This is pretty funny...

From commenter "Invisible Backhand": a generic Don Boudreaux letter to the editor.

This looks like it's just a swipe, but it's actually a very good insight about the weakness of reductio ad absurdum arguments when they are applied to optimizing behavior. When agents optimize, the idea is they can do too little of something, just enough of something, and too much of something. Don often takes a "just enough of something" when that "something" is something he doesn't like, extrapolates to a "too much of something", and then acts like he's refuted an argument. He often proceeds as if economics has nothing to do with rational optimization. I don't care what you think of Kirzner, or "surprise", or equilibrium, or search - rational optimization is something to be qualified and caveated - not abandoned.

Reductio ad absurdum arguments are usually badly implemented when they're applied to economics. If you see them, be suspicious and think about whether the claim makes sense.