State of the field: Why novel prediction matters

There is considerable disagreement about the epistemic value of novel predictive success, i.e. when a scientist predicts an unexpected phenomenon, experiments are conducted, and the prediction proves to be accurate. We survey the field on this question, noting both fully articulated views such as weak and strong predictivism, and more nascent views, such as pluralist reasons for the instrumental value of prediction. By examining the various reasons offered for the value of prediction across a range of inferential contexts (including inferences from data to phenomena, from phenomena to theory, and from theory to framework), we can see that neither weak nor strong predictivism captures all of the reasons for valuing prediction available. A third path is presented, Pluralist Instrumental Predictivism; PIP for short. Imagine three scientists. The first uses her theory to make a novel prediction, tests it, and the prediction is successful, garnering support for her theory. The second collects some evidence, and then alters her theory to accommodate the evidence. The third secretly gathers some evidence in her lab, accommodates her theory to the evidence, then publicly makes a prediction using her new theory that the evidence she already gathered will obtain (which it does again). The third scientist is obviously deceiving people about something; more than that, the deception seems to undercut the legitimacy of her results. As Gardner writes, it seems that lying about whether one had actually predicted or only accommodated evidence “would be a form of scientific dishonesty akin to fabricating experimental results” (Gardner, 1982, p. 11). But why should the third scientist’s misbehaviour seem relevant to the evidential status of her results? A natural answer is that she is 1 This paper was entirely collaborative, and authors are listed in alphabetical order.