Second Language Learning in the Early School Years: Trends and Contexts. Victoria A. Murphy

Chapter 2 on infant speech processing) not two years, and that the offset is likely to be far earlier than puberty, possibly around the age at which typical L1 development is (almost) complete, around 5 or 6 years. Other researchers have argued that age effects in language learning should not be viewed in the constraints of a ‘critical’ period, but rather, opt for the term ‘sensitive’ period to identify that there are no exact cut-off points (Herschensohn, 2007).

      There is a wide range of evidence considered supportive of the CPH for L1 development, ranging from neurological studies regarding lateralization of brain function (Penfield and Roberts, 1959), to research on age-of-acquisition effects in deaf children learning to sign, showing that children who learn to sign at younger ages tend to have higher scores on various morphological tasks (Emmorey, 2002; Newport, 1990). There are also compelling and disturbing cases of children who mature in extreme isolation and who fail to develop native-like linguistic competence (Curtiss, 1977; Curtiss, Fromkin, Krashen, Rigler, and Rigler, 2004). Despite research showing these age-of-acquisition effects, the research community has not universally accepted the view that there is a critical period for language development – for either L1 or L2 learning (Bortfield and Whitehurst, 2001; Hakuta, 2001; Snow, 2002). This is because there are a number of variables which are typically confounded with age – such as the nature of the linguistic input (both in terms of quality and quantity), the motivation and aptitude of the learner, and whether the learner receives formal instruction in the L2 – where some researchers have argued that these other variables are more influential in determining differences between young and older learners (Hakuta, Bialystok, and Wiley, 2003).

      The evidence for a CPH in L2 acquisition is equally mixed and there are many excellent reviews that attempt to clarify some of the problems and findings inherent in this research (Birdsong, 1999; DeKeyser and Larson-Hall, 2005; DeKeyser, 2013; Hyltenstam and Abrahamsson, 2003; Moyer, 2004; Scovel, 2000; Singleton and Ryan, 2004; Muñoz and Singleton, 2011). As indicated earlier, there are many different variables that interact with the age of the learner and contribute to L2 outcomes. DeKeyser, Alfi-Shabtay, and Ravid (2010) also highlight the fact that the learning mechanisms underlying language development in childhood are themselves qualitatively different relative to those used in adult L2 learning. Children are argued to use largely implicit mechanisms, whereas older, adult learners, have been argued to use more explicit learning mechanisms. Minimally, therefore, age of acquisition seems to interact with a range of variables. Thus age, in and of itself, may indeed be associated with L2 outcomes, but age interacts with so many other variables that it is often difficult to tease apart the effect of age, independent of the interaction of age with these other variables. Age, as a variable, is very important to understand as fully as possible because it might be assumed that as long as the child is exposed to and learns the L2 before the offset of the CPH (whenever that is), they will acquire the L2 without difficulty. This is most striking in the context of children from immigrant families who are educated through the medium of an L2. Policy makers may believe that ethnic minority children simply need to be exposed to the majority language when they are young; thus obviating the need for the development of new policy which supports minority language learners or L2 provision to be incorporated in current policy to support L2 learning in these contexts. Correspondingly, understanding age effects does not necessarily translate into educational implications, in that the notion of ‘just teach earlier’ is ‘… almost certainly wrong’ (DeKeyser, 2013, p. 55). In other words, establishing that there are age-of-acquisition effects in L2 learning does not mean that we just have to teach or expose the child to the L2 when they are young to guarantee successful L2 outcomes.

      In this volume the older vs. younger question is not directly assessed, as comparisons will not be made with older learners. However, what is relevant for this volume is being able to describe the extent to which context is a predictive variable in determining ultimate attainment in the L2. For example, minority language learners are approximately the same age as some FL learners when they begin to learn their L2 in an educational setting. However, the amount and nature of the L2 learned in these two populations is dramatically different, underscoring that while age may be relevant, there are many other variables (for example, type and amount of exposure, type of educational experience) that predict L2 outcomes. Young FL learners, for example, typically do not have a sufficient amount of input (being in an input-limited L2 environment) and they therefore may not have the opportunity to use their implicit learning mechanisms (DeKeyser et al., 2010; Muñoz, 2006). Minority language learners, on the other hand, having had significantly more exposure to the L2, and who are learning their L2 in an input-rich environment, are more likely to be in a position to be able to invoke such mechanisms that could subsequently lead to better developed knowledge of L2. While both populations of L2 learners may be the same age, the context is the variable that leads to differential success in these two settings.

      All of the learners discussed in this volume are young learners and therefore may be operating in the ‘critical’ period, however that may be defined. If successful L2 learning is more likely to happen in childhood, then we can argue that all of the contexts in this volume could yield ‘successful’ L2 learners. Nonetheless, as Moyer stated, ‘Clearly age of exposure is but one factor in ultimate attainment, and not an especially informative one …’ (Moyer 2004, p. 138). She argues for the importance of ‘putting age effects in context’ and while for her context means either the psychological, social, or cognitive context (which undoubtedly are important contexts to consider in relation to age and L2 learning), another perhaps more obvious form of context considered in this volume is the linguistic environment and nature of experience the L2 learner has with the L2.