Merge position of floating quantifier "hame" in Persian

This paper investigates the External Merge position of the floating quantifier “hame” (all) in Persian. Quantifier floating has received two major analyses: stranding analysis and adverbial one. In this paper, the former approach is shown to be more explanatorily justified. The paper provides evidence to indicate that “hame” is adjoined acyclically to the argument DP after the DP moves from its θ-position, along the lines of Boskovic (2004). More specifically, quantifier floating is shown not to be possible from θ-positions. The evidence comes from the prosodic pattern of Persian unmarked sentences as well as the scope interaction of negation and the floating quantifier “hame”. In regard to prosodic pattern of Persian unmarked sentences, main sentence stress has been argued by Kahnemuyipour (2009) to mark the left edge of vP. Floating quantifier is shown to occur before the element bearing main sentence stress in unmarked sentences, so it is claimed to be outside of vP, and therefore outside of the θ-position. As regards the second evidence, i.e. scope interaction, in negative sentences containing floating “hame’ when negation is not focused, ‘hame’ is constantly out of the scope of negation, and this means that in such sentences, neither the floating quantifier nor its copies are within the c-commanding domain of NegP. To put it another way, the base position of the floating quantifier is higher than NegP, which is argued to be between TP and vP. To determine the position of NegP in Persian, evidence is provided from negation in gerund phrases in Persian, and the scope interaction between manner and speaker-oriented adverbs on one hand and between these adverbs and negation on the other. Showing that NegP is located between vP and TP within the scope of “hame”, it is concluded that the External Merge position of quantifier is higher than vP, and therefore is outside the θ-domain.