This work critically addresses the age debate in second language acquisition studies, presenting an in-depth study of factors that predict foreign accent. Quantitative and qualitative analyses confirm that cognitive, social, and psychological factors contribute to attainment, and that biological influences must therefore be considered alongside these essential aspects of learner experience.
Related collections and offers
About the Author
Read an Excerpt
Contextualizing Critical Period Inquiry
Coming into contact with a foreign language means hardship which usually brings euphoria quickly to an end, making it a short episode for the individual – and also for groups and nations.
Konrad Ehlich, 1994
[A second] language emerges through necessity ... because I want it, urgently want it, and because I urgently attend to it.
Peter Bichsel, 1995 [translation mine]
The Problems of Scope and Classification
The field of second language acquisition (SLA) has long sought relevant factors to explain differential attainment for early and late learners. Since the appearance of Lenneberg's Biological Foundations of Language (1967), the idea of a critical period for language learning has guided a great deal of the research on second language acquisition. Though Lenneberg made little specific mention of SLA, his critical period hypothesis has evolved into a full-blown theory for the field, often assumed to be a 'unitary account of non-native like outcomes' (Birdsong, 1999: 9). Still, the question remains why late language learners typically perform in notably 'non-native' ways. In the search for answers, we too rarely recognize the highly individual and complex nature of the endeavor. This is an especially salient issue given that SLA is most often an uninstructed, i.e. non-classroom, process among immigrants to foreign lands who face harsh social and economic conditions, possibly remaining culturally isolated for years.
Given this context of great challenge, many fall far short of a native speaker ideal, while others succeed beyond all expectations. By all accounts, we cannot yet explain either extreme of the success scale: entrenched fossilization and exceptional learning. The empirical evidence points to no specific faculty or mechanism in either the neurological or cognitive realm to explain exceptional performance in L2 (Obler, 1989; Schneidermann, 1991 as cited in Birdsong & Molis, 2001). Why, then, do more learners not attain to this level? In a recent review of cognitive research on bilingualism, Gonzalez and Schallert (1999) confirm that language processing is a multi-level task, incorporating knowledge from structural, semantic, discursive and cultural levels of language. This suggests that barriers to new knowledge reside not only in linguistic transfer, but in negative transfer of cultural knowledge as well. If so, transcending barriers to native-like acquisition requires adaptation on a number of complex levels beyond the neuro-cognitive realm. Narrowly focused explanations for SLA outcomes do little to advance an appreciation for such complexity.
By and large, critical period research has emphasized differences in phonological and morpho-syntactic performance between native and non-native speakers with little reference to the individual learner (Flege et al., 1995; Johnson & Newport, 1989; Olson & Samuels, 1982; Oyama, 1976; Patkowski, 1980; Snow & Hoefnagel-Hoehle, 1982). Production tasks and grammaticality judgments are typically isolated, and language attainment is described in structural terms, i.e. emphasizing explicit errors or grammaticality judgments. Age of onset is then offered as the explanation for non-native attainment, with few notable exceptions (Birdsong, 1992; Bongaerts et al., 1997; Ioup et al., 1994; White & Genesee, 1996). What is striking about this research is what it lacks: an account of individual factors in context.
With a legacy of early work on bilingualism in Canada, socio-cultural perspectives in SLA, by contrast, emphasize the importance of contextual factors in understanding second language (L2) development (Clément et al., 1993, Clément et al., 1994; ; Clément & Noels, 1992; Clément & Kruidenier, 1983; Gardner, 1983, 1985b; Gardner&Lambert, 1972; Gardner, et al., 1997; Hamers, 1994; Lambert, 1977; Norton Pierce, 2000; Taylor, 1977). These factors include: identity, assimilation, psychological distance, and even instruction for second language attainment (Clément & Kruidenier, 1983; Gardner, 1983, 1985b; Giles & Johnson, 1981; Labrie & Clément, 1986; Lambert, 1977; Schumann, 1978). A fundamental contribution of this collective work is the understanding of SLA, not as an abstract phenomenon, but as a process that essentially connects the individual to a community, where language is closely tied to one's sense of self – an important consideration for long-term integration both in linguistic and in cultural spheres (Lantolf, 2000; Lantolf & Pavlenko, 1995; Norton Pierce, 2000; see Pavlenko, 2002 for review).
Early work on German as a second language (Zweitspracherwerb Italienischer und Spanischer Arbeiter, Heidelberger Forschungsprojekt ) also highlights the importance of personal contact with native speakers (Dittmar & Rieck 1977, as cited in Loeffler, 1985; Meisel et al., 1981). While much of this research sets out in search of universal, internally-driven, mechanisms in uninstructed SLA, the range of background factors tested points to the primacy of interaction as key to developing morpho-syntactic fluency. Similar results come from grammatical-functional research, where pragmatic and semantic processing mode preferences are seen to result from interactive experience (Broeder, 1991; Dietrich, 1995; Dittmar, 1992; Pfaff, 1987, 1992 – see Moyer, in press, for summary). If interactive experience is critical to long-term attainment, then opportunities for such engagement in the target language community must be accounted for.
