The volume offers a chronological compendium of the articles authored or co-authored by Bjorn Hammarberg about the analysis of a longitudinal study of Swedish polyglot adult learners. Each of the chapters in this book has has been pivotal in the development of the study of third language acquisition and multilingualism at both the theoretical and methodological levels, which makes it essential reading in the field.
The longitudinal study was created as a database for researchers interested in the acquisition of Swedish from a comparative perspective, be it amongst learners, between learners and native speakers, or at the synchronic or diachronic levels. The book includes five chapters that have been previously published as independent articles. These chapters constitute a frame of reference for a sixth chapter which extends the analysis of the corpus towards the consideration of the learner's "perceived crosslinguistic similarity" (p. 127) as a determinant factor in the activation of previously acquired languages in the acquisition of Swedish.
In the Preface, the reader encounters a detailed description of the birth of the project in 1990, as well as a portrait of the main participant, Sarah Williams. It presents the uncommon human dimension of a case study, which reminds the research community that informants are distinct individuals. The Introduction provides us with the motivations behind the study, grouped into three categories:
i. the practical motives that include interracial mobility, the increasing tendency to use English as a lingua franca, the educational challenges related to immigrant populations, and the cognitive questions that arise within the examination of the linguistic nature of polyglots;
ii. the theoretical motives, where we are reminded that "humans are potentially multilingual by nature and that multilingualism is the normal state of linguistic competence" (p. 2); and
iii. the empirical motives, where the researcher maintains that adopting a multilingual, rather than a bilingual, perspective would offer the research community a more complete picture regarding interlanguage development, metalinguistic awareness, linguistic creativity, and acquisitional processes, to name just a few fields of linguistic research.
The Introduction also clarifies that the publication will use the term "third language" (L3) to cover the current language in acquisition, while L2 would be any other additional language acquired after the L1. For Hammarberg, this is different from the chronological criterion since establishing a clear-cut sequential order of acquisition of more than two languages not only may be difficult for a large number of participants who would have been exposed to multiple languages simultaneously, but also problematic at the methodological level, in terms of determining what type of competence in the language is being considered for such a hierarchy (i.e., reading knowledge, cultural knowledge, metalinguistic knowledge).
The first chapter is co-authored with Sara Williams, the subject of the study. While this potential conflict of interest may be seen as a reason for a biased analysis of the data, the article carefully describes how the methodology of the study followed a series of introspective analysis methods as advised by Poulisse, Bongaerts, and Kellerman (1987). In addition to the defence of this choice of methodology, the authors emphasise that meta-linguistic discussions were not performed during the gathering of the data and that the retrospective commentaries were all done in English, which allowed Williams to remain impartial as a researcher. This chapter serves as an introduction to a deeper analysis of the data in the chapters ahead. It classifies a number of observations about the data in four groups: pronunciation, lexicon, morphology, and Williams' introspective comments. Most of the contents of these observations are extended in subsequent chapters, and the final discussion leaves the reader looking forward to more insightful analysis. However, this chapter is just a recapitulation of an earlier report on the study, at a time when the depth of the analysis had not yet taken place.
Chapter 2, also co-authored by Bjorn Hammarberg and Sara Williams, is perhaps one of the strongest of the book. It addresses the occurrence of language switches in Swedish L3, German L2, and English L1, from a psycholinguistic perspective. The authors adopt De Bot's (1992) model of bilingual speech production as a departure point, along with Poulisse and Bongaerts' (1994) revision to...