APPLIED LINGUISTICS Volume 29 Number 4 December 2008 CONTENTS Articles Contesting ‘Language’ as ‘Heritage’: Negotiation ...
24 downloads
439 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
APPLIED LINGUISTICS Volume 29 Number 4 December 2008 CONTENTS Articles Contesting ‘Language’ as ‘Heritage’: Negotiation of Identities in Late Modernity ADRIAN BLACKLEDGE and ANGELA CREESE Routine Trouble: How Preschool Children Participate in Multilingual Instruction ¨ RK-WILLE´N POLLY BJO Symmetries and Asymmetries of Age Effects in Naturalistic and Instructed L2 Learning CARMEN MUN˜OZ Textual Enhancement of Input: Issues and Possibilities ZHAOHONG HAN, EUN SUNG PARK and CHARLES COMBS
533
555
578 597
The Cultural Productions of the ESL Student at Tradewinds High: Contingency, Multidirectionality, and Identity in L2 Socialization STEVEN TALMY Language Ecology in Multilingual Settings. Towards a Theory of Symbolic Competence CLAIRE KRAMSCH and ANNE WHITESIDE
645
A Register Approach to Teaching Conversation: Farewell to Standard English? ¨ HLEMANN CHRISTOPH RU
672
Form-focused Instruction in Second Language Vocabulary Learning: A Case for Contrastive Analysis and Translation BATIA LAUFER and NANY GIRSAI
694
FORUM The Significance of First Language Development in Five to Nine Year Old Children for Second and Foreign Language Learning ROBERT VANDERPLANK
717
REVIEWS H-Dirksen Bauman, Jennifer Nelson, and Heidi Rose: Signing the Body Poetic. Essays on American Sign Language Literature MARION BLONDEL
723
619
Ivana Markova´, Per Linell, Miche`le Grossen, and Anne Salazar Orvig: Dialogue in Focus Groups: Exploring Socially Shared Knowledge GREG MYERS Simon Gieve and Ines Miller: Understanding the Language Classroom ASSIA SLIMANI-ROLLS Sandra Mollin: Euro-English: Assessing Variety Status. Jennifer Jenkins: English as a Lingua Franca: Attitude and Identity LUKE PRODROMOU
726 728
732
NOTES ON CONTRIBUTORS
737
INDEX TO VOLUME 29
741
Applied Linguistics 29/4: 533–554 ß Oxford University Press 2008 doi:10.1093/applin/amn024 Advance Access published on 21 July 2008
Contesting ‘Language’ as ‘Heritage’: Negotiation of Identities in Late Modernity ADRIAN BLACKLEDGE and ANGELA CREESE University of Birmingham, UK
with TAS° KIN BARAC¸, ARVIND BHATT, SHAHELA HAMID, LI WEI, VALLY LYTRA, PETER MARTIN, CHAO-JUNG WU and DILEK YAG˘CIOG˘LU In this paper we question key terms which appear frequently in discussions of language teaching and learning: ‘language’ and ‘heritage’. The paper draws on empirical data from one of four linked case studies in a larger project funded by the UK Economic and Social Research Council (ESRC), ‘Investigating Multilingualism in Complementary Schools in Four Communities’ (RES-000-23-1180). In our analysis we argue that the relationships between ‘language’ and ‘heritage’, far from being straightforward, are complex in the way they play out in classroom interactions. The data raise a number of questions in our attempts to understand how the linguistic practices of students and teachers in Bengali schools are used to negotiate young people’s multilingual and multicultural identities. First, participants articulate attitudes and values which raise questions about what constitutes ‘language’. Second, participants express views and attitudes, and perform interactional practices, which raise questions about what constitutes ‘heritage’. Our analysis finds that multilingual young people in complementary school classrooms use linguistic resources in sophisticated and creative ways to negotiate subject positions which appear to contest and subvert schools’ attempts to impose upon them ‘heritage’ identities (Creese, A., A. Bhatt, N. Bhojani, and P. Martin. 2006. ‘Multicultural, heritage and learner identities in complementary schools,’ Language and Education 20/1: 23–44).
INTRODUCTION This paper reports part of a research project which investigates how the linguistic practices of students and teachers in Bengali schools in the United Kingdom are used to negotiate young people’s identities. The analysis of data presented here raises questions in two areas. First, participants appeared to differ from each other in their views of what constitutes (a) language, and what ‘counts’ as (a) language. Second, we are led by our participants to consider the relationships between ‘language’ and ‘heritage’, and the role of teaching and learning language(s) in the reproduction of that ‘heritage’. The complexity and sophistication
534
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
of the young people’s responses to the teaching and learning of ‘language(s)’ led us to question by what means ‘the legacy of history is appropriated’ (Bourdieu 2000: 151). Is ‘heritage’ straightforwardly reproduced where the learner is born to linguistic, social, and environmental norms which are typical of urban late modernity, whereas the ‘heritage’ was associated with rural poverty? These questions about social reproduction for young people in the United Kingdom raise broader questions about what constitutes ‘(a) language’, and what counts as ‘heritage’ in late modernity. Before discussing the data, we reflect theoretically on these questions relating to ‘language’ and ‘heritage’.
‘LANGUAGE’ Heller (2007) proposes four sets of concepts in the critical analysis of languages in society. First, she argues that rather than treating notions of ‘community’, ‘identity’, and ‘language’ as though they were natural phenomena, they should be understood as social constructs. Specific or single categorizations therefore cannot be attached to an individual based on their ‘ethnicity’, or ‘language’. Second, Heller refers to the work of Giddens (1984) to consider language as a set of resources which are socially distributed, but not necessarily evenly. The third set of concepts holds that this uneven distribution of resources is the product of political and economic processes, enabling us to ask questions about what linguistic resources are assigned what value, and with what consequences (Gumperz 1982). The final set of concepts considers the discourses which inscribe value (or its lack) to particular linguistic forms and practices. In summary, Heller views language(s) as: sets of resources called into play by social actors, under social and historical conditions which both constrain and make possible the social reproduction of existing conventions and relations, as well as the production of new ones. (Heller 2007: 15) What Heller calls ‘the messiness of actual usage’ (2007: 13) can only be understood in relation to histories, power, and social organization. Conversely, structural analysis must include accounts of actual linguistic practices, which at times may differ from those we might expect. Our multilingual participants’ beliefs, attitudes, and practices in relation to ‘language’ resonated with Garcia’s (2007: xii) account that languages are not hermetically sealed units. The linguistic practices of Garcia’s students in New York bore very little relation to the ‘standard English’ of school texts or the ‘standard Spanish’ that was supposed to be linked to their ‘identity’. Rather, our data suggest, in line with the recent proposition of Makoni and Pennycook (2007), that the notion of languages as separate, discrete entities, and ‘countable institutions’ (2007: 2) is a social construct. Bourdieu (1991) proposed that: language is itself a social artefact invented at the cost of a decisive indifference to differences which reproduces on the level of
ADRIAN BLACKLEDGE and ANGELA CREESE
535
the region the arbitrary imposition of a unique norm. (Bourdieu 1991: 287) Makoni and Pennycook argue for a critical historical account which demonstrates that, through the process of classification and naming, languages were ‘invented’ (2007: 1). They add that, in direct relation with the invention of languages, ‘an ideology of languages as separate and enumerable categories was also created’ (2007: 2). Makoni and Pennycook point in particular to the naming of languages such as ‘Bengali’ and ‘Assamese’ as the construction of ‘new objects’ (2007: 10). Thus languages cannot be viewed as discrete, bounded, impermeable, autonomous systems. Our research participants, all at first glance of the same ‘ethnic’ and ‘linguistic’ group, not only disagreed with each other about what constituted a ‘language’, they also disagreed with each other about where a ‘language’ began and ended, and about the value that could be assigned to a particular set of linguistic resources. Makoni and Pennycook propose that such ‘local knowledge’ is crucial to our understanding of language: We are arguing for an understanding of the relationships between what people believe about their language (or other people’s languages), the situated forms of talk they deploy, and the material effects—social, economic, environmental—of such views and use. (Makoni and Pennycook 2007: 22) This interrelationship between what people believe about language and languages, and the way they access and make use of linguistic resources, provides a further focus to our analysis. If languages are invented, and languages and identities are socially constructed, we nevertheless need to account for the fact that at least some language users, at least some of the time, hold passionate beliefs about the importance and significance of a particular language to their sense of ‘identity’. It is now well established in contemporary sociolinguistics (Harris 2006; Rampton 2006) that one ‘language’ does not straightforwardly index one subject position, and that speakers use linguistic resources in complex, sophisticated ways to perform a range of subject positions, sometimes simultaneously. However, whilst accepting this, May (2001, 2005: 330) argues that ‘historically associated languages continue often to hold considerable purchase for members of particular cultural or ethnic groups in their identity claims’. For some of the people we spoke to in the course of our research, a ‘language’ held powerful connotations in terms of their sense of belonging and selfhood. It is evident from our data that for some people, in some circumstances, ‘particular languages clearly are . . . important and constitutive factor of their individual, and at times, collective identities’ (May 2005: 330). In the context of research on ‘heritage’ language education, it is almost a truism that learning the ‘heritage’ language ‘plays a critical role in the process of children’s identity formation’ (Nicholls 2005: 164). Whilst it is certainly an oversimplification to treat certain languages as ‘symbols’ or ‘carriers’ of ‘identity’, we are obliged
536
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
to take account of what people believe about their languages, to listen to how they make use of their available linguistic resources, and to consider the effects of their language use—even where we believe these ‘languages’ to be inventions.
‘HERITAGE’ Bourdieu and Passeron (1979) proposed that all teaching implicitly presupposes a body of knowledge, skills, and modes of expression which constitute the heritage of the cultivated classes. In our classroom observations and recordings, and in our participants’ statements in interviews, there was a clear sense that the teaching of ‘language’ was inexorably intertwined with the teaching of ‘heritage’. Many of our participants used the term ‘culture’ to refer to those elements of Bengali/Bangladeshi life and history which they wished to transmit through complementary schooling. In our analysis we interpret this as ‘heritage’, distinguishing ‘heritage’ from ‘culture’. Whereas ‘heritage’ refers to elements of past experience which a group deliberately sets out to preserve and pass on to the next generation, ‘culture’ is ‘reproduced and emerges in people’s activity together—it exists in the processes and resources involved in situated, dialogical, sense-making’ (Rampton 2006: 20). In recent times the scope of definitions of ‘heritage’ has broadened considerably from concern for the preservation of buildings and historical sites to include historical areas, towns, environments, social factors, and ‘intangible heritage’ (Ahmad 2006: 299; Smith 2006: 54). UNESCO’s Convention for the Safeguarding of the Intangible Cultural Heritage defines ‘intangible cultural heritage’ as: The practices, representations, expressions, knowledge, skills—as well as the instruments, objects, artefacts and cultural spaces associated therewith—that communities, groups and, in some cases, individuals, recognise as part of their cultural heritage. This intangible cultural heritage, transmitted from generation to generation, is constantly recreated by communities and groups in response to their environments, their interaction with nature and their history, and provides them with a sense of identity and continuity, thus promoting respect for cultural diversity and human creativity. (UNESCO 2003: Article 2:2) Patrick (2007) points out that appeals for the protection of forms of ‘intangible heritage’ have played an important role in campaigns for language rights. Whether we are dealing with traditional definitions of ‘tangible’ or ‘intangible’ heritage, we are engaging with sets of values and meanings, including emotion, memory, and shared knowledge (Smith 2006). ‘Heritage’ describes sets of shared values and collective memories that are ‘constructed as a ‘birthright’ and are expressed in distinct languages and through other cultural performances’ (Peckham 2003: 1). Pearson and Sullivan (2007: 208) suggest that heritage resources may have a ‘special value to minority groups in the
ADRIAN BLACKLEDGE and ANGELA CREESE
537
community’, who have a particular interest in their own history. Regardless of the ‘management’ of heritage resources, subordinate groups ‘often choose to mobilise a ‘‘strategic essentialism’’ as a political tool’ (Stanton 2005: 416). From our participants we heard at times that certain sets of linguistic resources were believed to function as threads of association with historic contexts. Sets of resources often come to represent abstract notions such as sense of place, community, or belonging (Smith 2004). ‘Heritage’ can be thought of as the preservation of a potential loss (Peckham 2003: 1), ‘anything that someone wishes to conserve or to collect, and to pass on to future generations’ (Howard 2003: 6). Bourdieu and Passeron (1979: 25) suggest that ‘inheritance always implies the danger of squandering the heritage’. However, it cannot be assumed that the preservation and transmission of ‘heritage’ is straightforward. Simply the process of ‘passing on’ resources will alter them. Tunbridge and Ashworth (1996: 92) argue that there is rarely a simple relationship between a group of people and ‘heritage’ resources: ‘The same piece of heritage can be interpreted and received by different groups in quite different ways’. Rather than being a static entity, ‘heritage’ is a ‘process or performance that is concerned with the production and negotiation of cultural identity, individual and collective memory, and social and cultural values’ (Smith 2007: 2). Heritage as a process of meaning-making may ‘help us bind ourselves, or may see us become bound to, national or a range of sub-national collectives or communities’ (Smith 2006: 66) as particular resources come to act as powerful symbols of, or mnemonics for, the past (Lipe 2007). Smith (2006: 3) proposes that the idea of ‘heritage’ is ‘used to construct, reconstruct, and negotiate a range of identities and social and cultural values and meanings in the present’. She argues that ‘heritage’ is a set of practices involved in the construction and regulation of values, a discourse about negotiation, about using the past, and collective and individual memories, to negotiate new ways of being and to perform identities. People engage with ‘heritage’, appropriate it, and contest it (Harvey 2007). ‘Heritage’ may become a site at which identities are contested rather than imposed unproblematically. That is, those who seek to preserve and pass on certain sets of resources may find that the next generation either rejects imposed subject positions, contests the validity or significance of resources, or appropriates them for other purposes. For Bourdieu ‘heritage’ is reproduced through ‘class’ and ‘education’, in the reproduction of ‘distinction’, ‘an unacquired merit which justifies unmerited attainment, namely heritage’ (Bourdieu and Darbel 1991: 110). Bourdieu (1993: 299) argues that in education there is an assumption of a community of values between pupil and teacher which occurs where the system ‘is dealing with its own heirs to conceal its real function, namely, that of confirming and consequently legitimizing the right of the heirs to the cultural inheritance’. He further argues that: Only when the heritage has taken over the inheritor can the inheritor take over the heritage. And this appropriation of the
538
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
inheritor by the heritage, the precondition for the appropriation of the heritage by the inheritor (which has nothing inevitable about it), takes place under the combined effect of the conditionings inscribed in the position of inheritor and the pedagogic action of his predecessors, themselves possessed possessors. (Bourdieu 2000: 152) In our data the teaching and learning of ‘heritage’ or ‘community’ languages (Hornberger 2005) act as sites at which ‘heritage’ values may be transmitted, accepted, contested, subverted, appropriated, and otherwise negotiated. These are sites for the negotiation of identities, for the acquisition and performance of sets of linguistic resources which are called into play by social actors under very particular social and historical conditions (Wiley 2005, 2007). These conditions may both constrain and make possible the reproduction of existing conventions and relations, as well as the production of new ones (Heller 2007).
METHODS The research project consisted of four interlocking case studies with two researchers working in two complementary schools in each of four communities. The case studies focused on Gujarati schools in Leicester, Turkish schools in London, Cantonese and Mandarin schools in Manchester, and Bengali schools in Birmingham. The present paper focuses on data collected in and around the Bengali schools in Birmingham. Complementary schools, also known as ‘supplementary schools’, ‘heritage language schools’, or ‘community language schools’, provide language teaching for young people in a nonstatutory setting. Bengali complementary schools in Birmingham are managed and run by local community groups on a voluntary basis, usually in hired or borrowed spaces, with few resources. They cater for children between 4 and 16 years of age, and operate mainly in the evenings and at weekends. The students’ families had migrated from the Sylhet region of Bangladesh. One of the specific aims of the research project was to investigate how the linguistic practices of students and teachers in complementary schools are used to negotiate young people’s multilingual and multicultural identities. Each case study identified two complementary schools in which to observe, record, and interview participants. The classes ran for between two and three hours, either in the evening or at the week-end. After four weeks observing in classrooms using a ‘team field notes’ approach, two key participant children were identified in each school. In the Bengali schools the key participant children were all 10 years old. These children were audio-recorded during the observed classes, and also for 30 minutes before coming to the class and after leaving class. Stakeholders in the schools were interviewed, including teachers and administrators, and the key participant children and their parents. In all we collected 192 hours of audio-recorded interactional data, wrote 168 sets of field notes, made 16 hours of video-recordings, and interviewed 66 key stakeholders.
ADRIAN BLACKLEDGE and ANGELA CREESE
539
A more detailed account of the methods used to collect documentary and home-based data are outlined in Creese et al. (2008).
‘LOCAL KNOWLEDGE’ ABOUT ‘LANGUAGE’ AND ‘HERITAGE’ The founder and administrator of one of the two schools (school A) made a forceful and emotional statement following an interview question which queried the rationale for teaching Bengali to children in Birmingham: ei bhaashar jonno 1952 te amaar theke dosh haath dure Barkat, Salam maara jaae 5 because of this language in 1952 ten yards away from me Barkat and Salam were killed4 1952 te 5in 19524I was also a student in year 10. From Sylhet to Dhaka was 230 miles we marched there Sylhet to Dhaka 230 miles with slogans. We want our mother language it is a raashtro bhasha 5state language4. How I will forget about my mother language? My brothers gave their life for this language. I will never forget it while I’m alive. (administrator interview)1 Throughout the paper we are mindful of Pavlenko’s (2007: 176) argument that interview or narrative data can not be treated as ‘truth’ or ‘reality itself’. Rather, in line with Pavlenko, we are ‘sensitive to the fact that speakers use linguistic and narrative resources to present themselves as particular kinds of individuals’ (2007: 177). We constantly saw individual participants positioning themselves in relation to the ‘ethnic, linguistic, and cultural loyalties’ (2007: 177) which they chose to emphasize. For the school administrator the ‘mother language’ was a vital symbol of the founding of the Bangladeshi nation. More than 50 years earlier he had witnessed the incident in which the ‘language martyrs’ were killed while demonstrating against the imposition of Urdu as the national language by West Pakistan, and these events seemed to have informed his view that British-born children of Bangladeshi heritage should learn and maintain the Bengali language. The historic incident which marks the Bangladeshi calendar as ‘Ekushey February’, continues to be celebrated as a key moment in the collective memory of the Bangladeshi nation, and in the Bangladeshi community in UK (Gard’ner 2004). One of the senior teachers in the same school argued that learning Bengali was associated with maintaining a knowledge of Bangladeshi ‘roots’: ‘We may have become British Bangladeshi or British Indians but we don’t have fair skin and we cannot mix with them. We have our own roots and to know about our roots we must know our language’. For both of these Bangladeshi-born men, teaching and learning Bengali was an important means of reproducing their ‘heritage’ in the next generation. Many of the students’ parents agreed. One mother typically told us that it was important that her children should be able to speak Bengali because: Bengali is our mother land, where we come from; really we come from Bangladesh. Even if you are born in this country it doesn’t matter, we need to know our mother language first. (parent interview)
540
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
Asked why it was important to learn the language of the ‘mother land’, she said ‘you need to know your side of the story, where your parents come from, you’ve got to know both from this country and the other one’. For her learning the ‘mother language’ was closely associated with learning about the ‘mother land’, and both represented her ‘side of the story’. We heard an explicit rationale from administrators, teachers, and parents that a key aim of the school was for the children to learn Bengali because knowledge of the national language carried features of Bangladeshi/Bengali ‘heritage’.
TEACHING ‘LANGUAGE’ AND ‘HERITAGE’ The rationale of the schools was put into practice in the classroom through a pedagogy which frequently introduced ‘heritage’ content in the context of teaching Bengali. Here ‘heritage’ included narratives of national belonging, and the introduction of national symbols of Bangladesh. In the first example the teacher (T) engages with historical events in the making of the Bangladeshi nation: T:
S: T:
S: T:
S: T: S: T: S: T:
Bangladesher teen taa national day aache, jaatio dibosh 5Bangladesh has three national days, national events4 National day not national anthem independence day etaa Banglae ki bolbe shaadhinota dibosh 5in Bangla it is shaadhinota dibosh4 Ekushey February shohid dibosh 521st February is shohid dibosh4 aage bolo Ekushey February shohid dibosh 5first say 21st February is shohid dibosh4 ekushey February shohid dibosh er pore aashlo shaadhinota dibosh 5after that comes shaadhinotaa dibosh4 independence day, independence day is not Bangla, it is English. Banglae holo 5in Bangla it is4 shaadhinota dibosh chaabbish-e March 526th March4 shaadhinota dibosh chaabbish-e March 526th March4 lastly nine months we fought against Pakistani collaborator language day language day holo ekushey February. Chaabbish March independence day. Sholoi December, after nine months bijoy dibosh 5victory day4 Pakistani occupied army ke aamraa surrender korchi. 5we made the occupied forces of Pakistan surrender their arms4 Al Badr against our independent war ke aamraa chutaaisi
ADRIAN BLACKLEDGE and ANGELA CREESE
S: T: S: T:
541
5 we chased them out 4 How many national days in Bangladesh? three Bangladesher jaatio dibosh koiti? 5how many national days in Bangladesh?4 teen ti 5three4 Shaadhinota dibosh ebong bijoy dibosh chilo 1971. Bhasha dibosh chilo 1952. Aar bhaasha dibosh kon din chilo 52. Tokhon amraa choto 5independence day and victory day was in 1971. Language day was in 1952. Language day was 52 when we were young4 Inshaallah eta every day jodi aamraa every day discuss kori taahole bhaalo 5by the grace of God it is good if we discuss this every day4 (classroom recording)
The curriculum content here is strongly nationalistic, and appears to have the aim of instilling in the young language learners an understanding of key dates and events in the making of the Bangladeshi nation. The student (S) seems to have some pre-existing knowledge of the historical context, although she is rather tentative in volunteering this. Here the teacher moves comfortably between Bengali and English, translanguaging within and between sentences, and making his final statement in the common Islamic expression ‘Inshallah’, derived from Arabic, together with Bengali and English. A second common feature of teaching ‘heritage’ in the Bengali complementary school classrooms was the introduction of national symbols associated with Bangladesh. In a typical example, the teacher asked a child to draw the English and Bangladeshi national flags on the whiteboard. After playing the Bangladeshi national anthem on his mobile telephone, the teacher continued: T:
Ss: T:
Ei. Eitaai aamaader jaatio shongeet othobaa national anthem. Ekhon aamaader Bangladesher ko-e ektaa jinish aache jaatio bol-e. 5This, this is our national song or national anthem. Now, we have a few things in Bangladesh which are our national symbols4 jaatio shongeet 5national anthem4 jaatio kobi 5national poet4, jaatio ful 5national flower4 baa jaatio, baa national fol 5or national, or national fruit4 baa national paakhi 5or national bird4 Bangladesher jaatio ful ki? 5What is the national flower of Bangladesh?4 [no response] water lily, water lily, water lily Bangla, water lily, shapla. Etaa aamaader jaatio ful 5shapla. This is our national flower4 (classroom recording)
Here the process of teaching Bengali is intimately interwoven with the process of teaching symbolic representations of Bangladesh, as knowledge of the national/cultural symbols, like knowledge of the Bengali language, comes to represent Bengali ‘heritage’.
542
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
‘LOCAL KNOWLEDGE’ CONTESTING ‘LANGUAGE’ AND ‘HERITAGE’ Despite these powerful discourses and practices which evidenced the teaching of ‘heritage’ through the teaching of ‘language’, the notion that the discourses and practices of the Bengali schools were homogeneous in their ideological orientation to ‘language’ and ‘heritage’ was not borne out in the data. Rather, what people believed about their language (or other people’s languages), and the situated forms of talk they deployed, revealed divergent and contested views about the value and status of particular linguistic resources. Bengali is the language of education and literacy in Bangladesh, and is characterized by diglossia. The two standard varieties of Bangla are Sadhu Bhasha and Cholit Bhasha and regional varieties include Sylheti, from the north-east of Bangladesh. Sylheti is the variety spoken by the vast majority of Bangladeshi immigrants to Britain. Sylheti is often regarded as a modification of standard Bengali which is not intelligible to the people of other districts in Bangladesh (Hamid 2007). Chalmers (1996: 6–7) observes that Bengali and Sylheti are ‘near enough mutually unintelligible’. Chalmers does acknowledge that Sylheti and Bengali are very closely related and speakers of one language or dialect are often exposed to the other, even though they may not speak them. Whilst Bengali is the literate language of Bangladesh, Sylheti is a vernacular variety. When we interviewed the administrators and teachers in the schools they spoke emphatically about the need for children to learn the standard variety. This was frequently held to be oppositional to Sylheti. One of the school administrators was emphatic that Bengali was ‘completely different’ from Sylheti, and that Sylheti should not be allowed to ‘contaminate’ the standard form. He was concerned that Sylheti forms were beginning to appear in the spelling and grammar of Bengali newspapers in the UK, introducing ‘thousands of spelling mistakes—Bengali newspapers I have seen in many places the spelling was wrong, sentence construction was wrong’. For the administrator non-Standard resources were ‘contaminating the language’. He made this point about the necessity for children to learn Standard Bengali: I am always in favour of preserving languages and all these things. But it doesn’t mean that this should contaminate other languages and give this more priority than the proper one. We have to preserve the proper one first, and at the same time we have to encourage them to you know, use their dialect. But we shouldn’t make any compromise between these two. (administrator interview) This was a strongly articulated argument in the data. The administrator of the other school stated that: ‘Bhasha to bolle Bangla bhasha bolte hobe Sylheti kono bhasha naa’
ADRIAN BLACKLEDGE and ANGELA CREESE
543
5When you talk about language it means Bengali. Sylheti is not a language4 (school administrator interview) For several respondents ‘Bengali’ constituted a more highly valued set of linguistic resources than ‘Sylheti’, and was regarded as the ‘proper’ language. Pujolar (2007: 78), referring to a different socio-historical context, makes the point that language policy may operate to foster knowledge of some languages, ‘but delegitimise or ignore other languages and other forms of multilingual competence and performance’. Patrick (2007: 127) similarly finds that in arguing in support of a particular language, ‘speakers can be locked into fixed or essentialised notions of identity, ‘‘authenticity’’ and place, which provide no recognition of mobile, postcolonial speakers’. It was clear that for some of our respondents not all linguistic resources were equally valued, and while some sets of linguistic resources were considered to be ‘a language’, others were not. In this sense there was a constant re-invention of ‘language’ on the part of some participants. Those who spoke ‘Sylheti’ were often criticized by ‘more educated’ people who spoke ‘Bengali’. They were characterized by the administrator of one of the schools as members of the ‘scheduled’, or ‘untouchable’ caste: people without rights or resources in the Indian sub-continent: Publicraa ki dibe amar aapne especially bujhben amader desher je shob lok aashche ora kon category lok aashchilo, mostly from scheduled caste, gorib, dukhi krishokra aashchilo. oder maa baba o lekha pora interested naa oder chele meye raa o pora lekha interested naa. Oraa baidhitamolok school jete hoe primary school sholo bochor porjonto jete hoe, ei jonne school jaai. 5What will the public contribute? You [the researcher, Shahela Hamid] especially will understand what type of people came from our country. They belonged to the category of scheduled caste, they are the poor, the deprived, farmers. Their parents were not interested in education nor are the children interested. They go to school because it’s compulsory4 (school administrator interview) Here the Sylheti speakers are referred to as the ‘scheduled caste’. Regarded as the least educated group in society, with no resources of any kind, they are the lowest of the low (Borooah 2005; Kijima 2006; Borooah et al. 2007). Here linguistic features were viewed as reflecting and expressing broader social images of people. Irvine and Gal (2000: 37) suggest that ‘participants’ ideologies about language locate linguistic phenomena as part of, and evidence for, what they believe to be systematic behavioural, aesthetic, affective, and moral contrasts among the social groups indexed’. One of the teachers argued that children should learn Bengali for ‘moral reasons’. Irvine and Gal propose that a semiotic process of iconization occurs, in which linguistic features that index
544
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
social groups appear to be iconic representations of them, as if a linguistic feature depicted or displayed a social group’s inherent nature or essence. Bourdieu and Darbel (1991: 112) argue that some more powerful groups provide ‘an essentialist representation of the division of their society into barbarians and civilized people’. Here the fact of speaking ‘Sylheti’, rather than ‘Bengali’, appeared to index the Sylheti group in particularly negative terms, despite the relative similarities between the ‘Bengali’ and ‘Sylheti’ sets of linguistic resources. Whilst some speakers in our study considered ‘Sylheti’ to be quite different from ‘Bengali’, others regarded the two sets of resources as indistinguishable. As we have seen, there were several instances of participants commenting on the differences between Sylheti and Bengali in terms of social status and value, but not everyone agreed about the extent to which these sets of linguistic resources were distinct. While the administrator of one of the schools argued that Bengali and Sylheti were ‘completely different’, a student’s mother said they were ‘thoraa different’ 5a little different4, while other parents also held this view, saying they were ‘little bit different thaake 5only4’ and even ‘the same’. Here there was clear disagreement about the nature and extent of the differences between the sets of linguistic resources used by the students’ parents at home, and the literate version of the language taught in the complementary school classrooms. That is, there was disagreement about the permeability of the languages. These differences of perception were likely to be ideological. Those who argued that the ‘languages’ were completely different from each other were speakers of the prestige language, unwilling to allow the lower status language to contaminate their linguistic resources. Those who argued that the ‘languages’ were almost the same as each other were speakers of Sylheti, which was held to index the lower status, less educated group. On many occasions the research participants interactionally evidenced their awareness of differences (perhaps mainly in status and value) between ‘Sylheti’ and ‘Bengali’. There was also an awareness of Bengali as the higherstatus language on the part of teachers (‘I talk posh Bengali, and the children can’t understand me’), students, administrators, and parents. The following example was recorded at the dinner table in the family home of one of the students: Mother: Father: Student: Father: Student: Father:
khitaa hoise? Tanvir, khaibaani saatni? 5what is the matter? Tanvir, would you like some relish?4 aaro khoto din thaakbo 5how many more days is that [voice recorder] going be with you?4 aaro four weeks 5four more weeks4 ( ) No they said any. If you talk all English. . . ginni, oh ginni [calling his wife using a Bengali term of endearment]
ADRIAN BLACKLEDGE and ANGELA CREESE
Mother:
Father:
545
ji, hain go daakso kheno 5yes, dear why are you calling me?4 Tumaar baabaa shuddho bhasha bolen 5your father is speaking the standard language4 paan dibaa 5can I have some paan4 aapne aamaar biyaai kemne 5how are you my relation?4 (home recording)
Here the Sylheti-speaking parents play the roles of Bengali speakers, adopting the airs and graces which they see as characteristic of the Bengali-speaking group. The terms of endearment used here (‘ginni’, ‘hain go’) are forms of parody (Bakhtin 1973, 1984, 1986), exaggerations beyond common usage, as speakers of Bengali are mocked in discourse which represents an inflated sophistication. This brief interaction is situated in a whole hinterland of language ideological beliefs and practices, as the couple acknowledge differences between Bengali and Sylheti as sets of linguistic resources, and the conditions which differentially provide and constrain access to linguistic resources. In parodic discourse the parents introduce into their own voices the exaggerated voice of the Bengali speaker, and that voice clashes with its host, as ‘Discourse becomes an arena of battle between the two voices’ (Bakhtin 1994: 106). Here the impromptu role-play light-heartedly, but not half-heartedly, ‘parodies another’s socially typical . . . manner of seeing, thinking, and speaking’ (Bakhtin1994: 106). In this section we have seen that for some of our participants, some sets of linguistic resources were very considerably privileged above other, similar sets of linguistic resources. While linguistic resources which were described as ‘Standard’, or ‘proper’, or ‘real’, or ‘book’ Bengali had come to represent the ‘heritage’ of the Bangladeshi nation, sets of resources described as ‘Sylheti’ had come to be associated with the uneducated poor, who were held to be uninterested in schooling, and unmotivated. However, we also saw that these distinctions were contested by others, who denied that clear differences existed, and at times made fun of the assumption that these differences were constitutive of differences in social status. That is, our participants represented disagreements about what constituted (a) language, and about the ideological links between speakers and the sets of linguistic resources which they called into play.
NEGOTIATING ‘LANGUAGE’ AND ‘HERITAGE’ The contested nature of the ideological links between sets of linguistic resources and their assumed associations was frequently made visible in the interactional data recorded in the classroom. Here teaching of ‘heritage’ and ‘language’ became sites at which identities were negotiated in discourse (Blackledge and Pavlenko 2001; Pavlenko and Blackledge 2004). Here ‘negotiation of identities’ is understood as ‘an interplay between reflective positioning, that is self-representation, and interactive positioning, whereby
546
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
others attempt to position or reposition particular individuals or groups’ (Pavlenko and Blackledge 2004). Negotiations here take place interactionally where particular subject positionings are contested. Negotiable identities refer to all identity options which can be contested and resisted by particular individuals and groups, although of course not all identity positions are equally negotiable. How much room for resistance to particular positioning individuals may have depends on each individual situation, the social and linguistic resources available to participants, and the balance of power relations which sets the boundaries for particular identity options. In the following example a new teacher to the class is perhaps not acquainted with the usual norms and expectations of linguistic behaviour in the classroom: S1: T:
S2: S1: S2: T: S2:
miss why can’t we just go home? Bangla-e maato etaa Bangla class 5speak in Bengali this is Bengali class4 khaali English maato to etaa Bangla class khene 5if you speak in English only then why is this the Bengali class?4 miss you can choose I know English why? because tumi Bangali 5because you are Bengali4 my aunty chose it. She speaks English all the time. (classroom recording)
In this interaction the teacher argues that the language of the classroom should be Bengali, proposing a model of learning which is at odds with the children’s usual experience. One of the students (S1) argues that it should be possible to choose which language to speak in a particular context, and is backed up by her friend (S2). When S2 asks why it is necessary to speak Bengali in class, the teacher says ‘because tumi Bangali’. In this English and Bengali phrase the ideology of the school is summed up in the most succinct terms. Bengali should be spoken, and should be learned, argues the teacher, because the children are Bengali (here ‘Bangla’ refers to the Bengali language, while ‘Bangali’ refers to Bengali national and/or ethnic belonging). Ironically, the teacher uses Sylheti to make her point about speaking Bengali. S2 contests the teacher’s point, and in doing so contests the ideology of the school. Reiterating her argument that it should be possible to choose which language to speak, she cites her ‘aunty’, who has chosen to predominantly speak English. The student’s ‘aunty’, herself of Bangladeshi heritage, is offered as an example of someone who has resisted the notion of ‘one-language-equals-one ethnicity/culture’. For S2 language choice is flexible. For the school, in this example at least, language learning is tied to ethnic and national belonging, and is inflexible. We saw many examples in the classroom of students resisting teachers’ attempts to teach them Bengali. In more than one example students mocked their teachers’ pronunciation of English words. Here, though, the children
ADRIAN BLACKLEDGE and ANGELA CREESE
547
challenge the teacher’s pronunciation of the name of a new child when she arrives at school: S1: T: S1: T:
S1: S2:
She is coming through the front door Jaara [correcting teacher’s pronunciation] Zahra Tumaader aamaake shikhaate hobe naa 5you all don’t have to teach me4 Ektu chintaa korbaa aamaader theke onaara boishko 5you should think that he is much older than us4 Okay, look Aleha, how do you spell Zahra? Z-a-h-r-a. In school we call her Zahra, in school we call her Zahra. (classroom recording)
Here S1 corrects the teacher’s pronunciation. The different pronunciations of the name are significiant: in Bangladesh /z/ is pronounced as /j/, so the teacher is not pronouncing the name ‘Jara’ incorrectly, but is pronouncing it just as it would be in Bangladesh. The pronunciation of the name which the children use in school is an anglicized version, pronouncing the /z/. The students contest the ‘Bangladeshi’ (‘Sylheti’ and/or ‘Bengali’) pronunciation of the child’s name, and insist on the anglicized version. Here the students appear to use the teacher’s pronunciation of the Bangladeshi name as an opportunity to negotiate a subject position away from the imposed ‘heritage’ identity, and to use available linguistic resources in subtle, nuanced ways to occupy a position which is oppositional to ideologies which rely on the ‘purity’ of the Bengali language. The students’ complex and sophisticated response to the complementary schools’ ‘heritage’ positioning of them was also evident in interviews. In this excerpt two students were talking to the researchers about a drama activity, based on a story of new arrivals from Bangladesh. In talking to the researcher (R) they described this group as ‘freshies’ (cf. Martin et al. 2004): R: S1: R: S1: S2: S1:
R: S2: R:
What do you mean ‘freshie’, what does that mean? freshie as in a newcomer is that bad to say to somebody? yea it’s kind of like a blaze but it’s also a word to describe a new person coming from a different place. it’s not a good thing. it’s kind of both..if you say it as in trying to tease somebody, ‘freshie’, and we say it as in erm trying to say erm, as in they’re newcomers and they come from a different country for the first time could you tell if someone was ‘freshie’? well from Bangladesh it’s not always their skin colour, it’s sometimes how they talk. . . how do you talk ‘freshie’?
548
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
S1: S2: S1: S2:
it’s kind of like they don’t know that much English. they might just show off in their language but if you ask them a question in English they just. . . they’re like ‘what’, ‘what’, you know they say strange words in their language and if you ask them a question in English they just say ‘what’ in their language. (student interview, School B)
The students negotiate their identity in opposition to that of the newly arrived children, repeatedly referring to ‘their’ language, which they see as different from the language they speak themselves. Here ‘what’ ‘what’ is spoken with an intonation which suggests some confusion on the part of the newly-arrived group. Although the students speak the same language as the new arrivals in daily interactions with their parents, they nevertheless indicate that ‘how they talk’ is one of the defining ways in which the ‘freshies’ are different from them. The students’ linguistic repertoires were wide-ranging. In addition to making use of linguistic resources of English, Sylheti, and Bengali, they watched Hindi films, read the Qur’an in Arabic, and listened to popular contemporary music in varieties of American English, and also Indian and Bengali pop music. For some, listening to contemporary American music and watching DVDs was an important part of the way they viewed themselves. In the following example two siblings were asked what sort of Hindi songs they like: S1: R1: S1: R2: S1: R2: S1: R1: S2: R: S2: S1: R1: S1:
I like Bhangra really? I like Bhangra with rap oh they have all kinds of crossover Bhangra music now don’t they I like rap like Fifty Cent I mean do you like Eminem? yes he’s all right so is that OK? I mean rap and all that is all right? erm yea your dad doesn’t. . .? he doesn’t really erm if it’s in front of him he will shout but erm if we stopped it it’s all right. rap anyway I don’t hear rap at home I might just hear it a bit cos I hear it from my friend’s dad in his cars and everything because is your friend Pakistani or Indian? English (..) I mean Bengali (student interview)
Here ten-year-old S1 associates himself with firstly Bhangra, then ‘Bhangra with rap’, and finally ‘rap like Fifty Cent’. This appears to represent a negotiation of an increasingly risky, or perhaps sophisticated, subject position.
ADRIAN BLACKLEDGE and ANGELA CREESE
549
Whereas listening to Bhangra music may be regarded as relatively mainstream and conservative, ‘Bhangra with rap’ moves towards an increasingly American pop culture position, and ‘rap like Fifty Cent’ is likely to represent a ‘Gansta Rap’ identity. S1 is happy to be associated with ‘my friend’s dad’, and the researcher assumes that his friend must be of Pakistani or Indian heritage, as it may be surprising for a good Bangladeshi to listen to this kind of music. In the final utterance in this excerpt, S1’s pause between ‘English’ and ‘I mean Bengali’ may suggest that nationality, ethnicity, and even language are not the salient categories for him at this moment—rather, he is more interested in positioning himself as a cool, sophisticated consumer of contemporary, transnational music. In this section we have seen that students and teachers at times used the complementary school classroom as a space in which to negotiate identities. These negotiations were often focused on beliefs, attitudes and values relating to language(s), and played out in sophisticated deployment of linguistic resources (see also Harris 2006). We have also seen that for these students the ‘heritage’ identities which the schools set out to reproduce were often contested in subtle, sophisticated ways, as the students called into play sets of linguistic resources which positioned them as somewhat different from the imposed ‘heritage’ identities of the institution.
DISCUSSION What, then, can we say about negotiations which constitute, and are constituted by, the values, attitudes, beliefs, and practices of ‘language’ in and around these Bengali complementary schools? It is essential to any analysis that ‘the messiness of actual usage’ (Heller 2007: 13) should be understood in relation to histories, power, and social organization. In the course of our research we heard strongly articulated views, from parents and teachers, that Bengali should be taught as a mandatory part of the mainstream school curriculum. In the US context Wiley (2007: 254) refers to ‘the crisis of monolingualist ideology’, which proposes that English alone is of value in society. In British political, media, and other discourses a powerful ideology similarly proposes that minority languages other than English are a negative force in society. Languages which originate in the Indian sub-continent in particular come to be ‘racialized’ in the discourses of elite groups in UK, and associated with social segregation, family breakdown, and even terrorism (Blackledge 2005, 2006, forthcoming 2009). To some extent at least, the complementary schools are ‘safe spaces’ (Creese and Martin 2006: 2) in which young people are able to practise and extend their linguistic repertoires. In doing so they are ‘contesting the historical inequalities that have seen minority languages, and their speakers, relegated to the social and political margins’ (May 2007: 26). However, the process of teaching ‘heritage’ through ‘language’ is complex. First, for our participants the notion of what constitutes a ‘language’ is disputed. For some of the social actors concerned, one set of linguistic resources
550
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
(‘Bengali’) is heavily endowed with symbolic associations, and becomes a ‘social artefact, invented at the cost of a decisive indifference to differences’ (Bourdieu 1991: 287). This ‘Standard’ set of resources is regarded by some as that which should be ‘preserved’ and kept free from ‘contamination’. This particular set of resources accrues symbolic capital as it is perceived by some social agents as intrinsically superior to some other sets of linguistic resources (Bourdieu 1998: 47), and is ‘invented’ as a ‘language’ which is reified and immutable. The discourse of the school administrators in particular proposes that the ‘language’ should be transmitted to the next generation in its pure and natural form. However, this clear distinction between Standard and non-Standard sets of linguistic resources (‘Bengali’ and ‘Sylheti’) did not attract universal consensus. While the case for the purity of the ‘Standard’ was often argued in institutional discourse, other social actors, especially the parents of the students, and the students themselves, contested this view. Furthermore, we saw that some linguistic features came to be iconic representations of their speakers, as if a linguistic feature displayed the Sylheti group’s inherent nature. The fact of using certain sets of resources (‘Sylheti’), rather than others (‘Bengali’), appeared to index these speakers in particularly negative terms, despite the relative similarities of the two sets of linguistic resources. We saw that the parents of one child mocked the ‘Standard’ resources, and the ideological beliefs which were perceived as accompanying their speakers. In doing so they acknowledged the relations of power at work in the uneven distribution of resources, and the discourses which inscribe value (or its lack) to particular linguistic forms and practices. Second, we saw that the teaching of ‘heritage identities’ (Creese et al. 2006), through nationalist and historical content, was at times contested and subverted by students, in interactions which became sites for the negotiation of identities. We saw examples of language teachers teaching language through ‘heritage’ content with messages which were deeply rooted in Bangladeshi nationalism, invoking features of the collective memory of the nation. Also, we saw the repeated teaching of tangible and less tangible symbols of Bangladeshi heritage, from the national flag and national anthem to symbols such as the national flower, national fish, and national bird. Billig (1995: 174) argues that ‘nationalism’ is produced and reproduced in ‘daily, unmindful reminders of nationhood in the contemporary, established nation-state’, through everyday, ‘banal utterances’. These ‘self-evidences’ are those apparently commonsense misrecognitions which constantly construct and reinforce ideologies (Bourdieu 2000: 181). For the Bangladeshi-heritage group in our study, however, their nationalism was not produced and reproduced in everyday discourses in wider society. Perhaps in the face of the everyday ‘flagging’ of British/English nationalism, their approach to teaching their ‘heritage’ was explicit and often direct. The teachers appeared to impose on the students identities which were associated with Bangladesh and its history. Like the
ADRIAN BLACKLEDGE and ANGELA CREESE
551
institutional ‘language’ ideology, ‘heritage’ ideology was reified and naturalized. However, the students did not always accept the static, essentialized version of ‘heritage’ which the school was teaching. Howard (2003: 6) propose that ‘things actually inherited do not become heritage until they are recognised as such’. That is, while the teachers and administrators of the schools believed that teaching ‘language’ and ‘heritage’ was a means of reproducing ‘Bengali’ identity in the next generation, the imposition of such identities was often contested and re-negotiated by the students. Their apparent rejection of some ‘heritage’ symbols, their challenge of their teachers’ insistence on the use of ‘Bengali’ in the classroom, and their insistence on the anglicized pronunciation of a Bengali name, all became instances of students negotiating subject positions which contested those imposed by the institution. In the schools there was a perceived institutional need to fill what some teachers and administrators called the ‘cultural gap’ which had been created between the students and their parents. One of the teachers said of the students: to oraa 5they are4 British born, so they need to know Bangladesh, where their parents were born in Bangladesh, what is Bangladesh, where is Bangladesh . . . so many of them never express own self from own self [that] they are Bangladeshi, they always think, they always think they are British. Their mind perform, mind create that they are British. (teacher interview) This teacher was just one of several who argued that the students lacked something in their knowledge and understanding of Bangladesh, and in their sense of themselves as ‘Bangladeshis’. Bourdieu proposed that: The history objectified in instruments, monuments, works, techniques etc. can become activated and active history only if it is taken in hand by agents who, because of their previous investments, are inclined to be interested in it and endowed with the aptitudes needed to reactivate it. (Bourdieu 2000: 151 emphasis in original) It was this very endowment of aptitudes, this activation of history, which appeared to be the raison d’etre of the Bengali complementary schools. For the teachers, the process of teaching ‘language’, and teaching ‘heritage’, had the potential to invest their students with the aptitudes they required to inherit their heritage, because ‘only when the heritage has taken over the inheritor can the inheritor take over the heritage’ (Bourdieu 2000: 152).
CONCLUSION In this paper we have raised questions in relation to understandings of ‘language’ and ‘heritage’. The beliefs and practices of the participants raised a
552
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
number of questions in our attempts to understand how the language use of students and teachers in Bengali schools in one city in the UK was used to negotiate multilingual and multicultural identities. First, participants articulated attitudes and values which raised questions about what constitutes ‘language’. For some (a) ‘language’ should be preserved and kept free from the contamination of other sets of linguistic resources. For others there was no distinction in practice between resources ideologically framed as legitimate and illegitimate. Second, participants expressed views and attitudes, and performed interactional practices, which raised questions about what constitutes ‘heritage’. While teachers and administrators of the schools believed that teaching ‘language’ and ‘heritage’ was a means of reproducing ‘Bengali’/’Bangladeshi’ identity in the next generation, the imposition of such identities was often contested and re-negotiated by the students, as classroom interactions became sites where students occupied subject positions which were at odds with those imposed by the institutions. These young people were discursively negotiating paths for themselves which were in some ways contrary to the ideologies of the complementary schools, where teachers and administrators held the view that they ought to learn Bengali because to do so was a practice which carried with it knowledge of Bangladeshi history, nationalism, and identity. The young people’s attitudes to their languages, and their multilingual practices, constituted a sophisticated response to their place in the world, as they negotiated subject positions which took them on a path through language ideological worlds constructed by others. The young people were flexible and adaptable in response to their environment, as they negotiated identities which were more complex and sophisticated than the ‘heritage’ positions ascribed to them institutionally.
ACKNOWLEDGEMENT The paper draws on empirical data from one of four linked case studies in a larger, ESRC-funded project, ‘Investigating Multilingualism in Complementary Schools in Four Communities’, ESRC grant RES-000-23–1180.
NOTE 1 Throughout the article we adopt the following transcription conventions: plain font: bold font: italic font: bold font underlined: 5plain font enclosed4 ( )
Sylheti Bengali English other language (e.g. Arabic, Hindi) translation into English speech inaudible
ADRIAN BLACKLEDGE and ANGELA CREESE
553
REFERENCES Ahmad, Y. 2006. ‘The scope and definitions of heritage: From tangible to intangible,’ International Journal of Heritage Studies 12/3: 292–300. Bakhtin, M. M. 1973. Problems of Dostoevsky’s Poetics. Trans R. W. Rotsel. Ann Arbor, MI: Ardis. Bakhtin, M. M. 1984. Problems of Dostoevsky’s Poetics. Ed. and trans. C. Emerson. Manchester: Manchester University Press. Bakhtin, M. M. 1986. Speech Genres and Other Late Essays. Eds C. Emerson and M. Holquist. Austin, TX: University of Austin Press. Bakhtin, M. M. 1994. ‘Problems of Dostoevsky’s Poetics’ in Morris, P. (ed.): The Bakhtin Reader: Selected Writings of Bakhtin, Medvedev, Voloshinov. London: Arnold. Billig, M. 1995. Banal Nationalism. Sage: London. Blackledge, A. 2005. Discourse and Power in a Multilingual World. Amsterdam: John Benjamins. Blackledge, A. 2006. ‘The racialisation of language in British political discourse,’ Critical Discourse Studies 3/1: 61–79. Blackledge, A. forthcoming 2009. ‘Being English, speaking English: Extension to English language testing legislation and the future of multicultural Britain’ in G. Hogan-Brun, C. Mar-Molinero, and P. Stevenson (eds): Testing Regimes: Critical Perspectives on Language, Migration and Citizenship in Europe. Amsterdam: John Benjamins. Blackledge, A. and A. Pavlenko. 2001. ‘Negotiation of identities in multilingual contexts,’ International Journal of Bilingualism 5/3: 243–59. Borooah, V. K. 2005. ‘Caste, inequality, and poverty in India,’ Review of Development Economics 9/3: 399–414. Borooah, V. K., A. Dubey, and S. Iyer. 2007. ‘The effectiveness of jobs reservation: Caste, religion and economic status in India,’ Development and Change 38/3: 423–45. Bourdieu, P. 1991. Language and Symbolic Power. Cambridge: Polity Press. Bourdieu, P. 1993. Sociology in Question. London: Sage. Bourdieu, P. 1998. Practical Reason. London: Polity Press. Bourdieu, P. 2000. Pascalian Meditations. Cambridge: Polity Press. Bourdieu, P. and A. Darbel. 1991. The Love of Art. European Museums and their Public. London: Polity Press. Bourdieu, P. and C. Passeron. 1979. The Inheritors. Chicago: University of Chicago Press.
Chalmers, R. 1996. Learning Sylheti. London: Centre for Bangladeshi Studies. Creese, A. and P. Martin. 2006. ‘Interaction in complementary school contexts: Developing identities of choice—An introduction,’ Language and Education 20/1: 1–4. Creese, A., A. Bhatt, N. Bhojani, and P. Martin. 2006. ‘Multicultural, heritage and learner identities in complementary schools,’ Language and Education 20/1: 23–44. Creese, A., T. Barac¸, A. Bhatt, A. Blackledge, S. Hamid, Li Wei, V. Lytra, P. Martin, C.-J. Wu, and D. Yag˘cıog˘lu-Ali. 2008. Investigating Multilingualism in Complementary Schools in Four Communities. Final Report. Birmingham: University of Birmingham. (RES-000–23–1180). Garcia, O. 2007. ‘Foreword’ in S. Makoni and A. Pennycook (eds): Disinventing and Reconstituting Languages. Clevedon: Multilingual Matters. Gard’ner, J. M. 2004. ‘Heritage protection and social inclusion: A case study from the Bangladeshi community of East London,’ International Journal of Heritage Studies 10/1: 75–92. Giddens, A. 1984. The Constitution of Society. Berkeley: University of California Press. Gumperz, J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Hamid, S. 2007. Unpublished PhD thesis, ‘A study of language maintenance and shift in the Sylheti community in Leeds. A cross-generational survey of language behaviour, attitudes and perceptions on Sylheti ethnocultural migrants from rural backgrounds’. University of York. Harris, R. 2006. New Ethnicities and Language Use. Basingstoke: Palgrave Macmillan. Harvey, D. 2007. ‘Heritage pasts and heritage presents. Temporality, meaning and the scope of heritage studies’ in L. Smith (ed.): Cultural Heritage. London: Routledge. Heller, M. 2007. ‘Bilingualism as ideology and practice’ in M. Heller (ed.): Bilingualism: A Social Approach. Basingstoke: Palgrave. Hornberger, N. 2005. ‘Heritage/community language education: US and Australian perspectives. The International Journal of Bilingual Education and Bilingualism 8/2&3: 101–8. Howard, P. 2003. Heritage. Management, Interpretation, Identity. London: Continuum. Irvine, J. and S. Gal. 2000. ‘Language ideology and linguistic differentiation’ in P. Kroskrity (ed.): Regimes of Language. Ideologies, Polities and
554
NEGOTIATION OF IDENTITIES IN LATE MODERNITY
Identities. Santa Fe/Oxford: School of American Research Press. Kijima, Y. 2006. ‘Caste and tribe inequality: Evidence from India, 1983–1999,’ Economic Development and Cultural Change 54: 369–404. Lipe, W. 2007. ‘Value and meaning in cultural resources’ in L. Smith (ed.): Cultural. Heritage. London: Routledge. Makoni, S. and A. Pennycook. (eds) 2007. Disinventing and Reconstituting Languages. Clevedon: Multilingual Matters. Martin, P. W., A. Bhatt, N. Bhojani, and A. Creese. 2004. Final Report on Complementary Schools and their Communities in Leicester. Leicester/ Birmingham: University of Leicester/University of Birmingham. (ESRC R000223949). May, S. 2001. Language and Minority Rights: Ethnicity, Nationalism, and the Politics of Language. London: Longman. May, S. 2005. ‘Language rights: Moving the debate forward,’ Journal of Sociolinguistics 9/3: 319–47. May, S. 2007. ‘Language education, pluralism, and citizenship’ in S. May and N. Hornberger (eds): Language Policy and Political Issues in Education. Encyclopedia of Language and Education, Vol. 1, 2nd edn. New York: Springer. Nicholls, C. 2005. ‘Death by a thousand cuts: Indigenous language bilingual education programmes in the Northern Territory of Australia, 1972–1998,’ The International Journal of Bilingual Education and Bilingualism 8/2&3: 160–77. Patrick, D. 2007. ‘Language endangerment, language rights and indigeneity’ in M. Heller (ed.): Bilingualism: A Social Approach. Basingstoke: Palgrave. Pavlenko, A. (2007). ‘Autobiographic narratives as data in applied linguistics,’ Applied Linguistics 28: 163–88. Pavlenko, A. and A. Blackledge. 2004. ‘New theoretical approaches to the study of negotiation of identities in multilingual contexts’
in A. Pavlenko and A. Blackledge (eds): Negotiation of Identities in Multilingual Contexts. Clevedon: Multilingual Matters. Pearson, M. and S. Sullivan. 2007. ‘Looking after the past. An introduction to the management of heritage places’ in L. Smith (ed.): Cultural Heritage. London: Routledge. Peckham, R. S. 2003. ‘The politics of heritage and public culture’ in R. S. Peckham (ed.): Rethinking Heritage. Culture and Politics in Europe. London: I.P.Tauris. Pujolar, J. 2007. ‘Bilingualism and the nationstate in the post-national era’ in M. Heller (ed.): Bilingualism: A Social Approach. Basingstoke: Palgrave. Rampton, B. 2006. Language in Late Modernity Cambridge: Cambridge University Press. Smith, L. 2004. Archeological Theory and the Politics of Cultural Heritage. London: Routledge. Smith, L. 2006. Uses of Heritage. London: Routledge. Smith, L. 2007. ‘General Introduction’ in L. Smith (ed.): Cultural Heritage. London: Routledge. Stanton, C. 2005. ‘Serving up culture: Heritage and its discontents at an industrial history site,’ International Journal of Heritage Studies 11/5: 415–31. Tunbridge, J. E. and G. J. Ashworth. 1996. Dissonant Heritage. The Management of the Past as a Resource in Conflict. Chichester: Wiley. UNESCO. 2003. Convention for the Safeguarding of the Intangible Cultural Heritage. Paris: UNESCO. Wiley, T. 2005. ‘Discontinuities in heritage and community language education: Challenges for educational language policies,’ The International Journal of Bilingual Education and Bilingualism 8/2&3: 222–9. Wiley, T. 2007. ‘Heritage and community languages in the national language debate,’ The Modern Language Journal 91: 252–4.
