A Note on Parasitic Gaps and Specificity

IP‐internal topic and focus phrases

“…The ungrammatical variants given below are meant to show that every question word must move into the Focus position. 26 Further (and very interesting) confirmation is provided by Modern Persian, a SOV language, in which specific and non-specific DOs occupy different positions in the surface syntax (Karimi 1999; all examples below are from this source). A specific (definite or indefinite) DO precedes IO, and is invariably followed by a special marker ra Ã; a non-specific DO follows IO: We can readily fit these facts into our analysis if we say that ra à is a Topic marker generated as the head of a Topic Phrase; and that a specific DO in Modern Persian is obligatorily topicalized.…”

Ellipsis in Farsi Complex Predicates

“…4 4 I assume Karimi's (1999a,b) analysis of the position of objects in Farsi, in which bare object DPs and DPs bearing the differential object marker rā are assigned distinct structural positions within the VP or nonverbal element phrase. Bare DPs follow a PP goal (i), while rā ‐marked ones appear before the PP (ii). (i)rāmin be vis gol dād.RamintoVisflowergive.…”

Constraint cumulativity and gradience: Wh-scrambling in Persian

“…He characterizes scrambling as a non-A, non-operator position (later Saito, 1992 tries to unify their views). I adopt the assumption that local wh-scrambling into the CP domain in Persian is A'-movement based on evidence in Karimi (1999bKarimi ( , 2005: licensing of parasitic gaps and unbound anaphora presumably. 11 Karimi (2005) works with the distinction between languages with structural whmovement and languages with scrambling of the wh-element (henceforth, I refer to the first one as ''wh-movement'' and to the second one as ''wh-scrambling'').…”