Equally problematic is the fact that adult language performance is frequently described as substandard, in spite of an early emphasis on creative construction in interlanguage (Ellis, 1985; Selinker, 1972 and 1992). In essence, the focus on documented patterns such as simplification and free variation has perpetuated stereotypes of foreigner talk through an emphasis on reduced interlanguage features. Fennell summarizes the unfortunate result for German as a second language:
... All such [classificatory] behavior, whether exhibited by linguists, pedagogues, officials, or everyday people, serves to ethnicize both the minorities, and concomitantly, the Germans themselves ... Here we return to Bourdieu's contention that the act of naming and classifying is a very strong exercise in power ... Comparing Gastarbeiter German with standard German has stacked the cards against foreign workers, begging the labels 'inadequate,' and ultimately 'inferior.' (Fennell, 1997: 92f)
Indeed, it is all too rare to focus on what learners are doing well. Marinova-Todd et al. suggest that 'rather than focusing on the low probability that adults will acquire fluency in L2s ... it is more productive to examine the factors that typically lead to native-like proficiency ...' (Marinova-Todd et al., 2000: 10). To this end, the influence of age should be examined in terms of how it may relate to availability of authentic input, instruction, interaction, etc., as well as to the learner's intention to attain to a certain level. These aspects of late language learning can be ascertained only through careful measure of L2 experience.
Second language experience effectively constitutes a 'black box' for SLA; many of the relevant mechanisms may be unobservable. Yet, in spite of these empirical challenges, the nature of late language acquisition suggests the need to move beyond simplistic analyses and interpretations. To understand how new knowledge is acquired, we must discover how learners actively participate in the learning process itself. This requires an appreciation of the socio-psychological aspects of learning approach, including an examination of motives, beliefs, and sense of linguistic and cultural identity (Liebkind, 1999; Norton Pierce, 1995). Even the target language community's expectations for non-native speaker assimilation may play a significant role in how the learner engages with available resources and negotiates learning and feedback opportunities against the backdrop of perceived expectations for 'success' (see Genesee et al., 1983; Pavlenko, 2002). These issues deserve deeper investigation in applied linguistics research in general.
The problems of scope and classification may thus be due in part to fundamental incomparability between research on what could be termed 'privileged' learning, typical of classroom foreign language instruction, and that of naturalistic language acquisition. Social, instructional, and psychological conditions cannot be compared for such disparate experiences. The social and political status of immigrants affords them few opportunities to receive supportive feedback on their developing interlanguage (see Norton Pierce, 1998). Moreover, their experience with the target language may accumulate in a somewhat haphazard, or reactive, manner. This is quite another matter for the classroom learner, who is essentially guaranteed a comfort zone for linguistic revision and gradual L2 identity development. These internal and external differences are not addressed by structural descriptions of interlanguage development, much less by isolated statistical analyses. It is therefore necessary to empirically integrate critical period perspectives with data on individual access to input and instruction (see Moyer, 1999).
In this book, I will integrate and critically review current literature on age effects, and add new data to the discussion as I test the biological basis of the critical period hypothesis relative to the impact of socio-psychological and experiential variables. This integrated methodology relies on descriptive and statistical analyses of relationships between multiple biological, psychological, and experiential factors. The interpretation of these relationships is then strengthened through ethnographic data on the individual's perspective on the process. Accordingly, the significance of these factors becomes clearer as we understand how the learner responds and adapts to the target language and its community over time. Given the context for most SLA today, a more holistic understanding of personal and communal opportunities to build L2 experience is appropriate. To this end, the data in this study focus on immigrants to Berlin.
Contextualizing SLA: Germany as an Empirical Framework
For immigrants who must learn a second language, the complex conflation of maturation, socio-psychological orientation, and cultural identity is very real. Present-day Germany provides an excellent empirical basis for exploring issues of identity and milieu in the SLA process. Currently, Germany claims about 9% (7.3 million) of its population as 'foreign' (not including Aussiedler, or ethnic Germans seeking asylum from neighboring Eastern European countries). Fluency in German is critical to the potential of these people for long-term assimilation and acceptance. Skutnabb-Kangas points out that, while the mother tongue may be necessary for psychological, cognitive, and spiritual survival, the official language of the state is needed for 'social, economic, political, and civil rights ...' and may determine further education, job prospects, and participation in wider society (Skutnabb-Kangas, 1999: 58). Acknowledging the juxtaposition of forces toward social integration as well as isolation, the adult immigrant's difficulties with acquiring German go beyond possible cognitive and neurological constraints. By exploring the experience of the learner, we may better understand how language acquisition proceeds, or fossilizes, according to predictable influences.
