Definition:

Lydia White is Professor of Linguistics at McGill University, one of the leading researchers in generative approaches to second language acquisition. Her research investigates whether adult L2 learners retain access to Universal Grammar (UG) — the domain-specific language faculty hypothesized by Chomsky — and at what points UG-governed principles and their interaction with specific parameter values are accessible to learners acquiring a language with different settings from their L1.

In-Depth Explanation

Generative SLA asks a question distinct from most other SLA research: not “how can we facilitate acquisition” but “what is the initial state and what are the constraints on possible grammars in L2 acquisition?” This question presupposes the theoretical framework of generative linguistics — that language acquisition is guided by a biologically endowed language faculty (UG) containing universal principles and variable parameters, and that acquisition of any language involves setting the values of those parameters on the basis of input evidence.

The core question White investigates is: do adult L2 learners have access to UG? This question has three possible answers, representing three positions in the debate:

1. Full access, full transfer (the Schwartz & Sprouse, 1996 position, which White has been associated with): L2 learners begin with the full L1 grammar transferred as their initial state AND retain full access to UG principles and parameters. They can restructure their interlanguage grammar in response to input, guided by UG constraints. This explains both L1 transfer effects and the observation that L2 learners’ grammars respect UG constraints even when those constraints are not instantiated in their L1.

1. No access (strong environmentalist position): Learners can only rely on general learning strategies in L2; UG is no longer available after the critical period. L2 learners can only approximate native-like grammar through explicit learning and pattern extraction, not through UG-guided parameter resetting.

1. Partial access: L2 learners retain access to universal principles (the UG constraints that hold across all languages) but cannot reset parameters that differ between L1 and L2 — meaning that full native-like grammar is only possible when L2 parameter values overlap with L1.

White’s empirical research has examined a range of UG-governed phenomena in L2 — including Subjacency (constraints on extraction), binding theory (coreference conditions for pronouns and anaphors), and feature strength (the strength of functional features that triggers movement). A consistent finding is that L2 learners do show sensitivity to UG constraints, even when those constraints are not overt in the L2 input and do not exist in their L1, providing evidence against the “no access” position.

Her work on the Role of Instruction asked whether explicit teaching can help learners acquire UG-governed properties. Results suggest that instruction can speed up acquisition of some properties — particularly those with clear surface reflexes — but that instruction cannot induce L2 learner grammars to violate UG constraints. This supports the view that UG constrains the space of possible interlanguage grammars, whether or not the learner receives explicit instruction.

Key Contributions

• Full Access/Full Transfer position — associated with the Schwartz & Sprouse framework; L2 learners begin with L1 grammar transferred and retain full UG access
• Empirical research on UG constraints in L2 — binding, subjacency, feature strength, negative evidence
• Role of Instruction in generative SLA — what instruction can and cannot do given UG constraints
• UG and the initial state — what L2 learners bring to the task before encountering L2 input

Common Misconceptions

• White does not claim that L2 success is determined solely by UG. The generative approach addresses grammatical competence — the implicit knowledge of possible sentences — not all aspects of L2 usage, processing speed, or pragmatic competence.
• The Full Access position does not predict native-like attainment. Full access to UG principles and the ability to reset parameters does not eliminate the influence of prior L1 grammar or guarantee convergence with native speaker intuitions; it sets the outer bounds of possible interlanguage grammars, not the endpoint of learning.

Related Terms

• Universal grammar
• Interlanguage
• Language transfer
• Second language acquisition
• Applied linguistics

See Also

• Noam Chomsky
• Michael Long
• Susan Gass

Sources

• White, L. (2003). Second Language Acquisition and Universal Grammar. Cambridge University Press — the primary monograph synthesizing White’s generative SLA program; the essential reference for the UG access debate.
• Schwartz, B.D. & Sprouse, R.A. (1996). L2 cognitive states and the full transfer/full access model. Second Language Research, 12(1), 40–72 — the foundational Full Transfer/Full Access paper that White’s empirical work largely supports and extends.
• Google Scholar: Lydia White universal grammar L2 acquisition — full citation index for White’s generative SLA research.