Executive Summary

What this paper takes on is not the principles that Māori data sovereignty (MDSov) has already established. It is how those principles are supposed to operate inside the institutions that actually collect and store the data. It comes from Paul T. Brown and Te Taka Keegan at the University of Waikato, Kiri West and Hana Rapata at the University of Auckland, and Maree Sheehan at Te Wānanga o Aotearoa. Universities and crown research institutes have gathered data about Māori for a long time, the authors write, generally without much consideration of how that data should be handled, managed, and used, or how the benefits of the knowledge generated flow back to the communities.

The authors start from the position that research is not neutral. Western research methods and methodologies have often been extractive toward Indigenous communities, and while the datafication of the world has made research more efficient, those same systems continue to structurally exclude Indigenous peoples. Policies upholding Indigenous rights to the Māori research data that universities collect are scarce, and the paper presents itself as an attempt to fill that ideological and policy gap.

The sharpest diagnosis sits in the third section. Universities in New Zealand lack the internal mechanisms to track Māori research data within their data ecosystems, and therefore have no idea what data they have, or how much actually exists. The authors call this an information vacuum, a state in which neither the institution nor Māori can account for and govern that data.

The demand to count what you hold did not start with this paper. A 2023 communiqué from the Global Indigenous Data Alliance urged universities to recognise that all data concerning Indigenous peoples is Indigenous data, to identify and account for such data held within their systems or by research partners, and to implement policies that ensure Indigenous authority, access, and control over its present and future use. Building the inventory sits at the front of that list.

Findability here is not the ability to type something into a search box and get results back. The paper defines it as the ability to locate, identify, and understand the existence and nature of research data through appropriate descriptions, metadata, and discovery mechanisms. Māori research data are genealogically connected to people, their places, and their histories, and the authors hold that such data cannot be properly collected, or meaningfully governed, interpreted, and used, if their existence, origins, or conditions of use go unacknowledged.

The Māori Research Data Sovereignty the authors propose does not invent a new set of principles. It applies the existing Māori data sovereignty principles to the research data lifecycle, and what changes, the paper says plainly, is not the foundation but the domain of application. Three elements carry that application into practice: data findability, data governance, and community engagement.

The foundation treats Māori data as a taonga, a treasured possession that requires a level of mana, meaning respect, along with active protection. The principles document from Te Mana Raraunga sets out six principles and 16 sub-principles that spell out what follows from that, and this paper leaves those principles as they are while moving the domain of application to research data.

Governance is the mechanism that determines who holds the authority to make decisions about research data. The paper treats it as a system of collective authority, accountability, and stewardship rather than a matter of policy compliance, and it asks for structures that recognise Māori authority at appropriate scales such as iwi and hapū. That is where it parts company with generic data governance models, which tend to put legal ownership or administrative control first. Community engagement means reciprocal relationships between researchers, institutions, and Māori communities, and the paper describes it as an ongoing partnership that shapes research questions, data practices, interpretation, and outcomes rather than a one-off consultation.

On governance the paper is specific about the role it assigns universities. The sovereignty of Māori data is an exclusive right of Māori that institutions can support but never themselves claim to possess. The engagement side carries a demand pointing the same way. A 2023 report on the New Zealand research data landscape argued that universities need culture change, and that engagement with Māori has to be embedded in research data practice if the data are to be trusted, legitimate, and socially valuable.

The relationship among the three is a circle rather than a sequence. Findability enables governance by making Māori research data visible and traceable, governance gives findability meaning by defining how visibility translates into authority and responsibility, and engagement keeps both from floating free by grounding them in lived relationships and in tikanga, the culture, values, and protocols of Māori. Communities cannot exercise authority over research data they cannot locate or recognise, and that sentence appears in the paper itself.

Appendix B of the paper maps the three elements onto the Māori data sovereignty principles one by one. Findability answers most directly to whakapapa, the genealogical connection, because metadata records provenance, relationships, and context, keeping the data tied to people, places, and knowledge systems. The same findability also answers to rangatiratanga, because discovery mechanisms that signal Māori authority, governance expectations, and conditions of use let Māori decisions be recognised before the data is ever accessed.

One more principle hangs on the same row. Kaitiakitanga, the duty of care and stewardship, cannot be exercised over data that are invisible, poorly described, or disconnected from their whakapapa. Visibility is a precondition for that responsibility.

The common language of research data management is the FAIR principles, published in 2016: data that are Findable, Accessible, Interoperable, and Reusable. There is one point the paper leans on hard while introducing them. FAIR does not require the data to be open, but rather that conditions to its access are clearly defined. The principles exist to enable reuse under appropriate governance arrangements.

The reason FAIR was written sits in the same place. Research data lose their scientific and societal value because they cannot be reliably found, accessed, or linked once projects end. The original authors argued that storing and sharing data is not enough, and that unique identifiers, informative metadata, and standardised formats have to travel with it so that both humans and machines can understand and reuse it.

