ABSTRACT OF PAPER
Title: Expertise as Measurement Instrument
Author: Reiss Julian
Experts play an enormously important role in our daily lives, even if not always easily noticeable. Taking Menger’s tripartite expression of the aims of economics – prediction, explanation, control – as starting point and adding description, we can easily see that few scientific activities in pursuit of these aims get by without expert judgement. Experts make predictions about future political events on the basis of which we plan and decide; they offer informal causal models to explain economic outcomes such as the current financial crisis; they advise the government about the effectiveness of policy; they assess the state of the economy and influence the data that are published. But scientific and political expertise has received some bad press recently. Experts are overconfident (e.g., Angner 2006), subject to numerous fallacies of reasoning (e.g., Gigerenzer 1991), and their predictions are almost always inferior to those of so-called statistical prediction rules (e.g., Bishop and Trout 2005; Tetlock 2005). In my paper I argue that cognitive and motivational limitations such as these should be taken seriously but not as a justification for outright scepticism. To criticise expert judgement as deficient without offering a viable alternative is complacent. To the extent that expertise is here to stay, and there are no good reasons to believe that mechanical procedures will soon, or indeed ever, substitute human judgement in all areas of science and policy, we should learn to live with it and find strategies for improving it. The metaphor of 'expertise as measurement instrument' suggests some points of departure: empirical strategies of calibration, error-correction, result validation and others work in one area as they work in the other. Conceived of as an instrument of limited but epistemically accessible and improvable reliability, expertise can play important and useful roles in science and policy.
Registred web users only can download this paper - Go back
Please note that files available for download have not been checked for viruses. These files have been submitted by authors of the conference to this web site. Conference organisers can't accept any responsibility for damages caused to users by downloading such files.