Applied Linguistics 29/4: 555–577 ß Oxford University Press 2008 doi:10.1093/applin/amm051 Advance Access published on 10 January 2008
Routine Trouble: How Preschool Children Participate in Multilingual Instruction ¨ RK-WILLE´N POLLY BJO Linko¨ping University, Sweden This paper examines the turn-by-turn organization of social actions during educational activities at a multilingual preschool in Sweden. Specifically, it focuses on instructional exchanges within two commonplace activities: ‘sharing time’ and ‘Spanish group’. The study builds on earlier research arguing that interactional routines facilitate children’s participation in social activities, and therefore promote language learning. Several instances of interactional trouble are identified and discussed in terms of the teachers’ elaboration of some routine features of these activities, resulting in a mismatch between the teachers’ local aims and the children’s projections of relevant next actions. The analysis further highlights a range of interactional means through which the participants act to come to terms with the trouble. These findings are discussed in terms of the participants’ local concerns as well as the children’s orientations to the routine features of preschool activities. Some educational implications are finally proposed on the basis of these findings.
INTRODUCTION Issues of bilingualism, multiculturalism, and immigrant children’s L2-acquisition have been debated in Sweden for years. The pendulum has swung from the policy of assimilation to the policy of integration of nonnative speakers, emphasizing today the need for ‘good Swedish’ for everyone. A recent report from the Swedish National Agency for Education (prop. 2005) argues that education based on children’s mother tongue promotes their cognitive development and enhances their opportunities for learning. Notably, matters dealing with how bilingualism is accomplished and organized in children’s everyday talk are largely overlooked in the report. However, implications of these arguments may be found in the educational challenges for many schools, as well as for preschools, which need to provide education for children with different language experiences. But how exactly do preschool teachers go about educating young children with diverse language backgrounds, and—perhaps even more importantly—how do preschool children go about participating in various curricular activities? The present paper originates from a larger study conducted in a multilingual preschool in Sweden. The preschool at hand offers daily curricular activities in Swedish, English, and Spanish. Furthermore, all three languages are spoken on a daily basis, in mundane talk. The multilingual approach of the
556
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
preschool involves an institutional emphasis on language awareness and language learning. Children’s participation in organized activities in a preschool context may be seen as priming activities for schooling. Preschool is part of the cultural processes that ‘can be thought of as practices and traditions of dynamically related cultural communities across generations’ (Rogoff 2003: 77). However, theories of cultural practice and learning tell us nothing about what goes on in actual preschool activities and the practical actions through which participants take part in, and reflexively construct, educational practice. Hence, I wish to argue that, if our goal is to understand how children achieve interactional competencies required for effective interaction, it is towards such mundane practices we need to direct our empirical efforts. That is to say, if we are to treat culture as a set of intersubjective practices, where the world and people are in constant reflexive involvement (Mehan 1979), the practices must be the focus of our studies of children’s learning. Further, children’s learning, in school as well as in preschool, may be seen as local and practical organizations where knowledge and competence are ‘installed’ (Macbeth 2000), and where educational facts are collaboratively and contingently constituted in interaction. Thus, if we are to provide insight into the nature of learning as embedded in the wider process of schooling, we need to closely attend to the social organization of interaction in educational contexts. Focusing on two local educational activities—‘sharing time’1 and ‘language group’—the present study is based on a view of learning as situated and constituted through social interaction. This perspective on learning originates from Vygotsky’s (1986) work on the relationship between language and the development of thinking and learning, but this view highlights the social, rather than the cognitive nature of learning: Learning is seen as something taking place between people rather than strictly inside the individual mind (Lave and Wenger 1991; Macbeth 2000; Mehan 1979; Rogoff 2003; Sa¨ljo¨ 2000).
INSTRUCTIONAL ROUTINES IN MULTILINGUAL EDUCATION Ethnographic studies of language socialization have often stressed the role of language routines in children’s language development (e.g. Ochs and Schieffelin 1995; Peters and Boggs 1986; Watson-Gegeo and Gegeo 1986). This is no less true for second language learning, and Ervin-Tripp (1986) stresses the role of play in children’s second language development. She points out that play activities show a number of recurrent patterns of action, and proposes that such recurrences may enable children to participate in peer group activities in spite of minimal language resources (cf. Cekaite 2006). Such participation in turn allows them to explore and develop a more nuanced communicative repertoire in their second language. Furthermore, play activities allow children to incorporate action patterns
¨ RK-WILLE´N POLLY BJO
557
from the classroom into peer group interaction (see also Bjo¨rk-Wille´n and Cromdal, in press), hence enabling children to explore some normative patterns of instructional talk between teachers and students. Along somewhat similar lines, Willett (1995) examines the interaction between three 1st-grade second language learners, showing how their talk draws upon routine features of teacher–student interaction. Clearly, drawing on institutionally established routines for accomplishing recurrent, and hence recognizable, activities is an important resource in second language learning. In a study of language socialization of a young Moroccan learner of Italian, Pallotti (1996, 2001) shows how the girl continuously acquired ways of participating in classroom talk by building on the talk of previous speakers within the structure of recognizable activities. By gradually adding new elements to recycled turns she was able to extend her lexical, pragmatic, and wider communicative repertoire. Somewhat similarly, Kanagy (1999) shows how young preschool children with no initial knowledge of the L2 rely on scripted instructional routines for their development of early interactional competence. By highlighting the role of verbal routines in recurring social activities, language socialization studies manage to anchor the process of second language learning within the cultural orders and practices in educational settings (Saville-Troike and Kleifgen 1986; Willett 1995). A distinct line of inquiry into second language learning and instruction is represented by a growing number of interaction analytic studies influenced, in varying degrees, by central ideas within ethnomethodology (e.g. Cromdal 2005; He 2004; Markee 2000; Slotte-Lu¨ttge 2005; Steensig 2001). While clearly distinct with respect to modes and detail of analysis, interaction analysts and researchers into language socialization both share a strong orientation towards the local practices through which second language learning and instruction takes place. For instance, analysing a conversation between a native speaker of English enrolled in an academic course in German and a native speaker of German, Kasper (2004) focuses on the ways through which the participants invoke different language identities by orienting to trouble with the production talk in German. Her analyses demonstrate how the interaction moves back and forth between mundane conversation and institutional (instructional) talk brought off by the non-native speaker’s repair initiations. Kasper’s (2004) study demonstrates how detailed analyses of L2 interaction allow us to treat participants’ identities—for instance, as native or non-native speaker—as products or their own actions, as invoked, displayed, and oriented to in and through the talk. Clearly, conversational repair presents a range of issues relevant to second language acquisition, over and above issues of speaker identity, and conversation analysts have explored a range of repair phenomena in second language educational settings and their relation to the learning affordances set up by different practices of repair in L2 instruction (Juvonen 1989; Kurhila 2001; Lehti-Eklund 2002; Seedhouse 2004; Slotte-Lu¨ttge 2005).
558
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
Focusing on the organization of different types of instructional activities in L2 education, Seedhouse (2004) highlights the reflexive relationship between the local organization of turns at talk and the curricular focus of the activities in which that talk is embedded. Along this line of reasoning, the specific educational focus of an instructional activity provides learners with cues on how to proficiently take part in the ongoing interaction. Hence, the recurrent and recognizable structure of instructional activities enables L2 learners to meaningfully participate in the institutional business of the L2 classroom despite often very limited language skills. However, this need not imply that L2 students always follow the instructional agenda, and in a series of recent studies of a Swedish immersion classroom, Cekaite and Aronsson (2004, 2005) show how early learners of Swedish draw upon a range of instructional routines across different classroom activities, resulting in humorous ‘time-outs’ from the currently ongoing instruction. Although resulting in a temporary suspension of the ongoing instructional activity, such actions allowed novice learners of Swedish to take part in, and contribute to, the ongoing classroom discourse. These discourse contributions comprised the very early steps in the students’ learning of Swedish—as well as their ability to take part in formal educational events (Cekaite and Aronsson 2004). The studies above highlight, in different ways, the routine aspects of interaction in bilingual and second language education. As such, they form the backdrop for the present analysis which focuses on the different interactional resources through which preschool children participate in two distinct types of multilingual instruction: sharing time and ‘Spanish group’. More specifically, the aim is to investigate the in situ accomplishment of routine instructional activities through which the preschool’s trilingual program is put into daily practice. In a nutshell, the analysis will flesh out some of the interactional means through which preschool teachers and their students talk trilingual education into being.
A MULTILINGUAL PRESCHOOL The present analysis draws on data from fieldwork in a multilingual preschool in a mid-sized Swedish town. The preschool at hand features a multilingual programme, including Swedish, English and Spanish. Swedish is the main language of interaction, but English and Spanish are routinely used for mundane as well as formal educational activities throughout the day.
Multilingual activities Each day at the preschool entailed several daily scheduled activities, such as sharing time, organized group activities (including language groups), outdoor play, meals, as well as a daily nap for the youngest children. In addition, there is also scheduled time for ‘free’ play (cf. Bjo¨rk-Wille´n and Cromdal,
¨ RK-WILLE´N POLLY BJO
559
in press). Common activities, such as sharing time, may involve the use of several of the three languages available at the preschool, with Swedish serving as the common base language, or lingua franca, whereas other activities target one specific language, such as, for instance, the ‘language groups’ are more or less strictly conducted in Spanish or in English. The language group activities, the Spanish group (‘spanskagruppen’) and the English group (‘engelskagruppen’), take place four times a week, directly after sharing time. The curricular aim of the two language groups is to ensure the children’s familiarity with both English and Spanish in a playful manner. In these ‘language groups’, the Spanish teacher typically introduces Spanish vocabulary and simple phrases. Similarly, the bilingual Swedish/English teacher introduces some basics of English communication.
Children and staff There are a total of 36 children at the preschool, including (a) a group of toddlers (12 children) and (b) a group of children 3–5 years of age (24 children). The present study draws on recordings of the latter group. The study involves six children from Swedish-speaking homes, five children from Spanish-speaking homes, five from English-speaking homes and eight children from other language backgrounds (German, French, Arabic, Suryoyo, Latvian, Tamil, and Persian). In several cases, the children speak more than one language at home, for example Spanish and French. The staff members responsible for this group also represent different language backgrounds: Spanish as L1 (Irina), Arabic as L1 (Fatima), English as L2 (Susanne), and Swedish as L1 (Sofia).2 All staff members speak Swedish sufficiently for daily purposes at the preschool, but Susanne and Sofia are the only native speakers of Swedish.
Observations and recordings The children’s everyday activities at the preschool were observed and recorded during a period of 5 months. During data collection, the video camera served as the main tool for recording the interaction (a total of 37 hours). Detailed analyses of situated ‘classroom’ interaction allow us to explore the participants’ methods for accomplishing and participating in educational activities, that is, to investigate participants’ ways of exploiting and reflexively producing contexts for learning.
Taking attendance: teacher improvization as trouble source Sharing time is a daily morning activity, during which children and teachers typically gather in a circle on a big rug at the preschool. One goal with sharing time is to greet each other with ‘good morning’ (in different languages) and to take attendance. Another goal is to sing and talk in the
560
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
different languages, mostly Swedish, English, and Spanish. It is also a forum for gathering information, for conducting various thematically organised activities, and last but not least, it serves to distribute the morning fruit among the children. Sharing time is always led by one of the teachers, while the other teachers join in, seated among the children as group members, but also supporting the lead teacher. (Please see the online version for a more extensive introduction of the basic attendance taking routines at the sharing time, available online to subscribers at http://applij.Oxfordjournals.org/)
Language choice, repair, and the management of conflicting interactional projects Most of the children start attending the present preschool group at the age of 3 and finish when they are 6 years old. Hence, some of the children are more experienced than others in participating in routine activities such as sharing time. Children’s participation in these daily routines is assumed to result in changes in their knowledge of curricular matters, their social skills, and their discourse competence. During sharing time, the more experienced children are treated as old-timers (Lave and Wenger 1991). The language skills also differ between the children, not only due to age differences, but also to their differing language backgrounds. The latter holds true for the teachers as well. The first two transcripts illustrate the children’s participation in a specific activity that routinely took place during sharing time, namely taking attendance. When the transcript begins, the Swedish teacher Sofia is seated on a small chair within sight of all the children. There is a cotton tapestry hanging on the wall nearby, picturing a train with several trucks in different colours. As part of the tapestry, there is a box containing a number of rag dolls. Each doll represents one individual child. The typical procedure for taking attendance is that the teacher produces a doll from the box, and the particular child or/and the group of children identifies the doll. The child then receives the doll and is expected to put it in one of the trucks on the tapestry. Normally, children are free to pick a truck of their own choice. However, during this particular sharing time event (shown below in Excerpts 1 and 2), it is the teacher who designates the truck for each doll by naming the colour of the truck in English. The routine character of this activity provides for its recognisability for everyone at the preschool. It is not surprising, then, that the interaction between the teacher and the children most of the time runs very smoothly. However, this is not always the case, and the following sections will show in different ways how a local difficulty of interpretation may arise when teachers introduce new elements into routine multilingual activities. The following excerpt shows how the participants’ diverse language skills are interactionally displayed and acted upon within the flow of talk. When the transcript begins the teacher is just producing Gustavo’s doll.
¨ RK-WILLE´N POLLY BJO
Excerpt 1 (Video 3, 2002–11–12)
561
562
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
Instead of initially naming a colour, the teacher asks Gustavo (Gusta) what language he prefers: ‘English or Spanish?’ (lines 4–6). She thus turns the issue of language choice into a topic, thereby introducing a new element into the routine format of this activity. During a relatively long pause (line 7), Gustavo, who is standing in front of the teacher, does not respond. Nicolaus, one of the oldest boys in the group, appropriates Gustavo’s proposal that Spanish would be best (line 9). In offering this reply, Nicolaus not only recognises that an answer is expected, but he may be seen to orient to what he believes is Gustavo’s language preference. In so doing, Nicolaus acts upon and displays, partly on behalf of Gustavo, the type of knowledge that the present situation makes relevant. As Macbeth (2000) points out: ‘Classroom discourse is a grammar of activity in and through which students see and find the sense of their lessons, including things like the competence of their fellows . . .’ (p. 42). However, the teacher rejects Nicolaus’ (Nico) contribution by latching on to his utterance and rapidly declaring, in a mildly irritated voice, that she wants Gustavo to make the choice himself (lines 10–11). Nicolaus’ contribution is thus treated as a breach against the local organization of turn allocation. In other words, the teacher openly disqualifies Nicolaus’ contribution, because it is evidently inconsistent with her initiation move (Mehan 1979). However, in the absence of a response from Gustavo, she answers her own question, proffering the colour specification in both English and Spanish (lines 12–13), thus orienting to the absent—but expected— language choice on Gustavo’s behalf. This clearly terminates a failed IRE (Initiative–Response–Evaluation) sequence, as she, in fact, produces the expected response herself. One of the upshots of this move is to stress that an allocated initiating move has a normative feature. It is produced for a unique recipient, who cannot be replaced by other children. Hence, this brief exchange may be seen as a public lesson in the normative workings of the social and educational order of the preschool. However, the lesson at hand is not merely about the forms of participation in sharing time activities—it is also an instance of language instruction in practice. Hence, in answering her own question, the teacher offers both the English and the Spanish terms for the color ‘blue’. What is notable about the turn in line 15 is that teacher Sofia, one of the Swedish teachers at the preschool, pronounces the Spanish word incorrectly: ‘azoll’, and in the subsequent turn, a child immediately initiates a repair sequence by offering the correct pronunciation. In instructional exchanges, such other-initiated other-repair (Schegloff et al. 1977; Schegloff 2000), through which the recipient of talk locates a trouble source and offers a corrected version, are overwhelmingly carried out by the teacher (cf. McHoul 1990). In the second language classroom, repair phenomena may be more complex, as Seedhouse (2004) points out in his discussion of the indexical-reflexive relation between the pedagogical focus of the instructional activity and the organization of repair sequences.
¨ RK-WILLE´N POLLY BJO
563
In a sense then, we are dealing here with an instance of public language instruction in which the roles have been reversed. What makes this reversal possible is the specific domain of knowledge relevant to the situation at hand, namely knowledge of Spanish phonology. This allows some of the children in the group to display for the whole group a higher degree of competence than that held by the teacher. Upon the repair initiating turn in lines 16–17, the teacher latches on to the child’s correction, and produces a phonological repair of the word azul, followed by a repetition (lines 18–19), hence displaying a recognition of her own mispronunciation and the need for its remedy. Overlapping with the Swedish teacher’s self-repetition, the Spanish teacher Irina, produces the correct form of ‘azul’ in line 20. Given Irina’s authority as the Spanish teacher, this action can be seen as a validation of the repair sequence in lines 16–19, but it also seems to provide the public norm for the pronunciation of the colour term, as evidenced in line 21, in which a child silently repeats the word, thus orienting to Irina’s pronunciation as instructionally implicative. At the same time, Gustavo is slowly moving—doll in hand—from the teacher towards the tapestry with the train trucks where the dolls are to be placed. The Swedish teacher then repeats her earlier instruction (preserving the bilingual format, as well as the erroneous pronunciation; line 23) to Gustavo to place his doll in the blue truck. Gustavo rebukes this instruction, suggesting in Swedish that he prefers the grey truck: ‘nej jag vara gra˚ (.) den’ (‘no I be the grey (.) this one’; lines 25–27). Gustavo hereby shows an agenda that differs from the Swedish teacher’s, in that he intends to place his doll in the same grey truck as Irina, the Spanish teacher, has placed her doll earlier during this sharing time event. By claiming a right to chose a different truck than the one suggested by the Swedish teacher, Gustavo can be seen to act upon the standard organisation of the attendance taking routine, according to which it is up to each child to pick a truck for her/his doll. Gustavo’s choice of language at this particular point of interaction is interesting: by producing his reply in one of his weaker languages (as displayed by the incorrect verb inflection in line 25, as well as the teacher’s orientation to his language preference in line 23), Gustavo manages to decline the language choice options proffered by the teacher in the initial phase of this interaction. Thus, his disaffiliation with the teacher’s instruction is displayed both at the level of the content of talk as well as his choice of language (for a study of children’s use of code-switching in the construction of oppositional stances, see Cromdal 2004). The above analysis targets some difficulties arising from the organization of instruction during sharing time. It could of course be argued that some of the educational difficulties may be related to children’s varying experience of instructional events. Some children already know the routines, whereas others are still exploring the territory. However, we have seen that in spite of his ‘old-timer’ position, Nicolaus’ attempt to supply a noticeably absent reply to a preallocated teacher question (line 9) is treated as a violation of the
564
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
norms governing turn-taking in formal events at the preschool. The teacher’s handling of such norm infractions involves efforts to restore the institutional order—to bring the interaction back to a routine activity. This practice then becomes in itself a lesson for the children in the interactional workings of schooling in practice. However, in the current extract, the problems arise with the teacher’s deviation from the attendance taking routine. In particular, three nonstandardized elements were oriented to by the children: the Swedish teacher’s insistence on choosing the truck for the children, her delegating the issue of her own language choice to Gustavo, and her notably erroneous use of Spanish. The first deviation relates to Sofia’s attempt to integrate the routine activity of taking attendance with L2 instruction: her naming the colour of the target truck in English allows the students to practise and display their passive colour vocabulary. However, the teacher’s instructional project is clearly at odds with Gustavo’s local concerns, and we learn towards the end of the transcript that he insists on picking his own truck. The disparity of Gustavo’s and the teacher’s interactional projects has a bearing on the second deviation from the attendance taking routine, namely the teacher’s delegating her own language choice to Gustavo. Here, Gustavo’s noncompliance with the teacher’s request to display his language preference for this exchange (line 7) may be accounted for in terms of his reluctance to letting Sofia choose the truck for him. Finally, Gustavo’s refusal to go along with the teacher’s instructional project has profound implications for the subsequent interaction, as it leaves the issue of his language preference unresolved. Accordingly, the teacher chooses to specify the target truck in both English and Spanish, the latter resulting in a side-sequence in which her erroneous pronunciation in Spanish was collaboratively repaired.
Some collaborative techniques for accomplishing multilingual activities In the previous extract, it was shown how the teacher’s attempt to integrate the attendance taking activity with L2 instruction resulted in an observable deviation from the routine organization of the interaction during this activity. In this section, I will further explore some distinct, interactionally deployed techniques that the teachers employ to accomplish language instruction during sharing time. The excerpt below took place within the same sharing time event as the previous extract and further illustrates how colour terms are used as part of taking attendance.
Excerpt 2 (Video 3, 2002–11–12)
¨ RK-WILLE´N POLLY BJO
565
566
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
In rough outline, Sofia shows the ‘Henrik doll’ and Henrik picks it up. She then indicates that he should place the doll in the orange truck by saying orange three times (lines 4–5). However, rather than completing the expected course of action—placing the doll in the orange truck—Henrik keeps staring at the multicoloured trucks on the tapestry (line 9). It is impossible to tell whether he is hesitating because he does not recognize the English colour term or because he is just working out which of the two orange trucks to choose. However, the teacher seems to act upon the former interpretation—that he does not know the English colour term for orange—and repeats the colour term, first in Swedish, then in English (line 10). Latching on to the teacher’s last ‘orange’, Henrik ultimately puts his doll in one of the orange trucks (line 11). Note also that there is a small gap in Sofia’s utterance (between lines 4 and 10) where Irina, the Spanish teacher, tells Gustavo to pay attention. Her silent voice and choice of language suggests that this is a side activity, addressed only to Gustavo. Moreover, her use of Spanish contrasts with the current ‘official’ activity, and this choice of constitutive language may be a way of accomplishing a parallel activity, rather than disrupting Sofia’s role taking. On the contrary, it is clear that Irina is also cooperating with Sofia, as well as directing Gustavo’s attention to the main activity (line 8). Therefore, the gap Sofia leaves before she finishes her invitation to Henrik may be seen as a smooth interactional device, allowing Irina to finish her admonition and Sofia to regain full attention. Sofia’s way of repeating the colour term ‘orange’ in English several times is an instructional technique typical of language teaching (Chaudron 1988). In line 10, she also provides the Swedish equivalent and then translates it back into English. This turn shape, comprising repetition interspersed
¨ RK-WILLE´N POLLY BJO
567
with translation (repetition–translation–repetition), was recurrently observed at the preschool and may be seen as a teaching strategy with distinct cognitive overtones: in the present extract, it provides a way of making sure that Henrik and the other children have understood the word in a ‘correct’ manner (Lemke 1990).4 Notice that the teacher is using the local availability of several languages to deal with a problem of potential incomprehension on Henrik’s part, while Irina may be seen as using this availability to produce a non-distributive turn in the midst of the change. In lines 12–13, after Sofia has provided feedback to Henrik, it is Anna’s turn to place her doll. Anna is a 4-year-old girl who speaks both Swedish and Spanish at home; this is her second year in the group. The teacher says ‘brown’ in English using a soft and low voice, and simultaneously shows the ‘Anna doll’ (lines 12–13). In the next line, Anna takes her doll and, in accordance with the activity routine, puts the doll in a brown truck. After a positive evaluation of Anna’s action, Sofia proceeds to ask Anna to teach her the Spanish word for brown. Her request may be seen as an explicit educational strategy, where she highlights the situated import of the request as an educational routine: ‘kan du la¨ra mig det’ (‘can you teach me that’) (Mercer 1995). The utterance also construes Anna as more skilled in Spanish than the Swedish teacher (lines 16–18). Anna, however, does not respond; instead she looks down when passing Sofia on her way back to her place among the other children (line 19). In the absence of immediate compliance from Anna, Sofia begins to vocalize the Spanish word for brown. However, her vocalization of the first phoneme is incorrect: ‘na’ (line 18), and Sofia displays her own uncertainty by breaking off mid-word and turning to look at the Spanish teacher (lines 20–21). At the same time, Anna orients to the mispronunciation by looking up at Sofia with a swift headshake. Fixing her gaze at the Spanish teacher, Sofia produces a hesitant ‘na¨?’ (‘no?’, line 24), thus displaying her inability to produce the correct Spanish colour term. In this way, Sofia’s actions comprise a self-initiated other-repair (Schegloff et al., 1977), and in line 25, Irina begins to produce the correct term. However, rather than completing the correction, Irina uses a word-search format (Goodwin and Goodwin 1986) ‘ma::’, managing in this way to solicit the engagement of several children (Anna included), who produce the correct Spanish term for the colour brown (lines 26 through 27–29). Following up on this, Irina elaborates on the Spanish colour term, presenting a synonymous alternative, cafe´ (line 30). Receiving this as news, Sofia enthusiastically picks up on the latter term and presents the children with a mnemonic device, pointing out that coffee is in fact brown. Several children follow suit by repeating the term (lines 35 and 38). Hence, by highlighting the relation between the colour brown and its referent on behalf of the children group, Sofia regains the initiative as head teacher in the current activity.
568
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
This example shows that the teachers are using different educational techniques such as repetition, attention getting, guidance, elaboration, association, etc. (cf. Mercer 1995) to accomplish language instruction while taking attendance. It also shows how, in pursuing a multilingual instructional agenda, the Swedish teacher initially relies on the children’s cooperation in producing locally appropriate terms in Spanish. Failing that, she appeals to the expertise of the assisting Spanish teacher, who readily manages to accomplish what the Swedish teacher could not: to solicit Spanish talk from the children. This episode then clearly shows the children’s local sensitivity to matters of language choice and recipient design. Ultimately, the children seem to act upon an understanding of the teachers’ language preferences, such that they will respond to the Swedish teacher’s talk in Swedish or English, depending on local circumstances. Their use of Spanish, however, is largely restricted to exchanges with the Spanish teacher. Clearly, the children are facing a range of concerns related to their participation in a multiparty activity, which potentially includes more than one teacher. This involves making sense of the routine features of the unfolding activity and the local organization of different languages in multilingual instruction. Hence, the children’s successful participation in such events involves knowing the other participants’ language skills as well as designing their actions in accordance with this knowledge. Finally, I have argued that a potential trouble source for the children’s participation in this sharing time episode has to do with the institutional hybridity of the event, resulting from the Swedish teacher’s attempts to integrate two routine yet distinct activities: taking attendance and language instruction. Against this backdrop, let us now move on to consider one of the frequently occurring ‘language group’ activities, which explicitly focuses on aspects of language instruction and learning.
TEACHER FEEDBACK AS TROUBLE SOURCE IN THE SPANISH GROUP In moving from the daily sharing time activity of taking attendance to the activities of the Spanish group, I will highlight some features of teacher feedback in language instruction. In particular, activities in which the teacher questions the children for specific words in the target language entail children’s expectations for routine feedback.
Making sense of teacher feedback In the Spanish group, as well as during sharing time, there is a formal organization of talk, especially when the focus is on learning issues. This implies that the teachers present the topics and initiate the joint activities
¨ RK-WILLE´N POLLY BJO
569
and that almost any deviation from this pattern is handled as repairable (McHoul 1978). Part of the normative organization of the Spanish group is that the children are expected to participate in the activities using Spanish. To be successful and fully recognized by the teachers, the children must discover and adapt to this routine. However, the instructional procedure is not always easy to penetrate for the children, because teachers do not always stick to the routinized organization of the events. The final excerpt shows how children work at figuring out how to produce the answer that the teacher expects. (Another example of teacher feedback in the Spanish group is available online to subscribers at http://applij.Oxfordjournals.org/) Excerpt 3 shows the opening of a teacher-led play event, involving small beanbags. The bag distribution procedure that we will explore in the excerpt below involves the introduction of a physical activity. Here, the children are expected to use the bags for different purposes, such as throwing them up in the air and catching them, or walking around the room with a beanbag on their head, hand, foot, back, etc. These exercises are accompanied by music. The excerpt begins when the teacher shows the first beanbag.
Excerpt 3 (Video 3A, 2002–09–26)
570
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
¨ RK-WILLE´N POLLY BJO
571
This episode shows how the children respond to and coordinate their talk with their teacher during the Spanish activity. During this introduction part, Irma is very eager to get the first beanbag, jumping up and down saying: ‘den vill jag ha’ (‘I want that one’; line 2). The teacher gives Irma the bag (line 5) and then asks explicitly about its colour (line 6). Note that the teacher is using Swedish to present children with the procedural information for this activity. Thus, on the level of conversational organization, the question in line 6 results in a switch into Spanish. Irma aligns with the teacher’s language choice in line 15 and this exchange sets off the Spanish activity, the normative feature of which seems to be naming the colour of the beanbag in Spanish. In the next sequence the teacher shows a red bag, and Anna correctly identifies its colour in Spanish. What we may interpret as a confirmation of Anna’s answer in line 20, when the teacher repeats ‘rojo’, may also be interpreted as a collectively addressed turn made for all children. The repetition of rojo is followed by a translation to Swedish, where the teacher identifies rojo as the Spanish word for red. We see that the teacher orients to Swedish, which works as a lingua franca, when she provides explanations and sometimes even when she asks
572
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
questions. Nevertheless, the routine being established here requires that the children answer in Spanish. Hence, we see how the children continue to answer in Spanish, despite the teacher’s occasional switches to Swedish, thus orienting to the ‘Spanishness’ of the activity (lines 20 and 26). In other words, the children orient to perceived normative expectations regarding the choice of language in this instructional event. In spite of this, the relation between producing the correct answer and receiving a beanbag, which is the fundamental evaluative move in this activity, is not very clear for the children. For example, Anna produces the right answer in line 20, without receiving a bag, which is instead given to Evelina (line 22). However, as the turn moves to Michael in lines 30–33, the event takes a rather different direction, as he repeatedly fails to provide the answer the teacher expects. How can we account for his failure? An intuitively appealing answer might be that Michael’s failure to provide the Spanish word for blue has to do with his limited language proficiency. However, upon closer inspection of the unfolding organization of this event, the picture looks rather different, and we see that Michael is very consistent in his way of answering. At first, when Anna answers in Swedish ‘bla˚’ (‘blue,’ line 34), Michael follows suit, producing the Swedish colour term (line 35). Then, when Anna and Henrik, in lines 39–42 and 46–48, produce the ‘expected’ answer twice without the teacher producing the feedback to establish its correct status, Michael consistently does not follow suit. Instead, he tries to come up with an alternative, testing English in line 44, and finally settling for the French word for the colour blue in line 50, hence avoiding giving the answer in Spanish. I would thus argue that Michael’s failure to produce the expected answer is not related to his limited proficiency in Spanish. Rather, it should be seen as a result of the unfolding interaction, in which the teacher fails to furnish the expected feedback by acknowledging the ‘correct’ answer that has been repeatedly produced by several other children. Whatever the teacher’s reason for sidestepping the expected routine (not awarding Anna a beanbag in line 20 and not accepting the correct answer from Anna and Henrik in lines 39–42 and 46–48), the publicly visible result is an inconsistent orchestration of this educational activity, which, as has been argued, puts Michael in an interactional cul-de-sac. In light of this, it is interesting to see the teacher’s subsequent line of action, in which she simply provides the colour terms herself, distributing the bags to the remaining children (lines 58–62). The well-known routine is thereby suspended, and we can only speculate as to whether this is her way of getting around the impasse and getting on with the physical activity. It clearly brings about a radical change in the children’s way of participating, in that they no longer have to tackle issues of language choice and other concerns about turn organization. They simply have to catch their beanbags. This excerpt thus shows how children rely for their successful participation in structured instructional events on the recognizable, routine features of Spanish group activities. Having grasped the procedural routines of
¨ RK-WILLE´N POLLY BJO
573
language instruction, children design their actions in line with the expected—or projected—trajectory of interaction, and deviations from the expected course of action may comprise a source of interactional trouble.
CONCLUDING DISCUSSION The overall aim of this paper has been to highlight the in situ accomplishment of educational activities in a multilingual preschool. Two different types of activities have been explored: sharing time, led by a Swedish preschool teacher, and Spanish group, conducted by the Spanish teacher. As the analyses show, a fundamental concern for the children participating in these activities has to do with the routine organization of interaction, which provides for the predictability of relevant next actions. The significance of children’s expectations becomes clearly observable in the trouble arising with the teachers’ improvising, elaborating, or otherwise deviating from the established, routine performance of these activities. This general point may seem trivial, as several studies have argued for the educational upshots of activity routines in second language learning (e.g. Ervin-Tripp 1986; Kanagy 1999; Pallotti 2001; Willett 1995), highlighting the predictability of expected action trajectories as a resource for the children’s participation in educational practice. However, the interactional workings of activity routines—their local relevance for the children’s moment-to-moment accomplishment of participation—has not been widely explored in studies of multilingual or second language education (but see Cekaite 2007; Seedhouse 2004). The analyses, then, contribute to the understanding of multilingual instruction as a contingently sensitive practice, and it does so by examining, in some degree of interactional detail, the trouble arising with instructional routines being somehow modified or suspended, as well as the management and resolution of such trouble. One type of children’s procedural expectations highlighted in the data has to do with the teacher’s language preferences, as oriented to and acted upon by the children. Here, the analyses have shown how the Swedish/English teacher’s attempts at doing Spanish instruction resulted in elaborate repair work, initiated by the children (Excerpt 1). As a possible result of this experience, we have seen how the teacher later on consults the expert Spanish knowledge of one of the children (Excerpt 2). This reversal of the institutional roles and relationships between teacher–child constitutes another alien element in the registry taking activity, resulting in some procedural trouble, which was ultimately resolved through the authority and expertise of the Spanish teacher. Another normative feature of instructional exchanges at the preschool has to do with teacher feedback. More specifically, children overwhelmingly expected their correct responses to be assessed in positive terms by the
574
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
teacher and I proceeded to show in Excerpt 3 how the lack of teacher feedback following upon several correct answers from the group raised some acute problems for Michael in predicting the relevant response to the teacher’s prompts. Let us consider some educational implications of these findings. The above analyses of the interactional trouble arising with the teacher’s deviations from the expected course of events suggests that children monitor the unfolding activity for its routine features, and further, that they use the predictability of the routine to participate in the interaction. This is not to suggest that teachers should constantly carry out instruction according to a fixed format. For instance, the first two excerpts showed a somewhat modified procedure for registry taking, according to which the teacher would temporarily take over the task of nominating the truck in which to place the student dolls. This modification as such did not counter—or fail— the children’s expectations to the extent that they were not able to participate in the activity. This would suggest then that children’s interactional competence is flexible enough to allow for novel elements to be introduced into routine activities. However, such variations may need to be clearly introduced, so as to enable children to predict the expected trajectory of the interaction. The analyses therefore allow us to conclude that whereas young children need the routines to participate in preschool business, they show enough interactional skill to follow the teachers’ modifications of recurring activities, provided that the procedural upshots of these elaborations are made clear. Finally and in conclusion, the focus has been on children’s trouble in ‘reading’ some educational routines, particularly in those cases where the teachers would depart from the expected course of action. This might seem ironic, as our data show that some of the teachers’ modifications and elaborations of the standard activities are explicitly designed to enhance children’s engagement and active participation in the educational practice at the preschool. However, I want to point out that the repair exchanges through which routine-related procedural problems are handled provide opportunities for the children’s early learning of foreign languages. As Goodwin (2000) puts it: if a child grows up in an ideal world where she heard only wellformed sentences she would not learn to produce sentences herself because she would lack the analysis of their structure provided by events such as repair. (Goodwin 2000: 5) Clearly, such learning is inseparably embedded in the children’s development of social and interactional skills. Thus, recognizing and remedying interactional trouble may be a crucial way of learning to become a competent participant in social life, including the context of organized multilingual activities in preschool. At present, we are far from understanding in any
¨ RK-WILLE´N POLLY BJO
575
sufficient detail the educational activities taking place in multilingual preschools and the upshots of these local practices for children’s early learning. Hence, the present work contributes to our understanding of very young bilinguals’ ways of managing commonplace social activities in educational settings—a hitherto largely overlooked line of inquiry in Swedish research on bilingualism.
ACKNOWLEDGEMENTS I am very grateful to Karin Aronsson, Jakob Cromdal, Michael Tholander, Asta Cekaite and Mathias Broth for their support and many helpful comments. Financial support from the Committee for Educational Research of the Swedish Research Council is gratefully acknowledged.
TRANSCRIPTION KEY [ Underlining CAPITALS
(3) (.) (( text )) : 4word5 5word4 ¼ ! #" (. . .) (text) (x) (xx) he he text text text text
Square brackets mark the start of overlapping speech Signals emphasis; the extent of underlining within individual words locates emphasis Marks speech that is obviously louder than surrounding speech Quieter speech Pauses measured in seconds Micropause Transcriber’s comments Prolongation of preceding vowel Quicker speech Slower speech Immediate ‘latching’ of successive talk Utterance interrupted or ebbed away Highlights a particular feature discussed in the text Indicates falling or rising intonation in succeeding syllables(s) Talk has been omitted Uncertain interpretation Inaudible word or words Laughter Swedish Spanish English Translation into English
576
PRESCHOOL CHILDREN AND MULTILINGUAL INSTRUCTION
NOTES 1 ‘Sharing Time,’ which may also be known as ‘Circle Time’ or ‘Show and Tell,’ is an umbrella activity that is widely practised throughout the Western world in both L1 and L2 primary classrooms and in preschools. Normally the children sit in a circle and the focus is on creating a sense of community and developing social and interactional skills (Yazigi and Seedhouse 2005). 2 All personal names are pseudonyms.
3 ‘Azoll’ is the teacher’s pronunciation of the Spanish word ‘azul’ (blue). Because Swedish is a non-phonetic language, I have used standard written language throughout the transcripts, except when pronunciation has an impact on the interaction, such as in the case of ‘azoll.’ 4 Lemke (1990) shows how thematic patterns are communicated using the strategy of repetition-with-variation to explore semantic relations for a wider understanding.
REFERENCES Bjo¨rk-Wille´n, P. and J. Cromdal. In press. ‘When education seeps into ‘‘free play’’. How preschool children accomplish multilingual education,’ Journal of Pragmatics. Cekaite, A. 2006. Getting started. Children’s participation and language learning in an L2 classroom. Linko¨ping Studies in Arts and Science, 350. Linko¨ping: Linko¨ping University. (Diss.) Cekaite, A. 2007. ‘A child’s development of interactional competence in a Swedish L2 Classroom,’ The Modern Language Journal 91/1: 45–62. Cekaite, A. and K. Aronsson. 2004. ‘Repetition and joking in children’s second language conversations: Playful recyclings in an immersion classroom,’ Discourse Studies 6/3: 373–92. Cekaite, A. and K. Aronsson. 2005. ‘Language play, a collaborative resource in children’s L2 learning,’ Applied Linguistics 26/2: 169–91. Chaudron, C. 1988. Second Language Classrooms: Research on Teaching and Learning. Cambridge: Cambridge University Press. Cromdal, J. 2004. ‘Building bilingual oppositions: Code-switching in children’s disputes,’ Language in Society 33: 33–58. Cromdal, J. 2005. ‘Bilingual order in collaborative word processing: On creating an English text in Swedish,’ Journal of Pragmatics 37: 329–53. Ervin-Tripp, S. 1986. ‘Activity structure as scaffolding for children’s second language learning’ in J. Cook-Gumperz, W. Corsaro, and J. Streeck (eds): Children’s World and Children’s Language. Amsterdam: Mouton de Gruyter.
Goodwin, C. 2000. ‘Practices of seeing: Visual Analysis: An Ethnomethodological Approach’ in T. van Leeuwen, and C. Jewitt (eds): Handbook for Visual Analysis. London: Sage Publications. Goodwin, M. and C. Goodwin. 1986. ‘Gesture and coparticipation in the activity of searching for a word,’ Semiotica 62/1–2: 51–75. He, A. W. 2004. ‘CA for SLA: ‘‘Arguments from the Chinese Language Classroom’’,’ Modern Language Journal 88/4: 568–82. Juvonen, P. 1989. ‘Repair in second-language instruction,’ Nordic Journal of Linguistics 12: 183–204. Kanagy, R. 1999. ‘Interactional routines as a mechanism for L2 acquisition and socialization in an immersion context,’ Journal of Pragmatics 31: 1467–92. Kasper, G. 2004. ‘Participant orientations in German conversations-for-learning,’ Modern Language Journal 88/4: 551–67. Kurhila, S. 2001. ‘Clients or languages learners— being a second language speaker in institutional interaction’ in R. Gardner and J. Wagner (eds): Second Language Conversations. London: Continuum. Lave, J. and E. Wenger. 1991. Situated Learning. Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Lehti-Eklund, H. 2002. ‘Dictoglossamtal som o¨vningsform i svenska spra˚kbadsklasser’, in M. Saari (ed.): Samtal och Interaktion (Conversation and Interaction). Meddelande fra˚n institutionen fo¨r nordiska spra˚k och nordisk litteratur vid Helsingfors universitet Vol. B22. Helsinki: Helsinki University.
¨ RK-WILLE´N POLLY BJO
Lemke, L. J. 1990. Talking Science: Language, Learning and Values. New Jersey: Ablex Corporation Norwood. Macbeth, D. 2000. ‘Classrooms as installations. Direct instructions in early grades’ in S. Hester and D. Francis (eds): Local Educational Order Ethnomethodological Studies of Knowledge in Action. Amsterdam/Philadelphia: Benjamins. Markee, N. P. P. 2000. Conversation Analysis. Mahwah, NJ: Lawrence Erlbaum. McHoul, A. 1978. ‘The organisation of turns at formal talk in the classroom,’ Language in Society 7: 183–213. McHoul, A. 1990. ‘The organisation of repair in classroom talk.’ Language in Society 19: 349–77. Mehan, H. 1979. Learning Lessons, Social Organization in the Classroom. Cambridge, MA: Harvard University Press. Mercer, N. 1995. The Guided Construction of Knowledge. Talk amongst Teachers and learners. Clevedon-Philadelphia-Adelaide: Multilingual Matters. Ochs, E. and B. Schieffelin. 1995. ‘The impact of language socialization on grammatical development’ in P. Fletcher and B. MacWhinney (eds): The Handbook of Child Language. London: Blackwell. Pallotti, G. 1996. ‘Towards an ecology of second language acquisition: SLA as a socialization process’ in E. Kellerman, B. Weltens, and T. Bongaerts (eds.): Proceedings of EUROSLA 6, Toegepaste Taalwetenshap in Artikelen, 55. Pallotti, G. 2001. ‘External appropriations as a strategy for participating in intercultural multiparty conversations’ in A. Di Luzio, S. Gu¨nthner, and F. Orletti (eds): Culture in Communication. Amsterdam: Benjamins. Peters, A. and S. Boggs. 1986. ‘Interactional routines as cultural influences upon language acquisition’ in Schieffelin and Ochs (eds): Language Socialization across Cultures. Cambridge: Cambridge University Press. Proposition 2005/06:2 (2005). Ba¨sta spra˚ket—en samlad spra˚kpolitik (The best language—a collected language politics). Stockholm: Utbildnings- och kulturdepartementet (Swedish National Agency for Education).
577
Rogoff, B. 2003. The Cultural Nature of Human Development. New York: Oxford University Press. Saville-Troike, M. and J. Kleifgen. 1986. ‘Scripts for school: Cross-cultural communication in elementary classrooms,’ Text 6: 207–21. Schegloff, E. A. 2000. ‘When ‘‘Others’’ Initiate Repair,’ Applied Linguistics 21/2: 205–43. Schegloff, E. A., G. Jefferson, and H. Sacks. 1977. ‘The preference for self-correction in the organisation of repair in conversation,’ Language 53: 361–28. Schieffelin B. and E. Ochs (eds): 1986. Language Socialization across Cultures. Cambridge: Cambridge University Press, pp. 80–96. Seedhouse, P. 2004. The Interactional Architecture of the Language Classroom: A Conversation Analysis Perspective. Oxford: Blackwell. Slotte-Lu¨ttge, A. 2005. Jag vet int va de heter pa˚ svenska‘‘: Interaktion mellan tva˚spra˚kiga elever och deras la¨rare i en enspra˚kig ˚ bo Akademi klassrumsdiskurs. A˚bo, Finland: A University Press. (Diss.) Steensig, J. 2001. ‘Some notes on the use of conversation analysis in the study of bilingual interaction’ in J. N. Jørgensen (ed.): Multilingual Behaviour in Youth Groups: Scandinavian studies in the use of two or more languages in group conversations among children and adolescents. Copenhagen: The Danish University of Education. Sa¨ljo¨, R. 2000. La¨rande i praktiken. Ett sociokulturellt perspektiv (Learning in practice. A sociocultural perspective). Stockholm: Prisma. Watson-Gegeo, K. A. and D. W. Gegeo. 1986. ‘Calling-out and repeating routines in Kwara´ae children’s language socialization’ in Schieffelin and Ochs (eds): Language Socialization across Cultures. Cambridge: Cambridge University Press. Willett, J. 1995. ‘Becoming first graders in an L2: An ethnographic study of L2 socialisation,’ TESOL Quarterly 29/3: 473–503. Vygotsky, L. S. 1986. Thought and Language (trans. A. Kozulin). Cambridge, MA: Harvard University Press. Yazigi, R. and P. Seedhouse. 2005. ‘ ‘‘Sharing time’’ with young learners,’ TESL-EJ 9/3: 1–26.
Applied Linguistics 29/4: 578–596 ß Oxford University Press 2008 doi:10.1093/applin/amm056 Advance Access published on 31 January 2008
Symmetries and Asymmetries of Age Effects in Naturalistic and Instructed L2 Learning CARMEN MUN˜OZ University of Barcelona The effects of age on second language acquisition constitute one of the most frequently researched and debated topics in the field of Second Language Acquisition. Two different orientations may be distinguished in age-related research: one which aims to elucidate the existence and characteristics of maturational constraints on the human capacity for learning second languages, and another which purports to identify age-related differences in foreign language learning, often with the aim of informing educational policy decisions. Because of the dominant role of theoretically-oriented ultimate attainment studies, it may be argued that research findings from naturalistic learning contexts have somehow been hastily generalized to formal learning contexts. This paper presents an analysis of the symmetries and asymmetries that exist between a naturalistic learning setting and a foreign language learning setting with respect to those variables that are crucial in the discussion of age effects in second language acquisition. On the basis of the differences observed, it is argued that the amount and quality of the input have a significant bearing on the effects that age of initial learning has on second language learning. It is also claimed that age-related studies in foreign language learning settings have yielded significant findings that contribute to the development of an integrated explanation of age effects on second language acquisition.
The distinction between naturalistic language learning and foreign language learning is often ignored in discussions of maturational constraints in second language acquisition. However, it is the contention of this paper that the differences in the amount and quality of the respective input of the two learning settings may have a significant influence on the effects that the age of initial learning has on the outcome of second language learning.1 In simple terms, naturalistic second language learning may be characterized as learning through immersion in the second language environment, whereas foreign language learning may be characterized as formal learning in the classroom. To expand, a typical foreign language teaching/learning situation in the schools of many countries may be characterized by means of some or all of the following features: (i) instruction is limited to 2–4 sessions of approximately 50 minutes per week;2 (ii) exposure to the target language during these class periods may be limited in source (mainly the teacher), quantity (not all teachers use the target language as the language of
CARMEN MUN˜OZ
579
communication in the classroom) and quality (there is a large variability in teachers’ oral fluency and general proficiency);3 (iii) the target language is not the language of communication between peers;4 and (iv) the target language is not spoken outside the classroom.5 While a binary distinction serves the aim of this paper of pinpointing symmetries and asymmetries between second language learning in the two different settings, it goes without saying that the situation varies from country to country. Likewise, although the paper will be focused on the two opposite ends of the distinction, it is expected that the discussion will be relevant to second language learning that takes place in school immersion situations, in which input and use of the target language may be said to be limited but to a much lesser extent. In order to avoid confusion, in this paper the term second language learning will be used in its general sense to refer to both learning that occurs in a naturalistic setting and learning that takes place in a foreign, instructed or formal setting. When referring to the former situation, the term naturalistic learning will be used, in opposition to the terms foreign, formal, and instructed, which will be used in reference to the latter situation. However, the terms learning and acquisition will be used as synonyms. Having drawn this first general distinction, we move now to a distinction that is specific to the field of age effects on second language acquisition—that between rate of learning and ultimate attainment—and which is drawn from an extended review of research findings made almost three decades ago (Krashen et al. 1979). These early research findings, which have been corroborated by subsequent research, revealed that (i) older children, adolescents, and adults generally make more rapid progress in the first stages of the second language acquisition process than younger children, particularly in morphosyntactic aspects, and that (ii) the younger a L2 learner is when the L2 acquisition process begins, the more successful that process will be, that is, the more possibilities s/he will have of attaining native-like proficiency. Therefore, the distinction identifies two different agerelated advantages: an initial rate advantage on the part of older learners over younger learners, and an ultimate attainment advantage of younger starters over older starters. Evidence from the latter comes from studies of naturalistic second language acquisition, the so-called immigrant studies, in which both phenomena are widely observed. Studies of formal learning in foreign language settings to date have only provided evidence of the former advantage, that is, the older learners’ initial rate advantage, in situations in which older starters proceed faster than younger starters with the same amount of instruction time (Garcı´a Mayo and Garcı´a Lecumberri 2003; Mun˜oz 2006b). The initial rate advantage is also observed in situations in which older learners begin instruction in the foreign language later than younger learners. In these cases, as Krashen et al. (1979: 579) note, the common finding is that children who start learning the foreign language later eventually catch up to those who begin earlier.
580
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
It has been claimed that, from a theoretical point of view, the crucial notion in the distinction is ultimate attainment. The superior ultimate attainment of younger starters is seen as evidence of the Critical Period Hypothesis (CPH) (Lenneberg 1967) and, in general, for maturationally-based explanations of the biological constraints on second language acquisition. What is important for a theory of second language acquisition—according to Long (2005: 291)—is not short-term differences in performance, but rather long-term differences in capacity for acquisition, that is, ultimate attainment. Although the term ultimate attainment6 has often been used as a synonym for nativelike proficiency, Birdsong (2004, 2006) points out that the term properly refers to the final product of L2 acquisition, whether this is nativelike attainment or any other outcome, that is, irrespective of the degree of approximation to the native grammar.7 Although it is difficult to dissociate the two terms because of their strong association in the literature in this area, it is certainly desirable both methodologically and conceptually. In fact, more and more voices in the field of SLA now argue against the traditional view of nativelikeness and its characterization, claiming that a bilingual or multilingual competence cannot properly be compared with a monolingual competence (see Grosjean 1989, 1998; Cook 2002; Birdsong 2005, for a recent discussion of the nativelikeness standard in CPH studies). It could certainly be claimed that people who have acquired two languages simultaneously from birth constitute a better yardstick for comparison than monolingual native speakers. In the case of foreign language acquisition, where the proficiency levels attained are known to fall very short of nativelike proficiency, the association of ultimate attainment and nativelikeness is clearly inappropriate. More importantly, not even the concept of final product can be transported or adopted. The first reason for this is that the definition of the final product in ultimate attainment studies may be said to entail the cessation of learning (this is why we need a long period of residence to ascertain that L2 users are not still learning, as discussed below). As in studies of fossilization, or permanent stabilization (see Long 2003), a necessary methodological requirement seems to be that cessation of learning appears in spite of optimal learning conditions (see, among others, Selinker and Lamendella 1979; Han 2004; Han and Odlin 2005; Nakuma 2005). The requirement of the presence of optimal learning conditions (including input that is neither quantitatively nor qualitatively limited) is not met in foreign language settings. Failing this requirement, in the case of classroom learning the attribution of low achievement solely to starting age does not seem to be justified. In the same vein, the case for the crucial influence of intensity of exposure in second language learning has been recently reinforced by neurolinguistic research that has obtained evidence of the crucial role played by intensity of exposure in the neural representation of multiple languages (Perani et al. 2003).
