The purpose of this paper is to make it clear where an inverted locative or directional PP moves in inverted locative PP constructions. The main points of previous studies can be summarized into the two following questions: Does the inverted PP move to TP-Spec or not? and where is the left peripheral position of the inverted PP if it moves to the left of TP-Spec? This paper has shown that the disputes on these questions are attributed to too simple a generalization of different structures due to the verb types. With a lot of empirical evidence I have illustrated that the most adequate description about the PP position can be achieved by positing three different positions (TP-Spec, FocP-Spec, and TP-adjoined position) according to three classes of verbs (be, unaccusative, unergative).