Reasonable Doubt

How to know when science is trustworthy and when it is not

We’re often told to “trust the science”—but is science really trustworthy? In Reasonable Doubt, Edouard Machery and John Doris argue that, surprisingly, it is often untrustworthy, with findings that can’t be replicated, empirical support lacking, and self-correction failing. Drawing on their expertise in cognitive science and social psychology, they show that the vagaries of the human mind and scientific culture, along with bad luck, make the production of untrustworthy science inevitable in the “soft” social sciences as well as the “hard” natural sciences. Once publicized, scientifically suspect results too often continue to be cited, even if they have been retracted, forming “zombie literatures” full of poorly supported empirical results. But the existence of untrustworthy science does not mean that we should advocate science denialism and abandon science for conspiracy theories or astrology. Instead, Machery and Doris argue, we should embrace agnosticism about science, relying on science where we must, in the spirit of reasonable doubt rather than credulity, while acknowledging the pervasiveness of scientific uncertainty.