Abstract
This article explores the social imaginary in the context of data ownership and the (non-)delivery of the data sharing revolution in biomedicine. We contribute to this special issue on imaginaries by developing a method and paradigm of 'competing narratives'. Despite multiple initiatives to encourage health data sharing, and a strong 'open access' agenda, the data sharing revolution is not yet delivered. Ownership is persistently (though inconsistently) presented as a barrier to data sharing. However, existing literature does not reveal how far appeals to ownership are part of the problem. This paper reports original, interdisciplinary research asking: in health research, in what ways, if at all, do notions of ownership (broadly conceived) of health-related data impact on sharing practices? Doctrinal and empirical research methods are used to expose evidence of drivers behind appeals to ownership in health data sharing. The findings speak to how funders and data custodians can better tailor existing and potential data sharing initiatives to perspectives and behaviors. The concept of 'my data' is important: notions of reward, opportunity, control, and safeguarding establish legitimate, potentially competing 'ownership' interests in data. In particular, this research raises questions about the long-term effectiveness of an open access ideology that ignores these subtleties. In conclusion we find power in the social imaginary of ownership with respect to biomedical data; however, that power emerges and is enacted in unexpected ways by multiple actors within the ecosystem, often driven by competing narratives about what is at stake. Importantly, formal legal property-type appeals to ownership appear to have far less power in the narratives about data than the ethical and social concerns that underpin responsible biomedical research.