, such as data tied to genetic resources and the lands they come from.
The significance of the TK and BC Labels is that they make community authority legible to systems. Where a Creative Commons licence expresses a rightsholder’s permissions in a form repositories and search engines understand, the TK Labels express a community’s protocols in a comparable, attachable form — turning “authority to control” from a principle into metadata that travels with the data. They do not assert copyright; they assert protocol, the kind of governance conventional licensing was never built to carry.
A TK Label does not lock data away. It does something more subtle: it tells anyone who encounters the data what the community of origin expects of them, moving the default from “open unless restricted” to “governed by those it belongs to”.
The international architecture
Indigenous data sovereignty is a global movement with a coordinated structure, which makes clear this is settled community infrastructure, not a fringe concern. GIDA is the global coordinating body. Beneath it sit established networks: Te Mana Raraunga, the Māori Data Sovereignty Network in Aotearoa New Zealand; Maiam nayri Wingara, the Aboriginal and Torres Strait Islander Data Sovereignty collective in Australia; and the US Indigenous Data Sovereignty Network. Connected to this, the Nagoya Protocol on access and benefit-sharing, under the Convention on Biological Diversity, provides a legal anchor where genetic resources and associated traditional knowledge are involved.
Why this belongs in a metadata standard
It would be easy to file CARE under “ethics” and leave it out of the structured-data conversation. That would be a mistake. The whole point of the TK Labels is that governance can and should be expressed as metadata that systems carry and honour. A repository that holds Indigenous data, a CRIS that records a project engaging a community, an access-control policy in a DMP — each needs vocabulary for collective consent, community-controlled access, and benefit-sharing if it is to represent CARE faithfully rather than flattening it into the open-by-default assumptions of FAIR-only infrastructure. CARE is, among other things, a metadata requirement — and it sits within a broader reorientation that includes Indigenous research methodologies and approaches such as two-eyed seeing, which bring Indigenous and Western knowledge systems into view together without subordinating one to the other.
Where shared vocabulary fits
The terms here carry deep significance and are easy to misuse — FPIC is not the same as individual consent, Indigenous data sovereignty is a right not a courtesy, and a TK Label is protocol rather than copyright. A shared, federated vocabulary that defines these precisely, pointing to GIDA for the CARE Principles and to Local Contexts for the TK and BC Labels, lets infrastructure honour Indigenous governance rather than override it. Supplying that definitional layer, and placing CARE alongside FAIR rather than beneath it, is the role the CASRAI dictionary is designed to play.
What to do now
For researchers engaging Indigenous communities: work to FPIC and collective-consent standards, agree benefit-sharing, and apply TK or BC Labels where the community directs. For repositories and systems: support the TK and BC Labels and the vocabulary of community-controlled access, not just open licences. For standards work: treat CARE as a first-class complement to FAIR, federating to GIDA and Local Contexts for the authoritative content.