Null case in the developing L2 grammers of adult Persian speakers acquiring English

Following Chomsky and Lasnik (1995) and Martin (1992), Boskovic (1996) claims that subject control infinitivals are IP/TP complements whose head is specified with a null case feature. He argues that one difference between ECM and subject control (SC) constructions is the fact that while in ECM constructions the head I/T does not bear any case feature, in SC infinitivals it carries a null case feature which is checked against PRO in its specifier position.This study investigates the acquistion of SC constructions by adult Persion speakers.In Persion all verb complements other than NP and PP are tensed CPs. Data were collected from four groups (elementary, lower intermediate, upper intermediate and advanced) of adult Persian speakers learning English as a foreign language at Hsfahan University. Results of a grammaticality judgment test and a translation task showed that L2 learners initially selected CP complements instead of IP complements for both subject control and ECM constructions. In spite of this initial transfer, the developmental patterns observed in the data indicated that Persion speakers are able to acquire the mismatch between the two functional categories involved (IP instead of CP in L2). No overt subject/pro with nominative case appeared in the specifier position of their Ips. Neither were any [± Past] or [+Person, Number] features encoded in their Ips. These results present counter evidence to Maturational Hypothesis (smith and Tsimpli 1995, Tsimpli 1998) and its recent version, Failed Functional Feature Hypothesis (Hawkins and Chan, 1997; Hawkins 1998, 1999) If [± Past] and [+ Person, Number] of I were imprinted instrinsic feartures, as Hawkins assunes, they would be subject to maturation and persistently difficult to disappear.