Language attrition (review)

The study of language attrition is currently entering its fourth decade. The development that this field of research has witnessed since the early 1980s is, however, not straightforwardly described as exponential or incremental. While the early years were characterized by groundbreaking investigations, there were also diverging research foci and isolation between studies. This state of affairs diverted attention and resources from the pursuit of a research agenda with common ground. The past decade, however, has seen a remarkable surge in attrition research, and Monika S. Schmid’s book is highly representative of this development. In fact, S herself has played an instrumental role in the advancement of the field, both as an editor of several volumes and special issues, and as a prolific researcher and synergist.

The book is an introduction to the study of first language (henceforth L1) attrition and a research manual. People who are about to embark on an attrition research project will doubtless find this book very useful. However, the readership is not circumscribed to novice researchers: the book’s five main sections contain novel conceptual and methodological work that is of great importance to anyone working in the exciting field of language attrition and/or related areas.

Part 1 outlines the linguistic aspects of attrition: that is to say, it describes the way in which the language system may undergo attrition. After a general opening in which the basic terminology is outlined, the reader is introduced to the phenomenon of crosslinguistic influence (here the focus is of course on how the L1 is affected by the second language, that is, the L2). Here, S presents a taxonomy to characterize different types of crosslinguistic influence: borrowing, whereby elements from the L2 are (temporarily) integrated into the L1; convergence, whereby an element is used in the L1 in a way that is distinct from both the L1 and the L2; restructuring, whereby L1 elements are reanalyzed on the basis of their L2 counterparts; shift, which is similar to restructuring, with the exception that the L1 element is now completely remodeled on the L2; and finally attrition, here defined as the complete loss of an L1 element, both productively and receptively. This taxonomy was originally introduced by Pavlenko (2004) in an attempt to problematize and refine the deviating instances in attriters’ speech that have often just been conveniently labeled ‘attrition’. Given that this classification has not received a great deal of attention among attrition researchers so far, it is significant that S has chosen to give it such prominence, illustrating with authentic data how the different processes are manifested across the domains of lexis, syntax, morphology, and phonetics. The fact that Pavlenko’s taxonomy is given this recognition points to a need for attrition research to avail itself of more fine-grained conceptual tools in order to ferret out the internal (L1) and external (L2) mechanisms that govern the nature and selectivity of the attrition process.

Part 2 deals with so-called extralinguistic aspects of attrition, that is, independent variables that influence the extent to which a language may undergo attrition. S arranges these variables into three main sets: first, there are personal background variables, including age of onset of attrition and length of residence in the L2 environment; second, there is the variable of L1 input and output (sometimes called L1 contact or L1 use); and third, there is the set of attitudinal and emotional variables, including motivation to maintain the L1 and the role of L1 in identity construction. S’s choice to treat L1 use as a variable distinct from the other two sets is unusual, since many times this variable is lumped together with the variables in one of the other two sets. It is, nonetheless, a well-motivated choice. The variable of language use has had a somewhat Janus-faced status in attrition research: on the one hand, it has been assumed that without any reduction in L1 contact, it is unlikely that there will be much attrition at all (cf. Andersen...