The Cost of Specialisation

Every discipline keeps a private list of things it has agreed not to examine. Economists rarely interrogate the concept of a preference. Lawyers seldom ask what makes a rule a rule rather than a habit. Psychologists work with a notion of the self that most philosophers abandoned a century ago. These are not failures of intelligence. They are the ordinary cost of specialisation: to get anything done, a field must treat some questions as settled.

The cost of the settlement

The trouble is that the interesting problems tend to sit precisely where two such settlements meet. Financial regulation is the obvious case. A capital requirement is simultaneously a legal instrument, an economic incentive, and a statement about how much confidence a society places in its own forecasts.¹ Analyse it as any one of these and you will produce something defensible and incomplete.

The specialist knows more and more about less and less, until eventually he knows everything about nothing.

The remark is usually deployed as a joke against academics. It is better read as a description of an institutional problem. No individual scholar chose this; the incentives of the modern university produced it.

What a review is for

A review of this kind is not a substitute for specialist work, and it should not pretend to be. Its function is narrower and, I think, more useful: to put arguments from different fields in front of readers competent to judge them, and to let the friction do its work.

That requires a particular kind of reader — one who is expert somewhere and literate everywhere else. Such readers exist in far greater numbers than the publishing industry believes.

Footnotes

1. The point is developed at length in the Basel III literature, though rarely in those terms. ↩