In his inspiring dissertation "Quantification in the Theory of Grammar", Nishigauchi has argued that Japanese interrogatives can only be adequately analysed if we assume pied-piping for wh-phrases in logical form (henceforth called LF Pied-Piping). The idea has since become fashionable. Among others, Hasegawa (1986), Choe (1987), Bayer (1990), and Watanabe (1992) assume LF Pied-Piping. In this paper I will argue that there is no pied-piping at the level of "transparent LF", by which term I understand that level of representation which determines the interpretation (modulo context-dependency). An adequate LF requires either unbounded LF movement or reconstruction of the pied-piped material or an equivalent method. It is not possible to interpret the pied-piped material in SpecC.1 The organization of the representation is as follows: