To my ear, "full" technocracy would be purely empirical, without any reliance on ideology that cannot be shown to be critically necessary. In all other cases ab

To my ear, "full" technocracy would be purely empirical, without any reliance on ideology that cannot be shown to be critically necessary. In all other cases absent a clear and present danger, when empirical clarity is unavailable, the only action is to do nothing. And I'd be prepared to accept consistency with the legal code as a version of "empirical". Prohibition had some measure of empirical support, but was legally completely indefensible. Alleged "Soviet" technocracy just wasn't - it subsumed all science to the One True Theory which was Communism. Indeed, it's the poster child for what I'd be against. I mean Lamarckianism? It has to be real science. The Coase theorem explicitly excludes consideration of transaction costs, in the same way constants get dropped in differential equations. But my native tendency is to underplay externalities anyway, because it's too hard to get intersubjective agreement on them. This goes to this peculiar property of norms to get weird when attempts to stack them on top of each other occurs. How was it that "state's rights" led to lynchings anyway? Because they linked "Southern Honor" and States Rights in this incoherent manner to produce Jim Crow. Sixth-graders - most of them - could argue that away. Finally, what is the role of government? To enforce and modify law. The minimal role is to enforce land claims, prosecute criminal acts and enforce contracts. Actions of government should not create norms. Anything you can manage beyond that should be a source of skepticism.