Political memoirs do not get better than this. And Rumseld is as competent, ethical, and wise as top government officials come — the best of a generation. In this memoir Rumsfeld very effectively takes the reader into the policy making and administrative process, quoting from Rumsfeld owns insightful and well crafted official memos — memos which can be read in full on the author’s website.
It is in turns gratifying and irritating to find many of my own evaluations of historical situations and personalities confirmed in real time by Rumsfeld. One point of interest for Hayek folks is the fine sense Rumsfeld has of the problem of limited or distributed knowledge — a consistent theme of the book, not coincidentally titled Known and Unknown. Much of Rumsfeld’s policy approach and management style falls out of a keen sense of the conflicting tensions between the need for an adaptive use of local on-the-spot understanding, played off against the demand for clear lines of responsibility, accountability, and control within a top-down bureaucratic organization. Rumsfeld is also intensely sensitive to the problems of precisely communicating knowledge through language — constantly bringing attention to how imprecise language misleads people into misunderstandings and unclear thinking, i.e. into cul de sacs of false knowledge.
I’m writing this post, however, because of a remark Rumsfeld makes while discussing his turn-around of the G. D. Searle company. It’s an insight about distributed knowledge and bureaucratic control which I believes explains a lot about what the university system is producing in the top economic departments– the output of “economic science” or “economic scientists” that we are being given by the guild system of these bureaucratic institutions. In economics (as David Colander, Deidre McCloskey, Philip Miroski and others have explained) what what the economics departments are giving us are “math jocks” who produces massive piles of extremely other-worldly math constructs, and endless reams of deeply misleading statistical “tests”. In other words, the economic departments are giving us folks who care about and think about math problems, but it doesn’t give us people who care about and think about providing sound causal explanation of the real world. In short, the guild bureaucracy of the university produces people who care about producing math and statistical “tests”, but it doesn’t produce people who care about producing genuine science or a genuine understanding the real world.
Over my eight years at Searle, I became a believer in the rule that, “What you measure improves.” A corollary rule in the military is, “You get what you inspect, not what you expect.”
My professor Larry Wright — a keen student of Thomas Kuhn and Ludwig Wittgenstein — is a specialist in the literature on argument, inference, and explanation. Wright in conversation often pointed to the mistake of confusing formal tests or metrics for real and deep understanding, e.g. the sort of understanding which can be displayed only in genuine competence in an array of activities, such as competence working in a lab, or using a language as a native, etc. What you measure improves — for example if you are constantly measuring the ability to produce math results, you will improve the math skills of those you are testing. There’s a formal metric for that. What you test, you get more of. But you really can’t formally test competence in achieving a sound causal understanding of the mechanism of the market. You can’t test for that, so you don’t get that, and that aspect of economic science has not improved, arguably, in 50 years — and if Hayek’s understandings of issue are correct, it has actually markedly declined.
In top economic departments, what they inspect is the math and data — what they don’t inspect is the explanatory capacity of that math and data, or its causal relevance in providing a genuine understanding of the real world. Economic departments don’t inspect that, so don’t expect to get that from a professor of economics.