CARMEN MUN˜OZ
581
Returning to the generalization concerning the higher ultimate attainment of younger learners in a naturalistic language setting, we can say that this implies that the initial advantage of older learners must be progressively reduced to allow the younger learners’ eventual advantage to emerge. In a naturalistic immersion setting, research by Snow and Hoefnagel-Ho¨hle (e.g. 1978) provided evidence of the beginning of the process by which differences in test results between older and younger learners were being reduced in a period of approximately 1 year. When comparing children, adolescents, and adults who had been in the L2 community for 6 months, significant age differences were noticed in favour of the adolescent and adult learners. However, by the second and third tests, after 4- to 5-month intervals, the younger learners had begun to catch up with and, in some tests, overtake the older learners. These findings reinforced the generalization concerning a higher eventual attainment of younger starters in a naturalistic learning setting but, as observed by Harley (1998) twenty years later, no evidence has yet been found that the generalization of the higher eventual attainment of younger starters in naturalistic settings can be automatically extended to foreign language settings. This remains the case today. In other words, no evidence has yet been found that the proficiency of younger formal learners eventually becomes higher than that of older formal learners after the same amount of instruction. It has been suggested that the reason why younger beginners have not been observed to overtake older beginners is that, in school conditions, there is less exposure to the target language and the experience is less intensive (Ellis 1994; Singleton 1995). That is to say, because of the differences in input in both learning contexts, the period of time needed for younger formal learners must necessarily be much longer than that needed in a naturalistic learning setting. It then follows that the older starters’ initial advantage as well as the younger starters’ eventual advantage must imply longer periods than in a naturalistic situation. Following this line of reasoning, Singleton and Ryan (2004) argue that the study by Florander and Jensen (1969) offers a small amount of evidence that the same phenomenon of progressive reduction of older L2 learners’ lead over younger learners also occurs in formal instructional settings. In that study the older learners’ lead diminishes after 320 hours, although it still remains significant. This may indeed constitute evidence of the progressive reduction of the older learners’ advantage, which has been observed recently with respect to some language dimensions and after a longer period (more than 700 hours of instruction; see Mun˜oz 2006b), but it does not yet in itself constitute proof that younger learners will necessarily outperform older learners in the longer term. Indeed, if the older learners’ advantage is due mainly to their superior cognitive development, as is claimed in Mun˜oz (2006c), no differences in proficiency are to be expected when differences in cognitive development also disappear with age. To summarize, there is still a lack of empirical evidence to date confirming that, after the initial stages
582
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
of foreign language learning, younger starters overtake older starters in school settings. On the other hand, even in studies where instructional time is not constant and younger starters have had greater exposure than older starters, the former do not succeed in showing a significant advantage. In this respect, Harley (1998: 27) notes that no explanation has yet been provided for why in school settings ‘the additional time associated with an early headstart has not been found to provide more substantial long-term proficiency benefits’. In contrast, studies comparing early and late immersion students in Canadian immersion programmes have observed long-term benefits, particularly in oral communication. While it is true that late immersion students have been observed to catch up in aspects of proficiency such as reading comprehension and writing skills (Lapkin et al. 1980; Swain 1981; Cummins 1983; Harley 1986; Turnbull et al. 1998) and that early learners have sometimes failed to demonstrate the dramatic long-term advantage that was expected of them,8 it can be argued that the massive amount of exposure provided by school immersion has had a significant influence on the results. That is, an early starting age produces long-term benefits when associated with greater time and massive exposure, as in immersion programmes, but not when associated with limited time and exposure, as in typical foreign language learning classrooms.
THE REQUIRED LENGTH OF EXPOSURE In ultimate attainment studies, the length of time of residence (LoR) is the operationalization of the learners’ amount of exposure between age of acquisition (AoA) or initial age of learning and testing time. In foreign language learning studies, amount of exposure is operationalized either as the number of hours of instruction or as courses of instruction, although equating hours of exposure with hours of instruction is a gross generalization. Researchers have speculated about the length of time of residence or immersion in the L2-speaking environment that is required for a learner to have reached ultimate attainment in naturalistic learning settings. Importantly, the effects of the LoR variable appear to be limited to an initial period, after which they diminish or disappear. Krashen et al. (1979) suggest that length of residence plays little or no role after the first 5–10 years. This is in line with DeKeyser’s (2000: 503) argumentation that a minimum of 10 years should be required in ultimate attainment studies in order to ensure that it is ultimate attainment rather than rate effects that are being picked up. Likewise, using self-estimations of L2 proficiency from an extremely large sample of census data in the United States, Hakuta et al. (2003) found no evidence of the effect of LoR on English in individuals who had 10 or more years of US residence. Stevens (2006) claims, however, that large-scale empirical analyses have uniformly shown that the positive relationship between LoR and measures of English language proficiency among adult
CARMEN MUN˜OZ
583
immigrants in the United States extends well beyond 5 or 10 years of residence. This period of 10 (or even more) years would be a proxy—using Birdsong’s (2004) words—for ultimate attainment, or L2 end state, given the difficulty of determining when the end state has been reached. Because the issue of reaching the end state, as discussed in the previous section, loses its relevance in second language learning in a foreign language setting, deprived by definition of unlimited input, no conceptual symmetry can be traced here. In contrast, the length of instructional time is an important matter of contention in relation to the alleged long-term benefits of an early start, namely the higher eventual attainment of younger starters. The issue then is to identify the long-term in foreign language learning, that is, the amount of instructional time that may be considered sufficient for the study of long-term benefits. Unsurprisingly, studies that have aimed to observe the long-term effects of starting age in a foreign language classroom setting have looked for them at the end or close to the end of secondary education, although the long-term benefits sought have not been found (e.g. Oller and Nagato 1974; Burstall 1975). More recently, the BAF (Barcelona Age Factor) project has compared groups of learners after approximately 200, 400, 700, and 800 hours (Mun˜oz 2006b; Nave´s 2006). Other recent studies of foreign language learning have used similar periods (from 600 to 800 hours) in their longest-term comparisons (Garcı´a Mayo and Garcı´a Lecumberri 2003). Identifying the long term at the end of secondary education, the natural limit of most students’ foreign language learning process, has external validity because it represents the longest possible term for the majority of the population and enables the determination of relevant educational implications. In addition, it marks the point beyond which keeping a tight control of variables becomes more and more difficult. In this respect, studies whose informants are college students who began foreign language education in elementary or infant school might be credited with studying longer-term effects. In fact, the amount of exposure time observed may be more equivalent to the amount known to be required for younger starters to overtake older starters in naturalistic language acquisition (18 years, according to Singleton’s (1989) estimations on the basis of the findings made by Snow and Hoefnagel-Ho¨hle (1978, 1979)). However, the drawback of these studies is that they may depend too heavily on selfreports, and they are therefore deprived of reliable information about the individual learning trajectories. They also inevitably contain significant interindividual variability in exposure time in and out of the classroom. To complete this account of time requirements and age effects, it is also pertinent to note that experts in the area of L2 speech acquisition have suggested that simple estimations of LoR may not always constitute a valid index of L2 input. For example, the study by Flege and Liu (2001) shows that L2 proficiency increases with LoR only if the L2 speaker is participating in social settings, such as schools, in which they receive a substantial amount of input from native speakers of the L2. Thus, very young children who do
584
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
not yet attend school may not be exposed to the L2 immediately after arrival, while some immigrants may reside in L1 ghettos, or may not have a job outside the home that requires use of the L2. In addition, immigrants are likely to hear different dialects of the L2, compatriots from the home country who speak the L2 with a similar foreign accent, and individuals from other L1 backgrounds with different L1-inspired foreign accents (Flege, in press). From his review of previous research that has revealed small or non-existent LoR effects (and significant age of arrival (AoA) effects), Flege infers that LoR may provide a valid index of L2 input only for immigrants who receive a substantial amount of native-speaker input. Hence, the quality in the input to which all learners are exposed must also be examined alongside the quantity of that input or LoR. In a foreign language learning setting, learners do not usually receive native-speaker input and, when they do, it is not substantial. Moreover, learners are not exposed to the target language during all hours of instruction, and the exact proportion of lesson time varies according to the different educational systems, schools, and teachers. As a result, it may be argued that the length of instruction alone may not be a valid index of L2 input either, and that the quantity and the quality of the input to which instructed learners are exposed must also be examined.
INITIAL AGE OF LEARNING AS THE BEGINNING OF SIGNIFICANT EXPOSURE Further asymmetries are revealed when considering the initial point of learning. It has been argued that estimations of LoR should take as the initial point the beginning of significant exposure.9 In naturalistic L2 acquisition studies, this corresponds to the age of acquisition, also referred to as age of onset. Birdsong (2006: 11) notes that age of acquisition is understood as the age at which learners are immersed in the L2 context, typically as immigrants.10 This landmark is separated from the age of first exposure, which can occur in a formal schooling environment, visits to the L2 country, extended contact with relatives who are L2 speakers and so forth. Although age at immigration is taken as a proxy for age at onset of second language learning in studies of L2 acquisition among immigrants (see Stevens 2006), the age of acquisition may not coincide with the age upon arrival, because residence does not guarantee exposure to and use of the L2, as seen above. Therefore, precise estimations of the amount of L2 input are crucial because it is often observed to be confounded with age of acquisition, since early learners have typically used the L2 longer than late learners (Flege et al. 1999). The development of detailed questionnaires that measure amount of L2 (and L1) input (e.g. Flege and MacKay 2004) is an important recent contribution. In studies of instructed foreign language acquisition, initial age of learning is equated with the age at which instruction begins. In order to decide whether this is ever significant exposure it is wise to examine what is meant by
CARMEN MUN˜OZ
585
the term: significant exposure is widely conceived as immersion in the L2 context, which provides learners with a variety of contexts of use and interaction. In other words, it is considered that learners have significant exposure when they are able to carry out a variety of speech acts over a wide range of situations and topics, and to participate in social settings effectively dominated by the L2 (see Stevens 2006: 681). Following this argument, and because foreign language learners are deprived of significant exposure of this sort, initial age of foreign language learning may be equated with the age at which insignificant exposure begins. As in studies in naturalistic settings, however, it is also methodologically important to measure precisely the amount of L2 input because, in this type of setting, initial age of learning may be confounded with amount of exposure when groups of early and late learners with different amounts of accumulated instruction hours are compared. When learners move from one kind of language learning setting to the other, it may be possible to identify an age of first significant exposure that is different to the age of first insignificant exposure. As noted above, naturalistic learners may have had some instruction in the target language before moving to the L2-speaking community. In ultimate attainment studies there has usually been no role for prior insignificant exposure, and the age of initial exposure in classroom contexts has not usually been found to be strongly predictive. For example, in the study by Johnson and Newport (1989) the learners’ experience of formal learning before arrival in the L2-country did not correlate significantly with the test scores; see also Birdsong and Molis (2001). In contrast, the study by Urponen (2004) found that the age at which learning of English as a foreign language began prior to arrival in the L2-context, a significant predictor of ultimate attainment, was in fact better than age upon arrival. A different type of change of language learning setting occurs when learners in regular school programmes with limited exposure later join immersion programmes, as in the study by White and Genesee (1996) in which French-speaking Canadians who had studied English as a compulsory school subject for at least the period of secondary education joined an English-medium university. White and Genesee (1996: 242–3), arguably (see Long 2005), took this latter point in time as the age of onset, on the basis that this was the first significant exposure to the target language. A change of language learning setting may also take place for only a limited period of time, as when foreign language learners spend a few weeks in the target language community, usually attending language classes and sometimes living with a host family. In such circumstances, the stay-abroad experience may provide significant exposure even if only for a short period of time, and in some cases the experience is repeated periodically over the years. Clearly, research is needed that looks into the impact that these short stays make on learners’ L2 proficiency in relation to their initial age of learning (with limited exposure) and their age at the beginning of significant exposure.
586
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
Turning now to examine the relationship between age of acquisition and second language outcomes, in naturalistic studies there is a wide consensus about the importance of initial age of learning on second language achievement (for recent reviews of experimental studies, see Birdsong (2005) and DeKeyser and Larson-Hall (2005)). For many researchers who defend the existence of a maturationally defined critical or sensitive period, a necessary though not sufficient condition for nativelikeness (see Hyltenstam and Abrahamsson 2003) is to begin learning the second language during the sensitive period (see also Flege et al. 1999). For example, in Long’s (1990) formulation of the maturational state hypothesis, it is explicitly stated that even beginning to learn (my emphasis) the second language during that period is sufficient. In contrast, in foreign language learning studies the consensus concerning the relationship between age of acquisition and second language outcomes is that older starters have a faster rate of learning (e.g. Cenoz 2002, 2003; Garcı´a Lecumberri and Gallardo 2003; Garcı´a Mayo 2003; Lasagabaster and Doiz 2003; Mun˜oz 2003, 2006c; Nave´s et al. 2003; Perales et al. 2004; A´lvarez 2006; Miralpeix 2006; Mora 2006; Torras et al. 2006). As seen above, no evidence exists that an early start in foreign language learning leads to higher proficiency levels after the same amount of instructional time, and even younger starters with more instructional time have often failed to show a particularly substantial advantage in terms of long-term proficiency benefits. To discuss this asymmetry one can turn to the formulation of the maturational state hypothesis above. It can certainly be argued that the condition that it is sufficient to begin learning the second language during the sensitive period could only be relevant to foreign language learning if, during the privileged period, children have access to significant exposure, which marks the beginning of acquisition. Similar considerations have led researchers to argue that the CPH is irrelevant to formal language acquisition (Patkowski 1994, 2003; Lightbown 2000; DeKeyser and Larson-Hall 2005: 101). DeKeyser’s argument (e.g. 2000, 2003) is based on an interpretation of the CPH according to which it applies only to implicit language acquisition (acquisition from mere exposure to the language). Children necessarily learn implicitly, DeKeyser argues, and implicit learning requires massive input. But in regular school programmes children are not provided with the massive amounts of input which constitute a necessary condition for triggering the implicit learning mechanisms that characterize their language learning capacity (2000: 520). Further, while a lack of significant exposure in the classroom deprives young learners of the possibility of using implicit learning mechanisms, the older learners’ superior cognitive development allows them to make more efficient use of faster, explicit learning mechanisms (Mun˜oz 2001, 2006c). To conclude, turning again to the maturational state hypothesis and its applicability to foreign language learning settings, it is arguable that age at the initial point of learning has as much influence in that context as in
CARMEN MUN˜OZ
587
naturalistic learning settings. In fact, with low intensity of exposure, foreign language learning proceeds in a slow cumulative way through multiple and discontinuous short amounts of limited exposure. Therefore, in a lowintensity or distributed programme, the whole age range that extends during the period in which foreign language learning takes place may have a stronger influence on the process and the outcome of second language learning than the initial point of learning.
LEARNERS’ CHRONOLOGICAL AGE Learners’ chronological age at time of testing (AT) is a biographical variable that has drawn the attention of a number of researchers, most of whom have been critical of the methodologies used in studies of the effects of age on second language acquisition. It has been argued that AT may impact on ultimate attainment by confounding with cognitive factors, education, and other background variables (Bialystok and Hakuta 1999). Hakuta et al. (2003) attribute the steady decline across the life-span (from age 5 to 60) in their census data to cognitive declines of older learners, which ‘gradually erodes some of the mechanisms necessary for learning a complex body of knowledge, such as a new language’ (2003: 31). The learning abilities that are relevant to language learning include, according to Hakuta et al. (2003), the ability to learn paired associates (Salthouse 1992), the ability to encode new information (Craik and Jennings 1992) and to recall detail as opposed to gist (Hultsch and Dixon 1990). In a similar vein, Birdsong (2006) has highlighted the difference between the effects of aging on L2 learning (i.e. effects in the temporal-associative areas of the brain, particularly) from the effects of maturation on L2 learning, that is, the probability of nativelike ultimate attainment if learning begins before the end of the sensitive period. Other researchers, however, have argued against the claim that the confound between age of onset or age on arrival and age at testing may partly explain the inverse correlations found between AoA and test scores in some ultimate attainment studies. Such a claim would diminish the strength of the effects of maturational constraints. Specifically, DeKeyser (2000) reports a study in which the correlation between test scores and AoA is still significant with age at test partialled out, whereas that between test scores and AT is non-significant. A new perspective on the interplay of these variables has been brought to the field of SLA by Stevens’s (2006) discussion of the linear dependency of age at immigration, length of residence, and age at testing. The linear dependency among trios of variables has been extensively discussed by demographers in relation to what is known as the age–period–cohort problem or conundrum (see e.g. Wilson and Gove 1999). This linear dependency is expressed by Stevens (2006: 672) by means of an equation, which can be rewritten as AT ¼ LoR þ AoA. A methodological consequence is that the age–length–onset problem cannot be unambiguously addressed by correlational analyses,
588
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
and hence it requires a more careful consideration of the concepts and conceptual processes that these three variables index. Stevens (2006) uses Johnson and Newport’s (1989) landmark study to illustrate how difficult it is to reach unambiguous conclusions when investigating the relationship between age at arrival and L2 proficiency among adult immigrants. Although that classic study considers only age at arrival and length of residence, Stevens argues that the interdependence between age, length of residence, and age at onset entails a negative relationship between chronological age and L2 proficiency, which provides grounds for some ambiguity about the conclusion concerning the importance of age at onset, interpreted as the only predictive factor. In addition, Stevens claims that the negative relationship may be explained by the fact that the participants in Johnson and Newport’s study were faculty and students at a university; the older participants in the study were more likely to be faculty and the younger ones to be students. The types of experience the two groups have with the language may be very different and much more intense and demanding in the case of students than for the faculty, which may explain, at least in part, the relationship between age at immigration and success in L2 acquisition found in the study (see also Bialystok and Hakuta 1994, 1999). In instructed foreign language learning, AT may also be seen as confounding with age of onset, but in this case it has a negative effect on the performance of the youngest learners in comparison with older learners in school settings, and thus contributes to the positive relationship between L2 proficiency and older age of learning. Specifically, AT may have an impact on test-taking skills, favouring older children, adolescents, and adults over younger children when taking language tests. In consequence, the older learners’ higher scores in cognitively-demanding tests might be attributable not only to a language proficiency advantage over younger learners, but also to their superior cognitive development, which helps them achieve a better understanding of the task in comparison with younger learners as well as choose better strategies for accomplishing the learning task. This highlights the importance of the type of tasks that can be used with different-age learners in order to ensure that the effects of AT are not confounded with AoA effects (see Mun˜oz 2006c). Chronological age is not just an indicator of biological processes associated with senescence; it is also, as illustrated in the discussion above, an excellent indicator of life-cycle stage, strongly associated with motivations and opportunities to speak and to maintain or improve proficiency in an L2 (Stevens 2006: 684). In instructed foreign language learning chronological age may also be seen as associated with attitudes towards learning the language, the teacher, and the classroom. A confound between initial age of learning and chronological age may also be suspected in studies that analyse learner motivation in relation to age. To illustrate, it may be argued that adult beginners who choose to enrol on a foreign language course may possess stronger motivation than younger learners following compulsory education
CARMEN MUN˜OZ
589
courses, as it has been argued that students in late immersion programmes may be more strongly motivated than students in early immersion programmes (Turnbull et al. 1998). On the other hand, it has often been reported that younger school learners have a more positive attitude towards a foreign language than older school learners, as shown by their answers to questionnaires on motivation, and that this is a definite advantage of an early start (e.g. Hawkins 1996; Blondin et al. 1998). However, the young learners’ positive attitude may be seen as associated with chronological age rather than, or as well as, age at onset. It is well known that young children are more eager to please the adult than pubescent and adolescent learners and, consequently, the former may be more inclined than the latter to respond positively to motivation-related questions. Thus, the relatively common finding that younger starters have a more positive attitude towards learning a second language than older starters may be a result of their chronological age (they are younger) when they answer the motivation questionnaire rather than or in addition to their earlier start. This is clearly an empirical question since evidence could be provided by studies that asked attitudinal questions to learners with different AoA when they were the same chronological age. It was mentioned above that the amount of formal education at AT has been observed to be a very good predictor of second language learning among immigrants. Strong evidence has been provided by the large-scale study using census data in US conducted by Hakuta et al. (2003). Flege and coworkers have also considered the amount of education in the host country in their studies of L2 speech learning. For example, in the study conducted by Flege et al. (1999), AoA effects disappeared when education was controlled. In a more recent study, Flege et al. (2005) observe the influence on foreign accent of variables that were confounded with AoA and find that AoA was correlated with length of residence in Canada, which was correlated with years of education in the host country. In instructed foreign language learning, the type and characteristics of formal education may have a strong influence on learners’ proficiency levels. For one thing, in many countries, socioeconomic background may determine a child’s initial age of foreign language learning through the choice of schools (privately- or state-funded) as well as the amount of extracurricular exposure learners can enjoy (in the form of extracurricular lessons, technological devices in the home, and studies abroad, mainly). More importantly, the educational level of parents has a significant influence on children’s foreign language learning success (see for example the PISA report in Europe 2003), which suggests a parallel with the role played by the formal education of immigrant learners in studies of naturalistic L2 acquisition. Because children of parents from different socioeconomic backgrounds and with different educational levels are not homogeneously distributed in all types of school, the effects of AoA may be confounded with the effects of socioeconomic and sociocultural variables which result in varying availability of curricular and extracurricular exposure to the foreign language (Mun˜oz 2006a).
590
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
CONCLUSION This paper has presented an analysis of symmetries and asymmetries that exist between a naturalistic learning setting and a foreign language learning setting with respect to those variables that are crucial in the discussion of age effects in second language acquisition. The analysis has highlighted important differences that have largely been ignored in such discussion; the reason being that due to the higher theoretical relevance granted to ultimate attainment studies in the field of SLA, the findings of classroom studies have been interpreted in the light of the assumptions and priorities of the former. One notable asymmetry is derived from the different status of ultimate attainment or final product in the two types of learning setting. Studies of age effects on naturalistic second language learning compare younger and older starters in terms of the final product of their second language learning process, bounded by a minimum length of time (i.e. at least 10 years of residence). In contrast, age-related studies in a foreign language setting, in which the conditions for reaching the end state are not met, compare the gains of younger and older learners after different lengths of time with an emphasis on long-term benefits. In such studies, long-term benefits are often temporally bounded by the educational system itself, which does not usually require foreign language learning beyond secondary education. In consequence, the amount that can be learned in such a limited period and with such limited exposure—that is, rate of learning—becomes a crucial concern. In relation to the long-term benefits of an early start, it is important to underline the absence of empirical evidence to date confirming that younger starters overtake older starters in school settings with the same amount of input. Nevertheless, the generalization has been accepted in the field of SLA and extended beyond the field that, as occurs in a naturalistic learning setting, younger starters will overtake older starters in the long run. Likewise, it has been tacitly agreed that when younger starters show better outcomes than older starters in situations in which the former have had a greater amount of exposure or instruction, this is due to the learners’ earlier starting age and not to the greater instruction time to which they have had access due to their earlier start, ignoring this methodological flaw in variable control. Noticeable asymmetries are also found in the amount of exposure between the two types of learning setting. While in naturalistic second language learning settings input is generally unlimited, in foreign language settings input is, by definition, limited and it is usually distributed in very small doses. Therefore, it may be hard to find equivalent amounts of exposure, and research that takes into account such long periods of time (i.e. 18 years of uninterrupted instruction) is difficult to conduct with methodological rigour. Similarly, it has been pointed out in this paper that while in naturalistic second language learning studies initial age of learning is taken to be the age of first significant exposure, foreign language learning studies irremediably take as the initial age
CARMEN MUN˜OZ
591
of learning the age of first insignificant exposure. A consequence of these considerable differences in amount and type of exposure is that a foreign language learning setting cannot provide the optimal learning conditions (input that is neither quantitatively nor qualitatively limited) that constitute the methodological requirement for establishing the cessation of learning. Notwithstanding, the generalization has again been made that the comparatively lower levels of language proficiency in classroom settings are due to starting age, and that they will automatically be improved by lowering the initial age of learning, disregarding the crucial role played by intensity of exposure in language learning. In summary, for a number of years now in the field of SLA there has been a tacit acceptance of sweeping generalizations of findings from natural settings to classroom settings that have not been upheld by research into the latter.11 An inferential leap has been made in the assumption that learning age will have the same effect on learners in an immersion setting as on students of a foreign language, when the latter are exposed to only one speaker of that language (the teacher) in only one setting (the classroom) and for only limited amounts of time (Mun˜oz 2006b). However, recent studies conducted in foreign language settings have clearly illustrated the role of input and exposure in the equation: an early start leads to success but only provided that it is associated with enough significant exposure. Not only have findings been generalized; the aims and priorities of research in naturalistic settings have also been extended to research in classroom settings. As has been discussed in this paper, the aim of age-related classroom-based research cannot be the same as that of ultimate attainment studies, namely, to provide ‘evidence for the existence, scope or timing of maturational constraints on the human capacity for learning second (including foreign) languages’ (Long 2005: 288). This recognition frees age-related classroom research from endeavours which strictly belong to the area of CPH studies and provides it with a clearer focus. Similarly, legitimate priorities for research in foreign language learning settings that should have implications for guiding age-related classroom research may be: (i) to determine the amount of input that is required for an early start to be effective in promoting language learning; (ii) to focus on the relative gains of different-age pupils with different types of time distribution, that is the advantages of intensive learning programmes that may be plausible in schools; (iii) the distinction between short-term and long-term benefits of starting at different ages; and (iv) the comparative study of the learning rate of different-age learners, which should inform educators about what to expect after 4, 8, or n years of foreign language instruction from different-age learners. To conclude, the field of SLA should now be mature enough to recognize that it is ultimately the findings of studies in actual classrooms that are most relevant to decisions concerning the time and timing of second language instruction. It is these findings that have yielded a more complete picture of the factors that lead to success in second language learning, providing
592
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
evidence of the essential role of input, and thus contributing to the development of an integrated explanation of age effects on second language acquisition.
ACKNOWLEDGEMENTS This work was supported by research grants HUM2004–05167 and 2005SGR00778. The author thanks Teresa Nave´s, two anonymous reviewers, and the editors for their comments on the manuscript.
NOTES 1 The term second language will be used in this paper in a general sense to include all its different aspects (from phonology to discourse) notwithstanding the differences in the way they are learned and the speed and degree of success with which they are learned. 2 For example, in Europe the average number of foreign language classes is between 30 and 50 in primary education, and a little over 90 hours in secondary education (see Eurydice 2005). This amounts in many countries to around 800 hours by the end of compulsory education. 3 For example, in a study assessing pupils’ skills in English in eight European countries (Bonnet 2002) it is observed that ‘in most countries it is basically the teacher who does the talking’ (p. 93) and that teachers’ use of the target language varies among countries: only 15% of the Spanish teachers state that they always speak in English in their lessons, whereas 40% of teachers from Norway, Sweden, or Denmark usually do so. 4 For example, in the study referred to in note 4 (Bonnet 2002) only 8.5% of pupils state that they work in groups and talk in the target language most of the time.
5 Variation among countries is large and in the case of English the situation is more correctly depicted as a gradient or a cline (Berns 1990). See also Bonnet (2002). 6 Other terms used more or less interchangeably in the literature on naturalistic L2 learning are end state, final state, steady state, and asymptote (see Birdsong 2006: 11). 7 On the other hand, and on the basis of their recent research findings, Abrahamsson and Hyltenstam (2006) argue that nativelikeness alone may not be enough to assess the CPH, and claim that language aptitude seems crucial for reaching nativelikeness. 8 Cummins (1980, 1981) and Cummins and Swain (1986) argued that older learners show greater mastery of L2 syntax, morphology and other literacy-related skills, such as vocabulary and reading comprehension, due to their greater cognitive maturity. 9 MacWhinney (2005: 136) qualifies it as serious exposure. 10 In that respect, Long (2005) argues that estimations of LoR should take as the initial point the beginning of significant exposure even in cases in which periods spent in the L2 environment are interrupted or discontinuous, such as when learners spend different periods of residence in the L2 community.
CARMEN MUN˜OZ
11 This is not exactly parallel to the divide between pure and applied research in SLA (see, for example, Sharwood-Smith 2004: 5), although some of the features of the latter may be reminiscent of it, such as the
593
direct use of results from basic theoretical L2 research in dealing with applied questions generated in instructional language learning situations. I thank an anonymous reviewer for the suggestion.
REFERENCES Abrahamsson, N. and K. Hyltenstam. 2006. ‘The role of language learning aptitude for near-native L2 attainment.’ Paper presented in the Joint AAAL and ACLA/CAAL Conference, Montre´al, 17–20 June. ´ lvarez, E. 2006. ‘Rate and route of acquisition in A EFL narrative development at different ages’ in C. Mun˜oz (ed.): Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters, pp.127–55. Berns, M. 1990. ‘ ‘‘Second’’ and ‘‘foreign’’ in second language acquisition/foreign language learning: a sociolinguistic perspective’ in B. Van Patten and J. F. Lee (eds): Second Language Acquisition–Foreign Language Learning. Clevedon: Multilingual Matters. Bialystok, E. and K. Hakuta. 1994. In Other Words: The Science and Psychology of Second-language Acquisition. New York: Basic Books. Bialystok, E. and K. Hakuta. 1999. ‘Confounded age: Linguistic and cognitive factors in age differences for second language acquisition’ in D. Birdsong (ed.): Second Language Acquisition and the Critical Period Hypothesis. Mahwah, NJ: Erlbaum, pp. 161–81. Birdsong, D. 2004. ‘Second language acquisition and ultimate attainment’ in A. Davies and C. Elder (eds): The Handbook of Applied Linguistics. Malden, MA: Blackwell Publishers, pp. 82–105. Birdsong, D. 2005. ‘Nativelikeness and nonnativelikeness in L2A research,’ International Review of Applied Linguistics 43/4: 319–28. Birdsong, D. 2006. ‘Age and second language acquisition and processing: A selective overview,’ Language Learning 56/1: 9–49. Birdsong, D. and M. Molis. 2001. ‘On the evidence for maturational constraints on second-language acquisition,’ Journal of Memory and Language 44: 235–49. Blondin, C., M. Candelier, P. Edelenbos, R. Johnstone, A. Kubanek-German, and
T. Taeschner. 1998. Foreign Languages in Primary and Pre-School Education. London: CILT. Bonnet, G. (ed.) 2002. The Assessment of Pupils’ Skills in English in Eight European Countries. Project commissioned by the European Network of Policy Makers for the Evaluation of Education Systems. http://cisad.adc.education.fr/reva/pdf/ assessmentofenglish.pdf. Burstall, C. 1975. ‘Primary French in the balance,’ Foreign Language Annals 10/3: 245–52. Cenoz, J. 2002. ‘Age differences in foreign language learning,’ ITL Review of Applied Linguistics 125–42, 135–6. Cenoz, J. 2003. ‘The influence of age on the acquisition of English’ in M. P. Garcı´a Mayo and M. L. Garcı´a Lecumberri (eds): Age and the Acquisition of English as a Foreign Language. Clevedon: Multilingual Matters, pp. 77–93. Cook, V. 2002. ‘Background to the L2 user’ in V. Cook (ed.): Portraits of the L2 User. Clevedon: Multilingual Matters, pp. 1–28. Craik, F. I. M. and J. M. Jennings. 1992. ‘Human memory’ in F. I. M. Craik and T. A. Salthouse (eds): The Handbook of Aging and Cognition. Hillsdale, NJ: Erlbaum, pp. 51–110. Cummins, J. 1980. ‘The cross-lingual dimensions of language proficiency: Implications for bilingual education and the optimal age issue,’ TESOL Quarterly 14/2: 175–87. Cummins, J. 1981. ‘Age on arrival and immigrant second language learning in Canada. A reassessment,’ Applied Linguistics 11/2: 132–49. Cummins, J. 1983. ‘Language proficiency, biliteracy and French immersion,’ Canadian Journal of Education 8/2: 117–38. Cummins, J. and M. Swain. 1986. ‘Bilingualism in Education’. London: Longman. DeKeyser, R. 2000. ‘The robustness of critical period effects in second language acquisition,’ Studies in Second Language Acquisition 22: 499–533.
594
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
DeKeyser, R. 2003. ‘Implicit and explicit learning’ in C. J. Doughty and M. H. Long (eds): Handbook of Second Language Acquisition. London: Blackwell, pp. 313–48. DeKeyser, R. and J. Larson-Hall. 2005. ‘What does the critical period really mean?’ in J. F. Kroll and A. M. B. de Groot (eds): Handbook of Bilingualism: Psycholinguistic Approaches. Oxford: Oxford University Press, pp. 88–108. Ellis, R. 1994. The Study of Second Language Acquisition. Oxford: Oxford University Press. Eurydice. 2005. Key Data on Teaching Languages at School in Europe. European Commission. Flege, J. in press. ‘Give input a chance!’ in M. Young-Scholten and T. Piske (eds): Input Matters. Clevedon: Multilingual Matters. Flege, J. and S. Liu. 2001. ‘The effect of experience on adults’ acquisition of a second language,’ Studies in Second Language Acquisition 23/4: 527–52. Flege, J. and I. MacKay. 2004. ‘Perceiving vowels in a second language,’ Studies in Second Language Acquisition 26: 1–34. Flege, J., G. Yeni-Komshian, and S. Liu. 1999. ‘Age constraints on second language learning,’ Journal of Memory and Language 41: 78–104. Flege, J., I. MacKay, and S. Imai. 2005. ‘What accounts for ‘‘age’’ effects on overall degree foreign accent?’ Manuscript submitted for publication. Florander, J. and M. Jensen. 1969. Skoleforsøg i engelsk 1959–1965. (School experiments in English 1959–1965). Stencilled. Denmark’s Institute of Education. Cited in Ekstrand (1976) ‘Age and length of residence as variables related to the adjustment of migrant children with special reference to second language learning’ in G. Nickel (ed.) Proceedings of the Fourth International Congress of Applied Linguistics. Volume 3. Stuttgart: Hochschulverlag. Reprinted in S. Krashen, R. Scarcella and M. Long (eds). 1982. Child–Adult Differences in Second Language Acquisition. Rowley, MA: Newbury House. Garcı´a Lecumberri, M. L. and F. Gallardo. 2003. ‘English FL sounds in school learners of different ages’ in M. P. Garcı´a Mayo and M. L. Garcı´a Lecumberri (eds): Age and the Acquisition of English as a Foreign Language. Clevedon: Multilingual Matters, pp. 115–35. Garcı´a Mayo, M. P. 2003. ‘Age, length of exposure and grammaticality judgements in the acquisition of English as a foreign language’ in M. P. Garcı´a Mayo and M. L. Garcı´a Lecumberri (eds): Age and the Acquisition of English as a Foreign
Language. Clevedon: Multilingual Matters, pp. 94–114. Garcı´a Mayo, M. P. and M. L. Garcı´a Lecumberri (eds). 2003. Age and the Acquisition of English as a Foreign Language. Clevedon: Multilingual Matters. Grosjean, F. 1989. ‘Neurolinguists, beware! The bilingual is not two monolinguals in one person,’ Brain and Language 36: 3–15. Grosjean, F. 1998. ‘Studying bilinguals: Methodological and conceptual issues,’ Bilingualism: Language and Cognition 1: 131–49. Hakuta, K., E. Bialystok, and E. Wiley. 2003. ‘Critical evidence: A test of the Critical Period Hypothesis for second language acquisition,’ Psychological Science 14: 31–8. Han, Z.-H. 2004. Fossilization in Adult Second Language Acquisition. Clevedon: Multilingual Matters. Han, Z.-H. and T. Odlin. 2005. ‘Introduction’ in Z.-H. Han and T. Odlin (eds): Studies of Fossilization in Second Language Acquisition. Clevedon: Multilingual Matters, pp. 1–20. Harley, B. 1986. Age in Second Language Acquisition. Clevedon: Multilingual Matters. Harley, B. 1998. ‘The outcomes of early and later language learning’ in M. Med (ed.): Critical Issues in Early Second Language Learning. Scott Foresman-Addison Wesley, pp. 26–31. Hawkins, E. 1996. ‘The early teaching of modern languages. A pilot scheme’ in E. Hawkins (ed.): Thirty Years of Language Teaching. London: CILT, pp. 155–64. Hultsch, D. F. and R. A. Dixon. 1990. ‘Learning and memory in aging’ in J. E. Birren and K. W. Schaie (eds): Handbook of the Psychology of Aging 3rd edn. San Diego, CA: Academic Press, pp. 259–74. Hyltenstam, K. and N. Abrahamsson. 2003. ‘Maturational constraints in SLA’ in C. Doughty and M. Long (eds): The Handbook of Second Language Acquisition. Malden, MA and Oxford, UK: Blackwell, pp. 539–88. Johnson, J. and E. Newport. 1989. ‘Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language,’ Cognitive Psychology 21: 60–99. Krashen, S. 1981. Second Language Acquisition and Second Language Learning. Oxford: Pergamon. Krashen, S., M. Long, and R. Scarcella. 1979. ‘Age, rate, and eventual attainment in second language acquisition,’ TESOL Quarterly 13/4: 573–82.
CARMEN MUN˜OZ
Lapkin, S., M. Swain, J. Kamin, and G. Hanna. 1980. Report on the 1979 evaluation of the Peel County late French immersion program, grades 8, 10, 11 and 12. Unpublished report, University of Toronto, OISE. Lasagabaster, D. and A. Doiz. 2003. ‘Maturational constraints on foreign-language written production’ in M. P. Garcı´a Mayo and M. L. Garcı´a Lecumberri (eds): Age and the Acquisition of English as a Foreign Language. Clevedon: Multilingual Matters, pp. 136–60. Lenneberg, E. H. 1967. Biological Foundations of Language. New York: Wiley. Lightbown, P. 2000. ‘Classroom second language acquisition research and second language teaching,’ Applied Linguistics 21/4: 431–82. Long, M. 1990. ‘Maturational constraints on language development,’ Studies in Second Language Acquisition 12: 251–85. Long, M. 2003. ‘Stabilization and fossilization in interlanguage development’ in C. Doughty and M. Long (eds): The Handbook of Second Language Acquisition. Oxford: Blackwell, pp. 487–536. Long, M. 2005. ‘Problems with supposed counterevidence to the Critical Period Hypothesis,’ International Review of Applied Linguistics 43/4: 287–317. MacWhinney, B. 2005. ‘Emergent fossilization’ in Z.-H. Han and T. Odlin (eds): Studies of Fossilization in Second Language Acquisition. Clevedon: Multilingual Matters, pp. 134–56. Miralpeix, I. 2006. ‘Age and vocabulary acquisition in English as a foreign language’ in C. Mun˜oz (ed.): Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters, pp. 89–106. Mora, J. C. 2006. ‘Age effects on oral fluency development’ in C. Mun˜oz (ed.): Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters, pp. 65–88. Mun˜oz, C. 2001. ‘Factores escolares e individuales en el aprendizaje formal de un idioma extranjero’ in S. Pastor and V. Salazar (eds): Estudios de Lingu¨ı´stica. Universidad de Alicante, Anexo 1: 249–270. Mun˜oz, C. 2003. ‘Variation in oral skills development and age of onset’ in M. P. Garcı´a Mayo and M. L. Garcı´a Lecumberri (eds): Age and the Acquisition of English as a Foreign Language. Clevedon: Multilingual Matters, pp. 161–81. Mun˜oz, C. 2006a. ‘Investigating age-related differences in SLA: Methodological issues’. Paper presented in the Symposium Doing SLA Research: Theoretical, Methodological and Ethical Issues. AESLA Conference, Madrid, 31 March.
595
Mun˜oz, C. (ed.) 2006b. Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters. Mun˜oz, C. 2006c. ‘The effects of age on foreign language learning: The BAF Project’ in C. Mun˜oz (ed.): Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters, pp. 1–40. Nakuma, C. K. 2005. ‘Attrition: easy questions, difficult answers’ in Z.-H. Han and T. Odlin (eds): Studies of Fossilization in Second Language Acquisition. Clevedon: Multilingual Matters, pp. 21–34. Nave´s, T. 2006. ‘The long-term effects of an early start on writing in foreign language learning.’ Unpublished doctoral disertation. University of Barcelona. Nave´s, T., M. R. Torras, and M. L. Celaya. 2003. ‘Long-term effects of an earlier start. An Analysis of EFL written production’ in S. Foster-Cohen and S. Pekarek (eds): Eurosla-Yearbook, vol. 3. Amsterdam: John Benjamins, pp. 103–30. Oller, J. and N. Nagato. 1974. ‘The long-term effect of FLES: An experiment,’ Modern Language Journal 58: 15–19. Patkowski, M. S. 1994. ‘The critical period hypothesis and interlanguage phonology’ in M. Yavas (ed.): First and Second Language Phonology. San Diego: Singular, pp. 205–21. Patkowski, M. S. 2003. ‘Laterality effects in multilinguals during speech production under the concurrent task paradigm: Another test of the age of acquisition,’ International Review of Applied Linguistics 41/3: 175–200. Perales, S., M. P. Garcı´a Mayo, and J. M. Liceras. 2004. ‘The acquisition of English sentential negation by bilingual (Basque/Spanish) children: The age factor in an institutional setting.’ Paper presented at the Annual Meeting of the American Association of Applied Linguistics, Portland, OR, May 1–4. Perani, D., J. Abulatebi, E. Paulesu, S. Brambati, P. Scifo, S. F. Cappa, and F. Fazio. 2003. ‘The role of age of acquisition and language usage in early, high-proficient bilinguals: An fMRI study during verbal fluency,’ Human Brain Mapping 19: 170–82. PISA. 2003. Learning for Tomorrow’s World—First Results from Pisa 2003. OECD Publishing. Salthouse, T. A. 1992. Mechanisms of Age–Cognition Relations in Adulthood. Hillsdale, NJ: Erlbaum. Selinker, L. and J. Lamendella. 1979. ‘The role of extrinsic feedback in interlanguage fossilization: A discussion of ‘‘rule fossilization: A tentative model’’’, Language Learning 29/2: 363–75.
596
SYMMETRIES AND ASYMMETRIES OF AGE EFFECTS
Sharwood-Smith, M. 2004. Second Language Learning: Theoretical Foundations. London: Longman. Singleton, D. 1989. Language Acquisition: The Age Factor. Clevedon: Multilingual Matters. Singleton, D. 1995. ‘A critical look at the Critical Period Hypothesis in second language acquisition research’ in D. Singleton and Z. Lengyel (eds): The Age Factor in Second Language Acquisition. Clevedon: Multilingual Matters, pp. 1–29. Singleton, D. and L. Ryan. 2004. Language Acquisition: The Age Factor 2nd edn. Clevedon: Multilingual Matters. Snow, C. and M. Hoefnagel-Ho¨hle. 1978. ‘The critical period for language acquisition: Evidence from second language learning,’ Child Development 49: 1114–28. Snow, C. and M. Hoefnagel-Ho¨hle. 1979. ‘Individual differences in second language ability: A factor analytic study,’ Language and Speech 22: 151–62. Stevens, G. 2006. ‘The age-length-onset problems in research on second language acquisition among immigrants,’ Language Learning 56/4: 671–92.
Swain, M. 1981. ‘Time and timing in bilingual education,’ Language Learning 31: 1–16. Torras, M. R., T. Nave´s, M. L. Celaya, and C. Pe´rez-Vidal. 2006. ‘Age and IL development in writing’ in C. Mun˜oz (ed.): Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters, pp. 156–82. Turnbull, M., S. Lapkin, D. Hart, and M. Swain. 1998. ‘Time on task and immersion graduates’ French proficiency’ in S. Lapkin (ed.): French Second Language Education in Canada: Empirical Studies. Toronto, ON: University of Toronto Press, pp. 31–55. Urponen, M. I. 2004. Ultimate Attainment in Postpuberty Second Language Acquisition. Boston University: Boston. White, L. and F. Genesee. 1996. ‘How native is near-native? The issue of ultimate attainment in adult second language acquisition,’ Second Language Research 12/3: 233–65. Wilson, J. A. and W. R. Gove. 1999. ‘The age–period–cohort conundrum and verbal ability: Empirical relationships and their interpretation,’ American Sociological Review 64: 287–302.
Applied Linguistics 29/4: 597–618 ß Oxford University Press 2008 doi:10.1093/applin/amn010 Advance Access published on 6 May 2008
Textual Enhancement of Input: Issues and Possibilities ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS Teachers College, Columbia University The input enhancement hypothesis proposed by Sharwood Smith (1991, 1993) has stimulated considerable research over the last 15 years. This article reviews the research on textual enhancement of input (TE), an area where the majority of input enhancement studies have aggregated. Methodological idiosyncrasies are the norm of this body of research. Seven major issues appear to be limiting the generalizability of the findings and holding up further progress in the understanding of the efficacy of TE for learning: (1) noticing and/or acquisition; (2) TE and comprehension; (3) simultaneous or sequential processing; (4) TE and the nature of the enhanced form; (5) TE and prior knowledge; (6) TE and input flood; and (7) TE and overuse. The existing research has nonetheless offered some important insights that future research should seek to build on.
Over the last 15 years, there has been a proliferation of studies on ways to work with, and more importantly, against second language (L2) learners’ natural, meaning-exclusive tendency for input processing (VanPatten 1996, 2004). The studies were guided by a number of theoretical and/or pedagogical proposals, which include, but are not limited to, Sharwood Smith’s (1991) ‘input enhancement’, Long’s (1991) ‘focus on form’, Gass’ (1988) model of second language acquisition (SLA), and VanPatten’s (1991) ‘processing instruction’. The empirical findings so far have largely been inconclusive. In this article, we review the research conducted within the framework of Sharwood Smith’s input enhancement, delving, in particular, into 21 studies involving textual enhancement (TE, hereafter). Our goal is to explore and examine underlying issues, with a view to providing an interpretation of the conflicting findings, taking stock of insights, and identifying directions for future research.
WHAT IS INPUT ENHANCEMENT? The question that has served as an impetus for Sharwood Smith’s (1991, 1993) proposal of input enhancement can be formulated as follows: Why is it that L2 learners typically appear to ignore a vast mass of evidence and continue, obstinately, to operate with a system that is in contradiction with the target norms as manifest in the input? (adapted from Sharwood Smith 1993: 168)
598
TEXTUAL ENHANCEMENT OF INPUT
For Sharwood Smith, the answer to the question is multi-faceted. First, L2 learners, in general, lack sensitivity to grammatical features of target language input. Consequently, even when a large amount of input is available in their learning environment, they may not benefit much from it. Second, certain grammatical features in the input to which the learners are exposed are inherently non-salient, and hence their presence often escapes the learners’ attention. Third, learners’ first language (L1) may act as a hindrance to their ability to notice certain linguistic features in the input. Thus, failure to benefit from input may arise from a combination of lack of noticing ability on the learner’s part and poor input characteristics such as lack of perceptual salience or ‘noticeability’ (Schmidt 1990). Accordingly, Sharwood Smith hypothesizes that a way to stimulate input processing for form as well as meaning, and therefore language learning, is through improving the quality of input. Specifically, he proposes input enhancement, an operation whereby the saliency of linguistic features is augmented through, for example, TE (e.g. color-coding, boldfacing) for visual input, and phonological manipulations (e.g. oral repetition) for aural input. The underlying assumption is that noticing is a prerequisite for intake (Sharwood Smith 1981; Gass 1988; Schmidt 1990). Input salience, Sharwood Smith further suggests, can be created by an outsider (e.g. a teacher) or by an insider (i.e. the learners themselves). Learners possess their own natural learning and processing mechanisms which can, in and of themselves, generate input enhancement—so-called ‘internally generated input enhancement’, which may or may not coincide with ‘externally generated input enhancement’ (as by a teacher or researcher). The learner’s mind, as Sharwood Smith sees it, is not singular or global, but rather modular in character; the learner has many minds—to use his term—vis-a`-vis different linguistic domains and subsystems. Consequently, when exposed to externally enhanced input, learners (a) may or may not notice it, or (b) may notice it partially, contingent on whether or not they are ready for it or how much overlap there is between externally and internally generated salience. A mismatch may, therefore, arise ‘between the intentions lying behind teacher or textbook generated enhancement of the input and the actual effect it comes to have on the learner system’ (Sharwood Smith 1991: 130). Hence, ‘whether the enhanced input will ultimately trigger the relevant mental representation is . . . an empirical question’ (p. 120). Precisely, it is this question that has spurred a considerable amount of empirical research, to which we will turn in the next section. To facilitate an understanding of the empirical research, a further note on externally created salience is in order. Sharwood Smith (1991) posits two variables, elaboration (i.e. duration) and explicitness (i.e. metalinguistic depth), to allow permutations of a multitude of input enhancement strategies. Thus, one-time use of underlining would count as a non-elaborate, non-explicit strategy; one-time rule presentation would be a non-elaborate yet explicit strategy; and so forth. Sample strategies given in Sharwood Smith (1993)
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
599
include explicit discussion of the form, metalinguistic description, explicit or implicit error correction, and TE—the focus of the present article (for a review of first language research on TE, see Lorch 1989). Whatever its elaboration and explicitness, input enhancement is premised on the notion that learners must comprehend what they read or hear before their attention can be drawn to form within the input (cf. Gass 1988; VanPatten 1996, 2004; Wong 2005).
RESEARCH ON TEXTUAL ENHANCEMENT A survey of the L2 literature since the 1990s yielded 21 empirical studies of TE,1 most of which adopted a comparative approach whereby TE was pitted against another strategy, such as rule presentation (Doughty 1991; Alanen 1995), output production (Izumi 2002), or exclusively meaning-based communicative teaching (Leeman et al. 1995). The studies were typically controlled experiments under meaning-oriented task conditions, with a pretest–posttest design, the treatment period varying between 15 minutes (e.g. Alanen 1995) and 2 weeks (e.g. J. White 1998), with the posttest usually following immediately. On most occasions, the studies targeted one or more morphosyntactic elements2 such as locative suffixes in a semi-artificial form of Finnish (Alanen 1995), English relative clauses (Izumi 2002), and Spanish preterit versus imperfect forms (Jourdenais et al. 1995). These features were chosen largely based on considerations of difficulty, learnability, semantic content, communicative value, perceptual salience, and natural occurrence and frequency. Participants were mostly adult second language learners in the United States, and the sample size varied between 14 (Jourdenais et al. 1995) and 259 participants (Lee 2007). Some studies involved learners with little prior knowledge of the target form (Alanen 1995), and some had learners with considerable prior knowledge (Doughty 1991). The mode of exposure varied from reading printed texts (Alanen 1995) to reading computer-mediated texts (Doughty 1991). Although all studies employed TE, few did so singly. Rather, the majority of the studies employed additional means to augment the effect of TE, such as ‘an explicit mention to the learner to attend to the highlighted form’ (Izumi 2002: 543), a memory-based recall task (Williams 1999), or activation of prior knowledge (Shook 1994). Another notable feature of the studies is that the post-instructional measurements employed were mostly form-oriented. Examples are: sentence completion (Izumi 2002), grammaticality judgement (Alanen 1995), sentence combination (Doughty 1991), sentence production (Doughty 1991), sentence interpretation (Izumi 2002), picture-cued oral sentence completion (Doughty 1991), picture-cued written production (Jourdenais et al. 1995), multiplechoice (Bowles 2003), fill-in-the-blanks (Leow 2001), and error correction (Lee 2007). In most cases, statistical significance served as a prime, if not the only, indicator of whether or not TE was effective.
600
TEXTUAL ENHANCEMENT OF INPUT
The 21 studies have yielded a spectrum of effects for TE, as sampled below (for a complete listing, see Appendix): 1 2
3
‘TE promotes noticing of target L2 form and has an effect on learners’ subsequent output’ (Jourdenais et al. 1995: 208). ‘Input enhancement failed to show measurable gains in learning despite the documented positive impact of enhancement on the noticing of the target form items in the input’ (Izumi 2002: 542). ‘The results indicated no significant benefit of TE over unenhanced input for (1) the amount of reported noticing of Spanish present perfect or present subjunctive forms, (2) learners’ intake of the forms, or (3) learners’ comprehension of the reading passage’ (Leow et al. 2003: 12).
Thus, in some studies TE was highly effective in that it led to noticing as well as acquisition; in some it was moderately effective in that it led to noticing but not to acquisition; but in others, it did not appear to be effective. This lack of congruence in the findings is arguably a natural consequence stimulated by the numerous methodological idiosyncrasies characterizing the individual studies, some of which have been noted above. In the main, the studies differ in the following aspects: (1) employing simple versus compound enhancement; (2) employing isolated words versus sentences versus discourse as stimuli; (3) enhancing a meaning-bearing versus a nonmeaningful form; (4) employing learners with or without prior knowledge of the target form; (5) enhancing the target form many versus one or a few times; (6) using a longer versus a shorter text; (7) employing a single versus multiple short sessions over an extended period of time; (8) enhancing one form versus multiple forms; (9) providing (or not) comprehension support prior to the treatment; and (10) providing (or not) explicit instruction on what to focus on prior to the treatment. By way of illustration of (10), in Shook (1999), the following instructions were given to both the experimental groups and the control group: Read the following article through so that you understand the information presented. You will be asked to recall the information from the article, so focus on comprehending the passage. (Shook 1999: 47) Here the researcher explicitly oriented his participants to reading for meaning. In so doing, he was hoping to see if the groups exposed to the visually enhanced input would simultaneously focus on the forms as well. Such practice contrasted with that of Leeman et al. (1995) and J. White (1998) where there were explicit instructions to the experimental group to focus on the forms while reading the enhanced material. Given the wide array of differences, the studies are not comparable, hence the difficulty of making any extrapolations. Each difference contributes a piece of divergence, essentially a variable, to what overall appears to be a confusing picture.
