This study explores A-movement reconstruction and asserts that A-movement reconstructs as A’-movement does, providing semantic type-based accounts in accordance with semantic reconstruction proposed by Lechner (1998) and Ruys (2015). The current proposal is that quantificational subjects leave a higher type trace <<e,t>,t> at any immediate landing site and an individual type trace <e> at their theta-position. Based on this hypothesis, three sentences that show no reconstruction effects can be accounted for. The components in the three sentences such as negations, degree modifiers and non-neg-raising predicates induce weak island effects for higher type traces. As resumptive pronouns in island conditions, an individual type trace <e> can obviate island effects, but a higher type trace <<e,t>,t> cannot.

Since the early 1970s, it has been recognized that some movements can cross clause-boundaries with COMP as an escape hatch but others cannot. These two types of movement were called A(rgument)- and A’-movements, which have been established in generative grammar since the early 1980s. Deeply related with this distinction is the ban on improper movement, which means the combination of the two movement types. The only exception to the improper movement is where the entire movement occurs from an A-position ends in an A’-position, i.e. where A-movement feeds A’-movement. Since Chomsky (2008) proposed the parallel movement hypothesis, a route has been opened to dissociating A- and A’-movements. Then, we need not allow A-movement to feed A’-movement. Is what we need not do what we should not do? To answer this question, this paper shows that though separate, A-movement is still able to manifest its effects over A’movement.

Dongseok Kim. 1999. Cyclicity in A-Movement. Studies in Modern Grammar 18, 21-37. In this paper, we examine operations that eliminate uninterpretable features in the recent version of the Minimalist Program presented in Chomsky (1998), and propose that defective T lacks EPP feature, contrary to Chomsky`s assumption. It is shown that if defective T is given EPP feature, problems arise in deleting ψ -feature of defective infinitival T in expletive raising constructions, since ψ -feature of expletive there, being defective, is not capable of deleting the probe. Reviewing arguments for SPEC position of infinitival T, we show that EPP feature is not forced or necessary for tenseless T. Thus, we suggest that there is no successive cyclic A-movement through SPEC of defective T.