Third language acquisition in adulthood ed. by Jennifer Cabrelli Amaro, Suzanne Flynn, Jason Rothman (review)

Over the past decade it has increasingly been argued that learning several languages is a different activity from learning two: beyond second language (L2) acquisition research lies third language (L3) acquisition, or indeed Ln acquisition. This book looks at some of the properties of L3/Ln acquisition primarily through three distinct models, presented in summary in the first chapter by the editors: the cumulative enhancement model (CEM), proposed by Flynn, which sees all previous language learning as contributing to L3/Ln learning; the typological primacy model (TPM), put forward by Rothman, which takes the contribution to depend on perceived or real typological closeness between the languages; and the L2 status factor, suggested by Camilla Bardel, which sees the L2 as contributing more to L3/Ln learning than the L1.

Part 1, ‘Theory’, mostly surveys swathes of existing L3/Ln research. The first chapter, by Maria del Pilar García Mayo and Jason Rothman (unnumbered in the text like all of the chapters), ‘L3 morphosyntax in the generative tradition’, explains the models, looks at the issue of defining the initial state of L2 learning, and argues that ‘generative L3/Ln acquisition’ has general theoretical implications. The next chapter, ‘L3 phonology’ by Jennifer Cabrelli Amaro, reviews research into L3 phonology, particularly transfer between the other languages and L3, the relationship between L3 and CEM, and methodology such as the lack of studies in L3 perception and intonation. Next comes the contribution by Camilla Bardel and Ylva Falk, ‘The status factor and the declarative/procedural distinction’, relating the status factor to Michel Paradis’s version of the declarative/procedural distinction and suggesting that future research might use L1s with different typological relationships to the L3. Kees de Bot’s chapter, ‘Rethinking multi-lingual processing’, introduces another orientation, dynamic systems theory (DST), in particular its flexibility over time, describes its main characteristics, and hopes to go beyond ‘a purely metaphorical use of DST notions’ (92) in multilingualism research. In ‘Multilingual lexical operations’, David Singleton provides a history of crosslinguistic influence in L2 learning and sees the leakage between languages in L2 production as a sign of their close lexical integration. To conclude Part 1, ‘L3/Ln acquisition: A view from outside’, by Roumyana Slabakova, measures four L2 hypotheses (interpretability, interface, feature reassembly, and bottleneck) against three studies of the syntax of L3 users, none of which accounts for the data fully, whereas a fifth hypothesis, modular transfer, introduced in the conclusions, may be better.

The first chapter of Part 2, ‘Empirical studies’, is ‘Further evidence in support of the cumulative enhancement model’, by Éva Berkes and Suzanne Flynn. This illustrates the CEM through L3 relative clause acquisition, demonstrating L3 facilitation based not on ‘the last language learned’ but on the L1. Next, ‘Acquisition of L3 German’, by Carol Jaensch, evaluates the three models against article acquisition in L2 learners of English learning L3 German, in which results favor the TPM and status models. Valeria Kulundary and Alison Gabriele, in their chapter ‘L2 syntactic development in L3 acquisition’, find that the most influence on L3 relative clause acquisition comes from L2 Russian. Jean-Marc Dewaele explores via L3/Ln speakers nine hypotheses in ‘Variation in two local and two foreign languages’, revealing the many factors that contribute to self-perceived proficiency. Next, ‘Advanced learners’ word choices in L3 French’, by Christina Lindqvist, tests how lexical choice between equally valid L3 items may be affected by L1/L2 frequencies, showing that advanced learners use more general words and more words from the L1. Finally, in ‘Foreign accentedness in third language acquisition: The case of L3 English’, Magdalena Wrembel finds more influence from L1 Polish than L2 French in the perception of...