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
601
UNDERLYING ISSUES The methodological idiosyncrasies, however, were not random. Rather, they were underlain by conceptual differences over, inter alia, these issues: (1) noticing and/or acquisition; (2) TE and comprehension; (3) simultaneous or sequential processing; (4) TE and the nature of the enhanced form; (5) TE and learner prior knowledge; (6) TE and input flood; and (7) TE and overuse. Further advances in the general understanding of TE are to depend on how well these issues are to be resolved.
Noticing and/or acquisition This issue concerns what to expect of TE. The assumption that appears to have underpinned the design of most of the studies is, as Izumi (2002) put it, a two-step logic: First, the perceptual salience created by highlighting the input will draw the learner’s attention to the highlighted forms. Second, once the first step is successful, learning of the attended form will occur based on the premise that attention is what mediates input and intake. (Izumi 2002: 567–8) However, in reality, the first step has more or less been eschewed in the TE research, as many have noted. Jourdenais et al. (1995), for example, observed that ‘while [the] effectiveness of input enhancement has been evaluated in terms of subsequent acquisition of the target forms, few attempts have been made to investigate whether enhanced input is processed differently by learners’ (p. 183). Similarly, Leow et al. (2003) pointed out that ‘most of the studies . . . did not methodologically measure learners’ noticing while exposed to the experimental L2 data’ (p. 2). Hence, the issue of internal validity arises. One does not have to look beyond the abstracts of the 21 studies to notice a test bias in assessing the efficacy of TE: The majority of the studies solely invoked so-called acquisition measures for pre- and post-tests (cf. Wong 2003). Thus, most researchers have more or less equated the efficacy of TE with its ability to generate acquisition, where acquisition is associated mainly with improved accuracy in production. There are, nevertheless, studies, though few in number, that have sought to measure noticing and acquisition—using one or more tasks (e.g. Alanen 1995; Jourdenais et al. 1995; Leow 2001; Izumi 2002; Bowles 2003; Park 2004). For these researchers, the primary function of TE is to serve as a priming device for learners’ noticing of features in input, whose corollary may, then, be that what is noticed translates into acquisition (cf. Leow et al. 2003). This understanding is largely in accord with the tenor of the input enhancement hypothesis. ‘Noticing a form is a prerequisite for its acquisition’ (Leeman et al. 1995: 222), yet it is not enough, Sharwood Smith points out, for learners to just notice critical features in the input and/or to detect the anomaly in
602
TEXTUAL ENHANCEMENT OF INPUT
their output. What is crucial is whether they can act upon the noticed features, and this would depend on whether or not they are able to perceive, store, and apply the salient information. Accordingly, it is hoped that input enhancement will spark a chain of cognitive processes initiated by noticing.3 Whether these processes can be, and how much time is needed for them to be, set in motion and completed has yet to be empirically ascertained. This issue itself in need of independent research, many studies appear to have operated on the assumption that if TE has any effect, it should show up instantly in learners’ production. This is aptly seen in the practice of providing a short treatment and an immediate, output-based posttest to measure acquisition (e.g. Alanen 1995; Jourdenais et al. 1995). Just as it is problematic to assume, without establishing, that TE draws learners’ attention to the target form, which, in turn, leads to further processing of it, it is equally problematic to expect instant learning to arise from a single (typically, short) treatment session of TE (cf. Wong 2005). Both create validity and reliability concerns. In the latter case, notably, because the treatment does not allow learners time to pursue deeper processing of any of the attended input; changes (or lack thereof) shown on the posttest can be evidence of anything but learning. Although it may be a truism that nothing in input can become intake without noticing (Schmidt 1990, 2001; see, however, Truscott 1998), the TE literature offers important evidence showing that noticing is not a guarantee for acquisition. For example, Izumi (2002) found that more externally induced noticing was not correlated with more learning, concluding that noticing may not lead to acquisition. Similarly, Williams (1999) noted that ‘not everything that is registered by the senses is . . . encoded in long-term memory’ (p. 2). Likewise, Leeman et al. (1995) observed that ‘forms may be noticed perceptually, but not linguistically’ (p. 219; cf. Doughty 1998). In a nutshell, enhanced forms may attract attention but may fall short of further processing. Taking the possible pitfalls into consideration, Sharwood Smith (1991), in his critical elaboration of input enhancement, cautions: Externally generated input enhancement does not automatically imply the internalization of that enhancement by the learner . . . Input enhancement should take into account a modular view of the learner as a set of systems: signaling information to the learner is, in effect, signaling to one or more of many separate knowledge systems. Input enhancement may . . . work in ways unforeseen by the enhancer. (Sharwood Smith 1991: 131; emphasis added) Implied here is (a) that not all the cognitive effects triggered by TE will have external manifestations, and (b) that the efficacy of TE is not as much controlled by the researcher as by the learner (cf. Bardovi-Harlig and Reynolds 1995; Wong 2005). As we will argue later, the efficacy of TE is, in part, a function of the learner’s prior knowledge (or lack thereof) and of the nature of the linguistic element enhanced.
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
603
Evidence that learners are active controllers of the input enhancement process can be found in Leeman et al. (1995), reporting that ‘not all of the enhancements were noticed by all the participants’ and ‘at least some learners prioritize meaning over form despite the emphatic instructions they received to consider both’ (p. 248). What such findings suggest is that learners have their own agenda to pursue, an agenda that may not be susceptible to instructional guidance (cf. Allwright 1984). In sum, it appears that short-term TE research—both in terms of the duration of the treatment and the interval between the treatment and the posttest—should concentrate on measuring its effect at the level of noticing and refrain from measuring acquisition (cf. Shook 1994; Overstreet 1998), which we define here as mastery of form–meaning–function relations.
TE and comprehension As indicated earlier, the input enhancement proposal is premised on comprehension, its underlying thrust being to prompt occasional metalinguistic attention for the ultimate benefit of balanced development in comprehension and acquisition (cf. Doughty 2004; Wong 2005; Kim 2006). It thus follows that any validation of its efficacy should investigate comprehension as well as noticing and, in the case of longer-term studies, acquisition. However, most of the studies have overlooked the need to measure comprehension (cf. Leow 1997; Wong 2003; Lee 2007). Thus, in spite of accumulated evidence suggesting that TE can promote noticing of some linguistic features, it remains unclear whether or not it has simultaneously created a trade-off with comprehension.4 Preliminary evidence from Overstreet (1998), substantiated recently by Lee (2007), suggests that input enhancement may detract from learners’ attention to meaning. This should, accordingly, raise a red flag for future TE researchers. In most of the existing studies, researchers have made a conscious effort to establish a meaning-oriented environment. Some had their participants answer comprehension questions before the onset of the treatment (Alanen 1995), some provided vocabulary assistance (Williams 1999), some had their participants write recall summaries, often in their L1, of the content of the texts they had read during the treatment (Izumi 2002), and some even took care not to make the enhancement so salient that it would distract students from reading (J. White 1998).5 Still, the literature has not yet witnessed any study with a rigorous design for measuring comprehension. The most concerted effort yet has been seen in trying to obtain a single summative score from the participants’ recall summaries, as in Doughty (1991) and Izumi (2002), or putting participants through a post-treatment comprehension test, as in Leow et al. (2003). Inasmuch as these practices did not include a pre- and post-treatment comparison of comprehension, they shed little light on the effect of TE, if any, on comprehension. A more robust design to measure the latter, therefore, should treat comprehension as a dependent
604
TEXTUAL ENHANCEMENT OF INPUT
variable to be measured both before and after the TE treatment, at withingroup and/or between-group levels. When measuring comprehension in relation to TE, it also appears that the focus should be on measuring local, as opposed to global comprehension, an insight from Leow (1997). Local comprehension here concerns only the part of the text that contains the enhanced form, and hence is arguably a more valid indicator of the influence, if any, of TE on comprehension; it also appears to be a good predictor of global comprehension, as demonstrated in Leow (1997). Other studies such as Shook (1999) and Wong (2003), however, have provided evidence that the effect of TE is positively felt at the level of local comprehension, but not at the level of global comprehension, thereby suggesting a dissociation between local and global comprehension. Future research controlling for local and global comprehension should shed further light on this issue.
Simultaneous or sequential processing Many studies had as their underlying assumption that a focus on form integrated into an overall focus on meaning is beneficial to L2 acquisition (Lightbown 1991; Long and Robinson 1998). However, there was little consistency among them with regard to when such integration should occur. According to information processing theory, (a) processing of information is selective; (b) individuals can process two different types of information simultaneously and effectively only if the processing of one of the information types is automatized and requires little, if any, conscious attention; and (c) simultaneous processing of two different types of information that are not automatized can lead to inadequate processing of either or both types of information, and to a ‘trade-off’ effect (cf. Skehan 1996; VanPatten 1996; Han and Peverly 2007; see, however, Robinson 2003). From this theory it follows that sequential processing will be more effective than simultaneous processing of meaning and form. In other words, learners might benefit more from having their attention first directed to meaning decoding and then to grammatical encoding than to both tasks at the same time. The research on TE has employed both kinds of design, although most of the studies probed simultaneous processing. As an illustration, in Alanen (1995), participants in the TE group were invited to read two short passages in which the learning targets were italicized. At the outset of their reading, they were explicitly instructed to ‘try to understand its meaning’ (p. 271). Thus, an experimental condition was created to stimulate an intentional focus on meaning and an incidental focus on form. Such is also the thrust of Shook (1994, 1999). In Shook’s (1999) study, 73 participants were randomly assigned to three input conditions: (a) textually enhanced, (b) textually enhanced plus emphatic instruction,6 and (c) textually unenhanced. They subsequently performed a free written recall task in which they wrote down anything they could recall from the passages they had read. The recall task
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
605
sought to measure both comprehension and intake. The results were revealing: on the one hand, the unenhanced group recalled the most idea units, the enhanced group recalled fewer idea units, while the enhanced plus instruction group recalled the fewest, thus suggesting that the unenhanced group comprehended the reading passages better than both the enhanced and the enhanced plus instruction group. On the other hand, in terms of intake of the target items, the following picture emerged: participants who received TE recalled more tokens than did those who received TE plus instructions, while those who received no enhancement recalled the least tokens, thus evidencing a trade-off between comprehension and intake. The difference between simultaneous and sequential processing lies not just in the temporal order of meaning-based and form-based processing of input, but in that one calls for incidental processing and the other intentional processing. In other words, the researchers who implemented a simultaneous processing design assumed that in processing the input for meaning, the processing of the enhanced forms would occur incidentally, as a byproduct, an assumption similar, though not identical, to that underlying Krashen’s Input Hypothesis. On the other hand, the researchers who had a sequential design assumed that with the processing of input for meaning happening first, attentional resources could be freed up and reallocated to the processing of form, an assumption in line with the fundamental tenets of cognitive theory of information processing.7 Although research on TE to date has not seen much of a sequential design, Doughty (1991) has offered some useful preliminary evidence. In her study, 20 participants, randomly assigned to one of the three groups, (a) the meaningoriented TE group (MOG), (b) the rule-oriented enhancement group (ROG), and (c) the control group (COG), first had to undertake a comprehension task to achieve global comprehension of the content of a reading passage prior to receiving the treatment. Thus, the study roughly had a sequential arrangement for a focus on meaning and a focus on form, followed, then, by a more fine-tuned sequential arrangement during the treatment. The treatment itself comprised a series of lessons, one per day over a 10-day period, carried out as follows: during each lesson, the participants in the MOG received chunks of the reading passage aided first by lexical and semantic explanation and then by TE, and those in the ROG received the same chunks of text enhanced by a presentation of rules concerning the target item. In contrast, those in the COG, as expected, received the same chunks of texts with no enhancement whatsoever. Each lesson ended with the participants writing recall summaries based on their comprehension of the texts. Results from the post-treatment tasks indicated that the MOG and the ROG did almost equally well and better than the COG, with respect to acquisition of the target feature, but the MOG displayed an advantage over both the ROG and the COG in terms of comprehension. Despite the fact that this study did not employ a pre- and a post-test task to measure and compare the participants’ comprehension, the TE employed did
606
TEXTUAL ENHANCEMENT OF INPUT
not produce a trade-off, as evident in a twofold fact. First, the MOG and the ROG did equally well on the acquisition measures; and, second, the ROG and the COG did equally poorly on comprehension measures. If the TE had affected comprehension, the COG should have achieved greater comprehension than the ROG, because they were not ‘distracted’ by any sort of TE. What, then, explains the ‘double’ gains in comprehension and acquisition by the MOG, but not by either of the other groups? The answer is in fact simple: this group received assistance in both meaning and form processing, notably, sequentially. This combined and sequential treatment may have aided both comprehension and acquisition. While future empirical research should attempt to ascertain which grammatical items would benefit most from a combined meaning–form enhancement, evidence from studies by Doughty (1991) and Izumi (2002) suffices to motivate the understanding that when meaning is clarified before a focus on form, TE is likely to stimulate even growth in comprehension and acquisition. Izumi (2002) assessed the effects of an output-based approach versus those of a TE approach to integrating a focus on form with a focus on meaning. Sixty-one university level ESL learners served as participants in this study, and were divided into four groups: (a) the þoutput –enhancement group, (b) the þoutput þenhancement group, (c) the –output –enhancement group, and (d) the –output þenhancement group. The experimental design largely followed that of Doughty (1991). Thus, it began by having the participants perform content-based processing (i.e. skimming the text for overall comprehension) before subjecting them to an extended period of focus-on-form treatment, hence, similarly, a sequential design. The treatment procedures for the enhanced and unenhanced groups, although bearing some similarities to the Doughty study, are nevertheless different in two main respects: First, the participants were asked to take notes while reading chunks of the reading passage, a means to measure noticing. Second, exposure to TE was not preceded by a lexical and semantic explanation of the target item; instead, the participants were directed through instructions to focus on meaning throughout the treatment period. Results showed that TE was able to promote noticing without taxing comprehension. Whereas the two studies mentioned above have incidentally revealed an advantage for sequential over simultaneous processing, a study by VanPatten (1990) has purposely investigated the feasibility of simultaneous processing of meaning and form, albeit within a different theoretical framework8 than input enhancement. Two hundred and two university students of Spanish, representing three levels of proficiency, were randomly assigned to one of the four experimental conditions: (a) attention to meaning alone, (b) simultaneous attention to meaning and an important lexical item, (c) simultaneous attention to meaning and a grammatical functor, or (d) simultaneous attention to meaning and a verb form. Results showed that ‘[C]onscious attention to form in the input competes with conscious attention to meaning, and, by extension, that only when input is easily understood can learners attend to form as part
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
607
of the intake process’ (p. 296). The study thus provides direct evidence that processing input for form presupposes that the meaning of the input has been comprehended,9 and that the former, when occurring, is sequential to the latter (see, however, Wong 2001). This finding is replicated in the TE research. Overstreet (1998) and Lee (2007) both noticed a similar form– meaning trade-off. Apart from responding to the limited information processing capacity that characterizes all learners, sequential processing has an additional putative advantage, namely that it matches learners’ predilection for processing input first for meaning (Barry and Lazarte 1995; VanPatten 1996, 2004). Shook (1994, 1999) and Leeman et al. (1995) observed that some of their participants focused exclusively on meaning, in defiance of TE and the researchers’ admonition to focus on both meaning and form. In a case like this, sequential processing may take learners beyond such natural processing tendency through redirecting their attention to form. The importance of the latter aspect cannot be overstated, as research has shown that even when learners are able to reallocate attentional resources to form-based processing as a function of attainment of comprehension, they do not automatically do so (see, e.g. VanPatten 1990; Leeser 2004). Interestingly, however, J. White (1998) attributed this phenomenon to the fact that the input had been comprehended, noting that ‘the forms may not have been novel enough to attract the learners’ attention to the extent that was predicted’ (p. 103). Hence, there is an intriguing circular problem here: Does comprehension facilitate or hinder attention to form? Further research is clearly needed to tease out the intricacies between the two.
TE and the nature of the enhanced form A now well-established tenet in SLA research is that not all linguistic elements are created equal. Over the years, researchers have posited a number of parameters for describing inter-structural differences. They include, but are not limited to, communicative value, formal complexity, functional complexity, semantic load, perceptual saliency, and underlying rules. VanPatten (1996), for example, distinguishes between linguistic features of high, medium, or low communicative value, based, in turn, on their semantic value and structural redundancy. Accordingly, English -ing is deemed of high communicative value in that it carries semantic value and, syntactically, is non-redundant. In comparison, English 3rd person -s is of low communicative value, because it is structurally redundant. In addition to functional differences, linguistic features may differ according to their underlying rules. DeKeyser (1994) notes that some linguistic features follow categorical rules (e.g. English plurals), and some probabilistic rules (e.g. English irregular past tense verbs). In the instructed SLA literature, evidence abounds that not all forms are equal in terms of the effectiveness of instructional activities (Williams and Evans 1998). Form type can mitigate the effect of instruction through their
608
TEXTUAL ENHANCEMENT OF INPUT
mutual interaction as well as interaction with other variables such as task demands, learner proficiency, prior knowledge, familiarity, content complexity, and availability of pragmatic information. In the TE literature, two studies have specifically investigated the nature of the target form and its interaction with different enhancement strategies. One study is Leow et al. (2003) in which two structures, the Spanish present perfect and present subjunctive, were investigated in connection with the presence or absence of TE. Analyses of the participants’ think-aloud protocols and their performance on an intake-oriented recognition task revealed more noticing and learning of the present perfect than of the present subjunctive. Of importance to note is that the structural effect overrode that of TE: both the enhanced and the unenhanced group noticed one structure better than the other. Similar findings are reported in Shook (1999) examining the free written recalls produced by participants in the wake of TE in relation to the Spanish present perfect and relative pronouns. Both studies thereby lend empirical support in favor of Sharwood Smith’s notion of learner-created salience. In light of these findings, it is likely, then, that the positive effects of TE, as reported in some of the studies, have in part derived from the fact that the researchers happened to have chosen linguistic elements that are susceptible to this type of intervention, although it is not entirely clear at the present time what determines the susceptibility (cf. Wong 2003). Current speculations include form–meaning/form–function transparency (Wong 2005) and the relationship between cognitive processing and grammatical elements (Izumi 2002). While future research should test these speculations, it will equally be worthwhile to build on and extend the existing research by replicating Leow et al. (2003) and Shook (1994, 1999), and by investigating other forms, ideally from other languages as well. If future research indeed demonstrates that only some forms are amenable to TE, then instruction utilizing this type of strategy should selectively target certain forms as opposed to any forms indiscriminately. The question of what forms to (or not to) enhance is a complex one, and the existing evidence is meager but suggestive. Preliminary evidence from Wong (2003), corroborated by Shook (1999), showed that enhancing a nonmeaningful form does not lead to better intake. On the other hand, as noted above, results from Leow et al. (2003; see also Leow 2001) and Shook (1994, 1999) revealed that learners are able to notice, on their own, meaning-bearing forms, suggesting, therefore, that there is no need to enhance them (see, however, Leow 1997; Shook 1999 for a different interpretation). This reasoning would seem logical were it not for the results of Leow (1997) comparing the effects of TE and text length on comprehension and intake. One of the findings of the study, although not statistically significant, illustrates a meaningful effect for enhancement on the participants’ comprehension, while no such effect was discerned for their intake. Pending future, independent verification, this finding underscores that comprehension and intake are different entities (cf. Færch and Kasper 1986; Sharwood Smith 1986),
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
609
inasmuch as enhancing the same linguistic element can produce a differential impact on comprehension and intake (cf. Wong 2003, 2005).
TE and learners’ prior knowledge One approach to tackling the susceptibility issue is through examining variables that can interact with the target form to modulate the effect of TE. Learners’ prior knowledge is one such factor. Evidence thereof can be found in studies that employed participants with or without prior knowledge of the target form (e.g. Jourdenais et al. 1995 vs. Leow et al. 2003). Research to date has offered three main findings on the relationship between TE and learners’ prior knowledge. First, simple enhancement (i.e. TE alone) is more effective for learners with some prior knowledge of the form in question (Park 2004) than for learners without (Alanen 1995). Second, simple enhancement may induce noticing (i.e. low-level awareness, following Schmidt 1990) but not understanding (i.e. high-level awareness) in learners with little prior knowledge (Shook 1994); however, it may incite understanding as well as noticing in learners with some prior knowledge (Lee 2007). Third, compound enhancement (i.e. TE in combination with other attention-getting strategies such as corrective feedback) is more effective than simple enhancement in inducing noticing, and further processing of, the target form in both types of learners (e.g. Doughty 1991; Leeman et al. 1995; Williams 1999). Given the nature of simple versus compound enhancement, these findings are logical. According to Sharwood Smith (1991, 1993) and as Robinson (1997) has demonstrated, TE is more an implicit than an explicit attention-focusing device. As such, its underlying purpose may not always be transparent, even to learners with some prior knowledge of the target form. Several participants in J. White’s (1998) study reported, via a post-treatment debriefing questionnaire, that they were not certain about the purpose of the enhanced forms. For learners with little prior knowledge of the target form, TE could be even more of a puzzle. Participants in the Leow (2001) study provided the following comments in their think-aloud protocols: ‘I don’t know what that is (means)’, ‘I don’t understand these underlined ones’, ‘I am not sure’, ‘I don’t know why this is underlined’, etc. (p. 502). Likewise, as shown in the protocols from Alanen (1995), some participants who reported noticing the use of italics had not considered a reason for its use. Thus, the fact that TE alone (i.e. simple enhancement) may fail to have a noticeable impact on learners’ comprehension and intake, as reported in Leow (1997, 2001) and Leow et al. (2003), might be due to the learners having no prior knowledge of the target form. Had those learners been exposed to compound enhancement, they would likely have demonstrated superior intake to that shown in the comparison group (i.e. who received unenhanced input).
610
TEXTUAL ENHANCEMENT OF INPUT
TE and input flood Most studies in the TE literature exhibited varying degrees of conflation of TE and input flood, both being options of input enhancement. In the case of input flood, salience is created of the target of instruction by virtue of (usually, artificially engineered) frequency. As an illustration, compare Leow et al. (2003) with Shook (1999). In Leow et al. (2003), two passages were used as stimuli, one having 222 words and the other 227 words. In each passage there were 10 tokens of a target form, giving the frequency ratios of 1/22 for one passage and 1/23 for the other—meaning in one text, one out of every 22 words was an exemplar of the target form, and one out of 23 words in the other. Shook (1999), on the other hand, also utilized two written texts as stimuli, one containing 185 words and the other 217 words, but there were only 6 instances of the target item in each text, yielding frequency ratios of 1/30 words for one text and 1/36 words for the other. Thus, the two studies differ in the amount of exposure to the target forms created for the participants: There is more of an ‘input flood’ in Leow et al. (2003)10 than in Shook (1999). Neither, however, matches the degree of input flood found in J. White’s quasi-experimental study where the students were provided numerous instances of the target form over an extended period of time. These variations notwithstanding, problems may arise when input flood is used along with TE to control for the experimental text and at the same time to manipulate the control text—the so-called ‘unenhanced text’. For one thing, the results from the experimental group can be contaminated with ambiguity. For another, input flood may cancel out the effects of TE. As J. White (1998) noted in her study, ‘benefits resulting from the experimental treatment conditions were due to increased exposure through input flood of target forms and not to any other kinds of enhancement’ (p. 103). The available evidence, albeit limited, induces the hypothesis that input flood may have a greater attention-getting capacity than TE. Future research should examine this further, including ascertaining, in the case of input flood, how many exemplars would be optimal (Wong 2005).
TE and overuse Several studies have provided evidence that learners receiving TE oversupplied the target form; that is, their production of the target form extended to non-obligatory contexts. Close inspection of the experimental procedures underlying these studies indicates that not only was the relevant input typographically enhanced, but there was input flood as well, which, as mentioned, is itself a form of enhancement. Hence, the target forms were double-enhanced. In the Jourdenais et al. (1995) study, in which participants showed overuse of the target form, the stimulus text of 204 words contained 10 instances of the Spanish imperfect and 18 of the preterit, each enhanced
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
611
in three typographic ways (i.e. underlining, changing the font, bolding/ shadowing). Similarly, in Alanen (1995), participants’ overgeneralization of -lla and -ssa could be linked to the over-salience of the target forms which were not only highlighted but also exhibited a high frequency in the two stimulus texts. In one text of 87 words, they appeared 6 versus 5 times, and in the other text of 98 words, 7 against 8 times, as compared with the rest of the target forms which each made between 2 and 4 appearances in either stimulus text. Alanen herself recognized that ‘these errors could have been caused by instruction itself (Selinker 1972)’ (p. 289). Further evidence of overuse can be found in Leeman et al. (1995) where overgeneralizations may be attributed to the fact that the researchers overemphasized the target forms through enhancing them in and out of the classroom, including providing corrective feedback whenever possible and giving explicit instructions, in order to attract the learners’ attention. Clearly, over-enhancing the target form can be counterproductive. When the target form is unnaturally frequent and excessively salient in the input, it hurts, rather than facilitates, learners’ processing of meaning as well as form. A case in point is Overstreet (1998), where participants in the enhancement groups not only received an input flood via a short text, but each instance of the target form was highlighted in three ways: (a) font enlargement, (b) underlining, and (c) boldface or shadowing. Results obtained from measures of recognition, comprehension, and production revealed the absence of a positive effect for TE on the intake of the target forms but the presence of a negative effect on comprehension. Indeed, underlying the reported negative effect of TE, it may not only have been the fact that there was too much salience, but also that more than one form was enhanced in the same input text.11 The coalescence of these two factors might have confused, rather than enlightened, the learners. Enhanced as such, input may trigger learners’ noticing but may not lead to much understanding, as seen in Jourdenais et al.’s study. In sum, finding the right balance appears difficult but necessary between frequency and saliency (cf. Barcroft 2003). If TE is purported only to induce noticing, then, of course, the more frequent the target forms are in the input, the more salient they are likely to be (however, see Lorch 1989), and the more salient they are, the more likely it is that they will be noticed. This reasoning has driven most of the empirical studies to date. Nonetheless, it seems to be time to switch gears and investigate a different question: if TE is intended to stimulate higher-level awareness of form–meaning connections, to what extent should input be enhanced so as not to produce aberrant noticing to the detriment of acquisition?
CONCLUSION As Wong (2003) has noted, ‘The contribution of [TE] to SLA . . . is presently not clear’ (p. 18). The available literature has provided conflicting findings on
612
TEXTUAL ENHANCEMENT OF INPUT
its efficacy. Taken at face value, these findings may suggest that TE is either helpful or unhelpful. Such an understanding is, however, the least desirable: from a theoretical standpoint, it neither validates nor invalidates the theory, and from a practical standpoint, little can be extrapolated from the ambiguous findings to teaching in the classroom. Although the jury is still out on the efficacy of TE for learning, the research to date has nonetheless produced a number of valuable and, importantly, testable insights, such as the following:
Simple enhancement is capable of inducing learner noticing of externally enhanced forms in meaning-bearing input. Whether or not it also leads to acquisition depends largely on whether the learner has prior knowledge of the target form. Learners may automatically notice forms that are meaningful. Simple enhancement is more likely to induce learner noticing of the target form when sequential to comprehension than when it is concurrent with comprehension. Simple enhancement of a meaningful form contributes to comprehension. Simple enhancement of a non-meaningful form does not hurt comprehension. Simple enhancement is more effective if it draws focal rather than peripheral attention. Compound enhancement is more likely to induce deeper cognitive processing than simple enhancement, possibly to the extent of engendering ‘overlearning’.
Future research should seek to test the robustness of, and/or substantiate, these insights. This effort will lead researchers into ascertaining psycholinguistic contingencies under which TE is able to trigger a chain of cognitive processes that result ultimately in acquisition. En route, researchers will gain a better understanding of the key modulating variables such as learners’ prior knowledge, the nature of the target form, learner processing capacity, and frequency of the enhanced element. They will perforce strive to operationalize the type and process of attention and memory adequately and rigorously in order to find out how what is noticed is further processed, retained, and eventually becomes transferable. Moreover, they will likely be able to determine, and even set up a benchmark for, ‘the extent to which learner attention should be directed’ (Doughty 2004: 187). Current research has provided evidence showing that compound enhancement is not always necessary (see Williams and Evans 1998; Berent et al. 2007). Furthermore, future research on TE must include a longitudinal perspective. SLA researchers at large have paid excessive lip service to the need for carrying out longitudinal studies, but have not yet been able to break the status quo (cf. Ortega and Iberri-Shea 2005). This is particularly surprising for research on the effects of instruction, since it is here—more than anywhere else—where study after study has conceded that its methodological
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
613
shortcoming is the lack of a longitudinal perspective. Non-longitudinal studies of effects of instruction have an inherent bias in favor of explicit instruction (for discussion, see Doughty 2004). Explicit instruction is known to be quick at provoking instant yet superficial changes (see, e.g. L. White 1991) in behavior, in contrast to implicit instruction whose impact on learning—especially in the domain of morphosyntax—may take time to build and be felt (see, e.g. Robinson 1996; Hulstijn 1997; Doughty 2004; Long 2007; Mackey and Goo 2007). A related issue that researchers at large have overlooked, but that TE researchers should examine, is the scope of form–meaning connection (FMC) in acquisition.12 If FMC concerns linking one surface form with one meaning, then explicit (in particular, short and intensive) instruction might be more effective than implicit instruction in drawing learner attention and provoking changes in representation and behavior. However, if FMC entails connecting a form with its multi-faceted meaning in a variety of contexts, protracted implicit instruction might prove superior to explicit instruction (cf. Han 2007). Clearly, unless the time issue is taken into account, any comparison between explicit and implicit instructional strategies, TE included, may risk being invalid and misleading.
ACKNOWLEDGEMENTS We are grateful to the anonymous reviewers for their insightful comments and helpful suggestions. Any errors are exclusively our own.
APPENDIX General findings from 21 studies on textual enhancement 1
2
3 4 5
‘[A]ttention to form, either via detailed analysis of structure or highlighting of target language (TL) structures in context, promotes acquisition of interlanguage (IL) grammar, but that only the latter comes hand-in-hand with comprehension of input’ (Doughty 1991: 431). ‘A significant main effect was found for the attention condition across all of the data, indicating that subjects whose attention was drawn to the grammatical items gained more linguistic information about the items than the subjects whose attention was not called to the items. Also, the type of information called to the input was shown not to be a significant factor’ (Shook 1994: 79). ‘Visual enhancement seemed to have a facilitating effect on the learners’ recall and use of the targets, especially those of locative suffixes’ (Alanen 1995: 259). ‘Textual enhancement promotes noticing of target L2 form and has an effect on learners’ subsequent output’ (Jourdenais et al. 1995: 208). ‘Only participants in the Focus on Form group demonstrated significantly improved accuracy and suppliance of the target forms, thus suggesting that it is
614
6
7
8
9
10
11
12
13
14
15
16
TEXTUAL ENHANCEMENT OF INPUT
possible to increase accuracy within a content-based instructional setting’ (Leeman et al. 1995: 217). ‘Results revealed a significant main effect for text length on readers’ comprehension but not on intake, no significant main effect for input enhancement on either comprehension or intake, and no significant interaction’ (Leow 1997: 151). ‘The knowledge acquired during incidental and enhanced learning is also partially memory-based, but the enhanced learners in particular show evidence of the development of a generalizable rule-based representation. Access to this is slow and effortful, in contrast to decisions based on memory for prior experiences’ (Robinson 1997: 242). ‘Even though learners in all three treatment groups improved in their ability to use his and her in an oral communication task, the findings did not support the hypotheses of this study. Although accuracy ratios overall followed the predicted order, that is, Group Eþ44Group E44Group U, the within-group variance canceled out most of the predicted between-group effects at the two posttests’ (J. White 1998: 101). ‘No positive effect was found for either enhancement or content familiarity on either the production or recognition tasks, but a negative effect was found on the subjects’ comprehension of the texts’ (Overstreet 1998: 228). ‘[D]rawing attention to the input can benefit processing, and that salience/ meaningfulness of the input is a major component in determining the extent of that benefit’ (Shook 1999: 39). ‘In general terms, these improvements would seem to demonstrate the effectiveness of devices that draw the subjects’ attention to specific aspects of the input through highlighting’ (Williams 1999: 29). ‘Results indicated no significant benefits of written input enhancement over unenhanced written input for (1) the amount of reported noticing of Spanish formal imperatives, (2) readers’ comprehension, or (3) readers’ intake’ (Leow 2001: 496). ‘Input enhancement failed to show measurable gains in learning despite the documented positive impact of enhancement on the noticing of the target form items in the input’ (Izumi 2002: 542). ‘The results indicated no significant benefit of textual enhancement over unenhanced input for (1) the amount of reported noticing of Spanish present perfect or present subjunctive forms, (2) learners’ intake of the forms, or (3) learners’ comprehension of the reading passage. The study did indicate a significant benefit of more salient forms (present perfect) over less salient forms (present subjunctive) for (1) the amount of reported noticing of the targeted verb forms, but not for (2) learners’ intake or (3) learners’ comprehension’ (Leow et al. 2003: 12). ‘[Textual enhancement] is not effective as a form of input enhancement on the acquisition of the French past participle agreement in relative clauses . . . Drawing learners’ attention to form via [textual enhancement] does not interfere with comprehension when whole clauses that contain the target form are underlined and when the form is a form of no communicative value’ (Wong 2003: 35). ‘[T]he main findings of the present study on discrete-item vocabulary learning were (a) no effect for enhancing 9 out of 24 words on learning rates for the enhanced words, (b) no effect for enhancing 9 out of 24 words on learning rates
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
17
18
19
20
21
615
for the unenhanced words, (c) a positive effect for enhancing 3 out of 24 words on learning rates for the enhanced words based on some but not all dependent measures, and (d) a negative effect for enhancing 3 out of 24 words on learning rates for the unenhanced words based on some but not all dependent measures’ (Barcroft 2003: 70). ‘No statistically significant benefit was found for enhanced input as compared to unenhanced input for (1) amount of reported noticing of targeted forms, (2) readers’ comprehension, or (3) readers’ intake’ (Bowles 2003: 406). ‘All of the 12 learners (6 dyads) in the [input enhancement] group reported that they had noticed some kind of visual enhancement in the reading given between Task 1 and Task 2’ (Park 2004: 14). ‘[Textual enhancement] alone did not aid form and meaning recognition of vocabulary, [and] [lexical elaboration] and [typographic enhancement] combined did not aid form recognition of vocabulary’ (Kim 2006: 341). ‘98% of the students in the Input [Enhancement] Group and 78% of the students in the Dictogloss Group experienced a positive change between the First and Last Essay. In contrast, only 38% of the students in the Conventional Group changed in a positive direction’ (Berent et al. 2007: 19). ‘Textual enhancement aided the learning of the target forms while having unfavorable effects on meaning comprehension’ (Lee 2007: 87)
NOTES 1 The 21 studies all had TE either as an independent variable or as a level thereof. 2 Of the 21 studies, Barcroft (2003) and Kim (2006) are the only two studies that focused on discrete-item vocabulary learning. There is arguably a distinction between vocabulary learning and grammar learning, which we will not deal with in this paper. 3 The cognitive processes may or may not be accessible to consciousness (Sharwood Smith, personal communication). 4 Barcroft (2003) provides evidence of a different kind of trade-off, ‘bidirectional effects’, namely that TE can draw attention to the enhanced items but away from the unenhanced items. 5 In order to account for the possible detracting effect of enhancement, J. White (1998) enhanced a different form than the target one in the reading texts for the unenhanced group.
6 The researcher explicitly told the participants to focus on the enhanced form. 7 What we have in mind here are the mainstream single-capacity models (see Robinson, 1997 for discussion). 8 The theoretical framework that underlies much of VanPatten’s research is known as ‘input processing’, which examines the form– meaning connection process in relation to intake. 9 See Greenslade et al.’s (1999) replication using aural and written input. 10 Wong (2005) sheds additional light on the lack of difference between the enhanced and the unenhanced group in this study, noting that both groups received explicit instructions to focus on verbal morphology, and that the effectiveness of the instructions may have overridden the effectiveness of the TE. 11 There is evidence from Barcroft (2003) that the fewer the enhanced
616
TEXTUAL ENHANCEMENT OF INPUT
items, the more effective they are, suggesting that ‘distinctiveness’ can be a moderator variable to TE. 12 A related issue that is worth looking into concerns the nature of learning as a function of input enhancement. In discussing the research on input flood,
Wong (2005) insightfully notes that ‘input flood does not appear to be effective in showing learners what is not possible in the target language’ (p. 42) hence suggesting that the learning incurred by this particular strategy might be inadequate.
REFERENCES Alanen, R. 1995. ‘Input enhancement and rule presentation in second language acquisition’ in R. Schmidt (ed.): Attention and Awareness in Second Language Acquisition. Honolulu: University of Hawai’i Press, pp. 259–99. Allwright, D. 1984. ‘Why don’t learners learn what teachers teach? The Interaction Hypothesis’ in D. M. Singleton and D. Little (eds): Language Learning in Formal and Informal Contexts. IRAAL, pp. 3–18. Barcroft, J. 2003. ‘Distinctiveness and bidirectional effects in input enhancement for vocabulary learning,’ Applied Language Learning 13/2: 133–59. Bardovi-Harlig, K. and D. Reynolds. 1995. ‘The role of lexical aspect in the acquisition of tense and aspect,’ TESOL Quarterly 29/1: 109–31. Barry, S. and A. Lazarte. 1995. ‘Embedded clause effects on recall: Does high prior knowledge of content domain overcome syntactic complexity in students of Spanish?’ The Modern Language Journal 79/4: 491–504. Berent, G., R. Kelly, S. Aldersley, K. Schmitz, B. Khalsa, and J. Panara. 2007. ‘Focus-on-form instructional methods promote deaf college students’ improvement in English grammar,’ Journal of Deaf Studies and Deaf Education 12/1: 8–24. Bowles, M. 2003. ‘The effects of textual input enhancement on language learning: An online/offline study of fourth-semester Spanish students’ in P. Kempchinski and P. Pineros (eds): Theory, Practice, and Acquisition: Papers from the 6th Hispanic Linguistic Symposium and the 5th Conference on the Acquisition of Spanish & Portuguese. Somerville, MA: Cascadilla Press, pp. 359–411. DeKeyser, R. 1994. ‘How implicit can adult second language learning be?’ AILA Review 11: 83–96. Doughty, C. 1991. ‘Second language instruction does make a difference,’ Studies in Second Language Acquisition 13: 431–69.
Doughty, C. 1998. ‘Acquiring competence in a second language: Form and function’ in H. Byrnes (ed.): Learning Foreign and Second Languages. New York: Modern Language Association, pp. 128–56. Doughty, C. 2004. ‘Effects of instruction on learning a second language: A critique of instructed SLA research’ in B. VanPatten, J. Williams, S. Rott, and M. Overstreet (eds): Form-meaning Connections in Second Language Acquisition. Mahwah, NJ: Lawrence Erlbaum, pp. 181–202. Færch, C. and G. Kasper. 1986. ‘The role of comprehensible input in second-language acquisition,’ Applied Linguistics 7/3: 257–73. Gass, S. 1988. ‘Integrating research areas: A framework for second language studies,’ Applied Linguistics 9: 198–217. Greenslade, T., L. Bouden, and C. Sanz. 1999. ‘Attention to form and content in processing L2 reading texts: A conceptual replication study of VanPatten 1991,’ Spanish Applied Linguistics 3: 65–90. Han, Z.-H. 2007. ‘On the role of meaning in focus on form’ in Z.-H. Han (ed.): Understanding Second Language Process. Clevedon: Multilingual Matters, pp. 45–79. Han, Z.-H. and S. Peverly. 2007.’Input processing: A study of ab initio learners with multilingual backgrounds,’ The International Journal of Multilingualism 4/1: 17–37. Hulstijn, J. 1997. ‘Second language acquisition in the laboratory,’ Studies in Second Language Acquisition 19/2: 131–43. Izumi, S. 2002. ‘Output, input enhancement, and the noticing hypothesis: An experimental study on ESL relativization,’ Studies in Second Language Acquisition 24/4: 541–77. Jourdenais, R., M. Ota, S. Stauffer, B. Boyson, and C. Doughty. 1995. ‘Does textual enhancement promote noticing? A think-aloud protocol analysis’ in R. Schmidt (ed.): Attention and
ZHAOHONG HAN, EUN SUNG PARK, and CHARLES COMBS
Awareness in Foreign Language Learning. Honolulu: University of Hawai’i Press, pp. 182–209. Kim, Y. 2006. ‘Effects of input elaboration on vocabulary acquisition through reading by Korean learners of English as a foreign language,’ TESOL Quarterly 40/2: 341–73. Lee, S. 2007. ‘Effects of textual enhancement and topic familiarity on Korean EFL students’ reading comprehension and learning of passive form,’ Language Learning 57/1: 87–118. Leeman, J., I. Arteagoitia, B. Fridman, and C. Doughty. 1995. ‘Integrating attention to form with meaning: Focus on from in contentbased Spanish instruction’ in R. Schmidt (ed.): Attention and Awareness in Foreign Language Learning (Technical Report #9). Honolulu: University of Hawai’i Press, pp. 217–58. Leeser, M. 2004. ‘The effects of topic familiarity, mode, and pausing on second language learners’ comprehension and focus on form,’ Studies in Second Language Acquisition 26/4: 587–616. Leow, R. 1997. ‘The effects of input enhancement and text length on adult L2 readers’ comprehension and intake in second language acquisition,’ Applied Language Learning 82: 151–82. Leow, R. 2001. ‘Do learners notice enhanced forms while interacting with the L2?: An online and offline study of the role of written input enhancement in L2 reading,’ Hispania 84: 496–509. Leow, R., T. Egi, A. Nuevo, and Y. Tsai. 2003. ‘The roles of textual enhancement and type of linguistic item in adult L2 learners’ comprehension and intake,’ Applied Language Learning 13/2: 1–16. Lightbown, P. 1991. ‘What have we here? Some observations of the influence of instruction on L2 learning’ in R. Phillopson, E. Kellerman, L. Selinker, M. Sharwood Smith, and M. Swain (eds): Foreign Language Pedagogy Research: A Commemorative Volume for Claus Faerch. Clevedon: Multilingual Matters, pp. 197–212. Long, M. 1991. ‘Focus on form: A design feature in language teaching methodology’ in K. de Bot, D. Coste, C. Kramsch, and R. Ginsberg (eds): Foreign Language Research in a Cross-cultural Perspective. Amsterdam: John Benjamins, pp. 39–52. Long, M. 2007. Problems in SLA. Mahwah, NJ: Lawrence Erlbaum, pp. 75–118. Long, M. and P. Robinson. 1998. ‘Focus on form: Theory, research, and practice’ in C. Doughty and J. Williams (eds): Focus on Form in Classroom
617
Second Language Acquisition. Cambridge, UK: Cambridge University Press, pp. 16–41. Lorch, R. 1989. ‘Text-signaling devices and their effects on reading and memory processes,’ Educational Psychology Review 1: 209–34. Mackey, A. and J. Goo. 2007. ‘Interaction research in SLA: A meta-analysis and research synthesis’ in A. Mackey (ed.): Conversational Interaction in Second Language Acquisition: A Series of Empirical Studies. Oxford: Oxford University Press, pp. 407–52. Ortega, L. and G. Iberri-Shea. 2005. ‘Longitudinal research in second language acquisition: Recent trends and future directions,’ Annual Review of Applied Linguistics 25: 26–45. Overstreet, M. 1998. ‘Text enhancement and content familiarity: The focus of learner attention,’ Spanish Applied Linguistics 2: 229–58. Park, E. S. 2004. ‘Constraints of implicit focus on form: Insights from a study of input enhancement,’ Teachers College, Columbia University Working papers in TESOL & Applied Linguistics 42: Retrieved 7 September 2005, from http:// www.tc.edu/tesolalwebjournal/Park.pdf Robinson, P. 1996. ‘Learning simple and complex second language rules under implicit, incidental, rule-search, and instructed conditions,’ Studies in Second Language Acquisition 18: 27–67. Robinson, P. 1997. ‘Generalizability and automaticity of second language learning under implicit, incidental, enhanced, and instructed conditions,’ Studies in Second Language Acquisition 19: 223–47. Robinson, P. 2003. ‘Attention and memory during SLA’ in C. Doughty and M. Long (eds): The Handbook of Second Language Acquisition. Oxford: Blackwell Publishers, pp. 630–78. Schmidt, R. 1990. ‘The role of consciousness in second language learning,’ Applied Linguistics 11: 129–58. Schmidt, R. 2001. ‘Attention’ in P. Robinson (ed.): Cognition and Second Language Instruction. Cambridge: Cambridge University Press, pp. 3–32. Selinker, L. 1972. ‘Interlanguage,’ International Review of Applied Linguistics 10/2: 209–31. Sharwood Smith, M. 1981. ‘Consciousnessraising and second language acquisition theory,’ Applied Linguistics 7/3: 239–56. Sharwood Smith, M. 1986. ‘Comprehension versus acquisition: Two ways of processing input,’ Applied Linguistics 7/3: 239–56.
618
TEXTUAL ENHANCEMENT OF INPUT
Sharwood Smith, M. 1991. ‘Speaking to many minds: On the relevance of different types of language information for the L2 learner,’ Second Language Research 72: 118–32. Sharwood Smith, M. 1993. ‘Input enhancement in instructed SLA,’ Studies in Second Language Acquisition 15: 165–79. Shook, D. 1994. ‘FL/L2 reading, grammatical information, and the input-to-intake phenomenon,’ Applied Language Learning 52: 57–93. Shook, D. 1999. ‘What foreign language reading recalls reveal about the input-to-intake phenomenon,’ Applied Language Learning 10/1&2: 39–76. Skehan, P. 1996. ‘A framework for the implementation of task-based instruction,’ Applied Linguistics 17: 38–62. Truscott, J. 1998. ‘Noticing in second language acquisition: A critical review,’ Second Language Research 14/2: 103–35. VanPatten, B. 1990. ‘Attending to form and content in the input,’ Studies in Second Language Acquisition 12: 287–310. VanPatten, B. 1991. ‘Grammar instruction and input processing’. Paper presented at the The Special Colloquium on the Role of Grammar Instruction in Communicative Language Teaching, Concordia University and McGill University, Montreal. VanPatten, B. 1996. Input Processing and Grammar Instruction: Theory and Research. Norwood, NJ: Ablex.
VanPatten, B. 2004. Processing Instruction. Mahwah, NJ: Lawrence Erlbaum. White, J. 1998. ‘Getting the learner’s attention’ in C. Doughty and J. Williams (eds): Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press, pp. 85–113. White, L. 1991. ‘Adverb placement in second language acquisition: Some effects of positive and negative evidence in the classroom,’ Second Language Research 72: 133–61. Williams, J. 1999. ‘Memory, attention, and inductive learning,’ Studies in Second Language Acquisition 211: 1–48. Williams, J. and J. Evans. 1998. ‘What kind of focus and on which forms?’ in C. Doughty and J. Williams (eds): Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press, pp. 139–55. Wong, W. 2001. ‘Modality and attention to meaning and form in the input,’ Studies in Second Language Acquisition 23/3: 345–68. Wong, W. 2003. ‘The effects of textual enhancement and simplified input on L2 comprehension and acquisition of non-meaningful grammatical form,’ Applied Language Learning 14: 109–32. Wong, W. 2005. Input Enhancement: From Theory and Research to the Classroom. San Francisco, CA: McGraw-Hill (Second Language Professional Series).
Applied Linguistics 29/4: 619–644 ß Oxford University Press 2008 doi:10.1093/applin/amn011 Advance Access published on 29 April 2008
The Cultural Productions of the ESL Student at Tradewinds High: Contingency, Multidirectionality, and Identity in L2 Socialization STEVEN TALMY University of British Columbia Although the originators of the language socialization (LS) paradigm were careful to cast socialization as a contingent, contested, ‘bidirectional’ process, the focus in much first language LS research on ‘successful’ socialization among children and caregivers may have obscured these themes. Despite this, I suggest the call for a more ‘dynamic model’ of LS (Bayley and Schecter 2003), while compelling, is unnecessary: contingency and multidirectionality are inherent in LS given its orientation to socialization as an interactionally-mediated process. This paper foregrounds the ‘dynamism’ of LS by examining processes comprising ‘unsuccessful’ or ‘unexpected’ socialization. Specifically, it analyses interactions involving ‘oldtimer’ ‘Local ESL’ students and their first-year teachers at a multilingual public high school in Hawai’i. Contingency and multidirectionality are explicated through analysis of two competing ‘cultural productions of the ESL student.’ The first, manifest in ESL program structures and instruction, was school-sanctioned or ‘official.’ Socialization of Local ESL students into this schooled identity was anything but predictable, however, as they consistently subverted the actions, stances, and activities that constituted it. In doing so, these students produced another, oppositional ESL student identity, which came to affect ‘official’ classroom processes in significant ways.
INTRODUCTION The socialization of children or novices by adults or experts into particular roles, identities, and world views has been the topic of scholarship for decades across the social sciences. Concerning as it does ‘the activity that confronts and lends structure to the entry of nonmembers into an already existing world’ (Wentworth 1980: 85), the nexus of socialization research engages such longstanding problematics as ‘agent vs. structure’, ‘voluntarism vs. determinism’, and ‘macro vs. micro’. Depending on disciplinary origin and theoretical orientation, emphases have varied in the diverse socialization literature on the influence that, for example, society has on the individual, or genetics has over the environment. Notwithstanding differences in emphasis, however, early theories of socialization—from the psychoanalytic tradition of Freud (1939), to the sociological functionalisms of Durkheim (1997) and Parsons (1937),
620
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
to the cultural anthropology of Mead (1961)—shared a general concern with the outcomes of socialization, characteristic of what Wentworth (1980) calls a ‘socialization-as-internalization model.’ In contrast, theories within a ‘socialization-as-interaction model’ have focused on the processes of socialization, including symbolic interactionism (Blumer 1969), phenomenology (Schutz 1967), ethnomethodology (Garfinkel 1967), and structuration theory (Giddens 1979). Despite substantive differences, these latter theories all consider ‘the medium through which the ability to produce society is transmitted [sic] from member to novice’ (Wentworth 1980: 79), that is, ‘the interaction that constitutes socialization’ (1980: 83). In doing so, they work to transcend binarisms such as ‘individual vs. society’, the inevitability and unidirectionality of a totalizing socialization, and a conception of the individual so ‘malleab[le] and passiv[e] . . . in the face of all powerful social influences’ (White 1977: 5) that s/he could be considered ‘oversocialized’ (Wrong 1961). It is within this ‘process’ orientation that language socialization (LS), the theory of socialization most often adopted in applied linguistics, can be situated. LS originated in the 1980s (e.g. Schieffelin and Ochs 1986a, 1986b) as a response both to earlier anthropological studies of child socialization, which ‘proceeded as if language was irrelevant’, and to the ‘invisibility’, in psycholinguistic studies of first language (L1) acquisition, of culture ‘as a principle that organized speech practices and their acquisition’ (Kulick and Schieffelin 2004: 349; also see Ochs and Schieffelin 2008). Consistent with its orientation toward socialization as a process that is accommodated, contested, and transformed by agents in interaction (see, e.g. Schieffelin and Ochs 1986a, 1986b; Garrett and Baquedano-Lo´pez 2002), LS provides a rigorous analytic framework for examining socializing processes in situ. This allows LS analyses to be grounded in ways unavailable to other models of socialization. It also offers the means to demonstrate the fundamental contingency and multidirectionality of socialization as it is—or is not—collaboratively achieved. Despite these analytic means, there has been some recent debate, particularly among second language (L2) LS scholars, concerning the extent to which unpredictability and mutuality in socialization have been accounted for in LS research. Bayley and Schecter (2003), for example, have strongly argued for ‘a more dynamic model’ of LS (2003: 6), one that ‘emphasize[s LS] as an interactive process, in which those being socialized also act as agents rather than as mere passive initiates’ (2003: 3). Such a call is an important corrective, they argue, to the view in ‘traditional’ LS studies of socialization as ‘relatively static, bounded, and relatively unidirectional’ (Schecter and Bayley 2004: 605). Yet such assessments are at odds with early claims made by Schieffelin and Ochs (1986a: 165) that ‘socialization is an interactive process’ and that ‘the child or novice . . . is not a passive recipient of sociocultural knowledge’ who ‘automatically internalize[s] others’ views’ but is a ‘selective and active participant . . . in the process of constructing social worlds’ (cf. Duff and Hornberger 2008). I argue that the call for a more ‘dynamic model’ of LS, while compelling and important, is in fact unwarranted given the basic premise in LS of socialization
STEVEN TALMY
621
as an interactionally-mediated process. That notwithstanding, however, it certainly is incumbent on LS researchers to highlight the ‘dynamism’ of LS in their analyses, lest the image emerge that socialization, and cultural and social reproduction, are ‘inevitable’ (Kulick and Schieffelin 2004; cf. Willis 1977). In the study below, I join a growing number of LS researchers (e.g. Duff 1995, 2002; Siegal 1996; Cole and Zuengler 2003; He 2003; Schecter and Bayley 2004) whose work foregrounds contingency and multidirectionality in LS by examining processes comprising what might be considered ‘unsuccessful’ or ‘unexpected’ LS. Specifically, I undertake an analysis of socializing interactions involving ‘oldtimer’ (Lave and Wenger 1991) ‘Local English-asa-second-language (ESL)’ students and their ‘newcomer’ teachers at Tradewinds High,1 a multilingual public high school in Hawai’i. My argument concerning contingency and multidirectionality takes shape around two competing ‘cultural productions of the ESL student’ (cf. Levinson et al. 1996) at Tradewinds. The first, manifest in ESL program structures and instructional practice, I gloss as school-sanctioned or ‘official’. Socialization of oldtimer ‘Local ESL’ students into this particular schooled identity was anything but certain, however, as they consistently withdrew participation in and otherwise subverted the acts, stances, and activities that constituted it. In doing so, these students produced another, oppositional ESL student identity which came to affect ‘official’ classroom processes in significant ways, contributing, perhaps paradoxically, to the social reproduction (Willis 1977, 1981; Giddens 1979) of the marginality and ‘stigma’ of ESL (also see Talmy forthcoming-a, forthcoming-b). In the following section, I consider the debate concerning contingency and multidirectionality in LS. Afterwards, I introduce the 2.5-year critical ethnography the data below is drawn from. I then analyse four classroom interactions involving Local ESL students and their newcomer teachers using a critical pragmatic (Talmy 2007) microanalytic framework that combines LS, interactional sociolinguistics (Gumperz 1982), membership categorization analysis, and ‘applied’ (ten Have 2001) conversation analysis (Sacks 1992). I use the first two interactions to illustrate contingency in LS by focusing on the competing ‘cultural productions of the ESL student’ mentioned above. My analysis of the latter two interactions focuses more directly on multidirectionality in LS, specifically in terms of how oldtimer students’ actions were increasingly accommodated by newcomer teachers over the course of a school year, suggesting their socialization into an ESL teacher identity that was at once adversarial and infantilizing. I conclude by considering implications of the analysis for LS, agency, and cultural reproduction.
