Resisting Detached Datafication: What Toxic Prisons Teach Us about the Imperative of Restorative/Transformative Data Science for Environmental and Social Justice

Résumé analytique

In this essay, geospatial data scientist Ufuoma Ovienmhada draws on three years of partnered research with the grassroots Campaign to Fight Toxic Prisons (CFTP) to argue against what she terms "detached datafication" — the use of data applied to an issue without analysis of power and without the context of people with lived experience in the collection, methodology, and presentation of results. Prisons and other carceral facilities are exposed to environmental hazards including extreme heat, air pollution, poor water quality, proximity to hazardous waste, and exposure to wildfires and hurricanes, and a growing body of geospatial research treats this as a data gap to be filled.

The problem, she contends, is that such detached, data-centric, state-centric, technological framings yield "solutions" — mandating air conditioning, respite areas, evacuation policies, or siting new prisons in less hazardous locations — that are detached from lived experience and from analysis of the power dynamics that determine who a solution reaches. Drawing on interviews with formerly incarcerated activists (Sloane, Jack, and Justice), she shows how these technical fixes address structural problems inadequately and can even create new forms of toxic prison; one cited study finds evidence of environmental injustice in 40% of the prisons it examined.

As a hopeful path forward, Ovienmhada applies Catherine D'Ignazio's four-stage framework for restorative/transformative (RT) data science — resolving, researching, recording, and refusing/using data — to the domain of toxic prisons. Rather than optimizing for standardization and generalizability, RT data science analyzes power, deconstructs binaries such as "toxic/non-toxic," elevates survivor testimony alongside quantitative data, and connects data use to transformative goals such as decarceration and prison closure. The essay urges researchers, technologists, and decision-makers across environmental and social justice struggles to employ data with greater awareness and transparency about its limits and the limits of solutions that do not shift power imbalances.