Corrigibility and Trust in the Practices of Science

In the context of crises in science, understanding what virtue in science is requires attention to virtue studies and virtue science themselves. Within and beyond the study of practices in laboratories and in scientific collaborations held up as exemplary or noted for being less so are core beliefs about what trust in science is or should be. Prevailing conceptions of trust in science and scientists are at the root of the crises in science. Virtue science and studies can help to identify the noxious tendencies of some conceptions of trust and suggest possible new ways of thinking. This article reviews evidence for sciences in crisis before drawing on Bayesian thinking to propose ways of thinking about trust, reliability, and validity in relation to science. Corrigibility, rather than fallibility, is the defining feature of science, calling for a trust in the persons, systems, and institutions of science as ever error-probable-plus-corrigible.