With this context in mind, the current study (see next section) presents linguistic and non-linguistic data gathered from immigrants to Germany from the US, Britain, France, Russia, Poland, Slovakia, and Turkey. The analysis of language attainment is thus contextualized here through a broad accounting of experiential and socio-psychological influences on language attainment. Furthermore, these immigrants provide an interesting contrast to the uninstructed data presented in earlier studies depicting L2 learners/acquirers, with little or no access to formal, or instructional, experience in the target language. Because of their combination of both formal and informal exposure, these immigrants have had what might be considered an optimal acquisitional experience. They are also well-educated, and as such, social separation is perhaps less dramatic for them than for many Gastarbeiter and Asylbewerber. They also tend to see themselves as actively engaged with the language and its culture for both personal and professional reasons. Given these potential advantages, these learners may 'defy' the critical period, in the sense that age effects may be weaker when external opportunities (i.e. access to L2) are greater, and internal orientations are relatively positive.
The numerous studies on immigrant language acquisition have revealed surprisingly little regarding such personal and communal issues. The history of Gastarbeiterlinguistik goes back to the mid-1970s with the ZISA (Zweitspracherwerb Italienischer und Spanischer Arbeiter) and HFP (Heidelberger Forschungsprojekt) projects (for discussion, see Fennell, 1997; Larsen-Freeman & Long, 1991).
Objectives of the many published studies from these, and subsequent, databases include seeking evidence for: (1) universal developmental pathways to syntactic and morphological acquisition; (2) the temporary (or permanent) nature of reduced varieties, i.e. so-called Pidgindeutsch or Foreign Worker German; and (3) potential access to Universal Grammar in the acquisitional process. Overwhelmingly, the emphasis has been on structural descriptions of morphosyntactic simplification. A number of studies have led to hypotheses concerning universal routes of development, and the role of instructional intervention (see Jordens, 1996 for discussion). Recent research has focused on areas of transfer from L1 features (Felix, 1980; Kaltenbacher, 1994; Meisel, 1983; Pfaff, 1984), processing strategies, stages in syntactic development and verbal movement patterns (Beck, 1998; Eubank, 1992; Vainikka & Young-Scholten, 1996), and potential access to Universal Grammar for specific features (Clahsen & Muysken, 1986; Felix 1991). More recently, analyses of developmental strategies have focused on possible 'dual mechanisms' for language learning, combining memory and analysis in the acquisition of regular and irregular morphological markers (Clahsen, 1997; Elsen, 1997).
To be sure, this ongoing descriptive and theoretical work has provided rich data and thought-provoking hypotheses concerning the nature of early versus late language learning. However, it acknowledges social and cultural concerns of immigrant learners in minimal ways. Moreover, the resulting theoretical interpretations have been criticized, given the small number of participants in many studies, the relative paucity of methods reporting, and the vastly different access among participants to interactive learning opportunities (see Hudson, 1993; Jansen, 2000; Mellow, 1996; Rogers, 1995).
Alternative studies from sociolinguistic and functionalist perspectives have expanded our understanding of immigrant language acquisition and the social circumstances of guest workers, especially in terms of the purpose and function of their interactions in German (Broeder, 1991; Frischherz, 1997; Jordens et al., 1989; Perdue, 1990; Pfaff, 1985, 1987, 1992; Röhr-Sendlmeier, 1990). Summarizing results from a number of studies, robust patterns emerge across a range of methods. Factors correlating to higher levels of linguistic attainment include: motivation to integrate into German culture; consistent interaction in German beyond survival situations, sustained peer relationships with native Germans, institutionalized integration assistance; and participation in German education programs (Barkowski et al., 1976; Buss, 1995; Esser, 1982; Götze & Pommerin, 1988; Horn, 1996; Lalleman, 1987; Pfaff, 1985; Worbs, 1995). For early German language acquisition (GLA), parental expectations and support for education are correlated to success in language attainment and overall educational achievement (Götze & Pommerin, 1988; Röhr-Sendlmeier, 1990). These findings confirm that consistency and support in the learner's interactive community are key to ultimate attainment. Such close examinations of opportunities for L2 contact over time are rare in the empirical research.(Continues…)
Excerpted from "Age, Accent and Experience in Second Language Acquisition"
Copyright © 2004 Alene Moyer.
Excerpted by permission of Multilingual Matters.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
1 Contextualizing Critical Period Inquiry, 1,
2 Accounting for Universal and Individual Factors in Ultimate Attainment: Focus on Phonology, 14,
3 Verifying the Relative Strength of Maturation, L2 Experience and Psychological Orientation: The Quantitative Findings, 50,
4 Understanding Identity, Intention, and Opportunities for L2 Contact: The Qualitative Findings, 98,
5 Conclusions and Proposals for Future Research, 138,
Appendix 1: Umfrage/Survey, 149,
Appendix 2: Linguistic Tasks, 152,
Appendix 3: Rater Survey, 154,
Appendix 4: Rating Feedback Sheet, 155,
Appendix 5: Semi-Structured Participant Interview, 156,