The Social Construction of Age

Adult Foreign Language Learners

Patricia Andrew

• Publisher: Multilingual Matters
• Year: 2012
• ISBN:
  9781847696168
• Topic: Sprachen & Linguistik
• Subtopic: Soziolinguistik

This book explores the social construction of age in the context of EFL in Mexico. It is the first book to address the age factor in SLA from a social perspective. Based on research carried out at a public university in Mexico, it investigates how adults of different ages experience learning a new language and how they enact their age identities as language learners. By approaching the topic from a social constructionist perspective and in light of recent work in sociolinguistics and cultural studies, it broadens the current second language acquisition focus on age as a fixed biological or chronological variable to encompass its social dimensions. What emerges is a more complex and nuanced understanding of age as it intersects with language learning in a way that links it fundamentally to other social phenomena, such as gender, ethnicity and social class.

Part 1

Framing Age as Socially Constructed

The first part of the book analyzes the treatment of age in second language acquisition (SLA), sociolinguistics and cultural studies, and provides the conceptual foundation for the empirical study described in Part 2. Chapter 1 examines the broad literature on the Critical Period Hypothesis (CPH) as the primary source of information on age issues in SLA. After considering the current state of the debate, I argue that the CPH for SLA has not focused on the age factor in adults in any important way, at best providing a limited account of linguistic outcomes from a biological perspective. The impact of age as a social dimension and the nonlinguistic outcomes of adult language learning lie beyond the scope of the CPH and are better understood in light of sociocultural perspectives on SLA and by taking a social constructionist approach to age.

In Chapter 2, the research on which this book is based is positioned as a contemporary discourse-oriented sociolinguistic study. The key features of social constructionism that are relevant to the study of age from an interactional perspective are discussed. By tracing the links between contemporary sociolinguistics and social constructionism, I contend that they provide a coherent framework for the study of age and other social dimensions.

Following this, Chapter 3 takes up the question of how age is socially constructed in present-day Western society. I discuss the principal discourses that frame the way people give meaning to the experience of aging and create their age identity. Additional consideration is given to the existence of competing discourses which contribute to the multiple, changing and potentially contradictory nature of these processes in themselves and in relation to the other life experiences and identities with which they intersect. The fundamental role of narrative in discursive interaction and identity construction is a recurring theme throughout this chapter, furnishing a valuable link to the discussion of the empirical work which follows in the second part of the book.

1 The Age Factor and Second
Language Acquisition

Introduction

To ascertain what part age and the age identity of adult language learners play in the process of acquiring and using English as a second language (L2), it is imperative to begin by considering how age has traditionally been treated in the field and what has been learned up to now. Given that questions regarding the importance of age in SLA have been restricted almost exclusively to the existence and nature of possible biological constraints on learning, a substantial part of this chapter is devoted to a discussion of the CPH. Following that, I offer an assessment of the existing orientation to the issue of age and SLA and suggest an alternative direction to take in exploring this topic. It involves adopting a sociocultural approach to SLA at the outset, a theoretical perspective in greater accord with the focus of the book. The sociocultural strand of SLA is the subject of the final part of the chapter and sets the tone for those that follow.

The Critical Period Hypothesis for Second Language Acquisition: A Critique

Any discussion of the age factor in SLA must necessarily give major consideration to the CPH. This is the most clearly articulated theory concerning age constraints on first language (L1) acquisition, and reflection on its applicability by extension to SLA logically precedes the exploration of alternate explanations of the relationship of age to the acquisition process. Questions of whether we are neurologically programmed for language acquisition at a defined period of time during our childhood, whether it is possible to acquire an L2 after this critical period has ended, what kinds of limitations might be entailed and what the ramifications are for teaching and learning languages comprise some of the issues of crucial importance in understanding the role that age plays in SLA.

The controversy surrounding this theory has been an animated one among SLA researchers in the 40 years since Lenneberg (1967) posited the existence of a critical period for language acquisition in Biological Foundations of Language.¹ According to the CPH, children have a special capacity for language development that is supported by an innate language-learning mechanism. The critical period ends around puberty, after which time the innate mechanism is no longer available and language development is virtually halted. This phenomenon is ascribed to the loss of neural plasticity of the brain and the establishment of hemispheric lateralization.² Central to Lenneberg’s notion of age constraints on language acquisition is its biological bases, alluded to in the title of his book, namely, aspects of the child’s neural structure involved in a fundamental way in the development of language. These are the key elements of the hypothesis as originally set forth by Lenneberg.

