acquisition and that humans were predisposed to learn language more successfully in early childhood than later on in their lives. Lenneberg’s CPH was temporally contiguous with Chomsky’s work arguing that certain abstract aspects of linguistic knowledge were innate, given to us at birth (Chomsky, 1965; Hauser, Chomsky, and Fitch, 2002; Chomsky, 2007) – a Universal Grammar (UG) enabling children to readily acquire linguistic competence in the languages to which they are exposed. In contrast to the ease with which children learn their first languages, older individuals have long been identified as less or unsuccessful language learners in comparison to children. Adults in particular, in attempting to learn a second or foreign language, have been shown to have particular difficulties and typically fail to achieve native-like competence in the L2. Some researchers have even argued that the rare cases of successful adult foreign language (FL) learning are ‘pathological’ in the same sense as the rare failures to acquire a first language in children (Selinker, 1972). The very fact of being old(er) has been argued to explain the apparent difficulty adults have in learning an L2. The view is that an older learner (i.e. beyond Lenneberg’s critical period) will necessarily struggle to acquire a second language because the window of opportunity has closed and access to Chomsky’s UG is lost3. These views suggest then, that the critical variable in determining ultimate success in learning languages is the age of the learner – the younger the better.
If children learn languages easily, at least, more easily than adult (older learners), then it follows that they will learn second languages easily, and more easily in childhood than in adolescence or adulthood. In the past few decades the facility with which children can become bi- or multilingual has gained increasing prominence, both from theoretical and more applied perspectives. From a theoretical perspective, a more comprehensive understanding of how language is acquired (Ambridge and Lieven, 2011), the extent to which language development is a child-internal process (inside–out) relative to a process driven by the environment in which the child develops (outside–in) (Hirsh-Pasek and Golinkoff, 1996), and the extent to which L1 acquisition theories can account for bi- or multilingualism (Ellis, O’Connell, and Römer, 2013; Martin and Ellis, 2012; Ellis and Sagarra, 2011), has led to an increase in research investigating young children’s L2 development. From more applied perspectives, increased global migration (Davis, D’Oderico, Laio, and Ridolfi, 2013), developments in educational policy and practice (Baker, 2011; García, 2009), and greater intercultural sensitivity, recognizing the need to preserve indigenous languages at risk of dying out (Cortina, 2014), have all been forces that have led to greater interest in L2 learning in childhood. However, the context in which a child becomes bi- or multilingual can vary substantially, therefore, understanding L2 learning in children necessarily entails a better understanding of the effect of context on child L2 outcomes. Children might be raised in homes or societies where multiple languages are regularly spoken and where there is consistent exposure and opportunities to interact with more than one language, as in cites such as Montréal, Barcelona, and Mumbai. Alternatively, children might receive instruction in a foreign language at some point in their formal education. This latter context of foreign language instruction in formal education is becoming a more prevalent language learning context for young learners throughout the world (see Chapter 6). As there is variability in the contexts in which children could become bilingual, there is corresponding variability in the extent to which a young child actually does become bilingual, and/or the degree of proficiency in each language. It is timely, therefore, to examine the research literature with respect to child L2 learning through formal education in the early, formative years.
The purpose of this volume is twofold. The first aim is to bring together in one book a review of the major themes and issues that have emerged from research investigating the development of L2 learning in childhood, across a range of contexts. Many of these contexts are in a formal education setting4. Each chapter, therefore, presents a review of a given context. These reviews, however, are selective not systematic. Consequently, there will be omissions both in terms of level of detail provided in certain discussions and also in terms of which themes or issues are included. I have included discussions of those themes that I believe are most salient, relevant, and informative for the purposes of this volume. Despite the selective nature of these reviews, each chapter should minimally provide a good beginning for the interested reader who wishes to delve in greater detail into these areas and their related research.
The second major aim of this volume is to use these reviews to investigate the notion that ‘younger is better’ for L2 learning. As mentioned earlier, there is a prevailing view that younger children learn languages more easily than older learners. Increasingly, this ‘younger is better’ idea has led educational policy makers to lower the age at which children begin learning a second language as part of the primary curriculum. For example, in the UK Modern Foreign Language (MFL) is being introduced as a curricular subject in 2014. To be more precise, it is being re-introduced, because MFL used to be part of the primary curriculum but was removed in part due to an influential (but methodologically flawed) study which purported to show that older learners actually were better learners than younger ones (Burstall, 1975). Consequently, the then government of the UK decided to remove MFL from its already overcrowded primary curriculum. The current UK government, however, has more recently decided to reintroduce MFL to primary school learners and provide taught foreign language instruction as young as seven years old. The stated aims for this decision include ‘younger children learn languages more easily’ (Department for Education, 2013, p. 4); ‘young children have a natural disposition to learn languages’; ‘… pupils will benefit from a more global outlook and enhanced career prospects’; ‘… children benefit from learning a foreign language …’ (Department for Education, 2012, p. 1). We can see, therefore, that there are underlying beliefs about the benefits of learning a foreign language at primary school in these governmental documents, many of which centre around the notion that children have a propensity to ‘pick up’ languages more easily in the younger years. This belief is neatly encapsulated in a quote from Tony Blair in 1999 when he was Prime Minister of the UK. At one point in delivering a speech he said ‘Everyone knows that with languages the earlier you start, the easier they are’ (Sharpe, 2001). It is not clear, however, whether and to what extent these beliefs concerning ‘younger is better’ hold across different contexts in primary school-aged learners. In order to tackle this issue, therefore, an examination of the research reviewed in each chapter of this volume will help identify the extent to which these different cases of child L2 learning is in fact ‘easy’. But first, it is worthwhile taking a closer look at the ‘younger is better’ idea, as all the contexts included in this volume represent instances of L2 learning in childhood, a time at which many people believe language learning comes naturally. To what extent is this belief supported by evidence?
Age and L2 learning in primary school children
It is important to be clear that the focus of this book is not on evaluating the validity of the Critical Period Hypothesis (CPH), nor indeed is it intended to identify the most effective age at which to learn an L2. Nonetheless, the age of the learners is an important variable when discussing L2 outcomes and is obviously related to key discussions in the literature about both the CPH and issues of when learners should be exposed to, and taught an L2. Indeed, as indicated earlier, the notion that ‘younger is better’ seems to underlie a range of different policy decisions with respect to child L2 learning in primary education.
The Critical Period Hypothesis (Lenneberg, 1967) has been widely discussed in the context of both L1 and L2 development (see Bialystok and Miller, 1999; Birdsong, 1999; DeKeyser, 2000; Singleton, 2005; Nikolov, 2009; DeKeyser, Alfi-Shabtay, and Ravid, 2010; Muñoz and Singleton, 2011; DeKeyser, 2013). There is considerable variability with respect to how the CPH has been defined and the types of learner and task types that have been researched in respect of the CPH. Consequently, the results are complex and somewhat mixed (Bialystok and Miller, 1999; Singleton, 2005). One of the issues that has been difficult for researchers to pin down is the exact onset and offset of the CPH – when the CPH begins and ends. Lenneberg (1967) originally proposed the onset was around two years old, with the offset around 12 years (corresponding to puberty). This claim initiated a considerable amount of research investigating the extent to