The Second Time Around: Minimalism and L2 Acquisition (review)

This book applies minimalist theory to second language acquisition, claiming that L2 acquisition in a sense replicates the learning of the first language in that UG circumscribes the learning process. Herschensohn assumes the availability of UG throughout the L2 acquisition process, together with the minimalist notion that syntax is acquired via morphological features and functional categories. The book contains seven substantive chapters and a concluding chapter.

The first chapter surveys the principal theoretical issues in L2 acquisition from a generative perspective including poverty of the stimulus, the modularity of language, universal grammar, and the compatibility of minimalist theory with L2 research. The second chapter, ‘Critical age and L1/L2 differences’, surveys theories on the existence of an inexorable critical period for complete L2 acquisition, as well as the ‘critical difference hypothesis’, according to which the learning of a second language differs substantially from the acquisition of the first language. H accepts both notions, albeit with some qualification, but also acknowledges the possibility of near-native L2 attainment within the normal range of second-language learning. The third chapter, ‘Generative theory from Aspects to minimalism’, surveys recent syntactic theories, with special attention to parameter setting and the accessibility of UG in first- and second-language acquisition. H asserts that the minimalist program provides a better account of L2 acquisition than does the principles and parameters model since the latter derives from UG only form, and not strategies. The minimalist theory proposes that all language variation is morpholexical, and all parametric variation arises from the lexicon. The ‘incompleteness’ of L2 acquisition follows from the fact that parameters are not ‘reset’ but rather gradually acquired as the second-language learner’s lexical store increases.

Ch. 4, ‘The stages of first and second language acquisition’, presents a model of language development within a generative framework. L1 acquisition follows the ‘coalition model’, which progresses successively through acoustic packaging, linguistic mapping, and complex syntactic analysis. The first stage involves phonemic contrasts but also the lexical-functional distinction, derived from acoustic cues. The second stage corresponds roughly to the ‘two word utterance’ period, in which functional categories are very scarce; H acknowledges the lack of consensus on whether functional categories are underlyingly present at this developmental point, suggesting that functional categories are projected only when essential. The final stage leads to adult-like syntax, including elaborate morphological paradigms and full use of functional categories. Parameters are set in L1 as morphology is acquired, with full morphology and set parameters emerging simultaneously. L2 acquisition follows the ‘constructionist’ approach in which parameters are slowly (re-)set as morphology—hence syntax—is acquired through the expanding L2 lexicon. Constructionism also involves three stages: L2 transfer; an intermediate period in which many morphological features are underspecified, and L1 parameter settings gradually disappear; and full L2 feature specification, thus parameter resetting.

Since the minimalist program asserts that cross-linguistic differences are morpholexical, L2 learners presumably are more erratic in their use of morpholexical features than in their acquisition of L2 syntax; the chapter ‘Parameter shifting in L2 acquisition’ explores this prediction, focusing on the process of acquisition (morpholexical expansion) rather than on the final outcome (reset parameters). H reviews the literature on three prominent parametric differences: verb raising, V2 in Germanic languages, and null subjects in Spanish; in the last case H postulates the unity of null subjects, that-trace violations, and VS word order among L1 speakers despite a considerable body of evidence which suggests that the three phenomena are not linked crosslinguistically (or among L2 speakers) as a single parameter. In all cases L2 parameter resetting is not crucially dependent on acquisition of morphology, as in initial L1 acquisition. The constructionist hypothesis claims that these acquisitional differences result from the intermediate stage of feature underspecification, between L1 transfer and the final L2 grammar. All interlanguage grammars are constrained by UG, and L2 learners ultimately acquire parametric settings (through grammaticality judgments) not available in the input data...