Why Most Experts Are Little Better Than Dart-Throwing Chimpanzees
The measure of any theory is its ability to accurately predict some event that must follow if it is true. Experts, especially those with one big idea or theme, tend to do quite poorly when their predictions are put to the test. See "Why Most Predictions Are So Bad" ht: MarginalRevolution.
To avoid the problem, clever experts try to construct opinions that aren't testable (at least not in their working lifetimes) so as to avoid the embarrassment (and financial loss) that would follow falsification of their ideas. Other experts, particularly those of the irrationalist school, deny that bad predictions undermine theories and may even argue that untestable theories can still be good science so long as an expert's subjective judgment convinces him of its truth. The Milward court couldn't find a problem with either approach.
So if henceforth our job will be to discover how firmly an expert is convinced of his opinion rather than the soundness of the foundations on which it rests how should we go about it? Immanuel Kant suggests the answer in "Critique of Pure Reason":
"The usual touchstone as to whether something asserted by someone is mere persuasion, or at least subjective conviction - i.e., firm belief - is betting. Often someone pronounces his propositions with such confident and intractable defiance that he seems to have entirely shed all worry about error. A bet startles him. Sometimes the persuasion which he owns turns out to be sufficient to be assessed at one ducat, but not at ten. For although he may indeed risk the first ducat, at ten ducats he first becomes aware of what he previously failed to notice, viz., that he might possibly have erred after all. If we conceive in our thoughts the possibility of betting our whole life's happiness on something, then our triumphant judgment dwindles very much indeed; we then become extremely timid and thus discover for the first time that our belief does not reach this far, thus pragmatic belief has merely a degree, which according to the difference of the interest involved may be large but may also be small."
The answer then is clear, I think. Let's put the fees of experts in escrow when they opine upon things not yet generally accepted. If they are proved correct, say within a decade (or at least not refuted - ties going to runners after all), then the expert claims his fees and all interest accrued. If on the other hand he is proved wrong he forfeits the fees and pays a 10% penalty with the total going to charity. Assuming the law retains some interest in the truth such a rule would undoubtedly work far better than any losing-party-pays rule.
