Fundamental Principles for Indigenous Data Sovereignty

From the AANZCA conference in Melbourne of the last few days I’ve moved on to the ACSPRI 2024 conference in Sydney for the rest of the week, which starts with a keynote by Maggie Walter, on methodologies for Indigenous statistics and quantitative research. Maggie is a Palawa woman from Tasmania. Data and population statistics have changed dramatically over the past decade or more; conventionally, Australian Indigenous people have been presented merely as average statistics that show what Maggie calls the Statistical Indigene: documenting prolonged disadvantage and inequality.

This is the case because these are the things we have data about: unemployment, imprisonment, health issues, etc. But these data are political: they are political artefacts that reflect a specific purpose, and position Indigenous people as hapless, helpless, and hopeless. This is a pejorative portrayal which is simplistic and undemanding of its audience; their presentation never advances beyond frequency tables and simplistic breakdowns (e.g. by gender or age). They define Indigenous people by the race they are not.

Maggie calls this ‘5D’ data: deficit, difference, disparity, disadvantage, and dysfunction. The aim may be to close the socioeconomic gap, but the aim is simply to bring Indigenous populations ‘up’ to a non-Indigenous level. And this pattern is not unique to Australia: the same is true for other (Anglo-)colonised nations, with many of the same deficits and dysfunctions identified – yet without ever acknowledging the underlying source of these patterns, which is Anglo-colonisation itself. This may be well-intentioned, but is nonetheless damaging.

The problems they identify are purposeful, therefore, and these data should be understood as ‘badd€r’ data instead: blameworthy, aggregate, deficit-based, decontextualised, and reductive. There are some moves towards recognising this, and the Closing the Gap reports in Australia have started to introduce some contextual information on the data they present, therefore. Yet still these data simply present a pulse-check on colonisation, and the problems they identify ultimately document that colonisation is working as intended: systematically disadvantaging Indigenous people.

Indigenous data needs instead require lifeworld, disaggregated, contextualised, Indigenous-priority, and available amenable data: Indigenous people need to own their data, and the data need to break down to the level of individual traditional owners, representing individual Indigenous communities. This is a question of Indigenous data sovereignty – and Indigenous data include data on Indigenous resources and environments; on Indigenous peoples; and from Indigenous peoples. Such data must then be governed in their collection, management, access, interpretation, dissemination, and reuse by Indigenous people; only this also enables Indigenous people to manage and enhance their own collective wellbeing.

This Indigenous data governance approach works in two directions, connecting governance of data with data for governance. It refutes the 5D data model, tells Indigenous stories in their own way, applies Indigenous data protocols, informs Indigenous programmes and policy delivery, develops Indigenous data infrastructures, and designs and deploys Indigenous data processes. In all of this, big and open data does not necessarily equal better data, unless these fundamental principles are respected: ‘big data’ can also simply further exacerbate the problems with 5D data, alienate Indigenous people from their own data, embed stigma, and thus marginalise Indigenous people. The well-known CARE principles for data governance are not enough here – this requires full-featured Indigenous data governance.

The simple question to ask here is whether statistical work done with Indigenous data would look the same if an Indigenous person were in charge of it. This also goes to the difference between method and methodology: the latter is about the central assumptions, values, and understandings of reality which underlie the conceptualisation and operationalisation of the methods being applied to the data. This includes our epistemological, axiological, ontological, and social-cultural positioning, all of which affect out chosen theoretical frameworks and thus our analyses of data.

From an Indigenous perspective, this lifeworld incorporates intersubjectivity within peoplehood as Indigenous Peoples, as well as intersubjectivity as colonised, dispossessed, marginalised peoples whose Peoplehood is denied. This ongoing conflict fundamentally affects all everyday experience, and methodology needs to make this visible. The conceptualisation of methodology needs to address Indigenous lifeworlds as well as epistemological (who are the knowers?), ontological (what is the problem?), and axiological (what are our values in relation to the issue?) positions.

Applying this to a practical example, Maggie now outlines a research project with the Larrakia Aboriginal Corporation in Darwin, which asked not simply about what Indigenous people did, but what they thought – showing for instance their perspectives on and experience with race relations (bad, and getting worse), and correlating this with personal attributes (showing that Indigenous people’s own personal circumstances did not strongly affect how they were treated by white people). This shows that interpersonal racism is an unavoidable part of life experience for Indigenous people in Darwin, and that nothing they could do would change this. The problem is colonial society.