Syntactic theory has played a role in second language acquisition (SLA) research since the early 1980s, when the principles and parameters model of generative grammar was implemented. However, it was the so-called functional parameterization hypothesis together with the debate on whether second language learners activated new features or switched their value that led to detailed and in-depth analyses of the syntactic properties of many different nonnative grammars. In the last 10 years, with the minimalist program as background, these analyses have diverted more and more from looking at those syntactic properties that argued for or against the various versions of the UG-access versus non-UG-access debate (UG for Universal Grammar) and have more recently delved into the status of nonnative grammars in the cognitive science field. Thus, using features (i.e., gender, case, verb, and determiner) as the basic units and paying special attention to the quality of input as well as to processing principles and constraints, nonnative grammars have been compared to the language contact paradigms that underlie subsequent bilingualism, child SLA, creole formation, and diachronic change. Taking Chomsky's I-language/E-language construct as the framework, this article provides a review of these recent developments in SLA research.