Abstract
This article investigates why statistical reasoning has had little impact on the practice of friction ridge (or 'fingerprint') examination, despite both interest and some modest scientific progress toward this goal. Previous research has attributed this lack of results to practitioner resistance and legal apathy. This article seeks to complement those explanations through interviews with experts with a variety of perspectives on contemporary fingerprint practice about practical and mundane obstacles to the belated statistical revolution in fingerprinting. Based on these interviews, we argue that a 'forensic device' is required to incorporate statistical reasoning into fingerprint practice. This device would consist of a robust statistical model fronted by accessible, usable software. These components, in turn, require other components, such as large research data sets, markets, early adopters, government clients, education, and training. We conclude that the statistical revolution has been delayed not just by grand debates over the probabilistic nature of fingerprint evidence, but also by the seemingly mundane problems posed by developing and maintaining the kind of forensic device that would make such a revolution possible and practical.