The CPH and L1 Acquisition

Evidence of the existence of a critical period begins with the commonplace observation that all normal children become fully competent in their L1, following a similar timetable through analogous developmental stages. Comparable critical periods are characteristic of different kinds of behavioral development in nonhuman species as well.

Studies of language recovery in adults and children who have suffered brain lesions indicate that children are clearly advantaged over adults. In addition, research into delayed L1 acquisition has tended to give substantial support to the concept of a critical period, in the sense that the linguistic competence ultimately attained by the subjects under study has proven to be deficient. Studies of feral children and others deprived of language input in early childhood as a result of abuse or neglect have offered valuable examples of the defective language that results (Curtiss, 1977, 1982, 1988). Another source of information has come from studies of congenitally deaf subjects whose first contact with American Sign Language, a fully functional language, occurred at different ages. Language development of Down’s syndrome children has also proved a useful way to look at delay in L1 acquisition. However, possibilities for carrying out this type of empirical work have necessarily been limited by ethical considerations and by the fact that incidences of delayed L1 acquisition are relatively rare.

While the neurobiological basis of L1 acquisition is largely accepted, the CPH as originally set forth has come under closer scrutiny as more is learned about the structure and working of the brain. The ensuing debate on issues such as the age at which the critical period begins and ends, or when hemispheric lateralization occurs and whether this is significant, has given rise to alternate versions of the CPH. Some researchers suggest that, given the complex nature of language, there may be multiple critical periods for linguistic competence, or that some aspects of language proficiency may be subject to critical periods but others are not.

Furthermore, in view of research findings with delayed L1 learners, the strong version of the CPH, which holds that no learning would be possible if a child were not exposed to language before a certain age, usually understood to be puberty, has tended to give way to a weak version in which, according to Long:

some learning would be possible beginning after that age, but that native-like abilities would be unattainable, and that the course of development would become more irregular and would fall further short of native levels the later the age of onset. (Long, 1990: 256–257)

The acceptance of the CPH for L1 acquisition, in either its strong or weak form, of necessity precedes any discussion of the more contentious issue of a neurobiologically based critical period for SLA, for it is unlikely that the latter could exist if the former did not. The foregoing considerations provide the necessary contextualization for looking at the CPH as applied to SLA, a task of more immediate concern here.

The CPH and Second Language Acquisition

Whereas L1 acquisition normally leads to full proficiency, SLA rarely does, and instead is characterized by a broad variation in outcomes. Divergent views regarding the potential of the CPH to explain this latter phenomenon have given rise to multiple ways of addressing a wide gamut of concerns about very specific aspects of language proficiency. This situation complicates a straightforward treatment of the CPH since there is only a minor area of overlap in the studies. Consequently, the present discussion will proceed along very general lines, with an attempt to tie in the various contributions where they seem pertinent.

The issues

Two broad claims about SLA furnish a point of departure:

• L2 learners rarely attain overall native-like language proficiency.

• Younger L2 learners are generally more successful than older ones.

On the whole, both these beliefs have a certain limited acceptance. Points of divergence derive primarily from considerations of what role the CPH plays, and whether it can be applied to all aspects of language proficiency and in all SLA contexts.

The fundamental questions to be addressed, then, are:

• Can the CPH account for the typically unsuccessful results of most L2 learners in achieving full mastery of the language?

• Does the CPH explain the better long-term achievement of younger learners in an L2 with respect to that of older learners?

A negative answer to either of the foregoing leads to the following question:

• What alternate theories are offered to account for these phenomena?

An affirmative response, albeit a qualified one, generates further questions, such as:

• How does the CPH explain the variability in SLA outcomes?

• To what extent can the CPH provide the sole explanation for this?

• Do distinct critical periods exist for different aspects of the SLA process?

• What other factors interact with the neurobiological ones in SLA, and to what extent?

These questions, in turn, will produce additional ones.

The viewpoints

The strong version of the CPH, that language acquisition can take place exclusively during the period of human development from infancy through puberty, maintains that once this period has ended, it is no longer possible, or exceptionally difficult, to learn a language. With regard to SLA, there is less support now for this extreme position, namely, that there exists an abrupt moment which ends any further language development. Were this true, postcritical period learners should not be able to learn an L2, patently not the case. Instead, supporters of a strong version of the CPH hold that after the close of the critical period, as a consequence of the loss of neural plasticity in the brain and because the biologically endowed faculty for language is no longer available, L2s are learned only with great difficulty. They point to findings that demonstrate a general tendency for younger learners to do better than older learners in the acquisition of an L2, claiming that these results are attributable precisely to the accessibility of the innate language-learning mechanism to younger learners but not to older ones (Johnson & Newport, 1989; Long, 1990; Oyama, 1976, 1978; Patkowski, 1980).

