This paper challenges the existence of PRO module in the grammar. Hornstein (1999) suggests that PRO does not exist and PROarb is identical to pro. Landau (1999), however, claims that PRO exists and PROarb is different from pro syntactically. The data provided here, along with the analysis to be presented, show that PROarb and pro function similarly in Persian impersonal constructions. Persian does not have any overt DPs with appropriate semantics such as impersonal one in English. Therefore, the only feature combinations that are compatible with the semantics are those for the covert equivalent of one.
Is PROarb the same as pro? Evidence from Persian Impersonal Constructions