Abstract

This essay, written for the symposium A Cross-Disciplinary Look at Scientific Truth, takes a close look at the two most significant structural problems with the use of expert evidence within the adversary system: partisanship and epistemic competence. Each of these has long been recognized as a central difficulty with our present system for using expert evidence. Because parties both select and pay their experts, there is an inevitable danger that the experts chosen are those specifically prepared to shape their testimony to please their employer. Because we permit experts precisely because their knowledge and insights go beyond that which a lay jury is expected to know without their assistance, there is an inevitable epistemic difficulty with asking a lay jury to assess competing claims by experts on opposing sides. Not only are these difficulties familiar, but so are a set of proposed solutions. To eliminate partisanship, many have called for court-appointed, or neutral, experts. To eliminate the problem of epistemic competence, many have called for expert juries or expert judges. In this essay, I attempt to show that even though these oft-lamented problems are indeed significant ones, these familiar (though rarely implemented) solutions offer less than meets the eye. The calls for neutral experts, I suggest, typically rely on a misunderstanding of the nature of many scientific disputes. In practice, neutral experts would often either reinscribe the same difficulty generated by competing experts (presenting multiple points of view among which a lay jury would not have a rational basis for distinguishing), or else would mask legitimate disagreement among the expert community. Making use of epistemically-competent decisionmakers also raises both practical and theoretical difficulties (for example, who is epistemically competent? Just how often could we ask, say, epidemiologists to be jurors? Would epistemically competent decisionmakers tend already to have made up their minds in advance of trial, or might they reduce diversity of the jury along numerous other dimensions separate from their epistemic competence?) In the end, I conclude that the real lesson is this: the best we can hope for is to manage imperfectly, rather than resolve, the deep structural tensions surrounding both partisanship and epistemic competence. This ought not to invite quiescence, but it does suggest the need for both realism and modesty about the use of expert evidence within our adversarial system.