However, a number of research studies have reported findings at variance with these in specific language domains, thereby casting doubt on the soundness of the strong version of the CPH. Studies reporting incidences of highly successful late-starting learners have raised serious questions about the existence of maturational constraints on SLA (Abu-Rabia & Kehat, 2004; Bongaerts, 1999, 2005; Ioup et al., 1994; Van Boxtel et al., 2005). In view of a growing body of counter-evidence and as a consequence of an increasing awareness of the inherent complexity of the acquisition process and the impossibility of tracing the variation in outcomes to a single origin, a weak version of the CPH now has considerable credence among SLA researchers. In this version of the hypothesis, the term ‘sensitive period’ is often used in preference to the more rigidly deterministic ‘critical period’, denoting an optimal interval of time in which circumstances are favorable for developing a particular type of behavior, and after which efficiency gradually declines.

While few researchers completely reject any biological basis of observed age-related differences in SLA outcomes, the contention that proficiency in an L2 is contingent entirely or principally upon maturational factors implies causality for which, Bialystok (1997) insists, evidence must be provided unequivocally linking the two. This standpoint, in essence, is that the burden of proof is on those affirming the validity of the CPH, and not the other way around.

The evidence

Age-related effects on SLA are generally studied in terms of either the rate of acquisition or ultimate attainment, a distinction with critical implications for interpretation. In the first case, the underlying assumption is that a faster rate of acquisition demonstrates a greater facility for learning languages. In the second, the supposition is that a higher degree of achievement in the final outcome corresponds to a greater fulfillment of the potential for language learning. Both types of focus have been used to look at age of onset (AO)³ or, less frequently, length of residence (LOR), as they relate to some measure of language proficiency. Interestingly, rate-of-acquisition studies, most of which were carried out in the 1970s and 1980s, found that older learners performed better than younger ones on measures of morphology, syntax and pronunciation. Long (1990) discusses this research in his survey of L2 investigation on age, and concludes that the initial advantage for adults is a transitory one. The consensus is that, in the long run, children outperform adults. Only two major studies of the rate of acquisition (Muñoz, 2006; Slavoff & Johnson, 1995) have been reported since the 1980s.

Ultimate attainment studies are considered more important in investigating maturational constraints on L2 development because of their longterm nature. Most of the earlier studies concluded that ultimate attainment declines as AO increases. These findings are interpreted as corroborating the strong version of the CPH. The most frequently cited are those of Oyama (1976, 1978), Patkowski (1980) and Johnson and Newport (1989). These studies found an earlier AO to be a strong predictor of greater proficiency in different aspects of L2 development.

Investigation that has attempted to disprove the broader claims of these earlier studies focuses on cases of highly successful older learners, adults who have achieved near-native proficiency in one or more language domains, despite a later AO. Studies carried out by Birdsong (1992), White and Genesee (1996) and Ioup et al. (1994) found subjects who performed at the same level as near-native speakers or native speakers of French, English and Egyptian Arabic, respectively. These studies have primarily explored the grammatical features of language. Several later studies of age looked at English as a third language in bilingual communities in the Basque Country and in Catalonia (García Mayo & García Lecumberri, 2003). Older learners were found to have an advantage over younger ones in formal instruction contexts in different language domains, including pronunciation. Another study reported attainment of native-like proficiency in Dutch by postpuberty learners, even those with a typologically distant L1 (Van Boxtel et al., 2005).

The question of phonetics and phonology is perhaps even more important to the discussion of adult learner success than morphosyntax, for the reason that pronunciation is generally believed to be more susceptible to age constraints than other aspects of language, presumably because it involves neuromuscular skills. The widely held view is that late learners will have a more marked foreign accent than their younger counterparts. In separate studies that corroborate this standpoint, both Flege et al. (1995) and Moyer (1999) showed that age correlates negatively with phonological performance. Their research is representative of other studies that have found an age-related decline in end-state attainment in pronunciation, although not necessarily in other language domains. Singleton points out that ‘the earlier an immigrant arrives in the host country and begins to be exposed to its language the more likely he/she is to end up sounding like a native’ (Singleton, 1995: 8–9).

Nonetheless, some studies, suc...