CONTINGENCY AND MULTIDIRECTIONALITY IN LS In earlier ‘first generation’ (Garrett and Baquedano-Lo´pez 2002) LS studies, the primary foci were on the ways that caregivers socialized children in and through the L1 to become ‘competent’ members of sociocultural groups.2
622
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
The focus on ‘successful’ cases of children’s L1 socialization, on recurrent language practices, universals, and monolingual communities, has, according to some, obscured the basic orientation in LS of socialization as contingent and multidirectional. In terms that evoke criticisms made of earlier functionalist theories of socialization, He (2003) maintains that LS research tends to emphasize the efforts made by the experts . . . to socialize the novices. . . . Less visible are the reactions and responses of the novices. Consequently, the process of socialization is often characterized as smooth and seamless, and novices are often presumed to be passive, ready, and uniform recipients of socialization. (128) Garrett (2004), for one, disputes He’s assertion, arguing that this ‘is simply untrue: Even the earliest [LS] studies stressed the agency (and the capacity for resistance and creativity) of the child or novice, and conceptualized socialization as a two-way process’ (778). This is indeed the case (see above); however, it should be noted that there is a difference between ‘stressing’ these themes, and demonstrating them empirically. In an apparent nod to the contention that LS has downplayed the ‘dynamism’ of LS by focusing on communicative processes involved in socialization that is realized rather than that which is not, Ochs (1999) herself has noted certain ‘undesirable effects’ deriving from the foci of earlier LS studies, including fixity, essentialism, and reductionism: We say . . . that Samoan caregivers communicate one way, EuroAmerican caregivers another way, Kaluli yet another, and so on. Our accounts . . . seem like fixed cameos, members and communities enslaved by convention and frozen in time rather than fluid and changing over the course of a generation, a life, and even a single social encounter. (231) Kulick and Schieffelin (2004) address the matter more directly: That the majority of [LS] studies have focused on [cultural and social] reproduction is a strength—they provide us with methodological and analytical tools for investigating and interpreting . . . continuity across generations. But the focus on expected and predictable outcomes is a weakness if there is not also an examination of cases in which socialization doesn’t occur, or where it occurs in ways that are not expected or desired. To the extent that [LS] studies only document the acquisition of normatively sanctioned practices, they open themselves up to the charge that they are merely behaviorism in new clothes. (355) The issue thus appears to come down to at least some combination of empirical focus in LS research (e.g. highlighting processes involved when L1 socialization is achieved) and historical moment (i.e. establishing and
STEVEN TALMY
623
elaborating the LS paradigm) (cf. Ochs and Schieffelin 2008), rather than some problem endemic to LS itself. For this reason, the focus of this paper can be considered part of an historically-situated effort to ‘push [LS’s] boundaries . . . further,’ as Garrett puts it, by examining ‘bilingual and multilingual settings . . . later stages of the lifespan . . . [and] contexts associated with those later stages, such as peer groups, schools, and workplaces’ (2004: 776; also see Duff and Hornberger 2008). Further discussion of this point is available online as Supplementary Material 1 for Subscribers at http:// applij.oxfordjournals.org
THE STUDY The data utilized here come from a larger study consisting of 625 hours of observation in 15 classrooms, including eight dedicated-ESL classes, over 2.5 years. Observational data were generated in field notes and supplemented by audio-recordings of 158 hours of classroom interaction. A total of 58 formal interviews were audio-recorded with 10 teachers and 37 students, and materials, classwork, and other site artifacts were collected for analysis.
First-year teachers in first-year ‘ESL-A’ classes This paper concerns the three first-year ‘ESL-A’ classes that I observed, each taught by an instructor in his/her first year at the high school.3 I observed Ms Cheney’s ESL-A (1X),4 Ms Ariel’s ESL-A (2W), and Mr Day’s ESL-A (2X) for 48, 68, and 64 hours respectively, with 26 hours of audio-recording in the ESL-A (2W) class, and 29 hours in ESL-A (2X).5 The ESL-A classes were the largest, most heterogeneous, and instructionally challenging classes in the Tradewinds ESL program. One reason for this was a policy that determined ESL placements based not on L2 proficiency but length of enrollment at the school. Thus, the ESL-A classes were for students in their first year at Tradewinds, regardless of age, grade-level, or L2 proficiency. The three classes I observed averaged over 30 students, aged 14–18, about one-third of whom were at early levels of L2 development and/or had interrupted formal educations; another third of whom had lived in the US for 3–10 years, many of whom I identified as Local ESL (see Table 1); with the remainder at levels of English expertise in between.
‘Local’ and ‘Local ESL’ ‘Local’ is an identity category in wide circulation in Hawai’i, signifying a racialized ‘sociopolitically constructed panethnic formation comprised of Asian and/or Pacific Islanders who were born and raised in the islands’ (Labrador 2004: 292).6 It is a relational category, defined as it is through a system of ‘categorical opposition[s] between groups considered [L]ocal and
624
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
Table 1: Local ESL students in ESL-A (2W) and ESL-A (2X) (age and length of residence at mid-point of school year; self-report data) Student pseudonym
Sex Age L1
Country of origin
US length of residence
Ms. Ariel’s ESL-A (2W) 618 F Ash M Barehand M China M Camp Kill Yourself (CKY) M Eddie M Maria F Benz M Nat M Raven M
14 14 16 15 14 15 14 15 14 15
Cantonese Cantonese Korean Cantonese Vietnamese Ilokano Tagalog Korean Marshallese Mandarin
China Hong Kong Korea Hong Kong Vietnam Philippines Philippines Korea Marshall Islands Taiwan
4.5 years 7.5 years 3.5 years 4.5 years 9.5 years 4 years 4.5 years 2.5 years 8 years 8.5 years
Mr. Day’s ESL-A (2X) CJ Computer Ioane Iwannafuckalot (IwannaFAL) Jennie
M M F M
17 17 16 14
Ilokano Korean Tongan, English Vietnamese
Philippines Korea US (Hawai’i) Vietnam
7 years 5.5 years n/a 9.5 years
F
14
Korean
US (California)
Laidplayer
M
15
Palauan
Palau
Mack Daddy Nina Joyleen
M F F
15 15 16
Chuukese Chuukese Chuukese
Chuuk, FSM Chuuk, FSM Chuuk, FSM
Shorty
F
16
Sky Blue
M
15
Samoan, Tongan, US (Hawai’i) English Cantonese Macau
2.5 years (returnee) 5 months (see below) 5.5 years 4.5 years 6 months (6 years in Guam) n/a 3 years
those considered non[L]ocal, including haole [i.e. ‘white people’], immigrants, the military, tourists, and foreign investors’ (Okamura 1994: 165). The production of ‘Local’ identity centrally involves participation in a range of social practices, including the use of Pidgin (Hawai’i Creole), the creole language of Hawai’i (see Sakoda and Siegel 2003).
STEVEN TALMY
625
‘Local ESL’, an analyst’s category, is conceptualized as a situationally contingent identity (Zimmerman 1998). It refers to students who were institutionally-identified as ESL by Tradewinds, yet who displayed the following: cultural knowledge of and affiliation with Local culture, cultural forms, and social practices; experience with US and Hawai’i school expectations and practices; difference from newcomer or lower-L2-proficient classmates, who many Local ESL students characterized as the ‘real’ ESL students; awareness of—and participation in—‘linguicist’ (Phillipson 1988) discourses about ESL in circulation in the wider social context; and, crucially, the L2 expertise and interactional competence to participate in these practices.7
School-sanctioned productions of ESL What I call the school-sanctioned productions of ESL can be explicated in terms of instructional practice and the ‘structural productions’ (Eisenhart 1996) of ESL. Drawing on Nespor (1990), Eisenhart (1996) uses the concept of ‘structural productions’ to analyse how the subject position ‘student’ is produced within and by particular institutional arrangements, that is to say, how program organization, intra-institutional relationships, and curriculum constitute certain preferred ways of being a student in those contexts. Such an idea is especially salient for programs such as ESL, which tightly control student enrollment and exit, promoting access to certain forms of learning and school experience, and denying it to others. Such structures have significant ‘implications for how students spen[d] their time (e.g. with whom, doing what) and what they learn’ (Eisenhart 1996: 172), and can result in the ‘produc[tion] of cohorts . . . with shared outlooks, ambitions, definitions of reality, and strategies for acquiring and using knowledge’ (Nespor 1990: 231–2). Analysing structures of the Tradewinds ESL program can therefore offer insight into how the category ‘ESL student’ was institutionally conceived. ESL at Tradewinds was structurally articulated as a hypernym, suggesting that students classified as ‘ESL’ were essentially the same, with negligible differences in L1 or L2 abilities, educational backgrounds, or needs. This was exemplified by the placement policy described above, whereby length of enrollment rather than L2 expertise determined what ESL classes students were to take. It was also signified by an undifferentiated curriculum: although the ESL-A classes were large and diverse, students were, with few exceptions, given the same assignments, activities, and tests, each with the same demands and deadlines. In addition to homogeneity, ESL at Tradewinds was structurally elaborated as an afterthought to pre-existing institutional arrangements. This was indicated by the status of ESL classes as ‘required electives’; by an absence of communication between ESL and subject-area departments; and a lack of administrative, curricular, and physical integration with the rest of the school. These arrangements were amplified by the assignment of ESL courses to new or junior faculty, most of whom had little background in ESL,
626
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
and who were provided minimal institutional support despite frequently overwhelming instructional circumstances. The ESL curriculum reflected, and helped to re/produce, the program’s institutional standing. This was most evident in the choice of L1-English juvenile fiction to form the basis of the ESL curriculum. Books such as James and the Giant Peach (Dahl 1961) and Dogsong (Paulsen 1985) were below grade-level and often had tangential relevance to high school academics or L2 learning. In addition to these books were assignments that evidently presumed students knew about and ‘automatically affiliated’ (Talmy 2007) with the cultures, customs, and languages of ‘their’ countries (Duff 2002; Talmy forthcoming-a).8 Assignments introducing initiates to cultures and customs of the US often appeared, as well, especially around holidays, as did other ESL mainstays such family tree activities and map exercises, which most Local ESL students maintained they had been assigned in earlier grades, often repeatedly. Thus, ‘ESL student’ as it was institutionally articulated at Tradewinds connoted a monolithic out-group of recently-arrived cultural and linguistic ‘Others’, that is, an iconic, stereotypical ‘ESL Student’, or ‘FOB’ (‘fresh off the boat’) (Talmy 2007, forthcoming-a, forthcoming-b). Local ESL students’ responses to these school-sanctioned productions of the ESL student, were, as might be expected, largely negative. Indeed, these responses became so pervasive in the ESL-A classes I observed that they ultimately affected what was taught, how it was taught, and who taught it, with the knowledge, orientations, and practices of an ‘institutionally disapproved interstitial community of practice’ (CoP) (Lave 1991: 78) coming to assume varying degrees of prominence in each of these classes.9 The ‘joint enterprise’ of this CoP, produced as Local ESL students ‘mutually engaged’ in a ‘shared repertoire’ of social practices (Wenger 1998), was the contestation, if not destabilization, of the school-sanctioned productions of ESL. Implicated in these efforts were Local ESL students’ prior experiences in ESL classes, as well as their connections to communities beyond the ESL program, where ESL was stigmatized, and English monolingualism, the ‘mainstream’, and ‘regular students’ were valorized as preferred ideals.
Competing productions of the ESL student The interactions I discuss next occurred between the third and sixth months of the school year. By this time, curriculum and classroom processes had begun to appreciably change in all three ESL-A classes as teachers contended with Local ESL students’ participation in practices, both subtle and overt, that worked to undermine ‘official’ curriculum and instruction. These included leaving assigned materials ‘at home’; not doing homework; completing assignments that required minimal effort (e.g. worksheets) but not others (e.g. writing activities); starting class late; and finishing early. The more overt practices included bargaining for reduced requirements on classwork and
STEVEN TALMY
627
extended time to complete it; refusal to participate in instructional activities; teasing students who did participate; and the often delicate negotiations with teachers that resulted (Talmy forthcoming-b). These in part constituted the shared repertoire of the Local ESL CoP described above, and the oppositional productions of the ESL student. The first extract I analyse involves Laidplayer, a 9th-grader from Palau. When I first met Laidplayer, I was surprised to learn he had only been in Hawai’i for a few weeks, due to his advanced expertise in L2 English and Pidgin, and his knowledge of Local culture. As it happened, Laidplayer had extended family in Hawai’i with whom he lived as he went to school, and who provided him with the resources to act and sound—to be—Local. Though he claimed to have had no idea what ‘ESL’ meant when he first came to Tradewinds, by the end of the year he was participating in a host of practices that were indexical of an ESL oldtimer. This interaction occurred five months into the school year, near the end of a ‘study hall’ class session, in which students were, in Mr Day’s words, to be ‘working behind or ahead’, that is, to be catching up on overdue work. Laidplayer had to this point spent 45 minutes on a short exercise from an audiolingual-era grammar book (Dixon 1956), then had talked with two Local ESL classmates, before finally falling asleep. Mr Day had woken him twice already to get him to finish an overdue summary from Shiloh (Naylor 1991). As the interaction begins, Laidplayer is fixing his watch and talking to IwannaFAL seated beside him (see Appendix for transcript conventions). Extract 1. So izi 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Mr. Day:
[ELA41XmdS10: 1185–1208]
((calls across room)) Laidplayer, you finish your summary. (0.4) IwannaFAL?: (no.) (0.5) ↑finish your summary it’s due today! do Mr. Day: you need help? (0.7) Mr. Day: I ↓kno:w, you want me to sit down and read the book with you. alri::ght. (0.7) ((Mr. Day approaches)) Mr. Day: >what’s wrong?< Laidplayer: I was: fixing my watch. (0.9) ((Mr. Day sits down)) Mr. Day: what’s wrong with your watch? Laidplayer: it’s too big. (2.9) IwannaFAL: oh ma:n, the bell’s gonna ring! (0.6) ?FS: ↑wha::t? (0.7) Mr. Day: u:h no you got about ten minu[tes before] we’re outta here.
628
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
Laidplayer:
Mr. Day: Laidplayer:
Mr. Day: Mr. Day: Laidplayer: Mr. Day: Laidplayer:
Mr. Day:
Laidplayer: Mr. Day: Laidplayer: Mr. Day: Laidplayer: Laidplayer: IwannaFAL:
[(sh:it.) ] (0.7) ((clicking sound of metal watch band)) why don’t you try get it done that way you won’t have to do it [at home. [SO IZI ai jas– IT’S SO EASY I just– (0.5) >so do it now!< (0.5) °>so do it now.<° bat ai neva rid da (h)as(hh)ai(h)n[men. but I didn’t read the assignment. [you gotta read this- the cha:pter. WEL AI DON RID DIS BUK wen ai get hom, >ai dono wai<. WELL I DON’T READ THIS BOOK when I get home, I don’t know why. (0.3) well why don’t you do it in class while I’m here making you, that way you can get it done. (0.5) pa:ge 66 to 88. (.) WELL, I read some, (0.4) [okay. [<when I> >go in the bathroom.< hhh huh? hey, I like to do that too. that’s the best time. just don’t get it dirty. HHHHH (0.4) hhhHHHHH. (1.0) ((Mr. Day gets up and leaves)) ho wipe my ass with (.) the book. ((clears throat)) oh yeah oh yeah.
In his first four turns, Mr Day enumerates or implies a series of acts and activities that constitute particular ESL student and teacher identities. Associated with the teacher identity are the rights, obligations, and competencies that warrant assigning ‘schoolwork’ to students, setting deadlines to complete it, monitoring classroom performance, making assessments about it, and if deemed problematic, pursuing some form of remedy, for example, in the form of ‘help’. The student is someone who has already completed the assigned reading of Shiloh so s/he can produce a summary of it, who is endeavoring to meet the due date of ‘today’, and is complying with Mr Day’s instructions in the larger activity of doing ESL coursework. This is, in other words, a student who is accommodating the production of a schoolsanctioned ESL student identity. It is clear that Laidplayer does not participate in the acts, stances, or activities associated with the production of this student identity: this has, in fact, occasioned Mr Day’s visit to him. Despite repeated ‘reminders’ to complete his overdue summary, during a ‘study hall’ class specifically intended for this purpose, Laidplayer has spent the period doing a 22-item grammar worksheet, talking with friends, sleeping, and is now fixing his watch. This is accountable
STEVEN TALMY
629
behavior and in his first turns, Mr Day displays an orientation to its problematicity. In lines 1–2, he calls Laidplayer by name, asking if he’s completed the summary as instructed. Although this may be a ‘real’ yes/no question, it is likely given the context that its polarity is reversed (Koshik 2002). Reversed polarity yes/no questions (RPQs), similar to rhetorical questions, are not information-seeking, but convey a ‘negative assertion’ (1852) about some form of social action and can thus ‘be heard as a complaint’ (1863). Whether this is an RPQ or not, the imperative, higher pitch, emphatic stress, and exclamatory intonation of Mr Day’s next utterance (lines 6–7) index a stance of unambiguous disapproval. Despite this, Laidplayer does not provide the sequentially-projected reply. In fact, it is not until more than 11 seconds after Mr Day calls his name that Laidplayer finally speaks. Even then, the account he provides, ‘I was: fixing my watch.’, is not what might be expected, since this provides denotative ‘proof’ that he has not been ‘finishing the summary’. Still, he offers no elaboration when one might, once more, be anticipated. In short, Laidplayer ‘does’ the same resistance interactionally that his utterance references denotationally; it is a mark of his L2 interactional competence that this utterance accomplishes both simultaneously. Interestingly, after a nearly 1-second delay, Mr Day accommodates rather than problematizes Laidplayer’s topicalization of his watch (‘what’s wrong with your watch?’). Although in line 16, Laidplayer (finally) provides an account (‘it’s too big.’), it is minimal. The silence that follows suggests that Mr Day rejects the account as inadequate (cf. Davidson 1984), and is awaiting further elaboration (cf. Pomerantz 1984). IwannaFAL’s sudden ‘oh man, the bell’s gonna ring!’ is thus particularly important, working as it does to reorient Mr Day from the pursuit of an elaboration to the matter of when class will end (lines 22–23). Coming as it does with more than 10 minutes left in the class (cf. the female student’s high-pitched ‘wha::t?’), this may be an instance of a Local ESL ‘oldtimer’ helping out a comparative apprentice during an interactional difficulty with his teacher. In lines 27–28, Mr Day returns to the matter of Laidplayer’s summary but in distinctly different terms. With the mitigated directive in question format (‘why don’t you try get it done’) and the added inducement (‘that way you won’t have to do it at home.’), Mr Day frames this utterance as a suggestion. When compared with Mr Day’s first two turns, it appears that he has oriented to Laidplayer’s disaffiliative actions and is accommodating them. With his line 29 utterance, ‘So IZI ai jas–’, Laidplayer at last provides the sought-after account about ‘what’s wrong’. However, the affective and epistemic stances indexed by the overlap, the amplified volume, the sentenceinitial placement of the intensifier ‘so’, the lack of hedging, and the sentence incompletion transform this utterance from an explanation, to a rejection that what is accountably ‘wrong’ is attributable to Laidplayer. It instead shifts the problem to the assigned material, that is, the award-winning children’s book Shiloh.
630
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
Significantly, Mr Day’s repeated ‘so do it now’ utterances in lines 31 and 33 do not challenge Laidplayer’s assessment, but implicitly concur with it. These can be seen as additional aligning moves, consistent with Mr Day’s earlier uptake regarding the watch, the mitigated directives in lines 27–28 and 40–42, and his final remark in lines 48–49. Indeed, the discourse marker ‘so’ that prefaces both line 31 and 33 turns marks a ‘fact-based result relation’ (Schiffrin 1987: 201–4), meaning the assessment is not only concurred with but used as an incentive for the student to ‘do it now!’ This transforms Laidplayer’s subsequent admissions in line 34 (bat ai neva rid da (h)as(hh)ai(h)nmen.) and lines 37–38 (WEL AI DON RID DIS BUK wen ai get hom, 4ai dono wai5.) into unaccountable actions. More significantly, it indexes broader pedagogical accommodations that have been implemented in this class to make it ‘easier’ for students to ‘finish’: the class time that has been allotted for students to complete overdue assignments, and the deadlines that have been extended. In other words, there is evidence here of this first-year teacher’s uptake to his students’ socializing practices. Laidplayer’s refusal to participate in the acts, stances, and activities that comprise the school-sanctioned ESL student identity, or in the interactional norms of a participation structure such as this (cf. Mehan 1979), are constitutive of the production of an alternative ESL student identity. As outlined in Table 2, this is a student whose disaffiliative actions index resistance to, difference from, and lack of investment (Norton 2000) in the ‘official’ productions of ESL.10 These points are further illustrated in Extract 2 from Ms Ariel’s ESL-A (2W). This interaction occurred early in the fourth quarter of the school year, by which time the Local ESL CoP in the class was also firmly established. It took place during a discussion activity for The Wanderer (Creech 2000), another award-winning book that by the publisher’s estimation was intended
Table 2: Acts and activities constituting productions of the ESL student at Tradewinds School-sanctioned ESL student
Local ESL student
Bringing required materials to class Reading assigned juvenile fiction Doing bookwork (summaries, questions, vocab) Meeting due dates Following instructions Working for full class session Not participating in proscribed activities (sleeping, talking with friends, playing cards, ‘off-task’ behavior)
Not bringing required materials to class Not reading assigned juvenile fiction Doing some bookwork (worksheets; little extended writing, e.g. summaries, questions)
Not meeting due dates Not following instructions Interactional resistance Participating in proscribed activities
STEVEN TALMY
631
for audiences aged 8–12. The reading group for this book was comprised almost entirely of Local ESL students: China, Ash, Raven, Barehand, Benz, 618, Sou Li (a low L2-proficient newcomer who had been paired with 618) and CKY, who was shuffling a deck of playing cards intermittently. For 7 minutes prior to the interaction, Ms Ariel had tried and failed to get a discussion started among these students about the following questions, which concerned Sophie, the 13-year-old protagonist of The Wanderer:
What is Sophie’s story? Why is she so attracted to the open ocean? Is it unusual for a 13-year-old to be the only female on the boat? What does this mean? (ELA41Wdoc: L). Extract 2. Cards 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Ms. Ariel:
Ms. Ariel: China: Ms. Ariel: Ms. Ariel:
China: China: Ms. Ariel:
Ms. Ariel: Benz?: Ms. Ariel:
China: Ms. Ariel: Ms. Ariel:
China: Ms. Ariel: China: Ms. Ariel: Benz: China: Benz: Ms. Ariel: Benz: China:
[ELA42WmdS11: 565-593]
you have to think about Sophie’s personality. (2.6) ((Ss talking in other group)) ((China opens recorder case, examines it)) (1.3) ((Ss talking in other group)) you have to think about what you’ve read. ((blowing into mic?)) SSSS[S [China. (0.6) why is she so attracted to the ocean. that is going to have a lot to do with your own (.) thoughts.= =I THINK IT’S QUIET AND PEACEFUL, (2.4) ((Ss talking in other group)) an:d dolphin is real[ly nice. [how many people have started on the second set of questions. (0.6) (as is us[ual.) [(nah.) (1.8) okay. start on them today, (0.3) finish the first three (0.7) before you start the se[cond set. [((heavy sigh)) (2.0) ((Ss talking in other group)) okay, and they’re essay style questions. (1.0) ((to CKY)) and don’t bring your cards in future please. (0.7) YAE CK↓Y. YEAH CKY. I’ll throw them away. are you looking at me, Miss? okay I’m gonna go see what they’re doing,= =kay. kay.= =bye. up to you,= =bye. hh
632
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
CKY: 618: China: Barehand: China:
Raven: China: Benz: China: Barehand: China: Benz: China: CKY:
hhh[h [(you’re cold.) ba’bye. >have a good day.<= =((smiley voice)) h-have a nice day. ((Ms. Ariel leaves)) (0.6)((Ss talking in other group)) heha hheha .h okay. (0.8) okay so= =get out= =so, cards. (1.3) ((CKY brings out playing cards?)) what- [okay. [cards. ((starts to deal the cards))
Similar to Mr Day in Extract 1, Ms Ariel here specifies and implies several acts constituting a particular ESL student identity: someone who has done the reading needed to participate in this activity, who has thought about what they’ve read and is prepared to discuss it, who has completed the first set of comprehension questions (in ‘essay style’) if not the second, and is following instructions. This is also a student who does not engage in ‘off-task’ behavior such as China’s in lines 4 and 7 (cf. Ms Ariel’s utterance in line 8), nor who brings cards to class, much less has them out to play with.11 It is obvious that the Local ESL students here, similar to Laidplayer above, do not engage in these activities, participating instead in the production of a different ESL student identity. None of the Local ESL students give any indication that they have done the reading, nor that they have started the first or second set of questions, even though the reading and both sets of questions were to have been ‘reviewed’ prior to this class. Also similar to Laidplayer in Extract 1, no explanations are forthcoming from the students for this behavior. Although China provides ostensible ‘answers’ in lines 13 and 15, the amplified volume, intonation, and nonsensical content indicate he is non-serious. In fact, it appears with these utterances, as well as the one in line 32, that China shifts footing here, mocking the school-sanctioned ESL identity by animating a child-like, compliant caricature of it. There is also his bald, on-the-record challenge in line 34, as well as the series of increasingly audacious utterances in lines 36–38 and 40–46, whereby these students essentially tell their teacher, in progressively more direct terms, to go away. Finally, after Ms Ariel departs for the other reading group, these students engage in the very activity that she explicitly proscribed a few turns earlier: playing cards. This interaction evinces many of the same characteristics of the two competing productions of the ESL student discussed above. What is additionally foregrounded here is the outline of what I have argued is an identifiable CoP. Returning to China’s utterance in line 53: ‘okay so’ signals a transition in discourse ‘from off-task or off-topic talk back to on-task or on-topic talk’, that
STEVEN TALMY
633
is, to ‘mutually understood and expected business at hand’ (Condon 2001: 509). Of note is that the ‘off-task talk’ in this interaction concerns The Wanderer, a manifestation of the ‘official’ ESL curriculum, while the ‘mutually understood and expected business at hand’ is the (forbidden) practice of card playing, itself an index of these students’ orientations toward if not histories of participation in (Local) communities beyond ESL (see note 11). The latched and overlapped utterances in lines 36–38 and 40–58 also highlight the cooperative, joint production of discourse here, and a shared orientation to the mutually understood task at hand. In Wenger’s (1998) terms, these students have engaged in a shared repertoire of practices for the larger enterprise of challenging the school-sanctioned productions of ESL. In doing so, they have produced an alternative ESL student identity that again indexes resistance to, difference from, and lack of investment in ESL. In this interaction as in the first, contingency in LS is evident in terms of these students’ refusal to accommodate the school-sanctioned productions of ESL. Despite power asymmetries that inhere in this institutional setting, and despite the power of those in hierarchically superior positions to prescribe to those in subordinate positions activities that constitute particular schooled identities, the Local ESL students’ actions above highlight how LS never proceeds ‘smooth[ly] and seamless[ly]’ (He 2003: 128) along lines of, for example, age or hierarchical standing (also see Jacoby and Gonzales 1991).
Teacher uptake to oppositional productions of the ESL student Although space constraints disallow consideration of the literature on teacher socialization (see, e.g. Zeichner and Gore 1990; Bullough et al. 1992; Kuzmic 1994; Kelchtermans and Ballet 2002), Local ESL students’ productions of the ESL student became increasingly accommodated by all three of the first-year teachers I observed, coming to affect classroom curriculum and instructional processes in significant ways. See Supplementary Material 2 online for Subscribers at http://applij.oxfordjournals.org. Particularly striking was how similar these accommodations were. As Local ESL students’ participation in oppositional practices intensified over the course of the school year, and as their academic performance simultaneously declined (see below), Ms Cheney, Ms Ariel, and Mr Day each began to routinely add ‘study hall’ sessions; extend deadlines on classwork; reduce the amount of and requirements for assignments, especially those involving writing; provide ‘alternative’ assignments; eliminate homework; and add expanded ‘review’ activities for quizzes and exams. Further discussion of this process is available online as Supplementary Material 3 for Subscribers at http://applij.oxfordjournals.org. Such adjustments were emblematic of multidirectionality in LS, in this case of the first-year teachers’ uptake to Local ESL students’
634
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
productions of the ESL student, and thus, of their socialization into a particular ESL teacher identity. I consider this in more detail next. The first interaction I consider involves an instructional accommodation that became more apparent over the course of the school year as increasing numbers of Local ESL students did not complete assigned schoolwork or bring to class the materials necessary for a lesson to proceed. To adapt to this circumstance, Ms Ariel, Ms Cheney, and Mr Day each began assigning ‘alternative’ classwork: spontaneous, unplanned assignments, which normally entailed little instruction or effort to complete (e.g. ‘freewriting’, reviewing for upcoming quizzes, looking up vocabulary words in a dictionary). This sort of classwork was frequent enough that another teacher in the ESL program had a specific term to refer to it: ‘filler’. The first interaction is from Ms Ariel’s class, and occurred weeks before winter vacation. On the whiteboard was the following agenda for this extended 95-minute class period: 1 2 3
Vocab test Reading/summaries and final book tests Extra credit and make Xmas cards—due today (ELA32Wfn: 1764–1766).
Below this information was a series of bulleted instructions detailing the ‘extra credit’ Christmas card-drawing activity, which was part of a schoolwide drive for charity. The interaction commences as the class was transitioning from the vocabulary quiz into ‘#2’: completing ‘readings/summaries’ for either Dogsong (Paulsen 1985) or Island of the Blue Dolphins (O’Dell 1960), and reviewing for the upcoming final book exam. Both the bookwork and exam review had been assigned as homework in the preceding class. CKY had just turned in his quiz, but was not doing the bookwork. Extract 3. Just do this 01 02 03 04 05 06 07 08 09 10
11 12 13 14 15 16 17 18
Ms. Ariel: CKY: Ms. Ariel: Ms. Ariel: Ms. Ariel: Eddie:
Ms. Ariel: Eddie: Eddie: Ms. Ariel:
[ELA32Wmd6: 1888-1899]
do you have your book. (3.9) ((Ss talking in b.g.)) ( leave it) at home. at home! (2.2) ((Ss talking in b.g.)) u::m. ((looks at whiteboard?)) (4.6) ((Ss talking in b.g.)) well. ((points at whiteboard)) Mis, kaen wi du dat?= Miss, can we do that? =yeah, you [can. [number three. (0.6) ((Ss talking in b.g.)) number three. ((to the class)) if- if you don’t have your book, (0.3) you can just do this::. ((points to Christmas card activity on whiteboard)) this is due today though,
STEVEN TALMY
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Ms. Ariel:
Raven: Ms. Ariel: Dannica: Ms. Ariel:
Dannica: Ms. Ariel:
it’s for (0.6) the card drive? (0.4) ((several turns clarifying instructions omitted)) bring your books next time! if you have them at home, bring them to class [next time. [( already. please. which book. (1.7) ((Ss talking in b.g.)) if you’re saying which book, then it probably doesn’t even matter. (0.6) ((Ss talking in b.g.)) oh. (0.5) ((Ss talking in b.g.)) the books (we’ve) been reading.
635
)
Similar to Extracts 1 and 2, there are several acts and activities here that entail a school-sanctioned ESL student identity. However, while CKY and Eddie have participated in at least one of these activities, the vocabulary quiz, they do not—cannot, since they do not have the requisite materials—participate in the reading/summary assignment or final exam review, the main instructional activities for the class (cf. Laidplayer’s completion of the grammar exercise above). In place of this work, Ms Ariel winds up assigning the ‘extra credit’ activity of Christmas card-drawing, suggested by Eddie in lines 9–10. Significantly, Ms Ariel offers this assignment not just to CKY and Eddie, but the entire class: ‘if– if you don’t have your book, (0.3) you can just do this.’ It is noteworthy that there is no check to see if anyone else has left their books ‘at home’ (cf. Talmy forthcoming-b). Rather, once CKY and Eddie indicate they have not brought their books, the apparent presumption is that other students have not done so either. What’s more, there is no check here or at any point during the period about whether anyone actually needs time to do the bookwork or exam review in class, which, as just mentioned, had been previously assigned as homework. This suggests that Ms Ariel has oriented to the likelihood that the homework would not have been done, necessitating class time to complete it. Further, the mild rebuke in line 4 (‘at home!’), the elaboration of the directions for the activity (omitted), and the mitigated directive in lines 23–25 (‘if you have them at home, bring them to class next time.’), which is followed by a politeness upgrade (‘please’), each in their own way index accommodation toward these circumstances, and by extension, Ms Ariel’s uptake to the acts, stances, and activities of a rather different ESL teacher: one who no longer requires students to complete reading/summaries homework, or review for the final exam; who accedes to, if not anticipates students’ leaving materials ‘at home’; who pursues no accounts for this behavior; and who assigns an alternative ‘filler’ activity instead (see Table 3).
636
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
Table 3: Acts and activities constituting productions of ESL teacher identity at Tradewinds School-sanctioned ESL teacher
Uptake to Local ESL students
Assigning official curriculum (e.g. juvenile fiction þ ‘bookwork’) Giving instructions Setting due dates Monitoring student performance Evaluating student performance Helping/remediating performance as necessary
No sanctions for noncompliance No pursuit of accounts Reducing assignments; requirements for them (especially writing) Extending deadlines for assignments Providing ‘alternative’ assignments Providing extra class time (‘study hall’) for overdue work Providing extensive ‘reviews’ for tests
Each of the preceding interactions took place during a ‘study hall’ class session. As noted earlier, ‘study hall’ involved setting class time aside so students could ‘catch up’ on late work. It was the most common accommodation deployed in the three ESL-A classes, with greater proportions of class time devoted to them as the school year progressed. This is hinted at in the next extract, which occurred at the beginning of a class, early in the final quarter of the year. Extract 4. Extra chance 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21
Mr. Day:
[ELA42XmdS13: 127-141]
shh. how many of you finished your work for today. Ioane: not me! Mr. Day: not [Ioane. Joyleen: [not me either! ?Ss: (3.3)((inaudible; overlapping talk)) Mr. Day: okay that’s good, that’s good. ( ) is good. Jennie: uh[:::. Mr. Day: [shhhhh. today I’m gonna give you guys a little bit of an extra chance to finish your work. (0.8) Mr. Day: how many of you rea::d (.) the (.) pages you were supposed to read in the book? Laidplayer: (don’t look at me, brah.) Mr. Day: no? Laidplayer: Mister. I don’t know where I put my book. Mr. Day: okay we’ll get you one, we’ll find it. shh! today we’re gonna be doing bookwork and grammar.
STEVEN TALMY
637
What is notable here is that Mr Day appears to have anticipated that students would not have finished work due this class. His question in lines 1–2 is designed such that negative responses like Ioane’s in line 3 and Joyleen’s in line 5 are not disagreeing or necessarily accountable (Ford 2001), even though a teacher could normatively expect that schoolwork assigned ‘for today’ would be complete. In fact, Mr Day does not pursue accounts from either Ioane or Joyleen, or from Laidplayer following his turn in line 18. Though it is unclear whether Mr Day’s repeated assessments of ‘good’ in lines 7–8 concern these students’ actions or what is inaudible in line 6, they just as well could: Mr Day has evidently planned this class for ‘bookwork and grammar’ (lines 20–21), with concomitant deadlines for the assignments that were ‘due today’ extended, and future in-class time set aside to work on them. Instructional accommodations such as these were arguably aimed at promoting academic performance, or at least, helping students improve their grades. However, despite the increasing provision of these accommodations, Local ESL students’ school grades in fact declined, alarmingly so, over the course of a school year. To give some indication of this, I present in Figure 1 the quarterly grades of 55 students for whom I have these data, 25 that I identified as Local ESL students, and 30 that I considered were not.12 Obviously, it is a mistake to draw too many conclusions from grades alone. However, they provide one (school-sanctioned) determination of classroom performance, and are of particular interest because the ESL-A classes actually got easier over the course of the year rather than more difficult. Groups by Quarter 2.6
Mean grade (4.0 scale)
2.4 2.2 2.0 1.8 1.6
Group Local ESL
1.4 1.2
Non-Local ESL 1
2
3
4
Quarter
Figure 1: Comparison of quarterly grades over one academic year (4.0 scale; Year 2 of study)
638
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
As the information in Figure 1 indicates, students in both Local ESL and non-Local ESL groups completed the first quarter of the school year with roughly similar grades in their ESL classes, and although the Local ESL group was slightly below the non-Local ESL group, this difference is not significant (t ¼ .679, df ¼ 53, n.s.). For the remainder of the year, however, the mean grade of the non-Local ESL group remained comparatively stable (ranging from 2.27 to 2.37), whereas for the Local ESL group, it declined: from 2.12 in the first quarter, to 1.88 in the second, and 1.4 in the third, though it recovered slightly in the fourth. Differences between groups were significant for the third quarter (t ¼ 2.355, df ¼ 53, p 5 .025) and fourth quarter (t ¼ 2.358, df ¼ 53, p 5 .025). A comparison of means within a repeatedmeasures analysis of variance showed a statistically significant interaction between Local ESL/non-Local ESL group membership and quarter [F(3,51) ¼ 2.939, p 5 .05]. Repeated-measures t-tests used as post-hoc tests indicated that for the Local ESL group, performance was significantly lower in Quarter 3 (t ¼ 2.688, p 5.05) and Quarter 4 (t ¼ 2.521, p 5 .05) than Quarter 1. No other differences between quarters were statistically significant. For the non-Local ESL group, none of the differences between quarters were statistically significant.13 In light of these data, the newcomer teachers’ instructional accommodations can be viewed as an attempt to mitigate such a substantial decline in Local ESL students’ academic performance over the school year.
Contingency and multidirectionality, cultural production and reproduction In this paper, I have foregrounded contingency and multidirectionality in LS by focusing on a population of oldtimer ESL students’ opposition to participating in the acts, stances, and activities that comprised a schoolsanctioned ESL student identity. This was, I have argued, reciprocally constitutive of an alternative ESL identity, one that indexed resistance to, difference from, and lack of investment in ESL, as well as orientations toward and affiliations with Local and mainstream communities beyond ESL. These oldtimer students’ refusal to accommodate the school-sanctioned productions of ESL illustrate the essentially contingent character of socializing processes, particularly those that involve older learners in a stratified institutional setting such as a high school, and that centrally concern the (compulsory) ascription of a contested, low-prestige identity such as ‘ESL student’. Multidirectionality was also evident as increasing numbers of students took up participation in what I have called a Local ESL CoP: denotatively (e.g. explicit, referential confrontation with teachers), interactionally (e.g. in terms of absence or delay of projected responses, or responses given as sequentially projected but otherwise out of line) and in embodied action (e.g. not bringing materials to class). As withdrawal from ‘official’ curricular and instructional business became broader and more sustained, and as the
STEVEN TALMY
639
newcomer ESL-A teachers increasingly accommodated this circumstance by extending deadlines, providing study hall sessions, eliminating homework, and so on, it became unclear ‘who . . . was socializing whom, and into what’ (Duff and Early 1999, in Zuengler and Cole 2005: 307; cf. Lave 1991: 79). That is to say, Local ESL students’ productions of the ESL student threw into question what precisely were the ‘target’ community, cultures, and practices—the ‘language, tools . . . symbols, well-defined roles . . . tacit conventions, subtle cues, untold rules of thumb’ (Wenger 1998: 47)—that students were being socialized into. These oppositional practices served as socializing resources through and to use language, for students, and significantly, for teachers as well, complicating normative assumptions that socialization ever proceeds ‘straightforwardly’ or ‘inevitably’, particularly along lines of hierarchical status, institutional role, or age. That notwithstanding, it should be noted that the Local ESL students’ productions of ESL were manifestly, relationally tied to the ‘official’ ones. Thus, there was already multidirectionality in socialization: Local ESL students may have opposed the school-sanctioned productions of ESL, and affected classroom processes and curriculum, but they were still ascribed the institutional identity of ‘ESL student’; there may have been a slowing down in the ESL curriculum, but there was still an ESL curriculum; classwork may not have been completed or turned in, but as a consequence, students received poor grades, earned the label of low-achieving, and remained ‘stuck’ in ESL. In this respect, it is important to note that although contingency and multidirectionality in socialization are processes that need to be foregrounded in any LS analysis, that ‘resistance’, for example, to hegemonic socializing forces (e.g. into an infantilized ESL student identity) may not necessarily be transformative, but culturally and socially reproductive (Willis 1977, 1981; Giddens 1979; see Talmy forthcoming-a, forthcoming-b). Indeed, as students and teachers in the Tradewinds ESL-A classes subtly yet progressively disengaged from doing much else beyond decontextualized grammar and vocabulary worksheets, the ESL program became precisely what the Local ESL students claimed throughout this study to dislike about it: an easy, academically inconsequential program that did little to meet their L2 learning or educational needs.
CONCLUSIONS LS research has considered how experts or oldtimers socialize novices to become ‘competent’ members of social groups, ‘by exposure to and participation in language-mediated interactions’ (Ochs 1986: 2). With the explicit focus on socialization ‘through language and to use language’, the originators of this research tradition were careful to cast it as a contingent, bi- or multidirectional process. However, these themes may have been obscured in earlier studies, concerned as they have often been with the ways
640
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
that caregivers socialize young children in and through an L1 into particular, culturally-specific ‘world views’. This paper has returned to the clearly asserted, but less often demonstrated commitments of early LS scholarship to foreground the contested, unpredictable, and reciprocal character of LS. I have done so by considering the ostensible socialization of older ‘learners’, who were generally uninvested, unwilling incumbents of a stigmatized identity category, studying an L2 in a multilingual, compulsory educational context. Although these themes were prominently featured above, given the focus of this study, they are inherent in LS, due to its orientation toward socialization as an interactionallymediated achievement. To return to a point made earlier, the call for a more ‘dynamic model’ of LS, while initially compelling, is in fact unwarranted; what is needed is analytic attention to the essential unpredictability, contestedness, and fluidity of socialization, as it is or is not achieved, in ways anticipated or not, in L1 and in L2, among younger and older ‘novices’ and ‘experts’, at earlier and later stages of the lifespan, across a range of monolingual, multilingual, ‘naturalistic’, and institutional contexts.
SUPPLEMENTARY DATA An appendix can be found at Applied Linguistics online
ACKNOWLEDGEMENTS Early versions of this paper were presented at AAAL/ACLA-CAAL 2006 in Montreal, Canada, and at NCTE 2007 in New York, where it received a 2007 NCTE Promising Research Award. Research was funded by grants from The Spencer Foundation and The International Research Foundation. I am grateful to both organizations for their support. I would also like to thank the participants of the 2006 AAAL/ACLA-CAAL colloquium, the three anonymous Applied Linguistics reviewers, and Gabi Kasper for insightful comments on previous drafts. All remaining errors are my own.
NOTES 1 The names of the school and the people who participated in this study have been changed. Please note that students provided their own pseudonyms, unless denoted at first mention by an asterisk (*). 2 Several first-rate LS reviews have appeared recently: Garret and Baquedano-Lo´pez (2002); Duff (2003); Watson-Gegeo and Nielsen (2003); Zuengler and Cole (2005). A major encyclopedic review has also appeared
(Duff and Hornberger 2008). To conserve space, readers are directed to these sources for overviews of LS research. 3 Although the teachers were in their first year at Tradewinds, all had previous teaching experience, though not in a public high school. 4 In order to preserve confidentiality in the small educational community in Hawai’i, I have omitted or altered many details about teacher participants,
STEVEN TALMY
5
6
7
8
9
including race/ethnicity, gender, and L1 background. I recognize that this limits analysis; though regrettable, I have accepted these limits as necessary. I observed ESL-A (1X) in the piloting year of research when I did no audiorecording. Thus, the data below come from ESL-A (2W) and (2X). However, my findings from all three classes were remarkably consistent. See Ohnuma (2002) for discussion of the ostensibly oxymoronic category of ‘Local haole’. The population that ‘Local ESL’ refers to has elsewhere been called ‘generation 1.5’ (see Harklau et al. 1999). However, 40% of the ESL students who participated in the Tradewinds study had lived in the US for over 2 years, 20% for 4–9 years, with several born in Hawai’i. This circumstance finds parallels across North America, where it has been estimated that in the US alone, 46% of the nearly four million school-age ‘limited English proficient’ students in 2000–01 had been born in the US, with an additional 15% having lived in the US for 5 years or more (Zehler et al. 2003). Although this study focuses on a Local ESL CoP, it was but one CoP in the Tradewinds ESL classes; it was not the
10
11
12
13
641
only one, nor was it always evident. Neither did students I identified as Local ESL always participate in it, nor was it their only CoP membership. Some might argue that these students display lack of investment to school in general rather than to ESL. However, Local ESL students consistently indicated (to teachers, classmates, and me) that the putative ‘problem’ with school was ESL; in fact, the public display of such a perspective was itself a constitutive practice of Local ESL CoP membership (see Talmy 2005, ch. 8; forthcoming-a). Furthermore, ‘school’ for most of these students predominantly was ESL, both in terms of their dedicated-ESL and shelteredcontent ESL class placements (see Talmy 2005, ch. 7; cf. Nespor 1990). Playing cards, especially Texas Hold ’Em, was a practice undertaken daily in the school by ‘mainstream’ students. Those in the ‘non-Local ESL’ group were Local ESL students’ newcomer and lower-L2-English-proficient classmates. My gratitude to Carsten Roever for his assistance with the statistical analysis.
REFERENCES Ahearn, L. M. 2001. ‘Language and agency,’ Annual Review of Anthropology 30: 109–37. Bayley, R. and S. R. Schecter. 2003. ‘Introduction: Toward a dynamic model of language socialization’ in R. Bayley and S. R. Schecter (eds): Language Socialization in Bilingual and Multilingual Societies. Clevedon, UK: Multilingual Matters, pp. 1–6. Blumer, H. 1969. Symbolic Interactionism: Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall. Bullough, R. V., J. Knowles, and N. Crow. 1992. Emerging as a Teacher. London: Routledge. Cameron, D. 2001. Working with Spoken Discourse. London: Sage.
Cole, K. and J. Zuengler. 2003. ‘Engaging in an authentic science project: Appropriating, resisting, and denying ‘‘scientific’’ identities’ in R. Bayley and S. R. Schecter (eds): Language Socialization in Bilingual and Multilingual Societies. Clevedon, UK: Multilingual Matters, pp. 98–113. Condon, S. L. 2001. ‘Discourse ok revisited: Default organization in verbal interaction,’ Journal of Pragmatics 33: 491–513. Creech, S. 2000. The Wanderer. New York: HarperCollins. Curtis, C. P. 1999. Bud, not Buddy. New York: Yearling.
642
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
Dahl, R. 1961. James and the Giant Peach. New York: Knopf. Davidson, J. 1984. ‘Subsequent versions of invitations, offers, requests, and proposals dealing with potential or actual rejection’ in J. M. Atkinson and J. Heritage (eds): Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press, pp. 102–28. Dixon, R. J. 1956. Regents English Workbook, Book 2: Intermediate–Advanced. New York: Regents Publishing Company. Duff, P. A. 1995. ‘An ethnography of communication in immersion classrooms in Hungary,’ TESOL Quarterly 29: 505–37. Duff, P. A. 2002. ‘The discursive co-construction of knowledge, identity, and difference: An ethnography of communication in the high school mainstream,’ Applied Linguistics 23: 289–322. Duff, P. A. 2003. ‘New directions in second language socialization research,’ Korean Journal of English Language and Linguistics 3: 309–339. Duff, P. A. and N. H. Hornberger (eds). 2008. Encyclopedia of Language and Education, Vol. 8: Language socialization, 2nd edn. Boston: Springer. Durkheim, E. 1997. The Division of Labor in Society, trans. W. D. Halls. New York: The Free Press (Original work published 1893). Eisenhart, M. 1996. ‘The production of biologists at school and work: Making scientists, conservationists, or flowery bone-heads?’ in B. A. Levinson, D. E. Foley, and D. C. Holland (eds): The Cultural Production of the Educated Person: Critical Ethnographies of Schooling and Local Practice. Albany, NY: SUNY Press, pp. 169–85. Ford, C. E. 2001. ‘At the intersection of turn and sequence: Negation and what comes next’ in M. Selting and E. Couper-Kuhlen (eds): Studies in Interactional Linguistics. Amsterdam: John Benjamins, pp. 51–79. Freud, S. 1939. Moses and Monotheism. New York: Vintage. Garfinkel, H. 1967. Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall. Garrett, P. 2004. ‘Review of Bayley & Schecter, 2003,’ Language in Society 33: 776–9. Garrett, P. and P. Baquedano-Lo´pez. 2002. ‘Language socialization: Reproduction and continuity, transformation and change,’ Annual Review of Anthropology 31: 339–61. Giddens, A. 1979. Central Problems in Social Theory: Action, Structure, and Contradiction in Social Analysis. Berkeley, CA: University of California Press.
Gumperz, J. J. 1982. Discourse Strategies. New York: Cambridge University Press. Harklau, L., K. M. Losey, and M. Siegal (eds). 1999. Generation 1.5 Meets College Composition: Issues in the teaching of writing to US-educated learners of ESL. Mahwah, NJ: Lawrence Erlbaum Associates. He, A. W. 2003. ‘Novices and their speech roles in Chinese heritage language classes’ in R. Bayley and S. R. Schecter (eds): Language Socialization in Bilingual and Multilingual Societies. Clevedon, UK: Multilingual Matters, pp. 128–46. Jacoby, S. and P. Gonzales. 1991. ‘The constitution of expert–novice in scientific discourse,’ Issues in Applied Linguistics 2/2: 149–81. Kelchtermans, G. and K. Ballet. 2002. ‘The micropolitics of teacher induction: A narrativebiographical study on teacher socialisation,’ Teaching and Teacher Education 18: 105–20. Koshik, I. 2002. ‘A conversation analytic study of yes/no questions which convey reversed polarity assertions,’ Journal of Pragmatics 34: 1851–77. Kulick, D. and B. B. Schieffelin. 2004. ‘Language socialization’ in A. Duranti (ed.): A Companion to Linguistic Anthropology. Malden, MA: Blackwell, pp. 349–68. Kuzmic, J. 1994. ‘A beginning teacher’s search for meaning: Teacher socialisation, organisational literacy, and empowerment,’ Teaching and Teacher Education 10: 15–27. Labrador, R. N. 2004. ‘Speech play, Hawai’i ethnic humor, and the myth of multiculturalism,’ Pragmatics 14/2–3: 291–316. Lave, J. 1991. ‘Situated learning in communities of practice’ in L. B. Resnick, J. M. Levine, and S. D. Teasley (eds): Perspectives on Socially Shared Cognition. Washington, DC: American Psychological Association, pp. 63–82. Lave, J. and E. Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Levinson, B. A., D. E. Foley, and D. C. Holland (eds). 1996. The Cultural Production of the Educated Person: Critical Ethnographies of Schooling and Local Practice. Albany, NY: SUNY Press. Mead, M. 1961. Coming of Age in Samoa: A Psychological Study of Primitive Youth for Western Civilization. New York: Morrow (Original work published 1928). Mehan, H. 1979. Learning Lessons. Cambridge, MA: Harvard University Press. Naylor, P. R. 1991. Shiloh. New York: Atheneum.
STEVEN TALMY
Nespor, J. 1990. ‘Curriculum and conversions of capital in the acquisition of disciplinary knowledge,’ Journal of Curriculum Studies 22/3: 217–32. Norton, B. 2000. Identity and Language Learning: Gender, Ethnicity, and Educational Change. London: Longman Ochs, E. 1986. ‘Introduction’ in B. B. Schieffelin and E. Ochs (eds): Language Socialization Across Cultures. Cambridge: Cambridge University Press, pp. 1–13. Ochs, E. 1999. ‘Socialization,’ Journal of Linguistic Anthropology 9/1–2: 230–233. Ochs, E. and B. B. Schieffelin. 2008. ‘Language socialization: An historical overview’ in P. A. Duff and N. H. Hornberger (eds): Encyclopedia of Language and Education, Vol. 8: Language Socialization, 2nd edn. Boston: Springer, pp. 3–15. O’Dell, S. 1960. Island of the Blue Dolphins. Boston: Houghton Mifflin. Ohnuma, K. 2002. ‘Local haole—A contradiction in terms? The dilemma of being white, born and raised in Hawai’i,’ Cultural Values 6/4: 273–85. Okamura, J. Y. 1994. ‘Why there are no Asian Americans in Hawai’i: The continuing significance of local identity,’ Social Process in Hawaii 35: 161–78. Parsons, T. 1937. The Structure of Social Actions. New York: The Free Press. Paulsen, G. 1985. Dogsong. New York: Bradbury Press. Phillipson, R. 1988. ‘Linguicism: Structures and ideologies in linguistic imperialism’ in T. Skutnabb-Kangas and J. Cummins (eds): Minority Education: From Shame to Struggle. Philadelphia: Multilingual Matters, pp. 339–58. Pomerantz, A. 1984. ‘Pursuing a response’ in J. M. Atkinson and J. Heritage (eds): Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press, pp. 152–63. Sacks, H. 1992. Lectures on Conversation, Vols I and 2. Oxford: Blackwell. Sakoda, K. and J. Siegel. 2003. Pidgin Grammar: An Introduction to the Creole Language of Hawai’i. Honolulu: Bess Press. Schecter, S. R. and R. Bayley. 2004. ‘Language socialization in theory and practice,’ International Journal of Qualitative Studies in Education 17/5: 605–25. Schieffelin, B. B. and E. Ochs. 1986a. ‘Language socialization,’ Annual Review of Anthropology 15: 163–91.
