On Empirical Equivalence and Duality

I argue that, on a judicious reading of two existing criteria--one syntactic and the other semantic--dual theories can be taken to be empirically equivalent. The judicious reading is straightforward, but leads to the surprising conclusion that very different-looking theories can have equivalent empirical content. And thus it shows how a widespread scientific practice, of interpreting duals as empirically equivalent, can be understood by a thus-far unnoticed feature of existing accounts of empirical equivalence.