The issue of cross-linguistic influence in non-primary language learning has long been an important topic. Studies conducted in this area suggest that a form of L1 or L2 transfer is evident in the language produced by third language learners (Cenoz 2001; Leung 2003; White 2003, among many others). On the nature of this transfer; however, there seems to be no consensus. Whether the acquisition of syntactic structures is a matter of mere parameter setting from a minus value in one language to a plus value in another or a rather painstaking process of reassembling the relevant features from the way they are conditioned and realized in the L1 to that of the L2 or L3 is not yet revealed. While, Hawkins (2003) supports a "Representational Deficit Hypothesis" (RDH) whereby interlanguage grammars are confined to L1 feature values and hence any problem in L2 or L3 initial state is attributed to problems in resetting syntactic features of L1, Lardier (1999, 2000, 2003) argues for a Missing Surface Tnflection Hypothesis (MSTH) in mapping lexical entries onto overt phonological forms. Still there is a third argument which claims that the entire L1 grammar (in the sense of all abstract properties) constitute the initial state in L2 acquisition. Based on this stance, Schwartz & Sprouse (1994, 1996) argue that changes to the initial grammar can take place to the effect that the L2 learners are not confined to representations based on L1 steady state. Finally, Hawkins proposes a moderate modulated structure building Hypothesis (MSBH) which claims for an incremental leaning mechanism for L2 and L3. Inspired by the given issue, the present study attempts to see what effects the interaction of Kurdish and Persian have on Kurdish EFL students' acquisition of syntactic and morphological properties of tense and passive structures in English. To this end, 120 Persian monolingual and Kurdish-Persian bilingual EFL Learners took part in this study. They were assigned to three proficiency bands (after taking a general proficiency test, OPT) and received a grammaticality Judgment and a contextualized functional production tasks. The results showed no significant difference between the performance of monolinguals and bilinguals at each level of proficiency beyond the initial state. Nonetheless, significant differences were found across the levels of proficiency. The findings, interpreted in terms of the viewpoints of the current generative models of L2A, revealed a dominant role for L1 settings in L3 interlanguage grammar especially at the initial state and were more in line with the claims of MSBH.