643
Schieffelin, B. B. and E. Ochs. 1986b. Language Socialization across Cultures. Cambridge: Cambridge University Press. Schiffrin, D. 1987. Discourse Markers. Cambridge: Cambridge University Press. Schutz, A. 1967. The Phenomenology of the Social World, trans. G. Walsh and F. Lehnert. Evanston, IL: Northwestern University Press. Siegal, M. 1996. ‘The role of learner subjectivity in second language sociolinguistic competency: Western women learning Japanese,’ Applied Linguistics 17: 356–82. Talmy, S. 2005. ‘Lifers and FOBs, rocks and resistance: Generation 1.5, identity, and the cultural productions of ESL in a high school’, unpublished doctoral dissertation, University of Hawai’i at Ma¯noa, Honolulu, HI. Talmy, S. 2007. ‘Reproduction and recursivity in high school ESL: A critical analysis of classroom language-in-use’. Plenary address to the 17th International Conference on Pragmatics and Language Learning, Honolulu, HI. Talmy, S. forthcoming-a. ‘Forever FOB?: Resisting and reproducing the Other in high school ESL’ in A. Reyes and A. Lo (eds): Beyond Yellow English: Toward a Linguistic Anthropology of Asian Pacific America. Oxford: Oxford University Press. Talmy, S. forthcoming-b. ‘Resisting ESL: Categories and sequence in a critically ‘‘motivated’’ analysis of classroom interaction’ in G. Kasper and H. T. Nguyen (eds): Talk-in-interaction: Multilingual Perspectives. Honolulu, HI: University of Hawai’i, National Foreign Language Resource Center. ten Have, P. 2001. ‘Applied conversation analysis’ in A. McHoul, and M. Rapley (eds): How to Analyze Talk in Institutional Settings: A Casebook of Methods. London: Continuum, pp. 3–11. Watson-Gegeo, K. A. and S. Nielsen. 2003. ‘Language socialization in SLA’ in C. Doughty and M. H. Long (eds): Handbook of Second Language Acquisition. Oxford: Blackwell, pp. 155–77. Wenger, E. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press. Wentworth, W. M. 1980. Context and Understanding: An Inquiry into Socialization Theory. New York: Elsevier. White, G. 1977. Socialisation. London: Longman. Willis, P. 1977. Learning to Labour: How Working Class Kids get Working Class Jobs. New York: Columbia University Press.
644
THE CULTURAL PRODUCTIONS OF THE ESL STUDENT AT TRADEWINDS HIGH
Willis, P. 1981. ‘Cultural production is different from cultural reproduction is different from social reproduction is different from reproduction,’ Interchange 12/2–3: 48–67. Wrong, D. H. 1961. ‘The oversocialized conception of man in modern sociology,’ American Sociological Review 26/2: 183–93. Zehler, A. M., H. L. Fleischman, P. J. Hopstock, T. G. Stephenson, M. L. Pendzick, and S. Sapru. 2003. Descriptive Study of Services to LEP Students and LEP Students with Disabilities, Vol I. Arlington, VA: Development Associates/ U.S. Department of Education.
Zeichner, K. and J. Gore. 1990. ‘Teacher socialisation’ in R. Houston (ed.): Handbook of Research on Teacher Education. New York: Macmillan, pp. 329–48. Zimmerman, D. 1998. ‘Identity, context, and interaction’ in C. Antaki and S. Widdicombe (eds): Identities in Talk. London: Sage, pp. 87–106. Zuengler, J. and K. Cole. 2005. ‘Language socialization and L2 learning’ in E. Hinkel (ed.): Handbook of Research in Second Language Teaching and Learning. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 301–16.
Applied Linguistics 29/4: 645–671 ß Oxford University Press 2008 doi:10.1093/applin/amn022 Advance Access published on 16 June 2008
Language Ecology in Multilingual Settings. Towards a Theory of Symbolic Competence CLAIRE KRAMSCH1 and ANNE WHITESIDE2 1
UC Berkeley and 2City College San Francisco This paper draws on complexity theory and post-modern sociolinguistics to explore how an ecological approach to language data can illuminate aspects of language use in multilingual environments. We first examine transcripts of exchanges taking place among multilingual individuals in multicultural settings. We briefly review what conversation and discourse analysis can explain about these exchanges. We then build on these analyses, using insights from complexity theory and interactional sociolinguistics. We finally outline the components of a competence in multilingual encounters that has not been sufficiently taken into consideration by applied linguists and that we call ‘symbolic competence’.
1. INTRODUCTION When Diane Larsen-Freeman gave her groundbreaking talk on chaos theory and SLA at the Second Language Research Forum (Larsen-Freeman 1997), few language teachers imagined how chaos/complexity science could possibly be relevant to their daily task of having to teach grammatical forms and functions, communicative strategies and cultural knowledge in language classrooms. Yet complexity theory was soon picked up by American and Dutch educators interested in language ecology (van Lier 2000, 2004; Kramsch 2002; Leather and van Dam 2002); it was connected with the work that had long been going on in Europe in ecolinguistics (Haugen 1972; Fill and Mu¨hlha¨usler 2001; Steffensen 2007), and was brought to bear on the way the teaching of foreign languages and cultures was being conceptualized (Larsen-Freeman 2003; van Lier 2004; Risager 2006; Kramsch and Steffensen 2007). Ten years later, the increasingly multilingual and multicultural nature of global exchanges is raising questions about the traditionally monolingual and monocultural nature of language education, and its modernist orientation. The prototypical communicative exchange, used by researchers to explore the processes of second language acquisition, and by teachers (and textbooks) to teach communicative or intercultural competence in a foreign language, usually includes two or three interlocutors, who all conduct the interaction in the same standard target language, all agree on what the purpose of
646
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
the exchange is and what constitutes a culturally appropriate topic of conversation, all have equal speaking rights and opportunities. But, as recent work in pragmatics and sociolinguistics has shown (e.g. Rampton 1995; Johnstone 1996; Blommaert 2005; Coupland 2007), in multidialectal and multilingual settings the reality can be quite different, especially in our late modern times. In the many places around the world where multiple languages are used to conduct the business of everyday life, language users have to navigate much less predictable exchanges in which the interlocutors use a variety of different languages and dialects for various identification purposes, and exercise symbolic power in various ways to get heard and respected (Rampton 1998, 1999). They have to mediate complex encounters among interlocutors with different language capacities and cultural imaginations, who have different social and political memories, and who don’t necessarily share a common understanding of the social reality they are living in (Blommaert 2005). This presents a double challenge. For researchers, the lack of a shared understanding, due to global migrations and deterritorialized living conditions in late modern societies, poses a problem because much of applied linguistic data only make sense on the basis of a shared understanding of reality between the researcher and a given speech community. For language teachers, it complicates the teaching of what has been traditionally called ‘communicative competence’. For, in such environments, as we shall see below, successful communication comes less from knowing which communication strategy to pull off at which point in the interaction than it does from choosing which speech style to speak with whom, about what, and for what effect. This paper draws on insights from complexity theory and post-modern sociolinguistics to explore how an ecological approach to language data can illuminate aspects of language use in multilingual settings. We first present transcripts of exchanges taking place among multilingual individuals. We then examine what various contextually-oriented approaches to discourse can reveal about these multilingual interactions. We extend these analyses by drawing on insights from complexity theory and recent work in interactional sociolinguistics. We finally outline the components of a language competence in multilingual encounters that has not been sufficiently taken into consideration by applied linguists and that we call ‘symbolic competence’.
2. AN EXAMPLE OF LANGUAGE ECOLOGY IN PRACTICE We first turn to data collected by Anne Whiteside (AW) as part of her research on Maya-speaking immigrants from Yucatan, Mexico, now living in San Francisco, California (Whiteside 2006). Attempting to understand their patterns of language use and the reasoning behind them, Whiteside spent over two years working closely with four focal Yucatecans, following them in
CLAIRE KRAMSCH and ANNE WHITESIDE
647
their daily lives, helping to organize community events, and exchanging English, Spanish, and computer literacy lessons for lessons in Maya. There are now an estimated 25,000 Yucatecans living in the greater San Francisco Bay area, and some 50–80,000 in California, many of whom left Yucatan over the last decade (INDEMAYA 2005). Like an increasing number of migrants crossing the Mexico/California border, many arrive without legal papers (Passell et al. 2004; Passell 2005), lured by service sector jobs that have replaced entry-level manufacturing jobs in California’s post-industrial economy. Their situation is typical of workers in a global economy that knows no national borders, no standard national languages, and thrives on the informal economic and social margins of national institutions. Whiteside found that since many work two and three jobs, and with longterm residence uncertain because of undocumented status, learning English often takes a back seat. Her informants worked in restaurants where as many as eight languages were routinely spoken, with English, if spoken, as the highly accented lingua franca. Spanish use was common, linking Yucatecans with other marginalized Spanish-speaking workers and allowing undocumented individuals to blend with Latino legal residents and citizens. Yet informants also complained of discriminatory treatment by speakers of other varieties of Spanish, and noted a tendency of fellow immigrants to disguise their Yucatecan accents. English provided an escape from such distinctions. And English is seen as portable capital, motivating some to learn it to teach future migrants back in Yucatan. By contrast, Maya can be a social liability, and speakers described a sense of ‘shame’ speaking Maya in public, inhibited by racialized colonial discourse and stereotypes linking Maya with poverty and ignorance (Gu¨emez Pineda 2006). Maya was used predominantly at home and among work teams, where it provided a safe code in which to vent about oppressive conditions. Whiteside collected data between January 2004 and June 2005, using participant observation, videotapes, interviews, and a language and literacy survey of 170 Yucatec Maya adults.1 The data presented below are taken from 12 multilingual conversations she recorded in stores located in a predominantly Spanish-speaking neighborhood. The speakers in these conversations are Yucatecans Bela Chan and Don Francisco Canche (DF) (pseudonyms), some local merchants, and the researcher. These conversations were transcribed and analyzed using conversation analysis, looking in particular at preferred and dispreferred responses, repairs, evaluations, alignments, and indexicalities. It was in the course of these analyses that the need for a more ecological approach to the data emerged, which might link the microanalysis of the conversational data to the broader ecological context. In the first set of data we discuss here, Don Francisco, 49, who runs an informal restaurant out of his apartment, is taking the researcher through his neighborhood as he shops for food. He has agreed to help her research project since she has been teaching him to read in Spanish, which he never learned. As DF chaperones AW around, he is regularly interrupted by
648
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
greetings in Maya and Spanish from fellow townspeople, now San Francisco neighbors, who know his status as a successful farmer in Yucatan. To local merchants DF is a preferred customer, one who makes frequent trips to supply his busy restaurant and who spends a lot of money. On this occasion he is stopping in to check out supplies and place orders, eager to show the researcher his routines and to demonstrate the Maya the merchants are learning. We next present the data with brief descriptions, followed by more detailed analyses in sections 3 and 4.
At the Vietnamese grocery The first excerpt occurs in a grocery story with Vietnamese writing on its awning. The Vietnamese owner, whom DF introduces as Juan, has been speaking to DF in English, who answers him in Spanish. Juan is busy loading meat from the freezer into the display case, and this exchange comes at the end of a short conversation about the meat DF needs.
Excerpt 1 1 2
Juan: DF:
3 4 5 6 7 8 9 10 11 12
Juan: DF: Juan: DF: Juan: DF: Juan:
DF:
how much panza you want? voy a comprar cinco libras de panza man˜ana OK man˜ana /\ma’ alob. _/OK! \/Dios bo dik _/bo dik _/saama @@, @@ _saama ah
(tripe) I’m going to buy 5 lb of tripe tomorrow. good thanks tomorrow
‘Juan’, who has adopted a Spanish name for his Spanish-speaking customers, and has demonstrated his understanding of DF’s Spanish earlier in the conversation, uses mixed utterances in #1 and #3. DF closes in Maya (#4) to which Juan answers in English, but DF persists in Maya (#6). Juan echoes him (#7), but his laugh in #9,10 marks an affective change. Juan’s final ‘saama’ (#11) indexes a willingness to let the customer have his way. We give a more extensive analysis of this excerpt in section 4.2.
At the Chinese grocery The next five excerpts take place in a Chinese-run grocery store, where DF has stopped to find out how much masa (corn-flour dough) his son had
CLAIRE KRAMSCH and ANNE WHITESIDE
649
picked up earlier in the day. DF, who is four foot ten, makes an odd pair with AW, Anglo-American and five foot nine.
Excerpt 2 1 2 3 4 5 6 7 8 9 10 11 12 13
DF: Butcher: DF: AW: Clerk: DF:
22 Clerk 23
((TO BUTCHER IN MAYA)) si si si ((TO CLERK)) buenas . . . vengo 5?4 mi maestra 5LO HI LO4 teacher OH [@] [ah] es mi maestra\ ah eh-noma´s, este, pase´ a preguntar/ la masa que agarro´ mi hijo\ ochenta y ahora/ ... si bien. le toco masa aca´ ahora
I’m with my teacher
She’s my teacher I just uh came to ask the masa that my son took, 80 and now yes good. he’ll take masa here now
DF first speaks with the butcher, a fellow Yucatecan, in Maya, then turns to the clerk in Spanish, introducing AW as ‘mi maestra’. He then asks about the order in #11–13. The clerk answers him in broken Spanish (Si bien. le toco masa aca´ ahora) which can be glossed as ‘OK good, he’ll come and get the masa here now’. In Excerpt 3, DF explains why he sent his son instead of coming himself to pick up the order.
Excerpt 3 31 32 33 34 35 36
DF: Clerk: DF:
37 38 39 40 41
AW:
Clerk:
DF:
estamos de paseo con la¼ maestra por eso yo no /\vine ah/ si\ ah ((to AW)) my Spanish is really limited but I try to understand him @[@@@@] [that’s good] [si ah ha]
we’re out walking with the teacher that’s why I didn’t come
650
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
In #36, the clerk addresses AW in English, aligning herself with the English speaker, now referring to DF in the third person, and laughs. AW evaluates this positively, as does DF. In the intervening lines, the clerk tries to explain to AW her routine with DF, but her English is not much clearer than her Spanish: ‘the masa I always send it there, he always pick it up there already’. In Excerpt 4, DF turns to an older woman—possibly the clerk’s relative— who is sorting beans.
Excerpt 4 74 75 76 77
DF: Older lady: ah @@@ DF: Clerk:
78 79 80
Older lady: AW: Clerk:
81 82 83 84 85
Older lady: Clerk DF:
86 87 88
Clerk:
DF: AW:
89 90 91 92 93 94
Clerk: AW: Clerk:
mucho trabajo.
a lot of work.
eso es el ticher. ((TO OLDER LADY: IN CHINESE)) hi @@@ hi ((TO DF)) man˜ana when you come [I give you no espan˜ol].. [@@@@@] [[solo English]] [[@@@]] NO,5@ no@4. Jose, tomorrow when you come in I don’t speak Spanish with you any more. [@@@@] [no, no, I’m]not teaching him English. I’m teaching him to read and writing in Spanish. I’m not teaching him English. oh, oh, read and write Span-[ish.] [yah,] read and write Span-ish that’s good ‘cause he like he not even recognize the numbers
this is the teacher.
The older woman laughs in response to DF’s Spanish, but he persists, introducing AW as ‘el ticher’, a mixed utterance that the clerk translates into Chinese for her. The Clerk then addresses DF as ‘Jose’, threatening to use only English with him. DF responds with a laugh (#87). In #88, AW jumps in, correcting what she perceives as an erroneous assumption on the part of the clerk that she is teaching DF English, which the clerk is enforcing by
CLAIRE KRAMSCH and ANNE WHITESIDE
651
addressing him in English. Her statement suggests that DF, who spends most of his time in his Spanish- and Maya-dominant restaurant and apartment building, has made learning Spanish literacy, not English, his priority. AW’s emphatic ‘no no’ can also be seen as an attempt to save DF’s face, which has been threatened by the clerk’s scolding tone, and by her use of a stereotypical name not DF’s own. AW’s repetition in #90 can be interpreted as ‘teacher talk’ to the NNS clerk, and her adoption of the prosody of the clerk’s English in #93 as ‘foreigner talk’. The clerk realigns herself with the revised teaching agenda in #91, noting that DF cannot read numbers. After a few other remarks the conversation returns to this topic. We shall return to this excerpt at greater length in section 4.2.
Excerpt 5 107
AW:
108
109
110 111
DF: Clerk:
112 113 114 115 116 117 118 119
AW:
Clerk: AW: Clerk:
120 121 122 123 124 125 126 127
AW:
we’re going to learn to read the numbers ((TO DF)) dice que vamos a aprender a leer los numeros para que [puedas . . .]
[hm] [that’s the] most important part first: one, two three four five six seven eight nine ten. that’s right yah where did YOU learn English? America [oh¼] [many] years ago 5HI you know I start from beginning I start from one, two three four five. HI4 I never know it in my life because my mother come when I come in 19 uh 80 I still went to ESL program I still learn that’s why he [can too] yah] yah yah
she says we’re going to learn numbers so that you can . . .
652
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
128 129 130 131 132 133
Clerk:
134
DF:
DF: Clerk:
((to DF)) when me aquı´ twenty years early nada speak English nada nada todo English aquı´
when I (was) here twenty years ago I spoke no English none none everything/all English here
ah
In #108, AW translates for DF, acknowledging that DF may not understand her exchange with the clerk, but the clerk persists in English, taking an authoritative stance, ‘that’s the most important part first’. As the maestra and English expert, AW then asserts her own authority to approve of this priority, but then turns the topic (#115) to the clerk’s status as English learner. The clerk takes this opportunity to launch into a ‘can-do’ story that draws on the immigrant frame ‘pulling-yourself-up-by-your-own-bootstraps’. Her story ends with a moral (‘that’s why he can too’) that draws a parallel between her and DF. However, by referring to him in the third person in his presence and talking ‘over’ him in a code he doesn’t understand, the clerk distances herself from DF. Considering that he is her preferred customer, this move can be taken as an affront. The clerk then turns to address DF in #128 in a mixture of broken English and broken Spanish or ‘foreigner English’ that further positions her in the English speaking camp (‘todo English aquı´’).
Excerpt 6 135 136 137 138 139 140 141 142
Clerk:
143 144 145 146 147 148 149 150
Clerk: AW: Clerk:
DF:
AW:
DF: Clerk: DF:
learn first ABCD todo aquı´ ah, entiendes Maya, an˜o ma´s ah entiendes Maya a lot of people speak Maya here, huh? yeah you’re learning some Maya? uh:: not much Latinos is 5LO??LO4 ahı´ esta? eh¼ en la tarde/[y¼/] [OK] bueno, (?)
everything here ah, you understand Maya (one) more year
that’s it? In the afternoon and Good.
CLAIRE KRAMSCH and ANNE WHITESIDE
151 152 153
Clerk:
nos vemos OK good to see you.
653
See you.
In #138, DF counters her suggestion that he ‘learn first ABCD’ by predicting that a year from now the clerk will be speaking Maya. Given the recent influx of Maya-speaking clientele in the neighborhood, DF suggests that her customers, not the all-English melting pot, will prevail through their buying power. We return to this interaction in section 4.2.
At the Vietnamese butcher In the next two excerpts, AW and Bela Chan, both women in their 50s, are out pricing meat in anticipation of an upcoming fiesta, for a typical Yucatecan pork dish, cochinita pibil. Raised in a Maya-speaking family, Bela stopped using Maya after being ridiculed for being ‘country’, code for backward/Indian, and has trouble speaking it. With a Maya and a Chinese grandmother, an Afro Honduran and a Spanish grandfather, all of whom spoke Maya, Bela has roots in four continents. After 15 years in San Francisco, Bela has made little headway in English, which she uses neither in the Spanish-dominant neighborhood where she lives nor at the Spanish Baptist church she attends. In the Yucatan, she held a managerial position in a Korean-run garment factory or maquiladora; by contrast, she has worked ‘on her knees’ in the USA, a consequence she attributes to her undocumented status. This butcher shop is advertised with a sign in Vietnamese; its customers speak English, Spanish, and several Asian languages, presumably including Vietnamese. The two Asian-looking butchers, whom we call Butcher A and Butcher B, both use some Spanish, although Butcher B is more proficient. The butchers stand behind a tall rectangular case full of meat, making fairly transparent what’s going on. Bela, who is under five feet tall, is hard to see from that height; AW, at five foot nine, is closer to them. Yet Bela doesn’t strain to make herself heard or understood: the implication is that if the butchers want a sale they will do what it takes to understand her. The whole interaction, 51 turns of talk, involves 27 instances of code-switching, during which everyone speaks everyone else’s language, with the exception of sideplay between butchers in their L1. Prior to the following exchange, Butcher B tells them the price in Spanish ($1.97 a lb) and they negotiate the amount needed, with Butcher A tactfully repairing Bela’s ambiguous ‘fifti pound’ (‘One five or five-0 you want?’). Having established the quantity, they negotiate the price:
Excerpt 7 60
Butcher A:
5you need two leg . . . 20 pound each4
654
61 62 63 64 65 66 67 68 69
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
AW:
Bela: Butcher A: Bela: Butcher A:
70 71
Butcher B:
72
AW:
73 74 75 76 77 78 79 80 81 82 83 84 85
86 87 88
Butcher B: AW: Bela: AW: Bela: Butcher A: Bela: Butcher A:
AW:
Butcher A: Bela: Butcher B:
so I order two. ((TO BELA)) 5HI pregu´ntele si es ma´s barato HI4 ¿si? oh ¿habla espan˜ol? es ma´s barato uh huh oh ((SPEAKS TO OTHER BUTCHER IN WHAT SEEMS TO BE CHINESE)). 5he said OK4 uno cincuenta y nueve la libra uno cincuenta y nueve si compramos ma´s? . . . (3) if we buy more? ok. (??) ¿quiere? ah.. [vamos a] [next week] [[vamos a..]] [[next week]] [ok] [next week] ¿cua´ndo quiere? next week ¿cua´ndo? ¿que dı´a? ah, todavı´a estamos pensando 5LO verdad Lo4 thinking, right? [ok] [ya] si bueno
ask if its cheaper yes? you speak Spanish? it’s cheaper
$1.59 a pound
you want it? we will . . . we will . . .
when when? what day? we’re still
do you want?
OK good
AW’s side play to Bela in Spanish in #63, to which the butchers respond with side play in Chinese (#69), is a common haggling strategy, the team huddle before the play. Butcher B comes back in #71 with a reduced price ($1.59) after which he asks for a commitment. This is followed by hedging in Spanish and English, at which point Butcher A in #82 switches to Spanish, in which
CLAIRE KRAMSCH and ANNE WHITESIDE
655
he is far from fluent, aligning himself with AW and Bela’s sideplay in Spanish. As they close the encounter in the next excerpt, Butcher A presents them with a Spanish name.
Excerpt 8 95 96 97 98 99
Bela: Butcher B: Bela: Butcher A:
What’s your name? Felipe Felipe (.) /\OK Felipe Felipe sabe espan˜ol muy bien
100
Butcher B:
this one with my name
Felipe knows good Spanish
((LEANING OVER THE COUNTER TO GIVE BELA A CARD)) 101 102
Bela: AW:
@@[@@] [thank you]
Bela responds to Butcher B’s introduction (#97) by repeating his name. The rising–falling tone in her next turn (#98) indexes surprise/acceptance at the butcher’s Spanish name, which she follows in #101 with a laughter of recognition. Bela is no stranger to Asians who adapt themselves to new linguistic and cultural contexts: her Chinese-born grandmother spoke Maya, and the Korean factory owners she worked for in Yucatan spoke Spanish and English. In an interview with AW, Bela expressed impatience with fellow townspeople in the USA who pretend to be from other parts of Mexico. She is clearly aware of how this game of masks is played to strategic advantage in such multicultural contexts.
3. CONTEXTUALLY ORIENTED DATA ANALYSES How can we analyze these data? Several approaches in applied linguistics take a situated, contextualized view of language use in social settings, in particular conversational analysis and mediated communication studies. We consider each in turn. As we shall argue, what becomes salient in these interactions is how problematic the traditional notions of context have become in a global environment.
3.1 Conversation analysis First we turn to conversation analysis (CA). Both in its ‘pure’ and in its ‘applied’ form,2 the branch of the sociology of language called conversation analysis provides an epistemological way of looking at what the participants in these encounters are doing. ‘Pure’ CA, represented by such foundational work as Sacks et al. (1974), Sacks (1992), Schegloff (2007), and inspired by
656
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
ethnomethodology (Garfinkel 1967), revolutionized the study of talk in interaction by rigorously confining its analysis to observable phenomena and the organization of sequences of turns-at-talk. Its strictly emic perspective has encouraged researchers to do close readings of conversational phenomena. It does not ask: what did the participants have in mind, or what larger forces prompted them to say what they said? but, rather: what do speakers orient to in their turn-by-turn contributions to the ongoing exchange? what normative expectations and assumptions inform and underpin the production of their conversational sequences (Schegloff 2007), their membership categorization devices (Sacks 1992)? In Excerpt 1, for example, by showing the unfolding of an unproblematic closing routine consisting of three adjacency pairs: good/OK, thanks/ thanks, tomorrow/tomorrow, CA can show evidence of the familiar accomplishment of the expected leave-taking between merchant and customer. In Excerpt 8, the identification routine in which Bela’s turn ‘What’s your name?’ elicits Butcher A’s turn ‘Felipe’ is expected to set up an environment that will facilitate future transactions. Bela’s repetition of the name Felipe momentarily flouts that expectation, as it offers a dispreferred response to the previous Q/A pair: ‘What’s your name?’—‘Felipe’. Of course, Bela’s surprise only makes sense to us because we share her expectation that a Chinese butcher should bear a Chinese name, not a Spanish one. In Excerpt 2, CA would find it significant that DF categorizes AW as ‘mi maestra’ while AW categorizes herself as ‘teacher’, and that DF, in Excerpt 4, picks up on AW’s self-categorization by referring to her as ‘el ticher’. Inferences could be drawn as to the kind of social structure these two participants are constructing through these categorization devices. The strength of CA as an epistemological approach rather than a mere tool of analysis lies in its constructionist view of the social world that emphasizes participants’ local, situated, ethnographic understanding of social reality. The analyst’s membership knowledge of this reality is crucial for the analysis. For, while the participants themselves might not be able to verbalize their orientation to this or that aspect of the interaction, the analyst can recognize and interpret it based on his/her shared understanding of the social world. While pure CA has dealt mostly with monolingual, symmetrical exchanges between native speakers in everyday life, it has been adopted and expanded by researchers in the broader field of microethnography and discourse analysis (DA), with varying degrees of adherence to ethnomethodological analytical principles. CA has been applied to the analysis of bilingual and multilingual interactions in cross-cultural settings (e.g. Moerman 1988; Gafaranga 2005; Torras 2005) and to second/foreign/lingua franca talk (e.g. Firth 1996; van Dam 2002). In particular it has informed analyses of the social symbolic meanings of code-switching, both the ‘on the spot observable’ and the ‘in the head’ meanings (Zentella 1997). The literature on the application of CA to bi/multilingual interaction and language alternation is
CLAIRE KRAMSCH and ANNE WHITESIDE
657
extensive (see, e.g., Auer 1984, 1998; Richards and Seedhouse 2005; Wei 2005) and it varies in its tolerance to interpretations that go beyond what is strictly demonstrably relevant to the participants themselves. The distinction between pure and applied CA has become somewhat blurred as applied CA has overlapped with much of DA. As Wooffitt (2005) puts it: ‘Whereas in CA the analytic focus is on people’s own sense-making practices as they are revealed in the turn by turn unfolding of interaction’ (2005: 84), in DA ‘the action orientation of language is located at a broader level, on the wider interpersonal or social functions served by a passage of talk’ (2005: 80). Our data cannot be understood without factoring in the broader societal language ideologies at work in the participants’ choice of this or that code, their exercise of this or that symbolic power.3 CA is not intrinsically incompatible with the analysis of power relations and other macrophenomena like ideology, history, and cultural values, but rather than assume that social interaction merely reflects preexisting power relationships, it shows how ‘the sequential structures out of which the differential distributions of resources emerge are not a natural but an oriented to feature of the interaction’ (Hutchby 1999: 90 cited in Wooffitt 2005: 194). In other words CA can serve as a reality-check for DA. For example, at the Vietnamese butcher in Excerpts 7 and 8, CA can show how Bela’s economic power gets interactionally generated through her words as well as through her silences and how these are taken up by the butchers and the researcher in the store. However, neither CA nor DA can deal with the multiple levels of the global context itself, which is not restricted to a Vietnamese store in San Francisco but now includes the reenactment of practices carried out in Yucatan or ventriloquated words by a Chinese clerk mouthing Anglo prejudice.
3.2 Mediated communication analysis Mediated discourse or sociocultural communication studies (e.g. Wertsch 1990; Scollon and Scollon 2001; van Lier 2004) focus less on the individuals than on the mediated action itself as a kind of social symbolic action. For example, in Excerpt 7, the accomplishment of the interaction is mediated through gestures like cutting, pointing, etc. and artifacts such as the tall rectangular meat case, the short stature of the customer, and the multiple languages used in the sideplays and the main track exchanges of butchers and customers. The relationships between the participants and their environment, or affordances, are seized upon and constructed as ‘opportunities for or inhibition of action’ (van Lier 2004: 4), as when butcher A overhears Bela and AW’s sideplay in Spanish and seizes this opportunity to switch from English to Spanish, negotiate a lower price with his colleague in Chinese, and have Butcher B return to the customer in Spanish. This exchange is also a good illustration of this approach’s notion of activity, grounded in physical, social, and symbolic affordances, among which
658
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
language plays an important part as ‘a system of relations’ (van Lier 2004: 5). In the encounters above, the participants are vying not only for economic goods and services but for symbolic power and recognition. We can see this particularly well in the dialectic between two opposite perceptions of DF in Excerpt 5: that of a lazy, reluctant learner of English presented by the Chinese clerk and that of an eager learner of Spanish literacy presented by AW. DF himself resolves this tension by reaching for a synthesis—a third identity, namely that of a proud and well-respected speaker of Maya who, moreover, casts himself as a teacher of his native language to non-native merchants in his neighborhood. But this dialectic itself is perhaps too neat; it does not account for the unstable play of perceptions and counterperceptions between a Chinese clerk who recognizes her former self in the Yucatec migrant and the Maya speaker who styles himself as the successful resident that the Chinese clerk has become. The approaches discussed above: conversation and discourse analysis, and the study of mediated communication provide a useful basis for understanding what is going on in our data. However, they presuppose a social reality bound by the usual constraints of time and space. Globalization has disrupted this social reality. The protagonists in these exchanges are physically and emotionally living on several axes of space and time that are embodied in their daily practices. Erickson (2004) makes the distinction between kronos, that is, ‘the quantitative aspect of time, time as continuous and thus as measurable’; and kairos, the subjective time of ‘tactical appropriateness, of shifting priorities and objects of attention from one qualitatively differing moment to the next’ (2004: 6). The use of multiple codes in the data at hand and the exploitation of their various subjective resonances by the participants require taking into account not only the measurable communicative time of turns-at-talk within an activity, but the subjective, embodied time of cultural memory (Damasio 1999; Gibbs 2006). The Yucatecans in these encounters may be objectively present in a store in San Francisco, but their bodies carry subjective traces of their experiences living in Yucatan, crossing the border, learning to negotiate the vicissitudes of daily life as undocumented residents of the Bay Area. On certain streets they may move discretely, like stage crew executing a scene change, but among people they trust, speaking Maya, they become lead actors. The Maya language is for them embodied memory that, while located in individual bodies, resurrects a collective memory of group practices in the present. DF’s voice speaking Maya with a fellow Yucatecan recreates the Yucatan in San Francisco. Without the notion of subjective time, it is difficult to understand the importance DF attaches to teaching Maya to so many of the merchants in his neighborhood or to explain Bela’s dispreferred response in Excerpt 8—two aspects to which we return in section 4.2. In the following we propose an ecological approach that combines insights from complexity theory and postmodern thought in sociolinguistics.
CLAIRE KRAMSCH and ANNE WHITESIDE
659
4. ECOLOGICALLY ORIENTED ANALYSIS 4.1 Aspects of complex dynamic systems Dynamic systems theory, also called complexity theory (Byrne 1997; LarsenFreeman 1997; Cilliers 1998; Larsen-Freeman and Cameron 2008; Mason 2008; Peters 2008) is, when taken to its logical conclusions, a late modernist theory that hails more from Bakhtin’s dialogism (Bakhtin 1981) than Marxist dialectics. Dialogism, the principle behind Bakhtin’s existentialist philosophy of the relativity of self and other, and of the openness of time, shares with complexity theory and with postmodern sociolinguistics some basic tenets that can be summarized as follows: 1 Relativity of Self and Other. In complex dynamic systems like human relations, both the self and the other are intrinsically pluralistic, and possibly in conflict with themselves and with one another. Because the I is not unitary, but multiple, it contains in part the other and vice-versa; it can observe itself both subjectively from the inside and objectively through the eyes of the other. Hence the frequency of stylization, parody, double-voicing in the discourse of everyday life observed by sociolinguists like Rampton (1995) and others. The researcher is part of this subjective/ objective observation game. His/her categories of observation and their relevance for the researcher are themselves relative to his/her subject position and to the perspective of the participants. 2 Timescales. A dynamic systems theoretical model of language shows that the meanings expressed through language operate on multiple timescales, with unpredictable, often unintended, outcomes and multiple levels of truth and fantasy, reality and fiction. Our memories are not in the past but live on as present realities in our bodies to be both experienced and observed (Hofstadter 2007). Blommaert (2005) refers to this phenomenon as ‘layered simultaneity’. ‘We have to conceive of discourse as subject to layered simultaneity. It occurs in a real-time, synchronic event, but it is simultaneously encapsulated in several layers of historicity, some of which are within the grasp of the participants while others remain invisible but are nevertheless present’ (2005: 130). Simultaneity does not necessarily mean congruence. Blommaert notes that the participants in verbal exchanges might speak from positions on different scales of historicity, thus creating ‘multiple and contradictory temporalities’ that may lead to different intertextual references and to communicative tensions (2005: 128), such as we have in Excerpt 5. 3 Emergentism. Complexity theory has in common with postmodern sociolinguistic theory the notion that any use of language, be it learning a language or using it to haggle, assert yourself, or exercise power does not derive from structures in the head—beliefs, rules, concepts and schemata—but are new adaptations that emerge from the seamless
660
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
dynamic of timescales. As Blommaert writes: ‘Meaning emerges as the result of creating semiotic simultaneity’ (2005: 126). 4 Unfinalizability. Complexity theory does not seek dialectical unity, or bounded analyses of discrete events, but on the contrary open-endedness and unfinalizability. It counts under ‘participants’ not only the flesh and blood interlocutors in verbal exchanges, but also the remembered and the imagined, the stylized and the projected, and the objects of identification (Hofstadter 2007). Similarly, sociolinguists have problematized the notion of bounded speech communities and focused our attention on openended, ‘deterritorialized’ (Rampton 1998) communicative practices rather than on the ‘territorial boundedness’ posited by the ‘one language—one culture assumption’ (Blommaert 2005: 216). 5 Fractals. Complexity theory, like postmodern sociolinguistics, is concerned with patterns of activities and events which are self-similar at different levels of scales, that is, which are fractal figures for larger or smaller patterns. In the encounters above, stereotypical names like Jose or Felipe are fractals of a whole Hispanic culture, Maya greetings and leave takings are fractals of a Maya culture, and the stigmatization of Maya speakers as poor Indians in Yucatan is refracted in the stigmatization of Maya speakers as illegal immigrants by the US immigration authorities. In the following analysis we draw on a complexity theory of language learning, as proposed by Larsen-Freeman (1997) and Larsen-Freeman and Cameron (2008: 115–161) and a sociolinguistic theory of language use as proposed recently by Blommaert (2005) to suggest an ecological reading of the data at hand.
4.2 An ecological analysis of the data An ecological perspective on the data can build on the other analytic approaches, and view the unfolding events as the enactement, re-enactment, or even stylized enactment of past language practices, the replay of cultural memory, and the rehearsal of potential identities.
The ecology of multilingual spaces By performing English, Maya, Spanish, or Chinese, rather than only learning or using these languages, the protagonists in these data signal to each other which symbolic world they identify with at the time of utterance. In the Vietnamese store in Excerpt 1, DF’s and Juan’s little pas de deux around the use of English, Spanish, and Maya indexes the various ways in which the protagonists wish to position themselves in the ongoing discourse. At the end of a transaction in which Juan has been speaking a mix of English and Spanish, and DF has been speaking exclusively Spanish, Juan and DF take leave—Juan in English, DF in Maya. Taking leave is always a delicate part of any verbal exchange as it has to sum up the exchange, make plans for future exchanges, and perform a recognizable and acceptable leave-taking routine.
CLAIRE KRAMSCH and ANNE WHITESIDE
661
But in multilingual exchanges like this one, it is doubly delicate, as language choice can always become foregrounded. Since Juan had addressed DF in English and had been responded to in Spanish, Juan’s OK in line 3 can be seen to be oriented not only toward the content of DF’s utterance, but toward the language that DF chose to speak in. A gloss of this ‘OK’ might be ‘I agree to sell you 5 lb of tripe tomorrow’ but also ‘I agree to respond to you in Spanish’ or ‘I acknowledge the legitimacy of Spanish in my store’. In #4, DF suddenly switches to Maya. Because the store is located in a predominantly Spanish speaking area of San Francisco, DF’s efforts to get Juan and other merchants to respond to him in Maya has been a form of public resistance to a Spanish colonial discourse which holds Maya in low esteem among Mexicans.4 Here, a Vietnamese clerk serves as an unwitting catalyst for DF’s efforts to provide a place for himself between the polarity Spanish–English that divides much of California today. Whereas speaking Maya can be a social millstone in Yucatan and marks speakers as belonging to a recent wave of migrants with dubious immigration status, in some neighborhoods of San Francisco Maya can be made to yield a different social capital vis-a`-vis third ethnic groups, that is, immigrants who are neither Mexicans nor Anglos, DF’s use of Maya gives him a prestige of distinction vis-a`-vis Mexicans, Spanish gives him a distinction on a par with Anglos. Juan’s laughter in #9–10 is both amused and slightly embarrassed at having to produce Maya sounds in front of the Anglo visitor. In the usual hierarchy of codes in this Hispanic neighborhood, English and Spanish would be the two unmarked codes, followed perhaps by Vietnamese as the storeowner’s language, but Maya is definitely marked. However, it has, in this case, acquired some historical presence due to DF’s repeated efforts to teach the local merchants some Maya, so we can interpret Juan’s chuckle as a sign that he is both willing to respect DF’s language and ambivalent about his own legitimacy as a Maya speaker. It is worth noting that DF does not administer his little Maya lesson in all stores. In the Chinese store, for example, he uses Spanish throughout even when admonishing the clerk that her ability to understand Maya is improving (Excerpt 6, #138–141). The Chinese clerk in Excerpts 2–6 also plays with the languages available in her store. She alternately speaks Chinese with her old relative, Spanish with putative ‘Mexicans’ like DF, and English with Anglos like AW. These three languages index respectively: her ethnic or cultural identity as a Chinese, the accommodating role that she wants to assume and cultivate with Spanish-speaking customers, and the public voice she feels appropriate to adopt with Anglos. But she clearly uses these languages to align herself symbolically with the shifting centers of power in her store. For her in these data, Chinese is the language of intimacy with fellow customers, family, and friends; Spanish is the useful service language of local transactions, but it also indexes for her the stigma of non-assimilated immigrants; English is her public transactional language but she can also use it as a way of distancing herself from Mexican newcomers.
662
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
Viewing these exchanges as dynamic complex systems enables us to see the various languages used by the participants as part of a more diversified linguistic landscape with various hierarchies of social respectability among codes, and added layers of foregrounding of the code itself rather than just the message. To the multiplicity of languages we must add their subjective resonances in the speakers’ embodied memories.
The ecology of embodied time It is important to note that the protagonists’ choice of language is not dictated by some pre-existing and permanent value assigned to each of these languages, rather, the meaning of these choices emerges from the subjective perceptions of shifting power dynamics within the interaction. It draws on multiple timescales of experience, for example, at the Chinese grocery, the clerk’s memories of learning English in America (Excerpt 5 #118 ff), DF’s reminders of past Maya lessons with the clerk and his prediction of her future progress (Excerpt 6, #138–141), and, as mentioned in the previous section, reenactments of similar transactions between DF and the Yucatecan butcher in their native Yucatan (Excerpt 2, #1). This last timescale is particularly important for an understanding of the social prestige accorded to DF in this neighborhood of San Francisco. His weekly tours of the grocers and butchers recreate the network of Maya-speaking connections he had in his hometown. They also show that social capital varies greatly at different scales, so that in Yucatan, DF can be a wealthy respected merchant, while at the Mexican national level he may be perceived as poor, Indian, and illiterate. The connections between these different timescales bolster the invisible symbolic power of his undocumented presence in the United States. They cast a halo around his words that cannot be captured by looking only at the utterances produced in the present. For example, DF’s broad smile and assertive posture when he turns to the Chinese clerk in Excerpt 2 #3 and proclaims ‘buenas . . . ’, carry evidence of the self-assuredness displayed a minute ago by a successful merchant chatting with his fellow Yucatecan in their common language. The conflation of timescales can be further exacerbated by imprecise tense markers in the various grammars used by the participants. For example, between Excerpts 2 and 3 in an exchange not presented here, DF asks the Chinese clerk about the remainder of the 80 lb of masa that he ordered earlier. But because Maya has no verb tense morphology, DF’s use of tense markers in Spanish is intermittent (DF: ochenta llevo ahora ¼ I take 80 now). Chinese does not have any verb tense morphology either, so the clerk, who tries to clarify things by responding in Spanish: aquı´ treinte (here thirty) does not help matters by using aqui (here) instead of ahora (now). When DF then answers: mas al rato (later), the temporal confusion is extreme. Will he or will he not take the 80 lb of corn flour? Will he do it now or later? This can only be disambiguated through reference to their prior arrangements. It seems to
CLAIRE KRAMSCH and ANNE WHITESIDE
663
suggest that the transaction might be in fact the reenactment of an exchange that took place earlier and is now being performed again for the benefit of the guest of honor, the researcher herself, who is being ‘toured around’ (cf. #31 ‘estamos de paseo con la ¼ maestra’). If that interpretation is correct, then the analysis has to take the words not as the spontaneous productions typical of natural conversations, but as a reflective replay for the benefit of a third party, a staging of sorts. Of course, this staging or styling serves also to nurture the human and commercial relations DF is keen on keeping up with the merchants in his neighborhood.5 Besides the conflation of timescales in the performing bodies of these social actors, we notice another aspect of embodied time. Spanish, Maya, English, Chinese, all acquire a subjective overlay of Mexican-ness, Maya-ness, etc. that makes uttering Spanish or Maya words more than the sum of their grammars or of the communicative roles they perform. Beyond haggling over the price of meat, the protagonists in these exchanges are performing not only themselves, but their cultures, their families, their countries of origin or the mythic and emotional memories that these historical realities have become. They are not just performing ‘being Maya’, they are maintaining alive an idealized or ‘de-territorialized’ kind of Maya-ness that transcends geographic boundaries and awaits to be reterritorialized in the subject positionings of individual speakers (Rampton 1998). Each of their utterances is less the performance of a language than the enactment of a performative speech act that creates the very reality it purportedly refers to (Pennycook 2007: ch. 4). As Blommaert notes: ‘The performance of identity is not a matter of articulating one identity, but of the mobilization of a whole repertoire of identity features converted into complex and subtle momentto-moment speaking positions’ (2005: 232). A good example of this is given at the Vietnamese butcher shop. As we described in section 2, Bela’s linguistic abilities include: conversational Spanish, limited English, and passive knowledge of Maya. At the end of the transactional encounter with the two Chinese butchers in Spanish (Excerpt 8), Bela asks Butcher B in English (#95) what his name is, presumably for future reference if she decides to buy meat in this store, since he is the one who earlier gave her—in Spanish—a good price. The reason for her switch to English is not immediately clear, but it makes the butcher’s response all the more striking. Like Juan, the Vietnamese grocer in Excerpt 1, this Chinese butcher has taken on a Spanish name for his Spanish-speaking customers. Bela’s choice of a dispreferred response to his name in #97—‘Felipe (.)’, a simple reiteration of the name rather than a vocative—draws attention to the name itself and what it connotes about the Mexican-ness of a Chinese butcher. Who says that a Chinese butcher cannot make himself into a Mexican butcher, since indeed, as the other Chinese butcher says: ‘Felipe sabe espan˜ol muy bien’?, glossed as: ‘Since he speaks Spanish well, he is entitled to give himself a Spanish name’ or ‘Felipe is not just any name he gives himself, it means that he also knows Spanish well’. With ‘/\OK Felipe’ Bela accepts
664
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
‘Felipe’s’ unexpected Hispanic identity. Hoping that these two customers come back, and offering a personal contact as incentive, Butcher B hands Bela his card, adding in English ‘This one with my name’. At the end of this exchange, the two butchers, who both know English, Chinese, and Spanish, make sure they cover all their bases with this elusive, multilingual customer: Butcher A addresses her in Spanish in #99, Butcher B in English in #100. If we take Butcher B’s adopted identity for the linguistic construction that it is, then we have to admit that in the multilingual and multicultural environment of immigrant communities, the symbolic dimension of interactions is as significant as their pragmatic one.
5. SYMBOLIC COMPETENCE An ecological analysis of these data reveals a much greater degree of symbolic action than is usually accounted for in applied linguistics. Social actors in multilingual settings seem to activate more than a communicative competence that would enable them to communicate accurately, effectively, and appropriately with one another. They seem to display a particularly acute ability to play with various linguistic codes and with the various spatial and temporal resonances of these codes. We call this competence ‘symbolic competence’. Symbolic competence is the ability not only to approximate or appropriate for oneself someone else’s language, but to shape the very context in which the language is learned and used. Such an ability is reminiscent of Bourdieu’s notion of sens pratique, exercised by a habitus that structures the very field it is structured by in a quest for symbolic survival (Bourdieu 1997/2000: 150). Here, however, we are dealing with a multilingual sens pratique that multiplies the possibilities of meaning offered by the various codes in presence. In today’s global and migratory world, distinction might not come so much from the ownership of one social or linguistic patrimony (e.g. Mexican or Chinese culture, English language) as much as it comes from the ability to play a game of distinction on the margins of established patrimonies. Because it depends on the other players in the game, we should talk of a ‘distributed’ symbolic competence, that operates in four different ways.
5.1 Subjectivity or subject-positioning Different languages position their speakers in different symbolic spaces (see, e.g., Weedon 1987). In the data above, speakers take on subject positions regarding the symbolic power of this versus that language, the respective social values of Maya, Chinese, Spanish, and English. In Excerpt 7, for example, Bela is linguistically at a disadvantage in English but she is commercially at an advantage, because she is the one who has the purchasing power. Because she is perceived as a powerful customer, the butchers and the clerk will go along with whatever language she wants to speak: English
CLAIRE KRAMSCH and ANNE WHITESIDE
665
at first when she looks like she prefers English; then Spanish in #66ff, when she is overheard speaking Spanish to AW. In turn Bela’s ambiguity serves to play one language against the other and, after the price has been brought down, to gain time until next week. This could be seen as strategic competence on Bela’s part, but strategic competence has been conceived up to now as an individual compensatory tactic (Canale and Swain 1980: 30), whereas the symbolic competence apparent here is a distributed competence that emerges from playing the game. Subject positioning has to do less with the calculations of rational actors than with multilinguals’ heightened awareness of the embodied nature of language and the sedimented emotions associated with the use of a given language. In Excerpt 2, the pleasure that the butcher and Don Francisco experience at using with each other the language of their common village in Yucatan is still visible in DF’s self-assured demeanor when he turns around and switches to Spanish in #3. The clerk’s volubility in English in Excerpt 5 indexes her pleasure at being able to converse in English with the researcher, something she cannot do with DF in Spanish. And AW’s switch to Spanish in Excerpt 5 #108 aligns her emotionally with DF, who may have felt affronted by the clerk’s use of English.
5.2 Historicity or an understanding of the cultural memories evoked by symbolic systems Throughout the data presented here, we have been confronted with the cultural memories carried by words, gestures, body postures, and scripts taken from a different timescale in a different place and reterritorialized in a Californian grocery store. We have noticed the timescale of Yucatan irrupting in the timescale of San Francisco, but there are other examples. During a visit to another Vietnamese grocery, AW and the clerk engaged in a comparative account of the ancient history of the Maya in Mexico versus the ancient history of the Chinese in Vietnam (Kramsch and Whiteside 2007). Neither the clerk nor the researcher were really teaching each other a history lesson; rather, each was lending weight to her words by performing ritualized utterances about the ancient nature of Maya and Chinese civilizations—an exchange of social symbolic power that put both parties on an equal footing. The utterances in these exchanges sounded formulaic because they were what Pierre Nora calls lieux de me´moire, realms or archetypes of social memory (1997: 3031). Any utterance or turn-at-talk can become a lieu de me´moire, formed by the sedimented representations of a people. Whether these representations are accurate or not, historically attested or only imagined, they are actually remembered by individual members and serve as valid historical models. As Blommaert writes: ‘The synchronicity of discourse is an illusion that masks the densely layered historicity of discourse’ (2005: 131). Indeed, symbolic competence is the ability to perform and construct various historicities in dialogue with others.
666
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
5.3 Performativity or the capacity to perform and create alternative realities Within an ecological perspective of human exchanges, utterances not only perform some role or meaning, but they bring about that which they utter, that is, they are performatives. We have seen how the utterances of the protagonists in our data recreate environments from other scales of space and time, produce fractals of patterns from one timescale to another. Multilingual environments can elicit complex relationships between speech acts and their perlocutionary effects. Take for example Excerpt 4. The clerk clearly devalues DF by ignoring that his utterance: ‘eso es el ticher’ (#76) names the researcher as ‘the teacher’, and by taking on herself the teacher role (#80). She puts down his Spanish by embedding it in her English: ‘Man˜ana when you come I give you no espan˜ol, solo English’, then calling him ‘Jose’ in #85. The cartoon-like foreigner talk is not lost on the older lady and on DF himself who bursts out laughing. But we understand that it was an insult and not just a joke from its perlocutionary effect on AW. Her immediate overlapping response in English in #88 (‘No no I’m not teaching him English’) seeks to cancel the potential perlocutionary effect of the insult by resignifying the ESL issue into a Spanish literacy issue (‘I’m not teaching him English. I’m teaching him to read and write in Spanish’)—a symbolic move that reestablishes DF at par with the clerk: in the same manner as the clerk learned English, DF is now learning Spanish literacy.6 Such a move exploits the time lag, materialized here by the general laughter in #81–87, between the illocutionary force of the clerk’s derogatory utterance and its perlocutionary effect on DF, and reconfigures the whole environment. The actors in the Chinese grocery store are quick to adapt to the alternative configuration introduced by AW in #88 and DF regains the symbolic space that was his at the onset of the exchange. Thus a third aspect of symbolic competence is the capacity to use the various codes to create alternative realities and reframe the balance of symbolic power.
5.4 Reframing Finally, the data highlight the importance of reframing as a powerful means of changing the context. In Excerpt 6, DF reframes the face threatening situation defused by AW’s intervention into one that reestablishes his legitimacy. For the Chinese clerk, legitimacy as an immigrant comes from having learned English, knowing how to count in English and the English alphabet. For DF, legitimacy comes from having money and clout from the old country, and influence in the neighborhood, even though he is illiterate. In Excerpt 4, by resignifying the clerk’s insult into an erroneous statement of fact (#88), AW reframed her relationship with DF (#88–90) from an ESL teacher to a Spanish literacy maestra. In turn, DF reframes his relationship with the clerk: at first it was the clerk who in #80 constructed for herself a ‘teacher’ role to ‘Jose’ the pupil. In Excerpt 6 #138–141, DF suddenly turns the tables as the Maya
CLAIRE KRAMSCH and ANNE WHITESIDE
667
‘teacher’. His insistence that she will end up understanding Maya is less a statement about her than about him contesting and reframing the view that ‘todo English aquı´’. Maya, he suggests, will be an increasing part of this world, as will Spanish. And, indeed, he gives leave in Spanish (#150–151), while the clerk closes the conversation in English (#152–153). Symbolic competence could thus be defined as the ability to shape the multilingual game in which one invests—the ability to manipulate the conventional categories and societal norms of truthfulness, legitimacy, seriousness, originality—and to reframe human thought and action. We have seen that this kind of competence is multiply distributed and that it emerges through the interaction of multiple codes and their subjective resonances. It is true that symbolic competence is not reserved to multilingual actors in multilingual encounters. Analyzing exchanges between monolingual speakers of English, Gumperz (1982) found that the meaning of utterances there too lie not only in the way participants orient themselves to the ongoing exchange, but in the way they implicitly ventriloquate or even parody prior utterances and thereby create affordances in ways that are favorable to them. Multilingual encounters increase the contact surfaces among symbolic systems and thus the potential for creating multiple meanings and identities. In the late modern stance offered by an ecological perspective, symbolic competence is both semiotic awareness (van Lier 2004), and the ability to actively manipulate and shape one’s environment on multiple scales of time and space. Symbolic competence in our view adds a qualitative metalayer to all the uses of language studied by applied linguists, one that makes language variation, choice, and style central to the language learning enterprise.
