Exploring the Subject Surface Position in Persian: A Psycholinguistic Study

The question of whether the subject stays in its base position within the vP or moves to its derived position at Spec TP/IP is difficult to answer with respect to free word order languages such as Persian, because the surface constituent orders in these languages do not necessarily provide sufficient information to determine syntactic positions (Koizumi & Tamaoka, 664:2010). The discussions in this regard has led to formulating two competing analyses: One is that the subject in canonically ordered sentences in Persian obligatorily moves from its base position within vP to its derived position at Spec TP/IP for the purpose of Nom Case or EPP checking (Anushe, 2008; Darzi, 1996; 2006; 2009; Tafakori R. 2007; Darzi & Beiraqdar, 2007). The other one, however, is that the subject does not have to move out of vP and stays in its base position within it (Karimi, 2005). In the present study we test these analyses by performing two off-line experiments, involving a sentence completion task and a grammaticality judgment task. Analyzing the data obtained from the sentence completion task revealed that in %90.89 of the experimental sentences, completed by participants, subjects were located in Spec TP/IP. It also revealed that in %9.11 of the experimental sentences, produced by participants, subjects were located within vP. Furthermore, studying the data gained from the grammaticality judgment task divulged that the sentences with their subjects within vP took longer to process than the sentences with their subjects in Spec TP/IP; It took 2.871 and 3.801 seconds for the participants to process these two groups of sentences respectively. These results present psycholinguistic evidence for the analysis according to which the subject must move to Spec TP/IP in canonically ordered sentences in Persian