I'm No Donald Rumsfeld Fan, But...

when people chortle about his "unknown unknowns," the laugh is on them:

DAVID DUNNING:  That's absolutely right.  It's knowing that there are things you don't know that you don't know. Donald Rumsfeld gave this speech about "unknown unknowns."  It goes something like this: "There are things we know we know about terrorism.  There are things we know we don't know.  And there are things that are unknown unknowns.  We don't know that we don’t know."  He got a lot of grief for that.  And I thought, "That's the smartest and most modest thing I’ve heard in a year."

 This is basically just the point people like Knight and Kirzner make on the difference between risk and uncertainty. It is why Kirzner notes that merely including search costs doesn't really fold Hayek into the mainstream: you can only evaluate search costs in any meaningful sense when you both know just what it is for which you are searching, and know with some exactitude the limits of where it might be found. But in cases where a true innovation is sought, we don't know either of those things.