In the recent literature, it has often been suggested that head movement may be a PF phenomenon. For instance, Chomsky (2001: 32) points out that head movement differs from core rules of the narrow syntax in several respects and that all of this would be unproblematic if the phenomenon in question is a phonological process reflecting affixal properties. (See also Boeckx and Stjepanovic 2001 for relevant discussion and references.) In this paper, based on a novel set of data from English, I suggest that in some specific contexts, it seems reasonable to assume that head movement can indeed take place in PF, which I argue provides partial evidence for Chomsky`s claim that head movement can be attributed to PF. The discussion here also has some implications for the nature of the coordinate structure constraint.