Soviet Science

Friday, August 14th, 2009

Mencius Moldbug disagrees with the progressive “senior public-health scientists and practitioners” at Effect Measure, and he explains his position:

I hate to break this to you, but post-WWII American academia is in practice an agency of USG [the US Government], because it (a) is funded by USG, and (b) drives USG policy. (Especially in fields such as yours, which exist largely for the purpose of telling USG what to do.)
[...]
Here’s an analogy that may help you understand my perspective. Perhaps you remember a country called the Soviet Union. Now, when you hear that Soviet science proved X or Y or Z, what do you think? You think: X or Y or Z might be true, or it might not. To know, you’d have to look into it.

Even if X or Y or Z was written in the Great Soviet Encylopedia. Even if it was endorsed by a unanimous vote of the Soviet Academy of Sciences. Even if Chairman Brezhnev himself declared it to be an ineluctable consequence of dialectical materialism. You say: I’d have to look into it.

And why is this? Because you don’t trust the process by which the Soviet system generated scientific results — despite the many excellent scientists who worked within it. And nor do I. We share (if I may assume) the belief that Soviet science could discover and amplify truth, and also discover and amplify error. Thus: “trust, but verify.”

Now, if we compare (a) the organizational structure of post-WWII Western science, to (b) the organizational structure of pre-WWII Western science, to (c) the structure of Soviet science, we see that (a) looks a lot more like (c) than like (b).

Namely: it is centrally funded and centrally planned, even when conducted in “private” universities. Consensus can be produced by a few bureaucrats — excuse me, public servants — who choose to fund the believers and defund the deniers. Moreover, since these public servants (eg, at NSF or NIH) are scientists themselves, it is terribly easy for one faction to exclude another. There is no effective independent supervision. There is certainly no way to shut down an entire field that has become pseudoscientific.

Whereas before WWII and Vannevar Bush, consensus actually meant something, because the (much smaller) funding of science was decentralized and independent, and most important depended far less on the results of that work. To retain their status and funding, scientists had to convince critical, intelligent, and independent nonscientists. They had far less incentive to exaggerate the public-policy importance of their work. Whereas nowadays, even in my own field (computer science) the typical grant application is richly marbled with preposterous claims of public importance. Everyone does it, so everyone has to.

Thus a reasonable person would expect the type of scientific malfeasance so frequently seen in the Soviet system to emerge in the West.

Leave a Reply