Abstract. In recent decades, scholars have presented several "meta-theorems" of preference aggregation, intended to unify and subsume celebrated impossibility results such as those of Arrow and of Gibbard-Satterthwaite. In this paper, I argue that several of these unification results are themselves implied by the classical theorems. My argument proceeds in two steps. First, I present a "meta-meta-theorem" that implies several existing unification results; second, I show that this theorem follows directly from Arrow.