That’s a question I’d like to ask of many different disciplines. Right now, for instance, I’m reading a Matt Yglesias piece about how poorly tax breaks for the wealthy work as stimulus. I believe this has been known for a very long time — I’m tempted to say since Keynes — though I’m not certain. It would be interesting to go around to lots of different disciplines and ask them which truths have been established for 50 years or more. (Mathematics, you’ll have to sit this one out; all your truths are permanent, and many are very old. You’d win by default.)

I’m reminded here of the famous story, which appears to come from “The Way of An Economist”, though I don’t have access to it, in which someone asked Paul Samuelson to name an idea from economics that is both true and not a tautology. It took him some years to respond with the doctrine of comparative advantage. That’s been accepted as true for 200+ years, so it seems to count as a disciplinary truth. I wonder what others might be.

I probably have to narrow this down, because I’d want truths that aren’t based on obviously unrealistic premises. For instance, there are a lot of results in economics that depend on constant returns to scale — for instance, that if you double the quantity of inputs, you double the quantity of outputs. But many phenomena in our world depend on *increasing* returns. Doubling the number of users in AT&T’s network more than doubles the value of that network. The very existence of cities is proof of increasing returns to scale: Silicon Valley isn’t the world capital of software development because it’s somehow better equipped to build software; it’s the capital because it had some initial burst of development, which led to a snowball effect that drew more developers to it. (The classic economic example here is Dalton, Georgia, which by historical accident has become the carpeting capital of the America.) If you can’t understand the existence of cities using easy economic assumptions, then you need to re-examine your assumptions. Yes, I realize that constant returns to scale are easier to model, but in any case: if we’re looking for accepted wisdom from a discipline, theorems about the real world cannot be accepted as true if they’re based on premises that are known to be far off the mark. (That gives another reason why mathematics has to sit this out: strictly speaking, mathematics isn’t *about* anything. It’s a collection of tautologies.)

I’ll start bugging people about this.