That distinction collapses often in practice. Read FAIR as a synonym for open and the goal shifts toward publishing more datasets, while the work of stating access conditions starts to look like an obstacle in the way of publishing. The CARE Principles, published in 2020 and cited in the paper, aim at exactly that spot. Collective Benefit, Authority to Control, Responsibility, and Ethics: where FAIR handles the technical aspects of reuse, CARE handles what the data means for the people it concerns.

The two are not in competition. The paper positions CARE as a complementary framework to FAIR, one that guides FAIR data infrastructures, metadata standards, and sharing policies to incorporate authority, benefit, and ethical stewardship. Findability is therefore not pointed at unrestricted openness. In the paper's own phrasing, it is pointed toward transparency on terms defined by the Māori individuals and collectives the data comes from.

A 2023 survey of the New Zealand research data landscape had already named this gap. Much of the data that researchers generate or use is not managed according to expectations, and data about, from, or connected to Māori frequently fails to meet the requirements of Māori data sovereignty frameworks or the CARE Principles. Among the 20 recommendations that report made to universities and the wider research sector is one to strengthen metadata standards.

The OECD identified findability and metadata in 2015 as prerequisites for research data to function as enduring scholarly and policy assets. Māori data sovereignty goes a step further. It requires that such metadata also signal Māori interests and responsibilities, not just technical descriptors. Recording what a dataset is and recording who it stands in relation to are two different jobs.

The paper borrows an older observation to make the point: data derive their value from the contextual information that allows others to understand how and why they were produced. For Māori data, that context includes tikanga, provenance, and collective authority.

The authors offer two pieces of prior work as evidence that this is more than a conceptual ask. One is a 2021 study showing how Indigenous data governance can be operationalised inside data systems through culturally informed metadata. The other, from the same year, describes a practical tool that gets researchers to attach labels and notices communicating Indigenous authority and expectations of use. Both make data visible while making the obligations attached to it visible at the same time.

From here on this is our observation rather than the paper's. The paper names no particular metadata standard. Still, picture the data catalogs that companies and public agencies run. The field that expresses ownership usually holds a person or a team address, and the field that expresses access usually holds a flag dividing public from private. Both record custody rather than relationship. Which collective the data came from, and under what conditions that collective allowed its use, tends to live outside the catalog in a contract or a review record.

A catalog running on the left column alone still finds data perfectly well. What it finds simply arrives without any statement of whose permission covers it and how far that permission reaches. This is why the paper casts findability as a technical function of discovery systems and a mechanism for upholding authority at the same time. When the right column is empty, access control moves downstream into case-by-case review, and those review records never make their way back into the catalog.

This paper is about New Zealand universities and Māori. It is not a study of public or health data catalogs in Korea, and this piece did not go through the metadata specifications of Korean portals field by field. So what carries over is a question rather than a conclusion. The three below can be held up against your own catalog right now.

• Is there a field for which collective the data came from? That is a different question from which department owns it.
• Can the terms of use take any value other than public and private? If a conditional use cannot be expressed as a value, the condition lives only on paper.
• Can the party that sets those terms sit outside the institution holding the data? If not, authority ends up belonging to the holder by default.

Rights over data in Korea have largely been designed at the level of the individual. The right to have your own information moved elsewhere is established in law, and consent is something each person grants. The paper works at a different level. It concerns the authority a specific community or region holds collectively over data that came from it, and that is a slot no amount of carefully collected individual consent fills automatically. Anyone who has worked with health data or with administrative data organised by region will recognise the distinction.

The Māori data sovereignty principles themselves have been refined over many years, so they are not news. What this paper adds is a route that brings those principles down into the daily operations of a research institution, and a clear statement that the first stop on that route is the inventory and its metadata. In the conclusion the authors write that this extends beyond symbolic policy commitments to require systemic change in how Māori research data are identified, classified, stored, accessed, and governed.

Three things are listed at the end as what that change will take. Digital research infrastructure capable of supporting culturally meaningful metadata, Māori-led governance processes, and long-term relational engagement strategies that let Māori communities, iwi, and hapū take part as decision-makers and research partners rather than as subjects of research.

Academic Papers

• 1.Brown, P. T., West, K., Sheehan, M., Rapata, H., & Keegan, T. T. (2026). "Data Findability, Governance, and Community Engagement for Māori Research Data Sovereignty." arXiv:2608.08905.
• 2.Carroll, S. R. et al. (2020). "The CARE Principles for Indigenous Data Governance." Data Science Journal, 19(43).
• 3.Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J. et al. (2016). "The FAIR Guiding Principles for scientific data management and stewardship." Scientific Data, 3, 160018.

Official Principles

• 4.Te Mana Raraunga. (2018). "Principles of Māori Data Sovereignty."

Korean Policy Portals

• 5.Ministry of the Interior and Safety. "Public Data Portal."
• 6.Ministry of Health and Welfare. "K-CURE Health Big Data Open System."