Matilde Carerra Dissertation defense

Matilde Carerra will be defending her dissertation, “Turning Data into Evidence in Human Origins Research.” All members of the philosophy department are invited to attend. Abstract: This dissertation sits at the intersection of the philosophy of historical sciences and the philosophy of data, measurement, and evidence. Focusing on the challenges of turning genomic data—both modern and ancient—into evidence about human population histories, each chapter explores philosophical topics in paleogenomics and demographic inferences related to human origins. The introduction notes that paleogenomics has received limited philosophical attention, which this dissertation addresses. I argue that the growing reliance on genomic sciences in human origins research, along with their data-driven character, risks conflating data collection with evidence production, with significant implications for narratives of human origins. The second chapter examines how paleogenomic data are credentialed as evidence through criteria of authenticity. Bringing together the philosophy of data and measurement validity, I argue that the robustness of the data production process—achieved through experimental and vicarious controls—allows researchers to consider ancient DNA data as an authentic signal from the past and, therefore, as reliable evidence, despite contamination risks. The third chapter examines cases of model-parameter underdetermination in demographic models located between authentic data and historical hypotheses. I argue that even in the presence of underdetermination, models can still support genuine knowledge. I defend a form of epistemic tolerance towards underdetermination, while also articulating constraints on its legitimacy, based on adequacy-for-purpose considerations. In the fourth chapter, I present a pluralist view of evidential assessment in human origins research, where data are regarded as reliable evidence only through a multimodal evaluative process that involves adequacy for purpose, contrastive force, and consilience. I demonstrate how background assumptions about classifying human populations are constrained by ethical considerations because they tend to reproduce or hide historically charged (e.g. racialized) ideas of human diversity. This offers a new perspective on how epistemic and ethical desiderata are connected in evidential assessment. Overall, the dissertation argues that data do not carry immediate evidential value, but acquire it through a complex, multistage, and temporally extended process of turning data into evidence. In human population research, this process depends on the interconnected practices of data production, modeling, and evidential assessment, where epistemic and ethical considerations are intertwined.

When 10:00 am - 12:00 pm on 6 July 2026
Building STH 525