Recent advances in linguistic theory within the principles and parameters framework have exerted considerable influence on the field of second language acquisition. SLA researchers working within this framework of syntactic theory have investigated the extent to which developing second language grammars are constrained by principles of Universal Grammar (UG). Much of the UG-based SLA research in the 1980s focused on adult L2 acquisition, but the role of UG principles in child L2 acquisition remained largely unexplored. More recently, however, this state of affairs has begun to change as SLA researchers are becoming more and more interested in child second language syntactic development. In this paper, I review recent and current developments in UG-based child SLA research, and I argue that child SLA has a valuable role to play in enabling us to arrive at a better understanding of the role of biological factors in language acquisition and in strengthening the links between SLA and linguistic theory. Specifically, I discuss the findings of child SLA studies with respect to the following issues: the role of UG parameters in child SLA, the status of functional categories and their projections in child SLA, and the nature of the evidence available to and used by child L2 learners. The overall picture emerging from these studies suggests that child L2 developing grammars are indeed constrained by Universal Grammar. While it is not fully clear at the present time whether the child L2 learners& knowledge is a result of direct access to UG or indirect access to UG (i.e., through the mediation of the L1), the evidence indicates that L1 transfer (at least in certain syntactic domains) cannot be entirely ruled out.

The focus of this study is the development of control over anaphoric reference and relativization by children learning ESL in a pull-out classroom employing little overt grammar instruction. Twenty-three children aged 6.5–14 representing ESL proficiency levels 3–5 on the Bilingual Syntax Measure (BSM)were given paper-and-pencil tests to determine comprehension of anaphoric (reflexive and pronominal)reference in English; in addition, they were tested on both comprehension and production of restrictive relative clause types (SS, SO, OO, and OS)in English. Scores were higher on reflexives than on pronominals, with length of residence significant in ambiguous references. Relative clause interpretation varied significantly with proficiency level, but production was not predictable from general proficiency except at BSM level 5, and then only on SO and SS relativization. Transfer from the L1 was minimal. It is concluded that exposure and overall L2 proficiency, rather than age or L1 background, are the most significant factors in the development of these generally untaught and untested “late-learned” rules.

Some earlier SLA research suggests that zero pronouns should be easily acquired by second language learners, yet this has not been carefully examined because studies of nominal reference are usually limited to the acquisition of languages that do not have zero pronouns. This study examines how speakers of languages with zero pronouns (Japanese) and without them (English) use zero pronouns when acquiring a language that has them (Mandarin Chinese). The findings show that second language learners do not use zero pronouns as often as native speakers and that their use increases with proficiency. When examined more closely, it can be seen that the speakers have no difficulty using zero pronouns when there is a syntactic or semantic restriction, but they do have difficulty at the discourse level. It is claimed that this underuse of zero pronouns corresponds with other research that shows that second language learners tend to avoid pronouns in favor of full noun phrases. Some explanations for these phenomena are postulated.

This is a report on a computerized experiment with a miniature linguistic system, consisting of five morphological rules and a lexicon of 98 words. Two hypotheses derived from the literature in cognitive psychology and psycholinguistics were tested: that explicit-deductive learning would be better than implicit-inductive learning for straightforward (“categorical”) rules, and that implicit-inductive learning would be better than explicit-deductive learning for fuzzy rules (“prototypicality patterns”). Implicit-inductive learning was implemented by pairing sentences with color pictures; explicit-deductive learning was implemented by means of traditional grammar rule presentation, followed by picture-sentence pairing. The findings were in the expected direction for both hypotheses, but only the first one could be confirmed through statistically significant results.