CONCLUSION An ecological analysis of multilingual interactions enables us to see interactions in multilingual environments as complex dynamic systems where the usual axes of space and time are reordered along the lines of various historicities and subjectivities among the participants. While the global economy has deterritorialized and dehistoricized the spaces of human encounters, participants find a way of reterritorializing and rehistoricizing them in their moment-by-moment utterances. Our analysis of their interactions has revealed the importance of taking into account embodied perceptions, portable cultural memories, and the power that comes from resignifying the illocutionary force of performatives. In environments where the boundaries of the distant and the proximal, the past and the present, the real and the imagined have become blurred, when names have become arbitrary, and signifiers are no longer transparent, multilingual exchanges require us to position ourselves as researchers in much more multidimensional ways than is usually done in applied linguistics. For language learners and educators, symbolic competence is not yet another skill that language users need to master, nor is it a mere component
668
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
of communicative competence. Rather, it is a mindset that can create ‘relationships of possibility’ or affordances (van Lier 2004: 105), but only if the individual learns to see him/herself through his/her own embodied history and subjectivity and through the history and subjectivity of others. Our symbolic survival is contingent on framing reality in the way required by the moment, and on being able to enter the game with both full involvement and full detachment. In this sense, the notion of symbolic competence is a late modern way of conceiving of both communicative and intercultural competence in multilingual settings.
ACKNOWLEDGEMENTS A version of this paper was given at the Annual meeting of the American Association for Applied Linguistics in Costa Mesa, CA in April 2007. We are grateful for the feedback received then, as well as for the generous comments received on other occasions from Alastair Pennycook, Diane Slade, and Ruth Wodak in Sydney, Constant Leung, Ben Rampton, Celia Roberts, and Brian Street in London. We are particularly indebted to the two editors of this journal and three anonymous reviewers.
APPENDIX: TRANSCRIPTION CONVENTIONS BASED ON DUBOIS (2006) Boundary Tone/Closure Terminative
Metatranscription .
Unintelligible
(??)1
Continuative
.
Comment
((WORDS))
Truncated intonation unit
-
Overlap
[]
Appeal
?
Tone shifts Rising tone
/
Breath (in)
(H)
Falling tone
\
Laugh Manner
@
Low to high tone High-low-high
_/ \/
Manner/quality
5MISC.
Voice tone
5VOX4
Vocalisms
NOTES 1 The interviews were conducted with the four focal participants and with 13 non-Yucatecan service professionals working with this population. 2 The distinction between ‘pure’ and ‘applied’ conversation analysis was made by ten Have (1999: 8) to distinguish between a focus on specific interactional situations and how interactants orient to these situations and their requirements (pure CA) on the one
hand, and a focus on the larger institutional arrangements as they pertain to the organization of interaction (applied CA) on the other hand. We apply this distinction to the two strands of CA we find in the literature today: the strictly local and the more ethnographically contextual. 3 Indeed, the study of talk in interaction has been associated with late modernist theories of structure–agency dialectic
CLAIRE KRAMSCH and ANNE WHITESIDE
(Giddens 1991) and the construction of intersubjectivity and identity (Antaki and Widdicombe 1998). 4 This interpretation follows from Whiteside’s general observation that the Maya speakers in this study tend to refrain from using Maya in nonMaya speaking contexts. DF’s deliberate flouting of this general practice must be seen as an act of defiance. 5 Rampton (1995) has convincingly shown on the language practices of multilingual adolescents in British schools the interpersonal effects of ‘styling’, i.e. performing a language or language variety other than one’s own for placating, teasing, mocking, or
669
playful purposes (see also Goffman 1959). Coupland (2007) points out that stylization, like staging, is a form of multi-voiced utterance originally theorized by Mikhail Bakhtin (1981). 6 What is perhaps at issue here is AW’s attempt to deflect another possible threat to DF’s face, a future-oriented one since, if DF is mistakenly perceived as receiving English lessons and then subsequently perceived as not making progress, he will be in danger of losing face as one who, in spite of instruction, is incapable of making progress in English. We thank an anonymous reader for this very plausible interpretation.
REFERENCES Antaki, C. and S. Widdicombe (eds). 1998. Identities in Talk. London: Sage. Auer, P. 1984. Bilingual Conversation. Amsterdam: John Benjamins. Auer, P. 1998. Code-switching in Conversation: Language, Interaction, and Identity. London: Routledge. Bakhtin, M. M. 1981. The Dialogic Imagination, ed. M. Holquist, transl. C.Emerson and M. Holquist. Austin, TX: University of Texas Press. Blommaert, J. 2005. Discourse. Cambridge: Cambridge University Press. Bourdieu, P. 1997/2000. Pascalian Meditations, transl. R. Nice. Stanford: Stanford Univerity Press. Byrne, D. 1997. ‘Complexity theory and social research.’ Social Research Update. Quarterly by the Department of Sociology, University of Surrey, Guildford, UK. http://www.soc.surrey. ac.uk/sru/SRU18.html. Canale, M. and M. Swain. 1980. ‘Theoretical bases of communicative approaches to second language teaching and testing,’ Applied Linguistics 1: 1–47. Cilliers, P. 1998. Complexity and Postmodernism: Understanding Complex Systems. London: Routledge. Coupland, N. 2007. Style. Language Variation and Identity. Cambridge: Cambridge University Press. Damasio, A. 1999. The Feeling of What Happens. Body and Emotion in the Making of Consciousness. New York: Harcourt.
DuBois, J. W. 2006. Basic Symbols for Discourse Transcription. Retrieved 15 December 2006 from http://www.linguistics.ucsb.edu/projects/ transcription/representing. Erickson, F. 2004. Talk and Social Theory. Cambridge: Polity Press. Fill, A. and P. Mu¨hlha¨usler (eds). 2001. The Ecolinguistics Reader. Language, Ecology and Environment. London: Continuum. Firth, A. 1996. ‘The discursive accomplishment of normality: On ‘lingua franca’ English and conversation analysis,’ Journal of Pragmatics 26: 237–59. Gafaranga, J. 2005. ‘Demythologising language alternation studies: conversational structure vs. social structure in bilingual interaction,’ Journal of Pragmatics 37: 281–300. Garfinkel, H. 1967. Studies in Ethnomethodology. New York: Prentice-Hall. Gibbs, R. 2006. Embodiment and Cognitive Science. Cambridge: Cambridge University Press. Giddens, A. 1991. Modernity and Self-Identity. Stanford: Stanford University Press. Goffman, E. 1959. The Presentation of Self in Everyday Life. New York: Anchor Books. Gu¨emez Pineda, M. 2006. Language, Culture and Indigenous Rights in Rural Yucatan. Paper presented at the Mayab Bejlae: Yucatan Today. 21–23 April 2006. Kroeber Hall, University of California Berkeley. Gumperz, J. J. 1982. Discourse Strategies. Cambridge: Cambridge University Press.
670
LANGUAGE ECOLOGY IN MULTILINGUAL SETTINGS
Haugen, E. 1972. The Ecology of Language: Essays by Einar Haugen, ed. A. S. Dil. Stanford: Stanford University Press. Hofstadter, D. 2007. I Am a Strange Loop. New York: Basic Books. INDEMAYA (Instituto para el Desarrollo de la Cultura Maya del Estado de Yucata´n). 2005. Diagno´stico de Migracio´n y Polı´ticas Publicas en el Estado de Yucata´n: Sı´ntesis Diagno´stica. INDEMAYA, Instituto Nacional de Antropologia e Historia. Johnstone, B. 1996. The Linguistic Individual: Selfexpression in Language and Linguistics. Oxford: Oxford University Press. Kramsch, C. (ed.). 2002. Language Acquisition and Language Socialization. Ecological Perspectives. London: Continuum. Kramsch, C. and A. Whiteside. 2007. ‘Three fundamental concepts in SLA and their relevance in multilingual contexts’. Modern Language Journal Focus Issue 91: 905–20. Kramsch, C. and S. V. Steffensen. 2007. ‘Ecological perspectives on second language acquisition and socialization’ in N. Hornberger and P. Duff (eds): Encyclopedia of Language and Education Vol. 8. Language and Socialization. Heidelberg: Springer Verlag. Larsen-Freeman, D. 1997. ‘Chaos/complexity science and second language acquisition,’ Applied Linguistics 18(2): 141–65. Larsen-Freeman, D. 2003. Teaching Language: From Grammar to Grammaring. Boston: Heinle/ Thomson. Larsen-Freeman, D. and L. Cameron. 2008. Complex Systems and Applied Linguistics. Oxford: Oxford University Press. Leather, J. and J. van Dam (eds). 2002. Ecology of Language Acquisition. Dordrecht: Kluwer Academic. Mason, M. 2008. ‘What is complexity theory and what are its implications for educational change? Educational Philosophy and Theory 40(1): 35–49. Moerman, M. 1988. Talking Culture. Ethnography and conversation analysis. Philadelphia: University of Pennsylvania Press. Nora, P. 1997. Les lieux de me´moire. Paris: Gallimard. Passell, J. S. 2005. Unauthorized Migrants: Numbers and Characteristics. Retrieved 10 November 2005 from http://pewhispanic.org/reports/ report.php?ReportID¼46. Passell, J. S., C. Randall, and M. E. Fix. 2004. Undocumented Immigrants: Facts and Figures.
Retrieved 15 July 2005 from http://www.urban. org/url.cfm?ID¼1000587. Pennycook, A. 2007. Global Englishes and Transcultural Flows. London: Routledge. Peters, M. A. 2008. ‘Editorial: Complexity and knowledge systems,’ Special Issue. Educational Philosophy and Theory 40(1): 1–3. Rampton, B. 1995. Crossing. London: Longman. Rampton, B. 1998. ‘Speech community,’ in J. Verschueren, J. O.Ostman, J. Blommaert and C.Bulcaen (eds): Handbook of Pragmatics. Amsterdam: John Benjamins, pp. 1–34. Rampton, B. (ed.). 1999. ‘Styling the other,’ Special issue. Journal of Sociolinguistics 3(4). Richards, K. and P. Seedhouse (eds). 2005. Applying Conversation Analysis. Houndmills, Basingstoke: Palgrave Macmillan. Risager, K. 2006. Language and Culture. Global Flows and Local Complexity. Clevedon, UK: Multilingual Matters. Sacks, H. 1992. Lectures on Conversation. Vols 1 and 2, ed. G. Jefferson, intro. E. A. Schegloff. Oxford: Blackwell. Sacks, H., E. A. Schegloff, and G. Jefferson. 1974. ‘A simplest systematics for the organization of turn-taking for conversation,’ Language 50: 696–735. Schegloff, E. A. 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis, Vol. I, Cambridge: Cambridge University Press. Scollon, R. and S. Scollon. 2001. Intercultural Communication. 2nd edn. Oxford: Blackwell. Steffensen, S. V. 2007. ‘Language, ecology and society: An introduction to dialectical linguistics,’ in J. C. Bang and J. Door. Language, Ecology and Society—A Dialectical Approach, eds. S.V. Steffensen and J. Nash. London: Continuum. Ten Have, P. 1999. Doing Conversation Analysis. A Practical Guide. London: Sage. Torras, M. C. 2005. ‘Social identity and language choice in bilingual service talk’ in K. Richards and P. Seedhouse (eds): Applying Conversation Analysis. Houndmills, Basingstoke: Palgrave Macmillan, pp. 107–23. van Dam, J. 2002. ‘Ritual, face, and play in a first English lesson: Bootstrapping a classroom culture’ in C. Kramsch (ed.): Language Acquisition and Language Socialization. Ecological Perspectives. London: Continuum, pp. 238–65. van Lier, L. 2000. ‘From input to affordance: Social-interactive learning from
CLAIRE KRAMSCH and ANNE WHITESIDE
an ecological perspective’ in J. P. Lantolf (ed.): Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp. 245–60. van Lier, L. 2004. The Ecology and Semiotics of Language Learning. A Sociocultural Perspective. Dordrecht: Kluwer Academic. Weedon, C. 1987. Feminist Practice and Poststructuralist Theory. Oxford: Blackwell. Wei, L. (ed.) 2005. ‘Conversational Codeswitching,’ Special Issue. Journal of Pragmatics 37(3).
671
Wertsch, J. 1990. Voices of the Mind: A Sociocultural approach to Mediated Action. London: Harvester Wheatsheaf. Whiteside, A. 2006. ‘We are the explorers’: Transnational Yucatec Maya-speakers negotiating multilingual California. Unpubl. PhD dissertation, University of California, Berkeley. Wooffitt, R. 2005. Conversation Analysis and Discourse Analysis. A Comparative and Critical Introduction. London: Sage. Zentella, A. C. 1997. Growing Up Bilingual. Oxford: Blackwell.
Applied Linguistics 29/4: 672–693 ß Oxford University Press 2008 doi:10.1093/applin/amn023 Advance Access published on 21 June 2008
A Register Approach to Teaching Conversation: Farewell to Standard English? CHRISTOPH RU¨HLEMANN Ludwig-Maximilians-University, Munich Owing to analyses of large spoken corpora the linguistic knowledge of conversation has grown in recent years exponentially. Up until now little of this knowledge has trickled down to the EFL classroom. One of the reasons, this paper argues, is the failure in the relevant literature to spell out clearly how teaching conversational grammar affects the role of what is the major variety in the EFL classroom, Standard English (SE). My aim in this paper is threefold. First, I briefly discuss some neglected conversational features in relation to SE, concluding that the contrast between the grammars of conversation and SE is so stark that the notion of SE is problematic in talking of the spoken language. Second, I consider what this contrast implies for EFL teaching, arguing that for authentic conversation to be taught effectively it is necessary to reduce the role of SE to ‘a core variety’ that has its place in teaching writing while conversational grammar might serve as the underlying model in teaching speech. I argue that such a redefinition of SE would best be implemented in a ‘register approach’ which shifts the emphasis from a monolithic view of language to a register-sensitive view thus acknowledging the fundamental functional diversity of language use. Third, I discuss some important issues arising from this approach and, finally, outline what may be gained by it.
INTRODUCTION It seems safe to say that analyses of corpora have fundamentally altered the ways in which linguists research and view language. Evidence of this ‘corpus revolution’ (Crystal 2003: 448) is not only the fact that all major publishers produce dictionaries these days which are corpus-based but also the fact that in recent years two major corpus-based grammars, the Longman Grammar of Spoken and Written English (LGSWE) (Biber et al. 1999) and the Cambridge Grammar of English (CGE) (Carter and McCarthy 2006) have appeared on the stage challenging dearly held conceptions about the nature of the grammar of English. While applied linguistics has seen lively debates about the usefulness, or otherwise, of using corpus data in EFL, the field of EFL itself has been slow to respond to these developments. Signs suggesting that the corpus revolution might ultimately reach the EFL classroom are coming from applied corpus linguistics. A growing body of research comparing corpus and classroom
CHRISTOPH RU¨HLEMANN
673
English suggests that the English taught is considerably at variance with the English spoken (e.g. Mindt 1997; Conrad 2004; Ro¨mer 2004, 2005). Also, principled attempts have been made to introduce and establish what has been termed, respectively, ‘spoken grammar’ (e.g. Carter and McCarthy 1995) and ‘conversational grammar’ (cf. Biber et al. 1999; Ru¨hlemann 2006) and to develop initial methodologies on how to teach it (e.g. Carter and McCarthy 1995; Timmis 2005). However, there are few signs, if any, that the notion of ‘conversational grammar’ is in effect being taken on board in classrooms (Timmis 2005: 117). This may be due to multiple reasons. I am concerned in this paper with one of these reasons: the fact that the relationship between conversational grammar and the model variety traditionally underlying EFL teaching, Standard English (SE), has to date not been made explicit, and what teaching conversational grammar implies in terms of the role of SE has not been discussed in detail. The aim of this article is therefore threefold. First, it aims to outline the relationship between conversational grammar and the grammar of SE. Some distinctive features of conversation to date that have not received much attention are discussed. The intention here is to illustrate that and how the language of conversation works largely by rules that deviate from the rules of SE. Second, considering what seems to be the most important implication of the difference between conversational grammar and SE, I argue that for conversational grammar to be effectively taught in EFL classrooms a rethinking of the role of SE is necessary: SE needs to be qualified in the sense that it can no longer be seen as ‘the one-and-only variety’ but should be reduced to ‘a core variety’. Given its intricate relationship with written registers, SE as a core variety would have its place in teaching the written language, while in teaching the spoken language the underlying model variety should be conversational grammar. I further argue that such a redefinition of SE as a core variety would best be implemented in a register approach, that is, within the context of shifting the emphasis in EFL from a monolithic view of language to a register-sensitive view which acknowledges the fundamental functional diversity of language use. In the final section, I discuss the issues arising from this approach and outline what may be gained by it. At first though it may be useful to characterize the notions of register and SE in some detail.
ON REGISTER AND SE Biber et al. refer to registers as ‘situationally defined varieties’ (Biber et al. 1999: 5). This characterization is useful in that it combines the two essential ‘ingredients’ of register: variation and situation. These are briefly explained in the following. Registers are social (Crystal 2003: 290 ff.) or functional varieties (Halliday 2004: 27), such as sports commentaries, legal discourse, academic discourse, and conversation, to name a few. Registers contrast with ‘dialects’,
674
A REGISTER APPROACH TO TEACHING CONVERSATION
or regional varieties, such as Indian or British English. In Systemic Functional Linguistics, registers are situated between the two poles of system and text: they occupy an intermediate location as ‘instance types’, that is, text types that ‘vary systematically according to contextual values’ (Halliday and Matthiessen 2004: 27). Central to an understanding of how context and register interact is the notion of situation type: ‘[l]ooking at how people actually use language in daily life, we find that the apparently infinite number of different possible situations represents in reality a very much smaller number of general types of situation’ (Halliday 1978: 29). That is, while texts, or instances of language, unfold within concrete ‘contexts of situation’ [a term Halliday borrowed from Malinowski (1923)] which are determined by myriads of situational details, the notion of register refers to an abstract level at which it becomes possible to discern what is shared by essentially similar situations. These shared properties of individual situations, referred to as ‘external determinants’ (Biber et al. 1999: 1041), ‘functional categories’ (Leech 2000: 694 ff.), or ‘situational factors’ (Ru¨hlemann 2006, 2007), constitute situation types. But registers are varied not only in terms of the situation types that give rise to them but also in terms of the language use that is characteristic of them. Being social in nature, a register is a ‘variety according to use’ (Halliday 1978: 35) and to ‘what a person is speaking, determined by what he is doing at the time’ (ibid.: 110). That is, a second characteristic of register is ‘the fact that the language that we speak or write varies according to the type of situation’ (ibid.: 32; emphasis added). Thus, the notion of register is best seen in a double perspective: an extralinguistic perspective, in which registers are intimately associated with certain situation types, and a linguistic perspective, in which registers are distinguishable on the basis of the linguistic features distinctive of them (cf. Crystal 2003: 290). The notion of SE is no less complex. In a definition which is as precise as it is concise, Crystal (2003: 110) points out five essential characteristics of SE. First, SE is ‘a minority variety’ in that it is not widely produced. In fact, Trudgill estimates that SE is ‘spoken as their native variety, at least in Britain, by about 12–15 per cent of the population’ (Trudgill 1999: 124). This low percentage suggests that SE is not a spoken variety at all. Second, although not widely produced, SE is very widely understood. Indeed, SE seems to underlie not only most serious writing but also the phenomenon of World Standard English (cf. Crystal 2003: 111 ff.), that is, the phenomenon underlying the fact that, for example, newspaper texts produced in New Zealand are as intelligible for British readers as academic papers written by a scholar from India are for readers in South Africa. Third, because those (few) speakers that come closest to actually speaking SE are found ‘at the top of the social scale’ (Trudgill 1999: 124) SE is the variety which carries most prestige. So, although historically a regional dialect in the southeast of England, SE today is a ‘clearly marked, socially symbolic dialect’ (Carter 1999: 163). Fourth, because it is the prestige variety, SE is associated, indeed even equated,
CHRISTOPH RU¨HLEMANN
675
with ‘educated English’ (e.g. Quirk et al. 1985: 18). It hence represents ‘a desirable educational target’ (Crystal 2003: 110). Finally, and most crucially in the present connection, probably because of its prestige and its central role in native-speaker education, SE has been the major model in EFL. Quirk et al. (1985: 7), for example, note that ‘[i]n countries where English is a nonnative language, the major models for both writing and speech have generally been the standard varieties of British and American English.’ That SE has been the major model in EFL for both writing and speech is of particular interest, for it is widely agreed that ‘there is a close relationship between standard English and the written language’ (Carter 1999: 158): Trudgill (1999: 118) notes that SE ‘is the variety of English normally used in writing, especially printing’, Cheshire (1999: 131) views SE ‘as primarily a written variety’, and Quirk et al. (1985: 18) note that SE is ‘almost exclusively the language of printed matter.’ An anonymous reviewer even went as far as to note that ‘the term SE can in any case only be referred to the written language.’ It is therefore little surprising that examination of corpus data shows that conversational language and SE display significant differences.
SOME NEGLECTED FEATURES OF CONVERSATION Analyses of large corpora strongly suggest that conversation works largely by rules different from those prescribed by SE grammar. As Biber et al. (1999: 1050) note: ‘Conversation employs a vernacular range of expression.’ A considerable body of research has been gathered in the past two decades or so highlighting in unprecedented detail those features by which the language of conversation is essentially distinguished from other registers and for which existing models of description, including SE, were found unsatisfactory. Beside the numerous differential analyses of features distinctive of conversation laid out in the LGSWE (Biber et al. 1999) and the CGE (Carter and McCarthy 2006), it may suffice to refer the reader to research, for example, on the GET-passive (Collins 1996; Carter and McCarthy 1999), conversational ‘speech reporting’ (McCarthy 1998; Ru¨hlemann 2007: ch. 6), vague language (Channell 1994; Overtstreet and Yule 1997; Cutting 2007), tails and headers (Aijmer 1989; Carter and McCarthy 1999), and the discourse marker like (Miller and Weinert 1998; Adolphs and Carter 2003; Levey 2003). In the light of this large body of evidence it is not necessary in the following to prove that distinctive features of conversation are largely at odds with SE; we may, instead, take this for granted. The intention is merely to illustrate that conversation is essentially distinguished from SE in order to base the subsequent discussion of implications for EFL teaching on ground that is shared with the reader. In the following, we will look at a few selected conversational features that have to date seen little detailed analysis: the phrase I says used in presentations of extended conversations and aphetic forms such as yeah and cos.
676
A REGISTER APPROACH TO TEACHING CONVERSATION
All illustrative examples in the following section are taken from the demographically-sampled subcorpus of the British National Corpus (BNC). This subcorpus comprises four million words (cf. Aston and Burnard 1998). It contrasts with the context-governed subcorpus, the second spoken subcorpus in the BNC, which comprises six million words: while the context-governed subcorpus contains data from a broad range of public speech events such as lectures, radio-phone-ins, sales-demonstrations, etc., it is widely agreed that the data contained in the demographically-sampled subcorpus is informal everyday conversation (e.g. Crowdy 1995: 224; Rayson et al. 1997: 133; Aston and Burnard 1998: 28; Biber et al. 1998: 14, 1999: 28).
I says The phrase I says is interesting for three reasons. To begin with, the use of the third-person -s morpheme with the first-person subject I is, from an SE perspective, a clear case of subject–verb ‘discord’. Further, I says is frequent in British conversation (cf. Biber et al. 1999: 191; Cheshire 1999: 138; Carter and McCarthy 2006: 823; Ru¨hlemann 2007: 169 ff.). The conversational subcorpus of the BNC attests the form in 911 occurrences altogether, corresponding to a normed frequency per million words of 228 occurrences; in the written subcorpus, by contrast, the form has a normed frequency of one occurrence only. I says is thus virtually restricted to conversation. Finally, I says seems not just a non-standard variant of SE I say. As is shown in a detailed quantitative analysis in Ru¨hlemann (2007), I says and I say occupy fairly different functional territories. I say can be observed to carry out a large number of different functions—ranging from its use as a response token, as in (1); a focusing discourse marker, as in (2); an emphatic discourse marker, as in (3); and a quotative, as in (4). With regard to this latter, quotative, function it is interesting to note that I say tends to introduce presentations not of actual utterances but potential utterances, that is, utterances which might occur in recurring situations (note the use of the zero conditional indicated by when and then): (1)
(2)
PS04U 4: PS04Y 4: PS04U 4: PS01A 4: PS01B 4: PS01A 4:
(3)
PS01F 4: PS03W 4:
(. . .).Now she gets them and she sells them for ten quid. I say. And they’re not, I mean they’re pretty but they’re not . . . (. . .) So I my opinion, what this government’s doing is stopping stopping the lower class, well I say lower classes, the lo the the [. . .] isn’t it. poorer people, they stop’em getting educated.That I’m I’m almost convinced. Yeah. I wonder how much it would cost the town, like?I know it sounds silly, but I say, the silly things like that are the ones that sometimes . . . are the
CHRISTOPH RU¨HLEMANN
(4)
677
PS000 4: Mm. PS03W 4: ones that are took seriously.(. . .). (. . .) when we put the rubbish out, then Paddy says is that everything and I say well you know it’s not, you know that it’s going to blow up again as soon as you’ve come inside, but don’t worry.
I says, by contrast, functions solely as a quotative. Moreover, I says has its place in a particular type of quotation, namely in the presentation of extended exchanges of a point–counterpoint nature. Consider (5): here, the speaker constructs a conversation she had with Steve. The utterances by Steve and the presenting speaker respectively are mostly very short and they are numerous, requiring frequent insertion of the reporting phrases he says and I says respectively to mark out whose speech is being presented: (5)
Cos he says, Steve says to me, is he in? I says, no. He says, he’s not in? I says, no. And a bit later on I says to him . . . I think he’s at Cadets. He says, he’s not, he’s in. I says, eh? He says, he’s in. And he’s just walked past me. I says, well you could of told me he were in. He says, he’s gone and done summat. I’ll tell you, he’s gone and done summat.
It seems that I says owes its high frequency in conversation to two aspects: unlike I say which is multi-functional, I says is mono-functional, serving solely as a reporting phrase. Second, being preferably used in presentations of multi-turn exchanges which require frequent shifts between the presenting speaker’s anterior speech and a displaced speaker’s anterior speech and, thus, between reporting phrases, the form I says is a welcome alignment to the forms he says and she says with which it frequently alternates. It is welcome both phonologically because the presenting speaker need not produce the two distinct vowel phonemes in says and say but only the one in the form says, and it is grammatically welcome, because the speaker need not mark the reporting verb for both first- and third-person by means of different morphemes. This phonological and grammatical reduction can be seen as an alleviation of the speaker’s processing and production load. The quotative form I says is thus a prime example of the principle of ‘economy of speech’, that is, of how conversational language use is adapted to needs arising from constraints set by the scarcity of planning and processing time in real-time conversation.
Conversational contractions Negative contraction and verbal contraction have been found to be significantly more frequent in conversation than in any other register (Biber et al. 1999: 1129). In addition to these two types of contraction a third type can be observed: the use of particular aphetic forms such as dunno (for don’t know), gonna (for going to), gotta (for got to), innit (for isn’t it), and yeah (for yes). I propose to refer to these forms as ‘conversational contractions’. While strictly avoided in all serious writing, they are very frequent in conversation.
678
A REGISTER APPROACH TO TEACHING CONVERSATION
Table 1: Comparison of raw frequencies of two ‘conversational contractions’ and their full forms
yeah/yes cos/because
Contracted
Full
Ratio
58,706 11,374
17,871 6,600
3.28 1.72
Among the most frequent such forms are unarguably yeah and cos. Table 1 presents the raw frequencies of these contracted forms and the respective noncontracted forms in the conversational subcorpus of the BNC. It is interesting to observe that the contracted forms are, by far, more frequent than the noncontracted SE-conformant forms. While cos is, roughly speaking, almost twice as frequent as its noncontracted counterpart, the ratio for yeah/yes is even more in favour of the contracted form: yeah, with its impressive raw frequency of 58,506 occurrences in the conversational subcorpus, is more than three times as frequent as yes. This very high frequency is no doubt due to the central role that yeah plays as a backchannel, that is, as a response token which is not intended to launch a bid for the turn but rather as a ‘carry-on signal’ (Stenstro¨m 1987) for the current speaker. Because of this ancillary role of backchannels as tokens of listenership, it is not surprising that backchannels are frequently found in overlap (McCarthy 2003: 59). Consider (6) (turns are numbered, overlapping speech is aligned and in curly brackets): (6) 1 2 3 4 5 6 7 8 9 10 11 12
PS1CX 4: PS000 4: PS1D1 4: PS000 4: PS1D1 4: PS000 4: PS1D1 4: PS000 4: PS1CX 4: PS000 4: PS1CX 4: PS000 4:
13 14 15 16 17
PS000 4: PS1CX 4: PS000 4: PS1CX 4: PS000 4:
So . . . yeah. Are you okay? Yeah, I’m fine thanks. {How’s} {So} married life? Smashing! Good! {Yeah.} {Good!} it is. Oh that’s {good!} {Yeah.} I wished I’d have met when I was [laughing] fifteen []! [laugh] It’s usually er case innit? Yeah. You feel you’ve wasted a lot of years {probably} {Yeah.}
CHRISTOPH RU¨HLEMANN
679
18 PS1CX 4: but 19 PS000 4: But only I can say I didn’t know before, I thought I was happy.
Yeah occurs six times in this short extract. While yeah in turn 1 seems to be used in a discourse marking function as a marker of topic transition and yeah in turn 2 is, in conversation analytical terms, a component of a second-pair part of a question–answer adjacency pair (cf. Sacks et al. 1974), all other instances of yeah in the extract are backchannels, three of them occurring in overlap. Casual inspection of conversational texts in the BNC suggests that yeah is indeed very commonly used as a backchannel. Thus, this short extract may suffice to highlight the importance of this item in conversation, an importance which it gains not only by virtue of being one of the 10 most frequent words in conversation (cf. http://www.kilgarriff.co.uk/bnc-readme.html) but also by virtue of its various functions in discourse, the most central being the backchannel function. This prominence both in terms of frequency and function seems to make the item a worthy candidate for inclusion in the ‘conversation’ classroom. The reduced form cos, too, is noticeably more frequent than the full form because. Also, it seems that the two forms are used in different ways and for different purposes. While because is mainly a subordinating conjunction, cos ‘often functions like a coordinating conjunction’ (Carter and McCarthy 2006: 58; see also Biber et al. 1999: 1078). Hence, cos may be placed ‘alongside and and but in their exemplification of the add-on strategy’ (Biber et al. 1999: 1079). Further, the two forms seem to carry out different discourse marking functions. While because mainly ‘introduces clauses of cause and reason’ (Carter and McCarthy 2006: 57), cos also serves in this function but more importantly as a discourse marker indicating that the speaker wishes to add background information (cf. Schleppegrell 1991; Stenstro¨m 1998). Consider: (7)
1
PS0BB 4:
2 3 4 5
PS0BA PS0BB PS0BA PS0BB
4: 4: 4: 4:
That sink must have got a leak in. I think we must have bunged it up with some gunge! [laugh] Cos it’s stopped leaking. [laugh] Cos I haven’t a emptied it for ages.
In turn 3, the speaker is giving a reason why s/he thinks they must have ‘bunged up’ the leak; thus, cos is indeed used as a conjunction coordinating a main clause (turn 1) and a subclause (turn 3) with the two clauses standing in an inverted cause–effect relationship. In turn 5, by contrast, no such relationship can be observed; rather, it seems, cos is used here to signal that the upcoming information is additional information about how the speaker arrived at the inference regarding the leak in turn 1. In that cos introduces sequences that mainly serve to expand utterances by adding thematically related information, this marker seems best explained as a ‘thematic marker’
680
A REGISTER APPROACH TO TEACHING CONVERSATION
contributing to ‘the thematic structure and cohesiveness of texts’ (Schleppegrell 1991: 131). Again, it would seem that the high frequency of cos and its function as a discourse marker suggest that it be covered in teaching the spoken language. The above list of clearly non-standard features of British conversation is obviously by no means exhaustive; it could most easily be extended. A wealth of other non-standard features of conversation have been noted in the literature (see above); for overviews see Biber et al. (1999: 1121–5), Carter (1999), and Cheshire (1999). Thus, there seems to be good reason to suggest that the grammars of conversation and SE do not map on to each other. As a result, the notion of SE, indispensable in talking of the written language, ‘is problematic in talking of the spoken language’ (Biber et al. 1999: 1121).1 What does this imply for EFL?
IMPLICATIONS FOR EFL TEACHING It would appear that this inadequacy of SE for conversation has far-reaching implications. It seems consistent to ask whether SE can still serve as the underlying model for both writing and speech, as has traditionally been the case (see above). It would seem rather that, in teaching the spoken language, which includes first and foremost conversation, SE is an inappropriate model because its grammar differs too much from the grammar of conversation. A more appropriate model would be what has been emerging from recent and current corpus analyses under the headings of ‘spoken grammar’ and ‘conversational grammar’ respectively (for a critique of the notion of ‘spoken grammar’, which equates mode with register, see Ru¨hlemann 2006). Although, as yet, no comprehensive description of conversational grammar has been elaborated, the advances made in this field are no doubt solid enough for teaching to be based on them. If this paper advocates the rejection of SE as a model for teaching the spoken language, this is not to imply that SE should disappear from the English classroom altogether. The fact that it is the model variety underlying serious writing and World Standard English renders SE indispensable in the EFL classroom. It seems, then, that what is called for is a redefinition of the role of SE from the-one-and-only model variety to, as Bex (1993: 261) suggests, ‘a ‘‘core’’ variety’ from which other varieties can be explored. This qualification of SE would give room to teach and explore language not only as it should be used in most writing but also as it is actually used in the core spoken register, conversation. Such a move toward allowing authentic conversation and its language into the classroom, it appears, is best implemented in a wider re-conceptualization of the English that we teach. Traditionally, the ‘E’ in EFL has stood for a kind of general English, that is, a largely SE-conformant monolithic block. This view informs, for example, Quirk et al.’s (1985) seminal Comprehensive Grammar of the English Language, whose aim is ‘to focus on the common core that is shared by standard British English and standard American English’
CHRISTOPH RU¨HLEMANN
681
(Quirk et al. 1985: 33, emphasis added). Corpus linguistic analyses radically diverge from this view. Indeed, one of the most significant outcomes of corpuslinguistic research is the conviction that grammatical patterns differ systematically across different registers (e.g. Biber et al. 1998: 35). As a consequence, the equally seminal Longman Grammar of Spoken and Written English is consistently register-specific exploring how grammar varies across ‘four major registers’ (Biber et al. 1999: 8): academic prose, fiction, news reportage, and conversation. Similarly, the recent Cambridge Grammar of English (Carter and McCarthy 2006) describes important differences between, not registers, but the spoken and written modes. Thus, corpus analyses have uncovered sufficient evidence to suggest that the assumption of a general English is, as Conrad (2000) argues, a ‘myth’. As a consequence, Conrad (2000: 549) predicts that corpus linguistics will revolutionize the teaching of grammar in that, inter alia, ‘[m]onolithic descriptions of English grammar will be replaced by registerspecific descriptions.’ I will refer to this shift in emphasis toward English as its registers as the ‘register approach’. We should be aware though that the problems entailed by implementing the register approach are considerable. I will outline some of them in the following discussion.
ISSUES ARISING Shifting the emphasis in the EFL classroom from general English to registersensitive descriptions of English with a smaller, yet still central, role given to SE and an equally central role awarded to the register of conversation no doubt constitutes a fundamental change; as such, it is small wonder that it entails difficulties that are by no means negligible. To start with, it would appear that a fundamental prerequisite to teaching authentic conversation in the EFL classroom is a change in attitude toward conversation. There is good evidence to suggest that the spoken language has traditionally been seen as ‘an ill-formed variant of writing’ (Hewings and Hewings 2005: 216) and that this tradition is still going strong not only in linguistics but also in EFL classrooms. Certain terminologies used widely in the description of conversational key features bear witness to this. The terms ‘dysfluency’ and ‘dislocation’ are prominent examples. ‘Dysfluency’ refers to speech management phenomena such as pauses (filled and silent), restarts, repetition, etc. while ‘dislocation’ refers to the use of noun phrases in clause-peripheral position, with a co-referent pronoun in the core of the clause (Biber et al. 1999: 956). Example (8) is an illustration of what is commonly referred to as ‘left dislocation’ (the ‘dislocated’ element is in bold, the co-referent pronoun in square brackets): (8)
Them on the bottom [they]’re bound to get their feet wet aren’t they?
The problem with these terminologies lies in the prefixes dis- and dys-. The negation they convey suggests that the concepts presuppose the expected
682
A REGISTER APPROACH TO TEACHING CONVERSATION
opposite—syntactic integration, in the case of ‘dislocation’, and uninterrupted delivery, or ‘eufluency’, in the case of ‘dysfluency’. Both syntactic integration and eufluency are obviously characteristics of written texts but very rarely, if at all, found in conversation. Thus, when used for the description of conversational features, the concepts of ‘dislocation’ and ‘dysfluency’ describe, less the features themselves, but rather the fact that the features fail to satisfy the expectations raised by written standards and by the situational factors that underlie writing (such as the abundance of planning and editing time which helps writers achieve eufluency). As a result, ‘dislocation’ and ‘dysfluency’ tacitly devalue the conversational features thus labelled (Ru¨hlemann 2006; cf. also Carter and McCarthy 1995; Miller and Weinert 1998). This ideological bias is likely to extend to perceptions of the public at large (Hughes 2002: 14). Thus, many teachers are likely to perceive the advent of conversational grammar as a threat to dearly held habits and convictions. To them, conversational grammar may simply be ‘bad grammar’ and, hence, not worth teaching. It would appear that, in order to remedy this basic prejudice, a functional approach in a Hallidayan sense can help, that is, an approach that ‘build[s] in the situation as an essential ingredient’ (Halliday 1978: 29; for a situationfunctional account of the particularly ‘bad’ quotative I goes see Ru¨hlemann 2008). For situation-sensitive functional analyses can show that, for example, the features subsumed under ‘dislocation’, far from merely being stranded in aberrant positions, as the terminology suggests, fulfil crucial functions in discourse and interaction benefiting both the speaker and the recipient in multiple ways (cf. Quirk et al. 1985; Carter and McCarthy 1995). Obviously, other conversational features also lend themselves to a situation-functional reading. Generalized I says, for example, can be seen as primarily serving to help speakers economize what is, in spontaneous conversation, scarce and hence sought, time to process and plan ahead (cf. Biber et al. 1999: 1041–52; Leech 2000; see Ru¨hlemann 2007 for a principled attempt to view conversational grammar as adapted to constraints arising from the conversational situation type). A second problem arising from the register approach is intimately linked up to this first problem in that it too concerns teacher attitudes. It may not only be hard to persuade teachers to accept, beside SE, conversational grammar as another model variety but at least equally hard to convince them that the notion of ‘correctness’ is, in a register approach, of limited use. Indeed, as Conrad (2000) argues, this notion, which is closely associated with SE, may have to give way to the notion of ‘appropriateness’. This notion refers to the set of contextual conditions that determine the use of alternative structures. That is, in a register approach, what is appropriate depends on the register and the specific set of conditions in that register constraining the use of the form in question. The problem arising is less that correctness may be a dearly held notion that is hard to dispense with than rather that appropriateness is more difficult to handle. A monolithic view of English whose underlying model variety is SE enables the teacher to teach language
CHRISTOPH RU¨HLEMANN
683
in a simple either/or fashion—either it conforms to the rules of SE and is hence correct or it does not and hence is incorrect. In this view, contextual factors reaching beyond the confines of the sentence are disregarded as irrelevant. In a register-sensitive view of English, by contrast, consideration of above-sentence and non-linguistic context is quintessential. To illustrate this point: if a learner uses the form I says in academic writing, it will have to be judged inappropriate because, as noted above, I says is virtually restricted to conversation and, thus, ill-placed in a serious written register such as academic writing; if, however, the learner uses the form as part of an informal account of a lively point–counterpoint conversation he or she had the day before, the form might be accepted and even appreciated by the teacher as being particularly idiomatic and adequate to the register. If the learner uses in the same context (an account of an anterior point– counterpoint conversation) SE I say in frequent alternation with he/she says or he/she said, this use might be considered at least questionable because, as noted above, the quotative function of I say is a marginal one in conversation and the form typically occurs in potential-speech presentations but not in actual-speech presentations. Thus, clearly, the notion of appropriateness is a complex one because it views language integrated into a wealth of context, while the notion of correctness views language largely out of context. As a result, both teaching and learning English in a register approach is a more complex task. For a good many teachers the notion of appropriateness will therefore be less attractive, particularly if they tend to regard variation first and foremost as a nuisance (cf. Conrad 2004: 68; Sinclair 2004b: 274). Others, it might be assumed, may welcome the register approach (and, thus, the notion of appropriateness that goes with it) because this approach is more likely to get EFL teaching anywhere close to reflecting the linguistic richness and functional diversity of real language use. The increased complexity that a register approach entails for the learners is no doubt a major issue that calls precisely for what is, according to Widdowson (2000), the raison d’eˆtre of applied linguistics: mediation. That is, the issue at stake is the issue of ‘how to simplify and stage the language presented to learners, and to simplify the rules used to explain it, in a way which will enable them to come gradually closer to native speaker use (if that is their goal)’ (Cook 1998: 61). Thus, it would seem that, for example, the intricacy of how I say and I says are used across registers demands that treatment of these features be postponed until advanced levels; on the other hand, it is hard to see why the backchannel yeah should not be introduced at a very early learning stage. Third, those teachers willing to familiarize learners with real conversation face the problem of how to illustrate this register. As Carter (1998) and Gilmore (2004) show, there is strong evidence to suggest that textbook dialogue is far removed from authentic interaction in that essential features of the latter are notably missing in the former, such as ‘spoken grammar’ features, filled and unfilled pauses, overlap, restarts, etc.; the type of ‘conversation’ most textbooks
684
A REGISTER APPROACH TO TEACHING CONVERSATION
present cannot serve as a reliable model for the teaching of conversation (see also Carter and McCarthy 1995: 154). Some recent textbooks, however, seem to be catching up (cf. also Gilmore 2004). Cambridge University Press’s very recent Touchstone series (e.g. McCarthy et al. 2006) is no doubt a major break-through in the history of English textbooks in that this series is consistently corpus-informed placing a major emphasis on conversational grammar. Still, however, apart from this ground-breaking textbook, and a few more modestly innovative textbooks, the serious dearth of corpus-informed teaching materials for use in the ‘conversation’ classroom persists. That it persists is convincingly shown in a recent survey of 24 EFL textbooks conducted by Cullen and Kuo (2007). The richest source for illustrating authentic conversation and its grammar in the classroom would unarguably be corpora. Here, much applied corpuslinguistic research has recently explored ways of using corpora in teaching (e.g. Wichmann et al. 1997; Burnard and McEnery 2000; Aston et al. 2004; Sinclair 2004a; O’Keeffe et al. 2007). Although intriguing approaches have been outlined there are serious practical problems involved in taking corpora to the classroom. For using corpora in EFL requires that costly computer facilities be in place, which in most parts of the world cannot even be thought of; however, their availability in classrooms seems to be spreading and has already become standard in some countries. Further, using corpora is not easy. On the contrary, it is agreed that asking the corpus sensible questions alone places high demands on the corpus skills of the teacher and the learners, not to mention the difficulties involved in making good sense of the data thus generated on the screen (cf. Aston and Burnard 1998; Mauranen 2004a: 98). It seems that before teachers and, thus, learners have generally acquired a reasonable degree of what Mukherjee (2002: 179) terms ‘corpus literacy’ there is still some way to go. One important help in that direction is no doubt the inclusion of corpus linguistics in initial teacher education (cf. O’Keeffe and Farr 2003). In spite of the difficulties, Mauranen’s (2004b: 208) dictum that ‘We need spoken corpora for teaching the spoken language’ has intuitive appeal. Finally, it is as yet quite unclear if, and how, real conversation should be taught at all. Influential voices in applied linguistics have suggested that invented language serves the language learner better than authentic corpus data given that in a pedagogic context authenticity is less a property of text than part and parcel of the learners’ ability to authenticate the text (e.g. Cook 1998, 2001; Widdowson 2000, 2003). The controversy somehow seems to rest on the tacit assumption that corpus data per se cannot be authenticated by language learners. This assumption can easily be countered. As Mauranen notes: Authenticity as a learner response is . . . like other classroom practices, socially negotiable and historically changeable. Educators can promote learners’ acceptance of corpora in the classroom as
CHRISTOPH RU¨HLEMANN
685
authentic data—or reduce it, but it is a matter of conscious pedagogic choice, not a law of nature (Mauranen 2004a: 93). Thus, authenticity in Widdowson’s sense does not depend on the text being invented by a materials designer or captured in a spoken corpus, but on the successful mediation through careful selection and motivating teaching. Another argument against using authentic conversational language in the classroom has been that it is ‘inarticulate, impoverished, and inexpressive’ (Cook 1998: 61). This is certainly a valid observation, at least from a written SE-informed perspective, with regard to many conversational extracts. It is, interestingly, also a valid observation with regard to textbook language. Schewe and Shaw (1993: 9), for example, deplore that ‘[m]any of the textbooks and coursebooks in current use are still full of banal, painfully obvious and often dull dialogues and such texts.’ Cook (2000: 157 ff.) himself criticizes the preference in language teaching materials for ‘the mundane transactional discourse of modern work, rather than the ancient playful discourse concerning intimacy and power’, arguing that it is the latter type of language which is likely to ‘stimulate interest in language learning’ (Cook 2000: 160). So, in terms of ‘dullness’, invented textbook language is easily on a par with authentic corpus language. Yet, while textbooks are limited in size thus presenting only a small number of texts to choose from, today’s spoken corpora are virtually infinite resources, in which it is indeed easy to retrieve texts that are more likely to stimulate interest and activate learning. [The reader might re-consider extract (7) above for illustration.] Further, among those who do advocate the use of authentic data in the EFL classroom there is still little agreement as to what methodologies are best suited to teaching conversational grammar, that is, what established methodologies can be taken on board and what methodologies must be constructed from scratch. Carter and McCarthy (1995), for example, suggest that the traditional P–P–P (presentation, practice, production) methodology cannot capture the subtleties of conversational grammar and should be replaced by an I–I–I methodology, based essentially on ‘observation, awareness and induction’ (Carter and McCarthy 1995: 155). The three Is stand for ‘illustration’, preferably through authentic data samples; ‘interaction’ through discussions about language features observed in the samples; and ‘induction’, that is, ‘making one’s own, or the learning group’s, rule for a particular feature’ (Carter and McCarthy 1995: 155). What is notably missing in this methodological framework is any explicit opportunity for the learner to produce the target language. However, as, for example, Mauranen (2004a: 98) notes, it is well known from research into second language acquisition that an analytic awareness alone does not lead to acquisition; rather, acquisition depends crucially on ‘opportunities for applying the incipient knowledge and skills to meaningful tasks’ (Mauranen 2004a: 103). It would appear questionable therefore whether the three Is methodology may indeed lead ‘to a more rapid acquisition by learners of fluent, accurate, and naturalistic conversational and
686
A REGISTER APPROACH TO TEACHING CONVERSATION
communicative skills’ (McCarthy and Carter 1995: 217).2 A plausible methodological proposal comes from Timmis (2005: 119), who argues with regard to authentic spoken texts, ‘that, wherever possible, the text should be listened to in the first instance’. Corpora, however, present text, written or spoken, only in written form, recent major corpora such as the BNC even present it in orthographic written form without prosodic annotation. Thus, before we can implement Timmis’s proposal, spoken corpora need to be complemented by sound concordancing in synchrony with the transcript (cf. Mauranen 2004a: 92), a technological innovation much to be desired particularly in a pedagogic context given the undisputed communicative value of prosody (cf. Brazil 1985). On the whole, it seems that discussion of what methodologies are best suited for the teaching of conversational grammar has only just begun and a lot of further applied linguistic enquiry, discussion, and development is needed.
POSSIBLE GAINS The problems involved in reducing the role of SE and teaching conversational grammar in the wider context of a register approach are no doubt daunting. Sceptical observers may therefore ask, why take the trouble at all? Why not ignore in the EFL classroom what corpus analyses have uncovered regarding conversation? To answer these questions we might want to consider the following possible gains. First, efforts to keep authentic conversation and corpus-linguistic knowledge of conversation out of EFL teaching would be unwise because conversation, both as a variety of native-speaker language use and a variety in the EFL classroom, is not just any register but, in either domain, the central one. As is agreed by a broad range of linguists, L1 conversation has a special status among registers, spoken and written, for a number of reasons. To begin with, conversation is ‘the most common, and, it would appear, the most fundamental condition of ‘‘language use’’ and discourse’ (Schegloff 1979: 283) in that, unlike any other register, it is received and produced by virtually every speaker (see also Goodwin and Heritage 1990: 298; Goodwin and Duranti 1992: 22). Further, conversation is regarded as the archetypical register from which other registers, including written ones, are departures (Halliday and Hasan 1989: 11; Goodwin and Heritage 1990: 298; Biber et al. 1999: 1038). Finally, conversation is the most ‘creative’ register—the ‘laboratory for linguistic innovation’, as Hughes (2002: 15) puts it—in that, here, ‘the semogenic potential of a language is most likely to get extended’ (Halliday 2006: 294). Similarly, L2 conversation in language learning has a special status in that (i) conversation-like ‘dialogue’ is unarguably a backbone of EFL textbooks, which, in turn, are the backbones of most EFL courses (cf. Ro¨mer 2004, 2005), and (ii) in the Communicative Approach, which is a good generation after its inception still the overarching methodology in EFL,
CHRISTOPH RU¨HLEMANN
687
the spoken language has strongly been upheld ‘as the primary source of language acquisition and is fostered and facilitated as a central activity in the language classroom’ (Hughes 2002: 49). Because of this centrality of conversation both in L1 and L2 contexts, it would seem consistent to argue that language pedagogy that aims to support the acquisition of speaking skills should take into account what we know about authentic conversation. Finally, excluding authentic conversation would be counterproductive because there is a chance that including it might make language learning more effective. Complaints about EFL teaching being ineffective are common. Mindt (1996: 232) critizes that ‘learners who leave their school surroundings very often find it hard to adapt to the English used by native speakers’ and Bex (1993: 257) deplores the fact that it frequently comes as a shock to non-native speakers that the use of SE, although widely intelligible, ‘does not always trigger equally intelligible responses’. The reasons for these shortcomings are no doubt manifold and complex. Space considerations prevent a comprehensive discussion. One reason, however, that is particularly pertinent to the present connection is worth elaborating. As noted earlier, a growing number of comparative analyses on certain lexicogrammatical features suggest that the English taught is in stark contrast with the English actually spoken. The evidence for this claim is massive. To economize on space, a few examples will have to suffice to illustrate this. Mindt (1996) scrutinizes the textbook representation of (i) modal verbs such as can, will, must, may, shall, etc., (ii) conditional clauses and (iii) future time orientation through will and going to respectively in juxtaposition to the (spoken) London–Lund Corpus. Mindt (1997) compares how (i) any, (ii) will and would, and (iii) irregular verbs are treated in textbooks and how they are used in a variety of corpora. Conrad (2004) studies the linking adverbial though as represented in textbooks and corpora. Ro¨mer (2005) investigates how forms, functions, and contexts of progressives are taught in German textbooks and used in spoken corpora respectively. All of these studies reveal a clear mismatch between the corpus evidence and what is covered in the textbooks: either the textbooks represent less frequent features at earlier stages than more frequent ones, that is, in an order which disregards their importance in actual discourse, or they fail to represent them at all. For example, Mindt (1996) found that textbooks postpone treatment of will in favour of the much less frequent modal verbs must and may; both Mindt (1996) and Ro¨mer (2004) found that the most frequent conditional, characterized by the simple present-simple present sequence of tenses, was by far the most frequent in the corpora consulted, but was not covered in the textbooks at all. These lexico-grammatical differences between school English and real English must be considered important. To make matters worse, the differences do not stop here; they extend to vital discourse features, such as discourse markers, backchannels, and what is commonly referred to as ‘speech reporting’. Treatment of conversational discourse markers is missing from most
688
A REGISTER APPROACH TO TEACHING CONVERSATION
published textbooks, although the ubiquity of discourse markers in actual conversation (cf. McCarthy 1998: 59) and the functional centrality of discourse markers as devices to create discourse coherence (e.g. Schiffrin 1987) would warrant coverage in good detail. A similar omission is the, arguably, complete lack of coverage in textbooks of backchannels, which are, like discourse markers, omnipresent in actual conversation and functionally central as tokens of active listenership and co-construction. Speech reporting, finally, is covered extensively in textbooks but, it seems, with an exclusive focus on written reporting (Ru¨hlemann forthcoming). That is, textbooks present it with an obsession for backshift and sequence of tenses with indirect speech thus utterly neglecting the characteristics of speech presentation in conversation, such as a preference for direct reports over indirect reports, the use of reporting verbs such as GO and BEþlike, reporting forms such as past -ing and I says, and utterance openers such as oh and well (cf. McCarthy 1998: ch. 8; Ru¨hlemann 2007: ch. 6). Since the very high frequency of presentations renders them an essential building block of everyday conversation, the preoccupation in EFL teaching with reporting features characteristic of writing is very unlikely to foster speaking skills and hence communicative success. Although detailed comparative studies on these higher-order phenomena are still scarce it may safely be assumed that none of them are covered in sufficient detail in the vast majority of coursebooks commonly in use in EFL classrooms–maybe the most striking exceptions being Cambridge University Press’s above-mentioned Touchstone series (e.g. McCarthy et al. 2006) and Carter et al. (2000), a corpus-informed grammar reference and practice book for intermediate to advanced learners, which covers spoken discourse markers in rich detail and also conversational speech presentation but not, for example, backchannels. Given these significant differences on both the lexico-grammatical level and the discourse level, it seems consistent to question whether it is advisable to continue giving learners the impression that by learning SE ‘they are learning English tout court’ (Bex 1993: 253). The SE-based variety of English learners are typically taught is obviously out of kilter with the variety native-speakers speak in everyday encounters; it thus fails to prepare the learners adequately for communication with them. This mismatch reveals another mismatch. As Widdowson (2000: 8) notes, ‘for many learners at least, the language as realized by its users is the goal to which they aspire and to which they will seek to approximate by the process of gradual authentication.’ So, the mismatch between school English and spoken English amounts to a mismatch between the end and the means deployed to reach it: SE-based school English fails to support learners in reaching their goal—to approximate to authentic English. To sum up this section, it would seem that, because the teaching of the spoken language is still rooted in SE-based descriptions of the grammar of writing (Carter and McCarthy 1995: 141) it produces (at best) ‘speakers of English who can only speak like a book’ (McCarthy and Carter 1995: 207). Exposure to authentic conversation, by contrast, can enhance fundamental
CHRISTOPH RU¨HLEMANN
689
speaking skills. It would appear that teaching learners how to use backchannels—such as yeah—and discourse markers—such as cos—appropriately might indeed help to make their speech more natural and idiomatic. Also, it seems safe to assume that learners who are familiar not only with SE but also other varieties, most notably conversational English, will have less difficulty adapting to a broader variety of interlocutors and situations and, hence, be more effective in communication. However, it is important to admit that claims to greater fluency, naturalness, and more communicative success through exposure to authentic language have to date not been substantiated by empirical evidence. Therefore, verifying (or falsifying) them would be a useful objective for future applied corpus-linguistic research.
CONCLUSIONS This paper aims to question the role of SE as the major model for both writing and speech. By way of briefly discussing a number of neglected features which are both distinctive of conversation and clearly at odds with SE, I concluded that authentic speech, particularly in its most common form, conversation, diverges from SE to such an extent that using SE as the underlying model for the teaching of speech and particularly conversation is problematic and that, hence, in teaching the spoken language, SE as the underlying model variety should be replaced by conversational grammar. I suggested that such a qualified farewell to SE is best implemented in a register approach, that is, in the wider context of shifting the emphasis from a monolithic view of English to a register-sensitive view of English. I outlined some issues arising from such an approach, such as the negative value judgements associated with conversational language, the need to replace the notion of correctness by the notion of appropriateness, the scarcity of corpus-informed teaching materials including the difficulties involved in using corpora in the classroom, and the as-yet unanswered question of which methodologies are best suited to the teaching of authentic conversation. I also sketched out some of the possible gains of the register approach, arguing that teaching authentic conversation is more conducive to the learners’ principal goal of approximating to the language as realized by its users. Undoubtedly, it should be stressed again, ‘applying’ in EFL teaching what corpus linguistic analyses have taught us about the spoken language and, in particular, conversation demands enormous applied linguistic efforts. Will the efforts pay? Given that the integration of conversational grammar in teaching materials and EFL courses is still in its very infancy, a definitive answer is at present not possible. The prospects are good, however, that the efforts will not be in vain. It appears that teaching the spoken language on the basis of conversational grammar as the underlying model while continuing to base the teaching of the written language on the model of SE may be conducive to learning because, by adopting a register-sensitive approach, we would not
690
A REGISTER APPROACH TO TEACHING CONVERSATION
only make an important contribution to bringing school English into closer correspondence with the language actually spoken but also get closer to redressing the balance in the EFL classroom between speech and writing, a balance which has traditionally been in favour of writing. Finally, by taking a register-sensitive approach, we would portray the target language in a more adequate way, throwing into profile one of the most fundamental properties of language, its heterogeneity (Stubbs 1993: 11). Before we will be able to reap these gains, however, much work indeed lies ahead: the work of applied linguistic mediation. Given that, as said earlier, corpus research has triggered fundamental, indeed ‘revolutionary’ changes in key areas of linguistic enquiry, the need for this work to be undertaken is possibly more pressing than ever.
NOTES 1 This is a somewhat generalized observation because, strictly speaking, it applies first and foremost to conversation, the only spoken register accounted for in Biber et al. (1999); other spoken registers such as sermons, lectures, or political speeches may well be much closer to SE and the notion of SE may hence be much less problematic in talking of these registers.
2 Once again, the Touchstone series indicates remarkable progress in that it gives learners ample opportunity, not only for noticing and induction tasks related to conversational grammar features such as reportive past –ing, but also for production (see McCarthy et al. 2006: 90 ff.).
REFERENCES Adolphs, S. and R. Carter. 2003. ‘And she’s like it’s terrible, like: Spoken Discourse, Grammar, and Corpus Analysis,’ International Journal of English Studies 3/1: 45–56. Aijmer, K. 1989. ‘Themes and tails: The discourse functions of dislocated elements’. Nordic Journal of Linguistics 12: 137–54. Aston, G. and L. Burnard. 1998. The BNC Handbook. Exploring the British National Corpus with SARA. Edinburgh: Edinburgh University Press. Aston, G., S. Bernardini, and D. Stewart. (eds). 2004. Corpora and Language Learners. Amsterdam/Philadelphia: John Benjamins. Baker, M., G. Francis, and E. Tognini-Bonelli. (eds). 1993. Text and Technology. In Honour of John Sinclair. Amsterdam: Benjamins. Bex, T. 1993. ‘Standards of English in Europe,’ Multilingua 12/3: 249–264.
Bex, T. and R. J. Watts. (eds). 1999. Standard English. The Widening Debate. London: Routledge. Biber, D., S. Conrad and R. Reppen. 1998. Corpus Linguistics: Investigating Language Structure and Use. Cambridge: Cambridge University Press. Biber, D., S. Johansson, G. Leech, S. Conrad, and E. Finegan. 1999. Longman Grammar of Spoken and Written English. Harlow: Pearson Education Limited. Brazil, D. 1985. The Communicative Value of Intonation in English. Birmingham: English Language Research, University of Birmingham. Burnard, L. and T. McEnery. (eds). 2000. Rethinking Language Pedagogy from a Corpus Perspective: Papers from the third international conference on teaching and language corpora. Frankfurt/M: Lang.
CHRISTOPH RU¨HLEMANN
Carter, R. A. 1998. ‘Orders of reality: CANCODE, communication, and culture,’ ELT Journal 52/1: 43–56. Carter, R. A. 1999. ‘Standard grammars, spoken grammars: Some educational implications’ in T. Bex and R. J. Watts (eds): Standard English. The Widening Debate. London: Routledge, pp. 149–166. Carter, R. A. and M. J. McCarthy. 1995. ‘Grammar and the spoken language,’ Applied Linguistics 16/2: 141–58. Carter, R. A. and M. J. McCarthy. 1997. Exploring Spoken English. Cambridge: Cambridge University Press. Carter, R. A. and M. J. McCarthy. 1999. ‘The English get-passive in spoken discourse: Description and implications for an interpersonal grammar,’ English Language and Linguistics 3/1: 41–58. Carter, R. A. and M. J. McCarthy. 2006. Cambridge Grammar of English. Cambridge: Cambridge University Press. Carter, R., R. Hughes and M. McCarthy. 2000. Exploring Grammar in Context. Cambridge: Cambridge university Press. Channell, J. 1994. Vague Language. Oxford: Oxford University Press. Cheshire, J. 1999. ‘Spoken standard English’ in T. Bex and R. J. Watts (eds): Standard English. The Widening Debate. London: Routledge, pp. 129–48. Collins, P. 1996. ‘Get-passives in English,’ World Englishes 15/1: 43–56. Conrad, S. 2000. ‘Will corpus linguistics revolutionize grammar teaching in the 21st century?’ TESOL Quarterly 34/3: 548–60. Conrad, S. 2004. ‘Corpus linguistics, language variation, and language teaching’ in J. McH. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam/Philadelphia: John Benjamins, pp. 67–85. Cook, G. 1998. ‘The uses of reality: A reply to Ronald Carter,’ ELT Journal 52/1: 57–62. Cook, G. 2000. Language Play, Language Learning. Oxford: Oxford University Press. Cook, G. 2001. ‘ ‘‘The philosopher pulled the lower jaw of the hen’’: ludicrous invented sentences in language teaching,’ Applied Linguistics 22/3: 366–87. Crowdy, S. 1995. ‘The BNC spoken corpus’. in Leech, G., G. Myers and J. Thomas (eds): Spoken English on Computer. Transcription, Mark-up and Application. Harlow: Longman, pp. 224–34.
691
Crystal, D. 2003. The Cambridge Encyclopedia of the English Language. 2nd edn. Cambridge: Cambridge University Press. Cullen, R. and I.-C. Kuo. 2007. ‘Spoken grammar and ELT course materials: a missing link?’ TESOL Quarterly 41/2: 361–86. Cutting, J. (ed.) 2007. Vague Language Explored. Basingstoke: Palgrave Macmillan. Duranti, A. and C. Goodwin. (eds). 1992. Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Gilmore, A. 2004. ‘A comparison of textbooks and authentic interactions,’ ELT Journal 58/4: 363–74. Givo´n, T. (ed.) 1979. Discourse and Syntax. New York: Academic Press. Goodwin, C. and A. Duranti. 1992. ‘Rethinking context: an introduction’ in A. Duranti and C. Goodwin (eds): Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press, pp. 1–42. Goodwin, C. and J. Heritage. 1990. ‘Conversation analysis,’ Annual Review of Anthropology 19: 283–307. Halliday, M. A. K. 1978. Language as Social Semiotic. The Social Interpretation of Language and Meaning. London: Edward Arnold. Halliday, M. A. K. 2006. ‘Afterwords’ in G. Thompson and S. Hunston (eds): System and Corpus: Exploring Connections. London/Oakville: Equinox, pp. 293–9. Halliday, M. A. K. and C. Matthiessen. 2004. An Introduction to Functional Grammar. 3rd edn. London: Edward Arnold. Halliday, M. A. K. and R. Hasan. 1989. Language, Context, and Text: Aspects of Language in a Socialsemiotic Perspective. 2nd edn. Oxford: Oxford University Press. Hasselgard, H. and S. Oksefjell. 1999. Out of Corpora. Studies in Honour of Stig Johansson. Amsterdam and Atlanta, GA: Rodopi. Hewings, A. and M. Hewings. 2005. Grammar and Context. London/New York: Routledge. Hughes, R. 2002. Teaching and Researching Speaking. Harlow: Pearson Education. Jucker, A. H. and Y. Ziv. (eds). 1998. Discourse Markers. Descriptions and Theory. Amsterdam/ Philadelphia: John Benjamins. Leech, G. 2000. ‘Grammars of spoken English: new outcomes of corpusoriented research,’ Language Learning 50/4: 675–724.
692
A REGISTER APPROACH TO TEACHING CONVERSATION
Levey, S. 2003. ‘He’s like ‘‘Do it now!’’ and I’m like ‘‘No!’’, some innovative quotative usage among young people in London,’ English Today 73, 19/1: 24–32. Malinowski, B. 1923. ‘The problem of meaning in primitive languages’ in Ogden, C. K. and I. A. Richards (eds): The Meaning of Meaning. London: Routledge, pp. 296–336 McCarthy, M. J. 1998. Spoken Language and Applied Linguistics. Cambridge: Cambridge University Press. McCarthy, M. J. 2003. ‘Talking back: ‘‘small’’ interactional response tokens in everyday conversation,’ Research on Language and Social Interaction. 36/1: 33–63. McCarthy, M. J. and R. A. Carter. 1995. ‘Spoken grammar: What is it and how can we teach it?’ ELT Journal 49/3: 207–18. McCarthy, M. J., J. McCarten, and H. Sandiford. 2006. Touchstone. Student’s Book 4. Cambridge: Cambridge University Press. Mauranen, A. 2004a. ‘Spoken corpus for an ordinary learner’ in J. McH. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam: John Benjamins, pp. 89–105. Mauranen, A. 2004b. ‘Speech corpora in the classroom’ in G. Aston, S. Bernardini, and D. Stewart (eds): Corpora and Language Learners. Amsterdam/Philadelphia: John Benjamins, pp. 195–211. Meijs, W. (ed.). 1987. Corpus Linguistics and Beyond. Amsterdam: Rodopi. Miller, J. and R. Weinert. 1998. Spontaneous Spoken Language: Syntax and Discourse. Oxford: Clarendon Press. Mindt, D. 1996. ‘English corpus linguistics and the foreign language teaching syllabus’ in J. Thomas and M. Short (eds): Using Corpora for Language Research, London: Longman, pp. 232–47. Mindt, D. 1997. ‘Corpora and the teaching of English in Germany’ in A. Wichmann, S. Fligelstone, T. McEnery, and G. Knowles (eds): Teaching and Language Corpora. London: Longman, pp. 40–50. Mukherjee, J. 2002. Korpuslinguistik und Englischunterricht: Eine Einfu¨hrung (Sprache im Kontext series 14). Frankfurt: Peter Lang. O’Keeffe, A. and F. Farr. 2003. ‘Using language corpora in initial teacher education: Pedagogic issues and practical applications,’ TESOL Quarterly 37/3: 389–418.
O’Keeffe, A., M. J. McCarthy, and R. Carter. 2007. From Corpus to Classroom. Language Use and Language Teaching. Cambridge: Cambridge University Press. Overstreet, M. and G. Yule. 1997. ‘On being inexplicit and stuff in contemporary American English,’ Journal of English Linguistics 25/3: 250–8. Quirk, R., S. Greenbaum, G. Leech, and J. Svartvik. 1985. A Comprehensive Grammar of the English Language. London: Longman. Rayson, P., G. Leech and M. Hodges 1997. ‘Social differentiation in the use of English vocabulary: some analyses of the conversational component of the British national corpus,’ Language 50/4: 696–735. Romaine, S. and D. Lange. 1991. ‘The use of like as a marker of reported speech and thought: A case of ongoing grammaticalization in progress,’ American Speech 66/3: 227–79. Ro¨mer, U. 2004. ‘Comparing real and ideal language learner input: The use of an EFL textbook corpus in corpus linguistics and language teaching’ in G. Aston, S. Bernardini, and D. Stewart (eds): Corpora and Language Learners. Amsterdam/Philadelphia: John Benjamins. Ro¨mer, U. 2005. Patterns, Progressives, Pedagogy: A Corpus-driven Approach to English Progressive Forms, Functions, Contexts and Didactics. Amsterdam/Philadelphia: John Benjamins. Ru¨hlemann, C. 2006. ‘Coming to terms with conversational grammar: ‘‘Dislocation’’ and ‘‘dysfluency’’,’ International Journal of Corpus Linguistics 11/4: 385–409. Ru¨hlemann, C. 2007. Conversation in Context: A Corpus-driven Approach. London: Continuum. Ru¨hlemann, C. 2008. ‘Conversational grammar—bad grammar? A situation-based account of I goes in the BNC,’ ICAME Journal 32: 157–77. Ru¨hlemann, C. (forthcoming). ‘I’m like aargh: A comparison of discourse presentation in authentic conversation and EFL textbooks’. Sacks, H., E. A. Schegloff, and G. Jefferson. 1974. ‘A simplest systematics for the organisation of turn-taking for conversation,’ Language 50/4: 696–735. Schegloff, E. A. 1979. ‘The relevance of repair for syntax-in-conversation’ in T. Givo´n (ed.): Discourse and Syntax. New York: Academic Press, pp. 261–86.
CHRISTOPH RU¨HLEMANN
Schewe, M. and P. Shaw. (eds). 1993. Towards Drama as a Method in the Foreign Language Classroom. Frankfurt am Main: Peter Lang. Schiffrin, D. 1987. Discourse Markers. Cambridge: Cambridge University Press. Schleppegrell, M. 1991. ‘Paratactic because,’ Journal of Pragmatics 16: 323–37. Sinclair, J. McH. (ed.). 2004a. How to Use Corpora in Language Teaching. Amsterdam/Philadelphia: John Benjamins. Sinclair, J. McH. (ed.). 2004b. ‘New evidence, new priorities, new attitudes’ in J. McH. Sinclair (ed.): How to Use Corpora in Language Teaching. Amsterdam/Philadelphia: John Benjamins, pp. 271–99. Stenstro¨m, A. 1987. ‘Carry-on signals in English conversation’ in W. Meijs (ed.): Corpus Linguistics and Beyond. Amsterdam: Rodopi, pp. 87–119. Stenstro¨m, A. 1998. ‘From sentence to discourse: Cos (because) in teenage talk’ in A. H. Jucker and Y. Ziv (eds): Discourse Markers. Descriptions and Theory. Amsterdam/Philadelphia: John Benjamins, pp. 127–46. Stubbs, M. 1993. ‘British traditions in text analysis—from Firth to Sinclair’ in M. Baker,
693
G. Francis, and E. Tognini-Bonelli (eds): Text and Technology. In Honour of John Sinclair. Amsterdam: Benjamins, pp. 1–33. Thomas, J. and M. Short. (eds) 1996. Using Corpora for Language Research. London: Longman. Thompson, G. and S. Hunston. (eds). 2006. System and Corpus: Exploring Connections. London/Oakville: Equinox. Timmis, I. 2005. ‘Towards a framework for teaching spoken grammar,’ ELT Journal 59/2: 117–25. Trudgill, P. 1999. ‘Standard English: What it isn’t’ in T. Bex and R. J. Watts (eds): Standard English. The Widening Debate. London: Routledge, pp. 117–28. Wichmann, A., S. Fligelstone, T. McEnery, and G. Knowles. (eds). 1997. Teaching and Language Corpora. London: Longman. Widdowson, H. G. 2000. ‘On the limitations of linguistics applied,’ Applied Linguistics 21/1: 3–25. Widdowson, H. G. 2003. Defining Issues in English Language Teaching. Oxford: Oxford University Press.
Applied Linguistics 29/4: 694–716 ß Oxford University Press 2008 doi:10.1093/applin/amn018 Advance Access published on 2 June 2008
Form-focused Instruction in Second Language Vocabulary Learning: A Case for Contrastive Analysis and Translation BATIA LAUFER and NANY GIRSAI University of Haifa The study investigates the effect of explicit contrastive analysis and translation activities on the incidental acquisition of single words and collocations. We compared three high school groups of learners of the same L1 and comparable L2 (English) proficiency. Each group represented one instructional condition: meaning focused instruction (MFI), non-contrastive form-focused instruction (FFI), and contrastive analysis and translation (CAT). The target items consisted of ten unfamiliar words and ten collocations in L2—English. The MFI group performed content-oriented tasks which did not require attention to the target items. The FFI group performed text-based vocabulary tasks which focused on the target items. The CAT group was assigned text-based translation tasks: from L2 into L1, and from L1 into L2. During the correction stage, the teacher provided a contrastive analysis of the target items and their L1 translation options. Time-on-task was kept constant in the three groups. After completing the tasks, the three groups were tested on the retention of the target items by two tests: active recall and passive recall. A week later, the participants received the same tests. The CAT (contrastive analysis and translation) group significantly outperformed the other two groups on all the tests. These superior results are discussed in light of the ‘noticing’ hypothesis, ‘pushed output’, ‘task-induced involvement load’, and the influence that L1 exerts on the acquisition of L2 vocabulary.
INTRODUCTION The pedagogical approach of form-focused instruction to second language teaching can be regarded as a modification of communicative language teaching, whose proponents believed that comprehensible input and meaningoriented tasks were necessary and sufficient for language acquisition. When it became evident that second language learners could not achieve high levels of grammatical competence from entirely meaning-centred instruction (Swain 1988; Lyster 1994; Lightbown and Spada 1999), applied linguists suggested that learners should also attend to form (Long 1991; De Keyser 1998; Norris and Ortega 2000; Ellis 2001; Housen and Pierrard 2005). The term ‘form’ includes the function that a particular structure performs. For example,
BATIA LAUFER and NANY GIRSAI
695
attention to the ‘form’-ed subsumes the realization that -ed signals an action performed in the past. Form-focused instruction, henceforth FFI, can be of two types: Focus on Form (FonF) and Focus on Forms (FonFs). The first is a pedagogical approach defined by Long as drawing learners’ attention to linguistic elements during a communicative activity. Focus on Forms, on the other hand, is an approach equated with the ‘traditional’ method, which entails teaching discrete linguistic structures in separate lessons in a sequence determined by syllabus writers. According to Ellis (2001), in a FonFs approach, students view themselves as learners of a language and the language as the object of study; in FonF, on the other hand, learners view themselves as language users and language is viewed as a tool for communication. The notion of FFI was developed in the context of grammar learning, but it can be extended to vocabulary as well. Thus, learners’ attention can be drawn to lexical items (single words and multi-word units) within a communicative task environment if these lexical items are necessary for the completion of a communicative, or an authentic language task. For example, when reading a text, or engaging in a group discussion, learners may come across unfamiliar words and look them up in a dictionary. This activity constitutes Focus on Form since the words attended to are necessary tools for task completion. Conversely, learners’ attention can be drawn to words in non-communicative, non-authentic language tasks, as in the case of matching words that were taught and are listed in column A to their definitions in column B, or filling in these words in given sentences, one word in each sentence. These are examples of FonFs in the sense that they entail teaching and practising discrete lexical items, which are treated as the objects of study and not as tools of language use. Empirical research of form-focused instruction has also been conducted mainly in the context of the teaching of grammar, not vocabulary. Only recently has FFI been explicitly linked to vocabulary instruction (Hill and Laufer 2003; Laufer 2005, 2006; Peters 2006). However, many earlier studies that examined the usefulness of various instructional techniques on vocabulary learning (dictionary use, negotiation of meaning in the input or output, writing original sentences, computerized exercises, etc.) investigated, in fact, the effect of FFI in its own right, or in comparison with meaning-based approaches thus providing indirect empirical support for the efficacy of FFI for lexical learning. For example, Luppesku and Day (1993) and Knight (1994) studied the incidental acquisition of new words, that is the acquisition of words without learners’ deliberate attempts to commit them to memory. They compared students who read a text and looked up unknown words in the dictionary with students who read the text without a dictionary. The dictionary condition fared better than the reading only condition. Ellis et al. (1994) found that interactionally modified input, that is input that included clarifications, was more beneficial for learning new words than pre-modified input (input prepared on the basis of prior student interaction before the experimental task).
696
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
De la Fuente (2002) found that learners exposed to input during negotiated interaction with or without production of pushed output learnt more words than learners exposed to non-negotiated input. The FFI conditions in the above examples could be classified as Focus on Form since they entailed attention to vocabulary during communicative tasks. For additional examples of FonF studies, see Hulstijn (1992), Newton (1993), Ellis and He (1999), and Watanabe (1997). There are numerous examples of studies that could be classified as Focus on Forms studies, where vocabulary was practised in isolation, or in minimal context. We will briefly describe two. Snellings et al. (2002) increased learners’ word retrieval speed through four types of exercises: sentence completion, a decision about which of two supplied words make a given sentence incorrect, a decision about which word will correct a given sentence, and a translation of a Dutch word into English within an English sentence context. Horst et al. (2005) investigated vocabulary learning through the use of word banks, on-line dictionaries, concordances, cloze exercises, hypertexts, and self-quizzes. The two studies found that many of the practised words were learnt both receptively and productively. For additional examples of studies that include at least one FonFs condition, see Prince (1996), Qian (1996), Paribakht and Wesche (1997), Zimmerman (1997), Groot (2000), Kitajima (2001), Boers et al. (2004), Mondria and Wiersma, (2004), and Webb (2005, 2007). The evidence for the effectiveness of FFI in vocabulary is indirect since the studies mentioned above were not designed to specifically contrast FFI and non-FFI conditions. Nor did the authors intend to specifically advocate decontextualized vocabulary learning of FonFs type. And yet, the overall conclusion of the studies is unambiguous: form-focused instruction, both of FonF and FonFs kinds, is beneficial to vocabulary learning. The results (in terms of the number of words retained after the FFI task) are usually superior to non-FFI conditions in comparative studies. In non-comparative studies, on average, two-thirds of the meanings of new words are recalled or recognized correctly on the post-tests (for a survey of FFI vocabulary studies, see Laufer 2005). The vocabulary studies mentioned above employ a variety of means to focus on the words targeted for learning: glosses, dictionary use, negotiation of meaning, vocabulary exercises involving comprehension or production of words, and computer assisted learning devices (hypertexts, concordances, on-line dictionaries). To our knowledge, no research has examined the value of contrastive FFI of vocabulary, such as interlingual comparisons with learners’ L1, or translation. By contrastive FFI, we do not refer to bilingual glosses which simply state the meaning of L2 words, but to the kind of instruction which leads to learners’ understanding of the similarities and differences between their L1 and L2 in terms of individual words and the overall lexical system. For example, in the case of French as L2, a bilingual gloss would translate the French word ‘savoir’ as ‘know’. Teachers using contrastive FFI, on the other hand, would point out that ‘savoir’ and ‘know’ do not overlap semantically and would provide explanation and practice of the difference between the two French
BATIA LAUFER and NANY GIRSAI
697
translations of ‘know’: ‘savoir’ and ‘connaıˆtre’. Lack of contrastive FFI studies is rather surprising. With the growth of interest in transfer studies and crosslinguistic influence since the 1980s (e.g. Gass and Selinker 1983; Odlin 1989), we could have expected to see an increased interest in researching the connection between overcoming learning difficulties and heightening the learners’ awareness to the differences between L1 and L2 that were causing them. Moreover, since providing cross-linguistic information is a clear case of focus on form, or forms (depending on whether it is provided within a communicative task, or not), it seems natural that FFI research should extend to cross-linguistic instruction.1 Such research has been conducted in the area of grammar. The results showed that providing learners with cross-linguistic information proves to be effective in the instruction of some selected structures (Kupferberg and Olshtain 1996; Sheen 1996; Kupferberg 1999; Ammar and Lightbown 2005). We hypothesize that, similarly to grammar, L2 vocabulary teaching will benefit from cross-linguistic form-focused instruction which entails comparison with L1 and translation. Such an approach to vocabulary teaching can be justified in terms of several hypotheses that explain effectiveness in L2 learning in general: ‘noticing’, ‘pushed output’, and ‘task-induced involvement load’. It can also be justified by work in contrastive semantics, psycholinguistics, error analysis, and corpus analyses of learner language, work that demonstrates the pervasive influence of L1 on L2 vocabulary learning. The ‘noticing’ hypothesis, which provided the theoretical underpinning of FFI, states that learners must consciously notice forms and the meanings these forms realize in the input in order to convert input into intake for learning (Schmidt 1990, 1994). Schmidt (2001) defines ‘noticing’ as the subjective correlate of what psychologists call attention, and concludes, on the basis of psychological literature, that although there may be some forms of learning without attention, in general, intentionally focused attention may be a practical necessity for successful language learning. Since many features of L2 input are likely to be infrequent, non-salient, and communicatively redundant, they may go unnoticed unless attention is drawn to them. According to Robinson (1995), noticing entails detecting and then rehearsing the linguistic feature in short-term memory before storing it in long-term memory. One way to make a foreign language feature noticeable or salient in the input is to enhance it by providing contrastive association with the corresponding L1 item. Such associations are the essence of explicit contrastive instruction. As mentioned earlier, empirical evidence can be found for the effectiveness of this approach in grammar. Since vocabulary can be enhanced in the same way, we can expect a similar beneficial influence of contrastive FFI on vocabulary learning. The ‘pushed output’ hypothesis claims that when learners produce language and stretch their linguistic resources in the process, they improve their language production and their language development (Swain 1985;
698
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
Swain and Lapkin 1995). There is empirical evidence that output tasks have been more effective than input tasks for learning new words (Ellis and He 1999; De la Fuente 2002). Translation into L2 is a manifestation of pushed output. In order to translate, the learner is required to produce language, but unlike in the case of free production, the learner cannot produce a good translation if s/he avoids problematic words or structures. Hence, translation should be at least as effective as other pushed output tasks for learning vocabulary. The ‘task-induced involvement load’ hypothesis (Laufer and Hulstijn 2001) postulates that the learning of words is best achieved by means of tasks with a high involvement load, that is tasks which combine three elements with regard to the words being practiced: ‘need’, ‘search’, and ‘evaluation’. The ‘need’ component is the motivational dimension of involvement. Search and evaluation are the two cognitive dimensions of involvement. ‘Need’ is present in the task when the word is perceived as necessary for task completion. ‘Search’ is the attempt to find the meaning of an unknown L2 word or, conversely, to find the L2 word form expressing a given concept. Examples of ‘search’ are: trying to find the L2 translation of an L1 word by consulting a dictionary, or trying to infer the meaning of an L2 word from context. ‘Evaluation’ implies some kind of selective decision about the word’s meaning, or form, in which the word’s context is taken into account. It entails a comparison of a given word with other words, a specific meaning of a word with its other meanings, or comparing the word with other words in order to assess whether a word does or does not fit its context. Two degrees of prominence have been suggested for ‘evaluation’: moderate and strong. A ‘moderate evaluation’ entails recognizing differences between words (as in a fill-in task with words provided in a list), or differences between several senses of a word in a given context (as in the decision of the meaning of a homonym in a particular text context). ‘Strong evaluation’ requires a decision as to how additional words will combine with the new word in an original, as opposed to given, L2 sentence. The involvement hypothesis was proposed as an attempt to operationalize the concepts that have been used in connection with good retention: depth of processing, degree of elaboration, quality of attention, richness of encoding (see, for example, Craik and Lockhart 1972; Baddeley 1997). Empirical evidence was found to support the involvement load hypothesis fully, or partially (Laufer and Hulstijn 2001; Hill and Laufer 2003; Webb 2005). Translation tasks embody the element of need since the words that have to be understood (when translating into L1), or produced (when translating into L2) are predetermined by the source text. The element of search is present as well. If an L2 word is unfamiliar, learners have to conduct a search for its meaning when translating into L1, or a search for its form when translating into L2. Most importantly, an element of evaluation is necessary to carry out a translation activity. There is usually more than one translation alternative for a given sentence. Therefore, when translating, learners have to make a decision as to how each alternative fits the text they create. When the
BATIA LAUFER and NANY GIRSAI
699
translation is into L2, this decision will be based on the way other words in L2 combine with the new word. Hence, according to the model of involvement, the evaluation is strong. Since translation is a task with a high involvement load, it can be assumed that it will be effective in vocabulary learning. Additional justification for incorporating translation tasks into vocabulary learning is provided by earlier research in contrastive semantics and error analysis, and later corpus studies and psycholinguistic experiments. Dagut (1977) suggested that learners do not attach new L2 words directly to the concepts that they represent, but to L1 words which represent L1 concepts that learners possess. A similar view was expressed by Blum and Levenston (1978), Ringbom (1983), and later by Ellis (1997), Hall (2002), and Jiang (2004). Sometimes L1 concepts are identical to the concepts represented by the new L2 words, resulting in a correct translation of the learner. However, since different languages do not have entirely identical conceptual systems, many L2 words based on L1 meanings may not be identical in all semantic properties, that is have no exact translation equivalents. For example, even though ‘home’ is translated as ‘maison’ in French, the two are not totally equivalent since ‘home’ has a specific feature of comfort and safety (as opposed to ‘house’) and this feature is not obvious in the French translation. In order to use the L2 word with its correct specifications, a process of semantic restructuring must occur in which the learner readjusts the semantic knowledge of the word that s/he possesses to that of the native speaker. Wolter (2001, 2006) extends the idea of relating L2 words to L1 concepts to the issue of lexical combinations. He suggests that learners draw upon L1 conceptual knowledge when making assumptions about connections between L2 words and that this knowledge will sometimes provide learners with misinformation about allowable combinations of L2 words. Thus, a learner who produces unusual collocations or combinations of words in the L2 is probably relying too heavily on L1 collocational knowledge. Collocational errors can occur even when learners are familiar with both of the words that comprise the correct collocation. Wolter gives an example of the Japanese learner who produces the collocation narrow room even though s/he knows the word small. Learning to connect words correctly requires restructuring of the existing network. Since word combinations cannot be easily predicted from the knowledge of individual words, the task of second language vocabulary learning is even greater than was thought when vocabulary was considered in terms of single words (Lewis 2000). It may be argued that semantic restructuring of single concepts and lexical networks may occur with continued exposure to L2 vocabulary in a variety of contexts. Even if we agree to this position, an important question is whether a foreign language classroom can provide enough exposure to L2 lexis for such restructuring to take place. Judging by the available empirical evidence on learner lexis, it is doubtful whether this is the case. Early error analysis studies showed that lexical errors resulting from L1 influence persisted with advanced learners, particularly where the concepts in L1 and
700
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
L2 did not overlap (Myint Su 1971; Laforest 1980; Ringbom 1982). Recent psycholinguistic studies confirmed this finding. Jiang (2002, 2004) found that advanced Chinese learners of English had difficulty distinguishing between pairs of English words that were translated by a single Chinese word, for example criterion-standard, accurate-precise, safe-secure, etc. He concluded that the semantic restructuring that is necessary for acquiring the meaning of words different from L1 is a slow process, which is relatively unaffected by the quantity of input that L2 learners receive in their learning context. Collocational errors persist in the learner language as well, thus showing resistance to language input and teaching. When encountered in the input, collocations are usually semantically transparent, for example make a decision, send a message, offer help, submit an application, hand in a paper, etc. Therefore they may not be noticed by learners and teachers as problematic. However, producing correct collocations is often difficult since ‘equivalent’ collocations in L1 may include at least one word which is different from L2. For example, English make a decision is lakaxat haxlata ‘take a decision’ in Hebrew; give examples is lehavi dugmaot ‘bring examples’, practice law is laasok be-mishpat ‘deal in law’. Biskup (1992) conducted an error elicitation study and found that advanced Polish-speaking learners of English made L1 based collocation errors. More recently, the same conclusion has been reached in a series of corpus studies (Bahns 1993; Granger 1998; Nesselhauf 2003, 2005). For example, Nesselhauf (2003) analysed the use of verb–noun collocations such as take a break or shake one’s head by advanced German-speaking learners of English in free written production and concluded that the learners’ L1 had a much stronger influence than earlier studies had predicted. Xiao and McEnery (2006) performed a cross-linguistic analysis of collocation, semantic prosody, and near synonymy, drawing upon data from English and Chinese. They concluded that teachers should compare the collocational behaviour in L1 and L2 since learners’ awareness of L1–L2 differences should considerably reduce the number of L1 interference errors. The pervasive influence that L1 has on the learner lexis and the persistence of L1-based errors at advanced levels of learning suggest that contrastive form-focused instruction, which raises the learners’ awareness of the L1–L2 differences and provides practice in the areas of these differences, may prove more effective than teaching methods that ignore the cross linguistic influence on lexical learning. We started investigating the effectiveness of contrastive FFI in a preliminary study (Laufer and Girsai 2008), where two groups of high school learners had been taught unfamiliar words and collocations in English as a foreign language. One group was exposed to the target items (the unfamiliar words and collocations) by reading a text and answering comprehension questions. Another group of the same L1 and the same English proficiency level read the text as well, but was also given two translation activities during which the target items were compared with their L1 equivalents. After the completion of the tasks, the two groups were tested on the retention of the target items by two
BATIA LAUFER and NANY GIRSAI
701
tests. The ‘reading þ translation group’ outperformed the ‘reading group’ on both tests. In this study we did not include a group performing a noncontrastive form-focused activity. Hence, we could only conclude that contrastive FFI was superior to message based instruction, but not that it was any different from FFI in general. The study in this paper was designed to find out whether this is the case.
THE STUDY Having shown in our preliminary work that vocabulary teaching will benefit from cross linguistic form-focused instruction, we now hypothesize that contrastive FFI will be at least as beneficial as, if not more than, other types of FFI for the learning of new vocabulary. To this effect, we set up three learning conditions for comparison: 1 MFI—message-focused instruction 2 FFI—non-contrastive form-focused instruction 3 CAT—contrastive analysis and translation Detailed description of the conditions will be presented in the ‘procedure’ section.
Research questions 1) Will contrastive form-focused tasks lead to the acquisition of a significantly larger number of lexical items than non-contrastive form-focused tasks and than message-focused tasks? (a) in the case of single words; (b) in the case of collocations. 2) Will the above differences (if any) be retained on a delayed test taken one week after performing the first post-test? Acquisition was operationalized as the ability to provide the form of the target words in response to their L1 translation equivalents (active recall), and the ability to provide the meaning of the target words (passive recall).2 Word knowledge consists of many components of knowledge: the word’s pronunciation, spelling, morphology, syntax, meaning, lexical relations, etc. (Nation 2001). Yet we contend that the most important component of word knowledge is the knowledge of the form–meaning relation, that is the ability to retrieve the meaning of a given word form, and the ability to retrieve the word form of a given concept. (For a justification of this position, see Laufer et al. 2004). This is why we chose to test recall of meaning and recall of form. Each main research question was subdivided into four secondary questions, depending on the type of vocabulary items tested (single words or collocations) and depending on the type of knowledge elicited (passive recall or active
702
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
recall). Thus, each secondary question asked whether there would be a difference between the three conditions (MFI, FFI, CAT) regarding the following results: 1 2 3 4
the the the the
number number number number
of of of of
the single target words recalled passively; the single target words recalled actively; target collocations recalled passively; target collocations recalled actively.
Participants Seventy-five 10th graders (aged 15–16) learning English as a foreign language participated in the experiment. They were Hebrew native speakers, who had studied English for six years prior to the experiment. Within the framework of the Israeli curriculum, such students are at the last stages of intermediate level. Native speakers of English, pupils who had lived abroad for a period of time, and students with learning disabilities were not included in the experiment. The participants studied in three parallel intact classes and were comparable in terms of proficiency in English. During their years of study, they followed the same curriculum prescribed by the Ministry of Education, had the same teachers, had been preparing for the same governmental exam and were all placed in the ‘5 point stream’ of English, that is the advanced stream of the Israeli curriculum.3 They were also from comparable socio-economic backgrounds. The teaching in these classes stresses communication, but occasional FonF happens as well. Teachers conduct the lessons in English and do not practise translation. Each class was randomly assigned to one condition: the message focused instruction (MFI) group, which consisted of 26 learners, the non-contrastive form-focused instruction (FFI) group, which had 23 participants, and the contrastive formfocused instruction (CAT) group, which comprised 26 learners.
Target items The target vocabulary to be taught and tested included single words and collocations. Collocations were defined here as habitually occurring set semantic phrases. The target items were selected after pre-testing all the participants.
Selection and presentation of target items Two weeks prior to the beginning of the treatment phase, two written pretests were administered to the groups during regular class time in order to determine whether the target vocabulary was indeed unknown to the participants. The tests consisted of 50 single words and 41 collocations. Some of these items were taken from reading material which we intended to use in the experiment. Others served as distracters. The distracters included
BATIA LAUFER and NANY GIRSAI
703
vocabulary which was supposed to be familiar to the learners as it had been previously taught. Each pre-test took about 10 minutes.
Pre-test 1 contained 50 single English words. The learners were asked to supply in writing the Hebrew translation for as many words as they knew. Pre-test 2 was intended to check active collocational knowledge. The test consisted of 41 Hebrew phrases to be translated into English. Passive knowledge of collocations was not tested, because, as mentioned earlier in the Introduction, many collocations are semantically transparent and therefore easily comprehended. The particular difficulty associated with collocational knowledge lies in their production. For example, the meaning of hit the headlines would probably have been clear to many participants. However, in production, we often observed a word-for-word translation from Hebrew ( ‘to break into headlines/titles’, for Hebrew ‘lifrotz la’kotarot’). On the basis of these pre-tests, we selected 10 single words and 10 verb–noun collocations that were unknown to all the participants. The single words were: Relish, glean, candid, laudable, opulent, plague, account, detractor, gregarious, lavish. The collocations were:
Settle scores, present a problem, derive pleasure, hold a vote, fulfil an ambition, launch a campaign, reclaim the trust, hit the headlines, meet the expectations, place orders. The target vocabulary was embedded in a reading passage compiled by the second author and entitled ‘Clinton’s biography’. The text was based on six articles which dealt with Clinton’s book My Life and were available on the Internet. It consisted of 536 word tokens. The grammar of the text was kept simple in order to minimize grammar-related comprehension difficulties of the text. Moreover, the authors inserted textual clues into the text so that the meaning of the target words could be inferred from context without too much difficulty.
Procedure The study adopted an incidental acquisition design, that is we investigated whether the target words and collocations were acquired without learners’ deliberate attempt to commit them to memory. Hence the learners were not told they were participating in an experiment at the end of which they would be tested. Each step of the study was conducted during regular class time and took the same amount of time in the three parallel classes that represented the three conditions. The first stage of the treatment was identical in the three groups. They were asked to read the passage ‘Clinton’s biography’ and to answer
704
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
13 True-or-False text-based statements. The learners had no dictionaries and no marginal glosses for the unknown vocabulary. However, whenever a learner wanted to ask the meaning of a word, the teacher would provide the necessary information.4 When the exercise had been completed, the teacher went over the answers with the class. Each true or false statement was checked against the relevant section of the text and any questions or comments from the learners were addressed by the teacher. This activity took about 40 minutes. All the materials were collected at the end of the session. The next stage of the study was conducted on the following day, during a double lesson (90 minutes). Each group received a different treatment, in accordance with the experimental condition it was assigned to. The treatment consisted of two tasks. The first one was performed with the text, which the learners received again, the second one without it. As in the first stage, the learners had no dictionaries and no marginal glosses for the unknown vocabulary. Whenever the new words were deemed necessary for a task by learners, they could infer them from context, or ask the teacher for the words’ meaning. Since the teacher went over all the answers in the three groups, the meaning of the target words used in the tasks was verified by her. The MFI (meaning-focused instruction) group received two communicative tasks: reading comprehension and pair/group discussion. Here are two examples of reading comprehension questions: 1 Why is Arkansas mentioned in the text? 2 Complete the sentence: Although Clinton finally felt free, the affair _______________________ Upon completion of the exercise, the teacher went over the answer with the active participation of the learners. Then the texts were taken away once more and the discussion task followed. Here is one of the two discussion topics: With your partner, discuss the following: A president’s love affair is his own personal business and should not concern the public; it is a private matter between him and his wife and has nothing to do with his position as president. Agree or disagree. First, learners worked in pairs. The teacher would join a different pair each time to provide feedback. Following pair work, the entire group discussed the topic. During these activities, some of the target words may have been used by some learners. However, they were not singled out for teaching. The language of interaction in this group was English only. The FFI (non-contrastive form-focused instruction) group received two form-focused tasks: meaning recognition of the target vocabulary (a multiple choice exercise), and a text fill-in activity with the target words provided in a ‘word bank’ at the end of the text. Here is an example of meaning
BATIA LAUFER and NANY GIRSAI
705
recognition, where learners were required to select the correct meaning of a collocation in the text. Hit the headlines (paragraph 1) (a) became hot news (b) caused headaches (c) was first in line Here are the first two sentences of the fill-in exercise. The missing phrases are ‘hit the headlines’ and ‘launched a campaign’ Bill Clinton has 1. _______________ again. He has 2. __________ for his autobiography. Upon the completion of each task, the teacher went over the answers with the class and provided the necessary feedback, including verifications and clarifications of the meaning of the target items. The interaction was in English only. The CAT (contrastive analysis and translation) group received two translation tasks and brief explicit contrastive instruction. Here is an example of the first task in which learners were asked to translate the given sentences from L2 to L1. The target items are in bold font. Will the book meet the expectations of the publishers? ____________ Scandals plagued Clinton’s administration. ______________________ The sentences were nearly identical to the sentences in the original text. In the second task, after the L2–L1 translations had been collected, learners translated the same sentences from Hebrew into English. They could carry out the two tasks alone, or with their peers. As in the other two groups, on task completion, the teacher checked the answers and provided corrective feedback. During the second task, she also supplied a brief explicit contrastive instruction. For example, in the case of collocations, she pointed out that although the nouns had equivalent translations in Hebrew, the verbs that collocated with them were totally different (e.g. hit the headlines is ‘break into headlines’ in Hebrew, meet expectations is ‘answer expectations’). She suggested that learners should be cautious not to provide automatic verb translations, as this would result in unacceptable English collocations.
Testing The day after the treatment, all the learners were unexpectedly tested on the target vocabulary. They were first tested on their ability to provide the words in response to their Hebrew translations, that is on active recall. Here is an example of two test items: a single word and a collocation: Translate the following Hebrew words and phrases into English: jb?l ywar (raui le-shevach) ______________________ twypych lu twnul (laanot al tzipiyot) _________________________
706
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
The test sheets were collected and a 20-minute delay followed, during which the learners were engaged in an activity that did not include the target vocabulary. After the delay, the second test was administered. The target words were given in a different order than in the first test, and the learners were required to provide their meaning in Hebrew, or in English, that is to demonstrate passive recall. Here is an example of two test items: a single word and a collocation: Translate the following English words and phrases into Hebrew or explain them in English: Meet the expectations _________________________ Laudable_________________________ A week later the same tests were administered in the same manner in order to check how much of the target vocabulary was still remembered. The answers were scored dichotomously. A correct answer received one point, an incorrect answer or a blank received a zero. In the test of active recall, minor spelling errors that did not distort the word’s pronunciation considerably were not taken into account. For example, the answer opulant for ‘opulent’ received a point, but gragirious for ‘gregarious’ did not. Since the focus of the test was the knowledge of the meaning–form link and such errors did not distort the word’s form greatly, the answer received one point. We decided not to accept non-target synonyms for the Hebrew prompts, for example ‘honest’ instead of ‘candid’. Therefore, learners were specifically told to provide words from the text they had studied and they did so. In the test of passive recall, any Hebrew translation or English explanation that reflected the semantic content of the word was considered correct. The maximum score for single words was 10 and 10 for collocations.
RESULTS Descriptive statistics We will present the descriptive statistics first. Table 1 presents the means and standard deviations (in brackets) of the tests measuring passive and active recall of the target vocabulary immediately after the instruction. Table 2 presents the results of the delayed tests. The figures in both tables show that the lowest scores were received in the MFI group and the highest in the CAT group. Moreover, the scores did not go down on the delayed tests 1 week after the instruction. Some tests even yielded slightly higher scores. This can be explained by the interest that the tests generated in learners. They reported that even though all the materials were collected after the immediate tests, some items could be remembered, particularly if they had previously been practised in form-focused activities, both contrastive and non-contrastive. Hence, some learners checked them with their peers, or in the dictionary. The only results of the MFI group that are not lower than 1 are those of the passive collocation tests. Yet this has very little to
BATIA LAUFER and NANY GIRSAI
707
Table 1: Immediate recall scores Condition
Passive recall
Active recall
Single words Max. ¼ 10
Collocations Max. ¼ 10
Single words Max. ¼ 10
Collocations Max. ¼ 10
MFI N ¼ 26
0.31 (0.68)
3.69 (2.40)
0.12 (0.33)
0.27 (0.53)
FFI N ¼ 23
4.04 (2.64)
5.70 (2.85)
2.30 (2.18)
2.91 (2.79)
CAT N ¼ 26
5.81 (2.48)
8.58 (1.72)
4.31 (2.51)
5.81 (2.61)
Table 2: Delayed recall scores Condition
Passive recall
Active recall
Single words Max. ¼ 10
Collocations Max. ¼ 10
Single words Max. ¼ 10
Collocations Max. ¼ 10
MFI N ¼ 26
0.15 (0.37)
4.54 (2.30)
0.12 (0.33)
0.35 (0.69)
FFI N ¼ 23
3.52 (2.94)
6.00 (3.01)
2.13 (2.71)
3.00 (2.84)
CAT N ¼ 26
6.27 (2.89)
8.73 (1.61)
4.12 (2.45)
6.12 (2.84)
do with L2 learning, but it is concerned with the ability to detect the meaning of collocations as they are often transparent, as mentioned earlier. The active test of the same collocations, however, which required demonstration of L2 knowledge, yielded a result close to zero. On all the tests, the scores of passive recall are higher than the corresponding scores of active recall, which is not surprising. Vocabulary learning is an incremental process and learners usually acquire passive knowledge of a word before they acquire its active knowledge (except in the case of cognates) (Laufer 1998; Laufer et al. 2004; Webb 2005).5
Inferential statistics In order to answer the research questions regarding the difference between the three instructional conditions, we carried out eight one-way ANOVAs,
708
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
four with the immediate tests, and four with the delayed tests. Each ANOVA compared the difference of means between the three conditions as follows: in the case of the single words–passive recall, single words–active recall, collocations–passive recall, collocations–active recall. The F values of the eight ANOVAs are presented in Table 3. Since the differences of all eight comparisons were significant, we carried out Tukey’s post-hoc tests to check the differences between each pair of conditions for each ANOVA. Table 4 presents the differences between the means (standard error is provided in
Table 3: Differences between the three conditions (ANOVA results) Test time
Passive recall F value
Active recall F value
Single words
Collocations
Single words
Collocations
Immediate
46.02 (df 2.72)
28.43 (df 2.72)
31.05 (df 2.72)
41.16 (df 2.72)
Delayed
43.65 (df 2.72)
21.27 (df 2.72)
23.70 (df 2.72)
39.76 (df 2.72)
p 5 .001.
Table 4: Differences between pairs of conditions (Tukey post hoc tests). Immediate recall Condition
MFI
Condition
FFI CAT
FFI
MFI CAT
CAT
MFI FFI
p 5 .01;
p 5 .001.
Passive recall Difference in means
Active recall Difference in means
Single words
Collocations
Single words
Collocations
3.74 (0.60) 5.50 (0.58) 3.74 (0.60) 1.76 (0.60) 5.50 (0.58) 1.76 (0.60)
2.00 (0.65) 4.88 (0.67) 2.00 (0.67) 2.88 (0.67) 4.88 (0.65) 2.88 (0.67)
2.19 (0.55) 4.19 (0.53) 2.19 (0.55) 2.00 (0.55) 4.19 (0.53) 2.00 (0.55)
2.64 (0.61) 5.54 (0.63) 2.64 (0.63) 2.89 (0.63) 5.54 (0.61) 2.89 (0.63)
BATIA LAUFER and NANY GIRSAI
709
Table 5: Differences between pairs of conditions (Tukey post hoc tests). Delayed recall Condition
MFI
Condition
FFI CAT
FFI
MFI CAT
CAT
MFI FFI
Passive recall Difference in means
Active recall Difference in means
Single words
Collocations
Single words
Collocations
3.37 (0.68) 6.12 (0.66) 3.37 (0.68) 2.75 (0.68) 6.12 (0.66) 2.75 (0.68)
1.46 (0.67) 4.19 (0.65) 1.46 (0.67) 2.73 (0.67) 4.19 (0.65) 2.73 (0.67)
2.65 (0.60) 5.57 (0.58) 2.65 (0.60) 3.12 (0.60) 5.57 (0.58) 3.12 (0.60)
2.64 (0.67) 5.54 (0.65) 2.64 (0.67) 2.89 (0.67) 5.54 (0.65) 2.89 (0.67)
p 5 .001.
brackets) of each pair of conditions in the immediate recall test; Table 5 presents the results for the delayed recall test. On the immediate tests, the differences between all pairs of conditions were significant. On the delayed tests, the only non-significant difference was between FFI and MFI in the passive recall of collocations. As mentioned earlier, this is a relatively easy test as most collocations are semantically transparent. But even here, the CAT condition yielded a significantly higher score than the other two conditions.
DISCUSSION The study investigated whether incorporating contrastive analysis and translation activities into a text-based communicative lesson would make a significant difference in acquiring new vocabulary by comparison with a reading comprehension task alone, and by comparison with other formfocused activities following the reading task. The results revealed that the CAT (Contrastive Analysis þ Translation) group scored significantly higher than the two other groups on all eight tests: four immediate tests—passive recall of single words and of collocations, active recall of single words and of collocations, and four identical delayed tests. The group that did not receive any form-focused instruction learnt almost no vocabulary. (As mentioned earlier, the scores on passive collocation show, at least in part, comprehension of
710
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
transparent meaning, not of learning new forms.) Non-contrastive formfocused instruction resulted in learning about 50 per cent of passive new vocabulary and 27 per cent of active vocabulary on both the immediate and delayed tests. The CAT group learnt approximately 72 per cent of passive and 51 per cent of active vocabulary, much more than the other two groups6. It is difficult to compare these results with the results of other studies which investigated FFI together with a communicative task. Different studies use different numbers of target words, different numbers of form-focused activities and use different ways to measure learning. (For description and comparison, see Laufer 2005). However, to our knowledge, almost none of the studies which investigated incidental vocabulary acquisition resulting from FFI, have yielded figures as high as our CAT group. It is remarkable that learners received these high scores when tested on the most difficult aspects of form–meaning knowledge: recall of word form and recall of word meaning. We did not test recognition of meaning or recognition of form. Nor did we ask for learners to provide a self-evaluation of whether they were familiar with the word or not. Since the acquisition of word knowledge is incremental and the more difficult aspects, such as active recall, are learnt last, we can say with confidence that since learners’ passive and active recall were enhanced by CAT, the easier aspects of word knowledge were acquired as a result of CAT activities as well. As for other aspects of word knowledge, which were not investigated here, our teaching experience and our experience as foreign language learners suggest that CAT could enhance them all. This is an empirical question worthy of further investigation. As we suggested earlier, in the introductory part of the paper, the effectiveness of cross-linguistic form-focused instruction can be explained by the hypotheses of ‘noticing’, ‘pushed output’, and ‘task-induced involvement load’ and further supported by findings which show the pervasive influence of L1 in vocabulary learning. Learners were forced to notice the target items by receiving cross-linguistic instruction. These items became salient in the input when learners were taught the corresponding L1 forms and received information about the particular difficulties resulting from L1–L2 differences. As James (2005) points out, in the CLI (cross-linguistic influence) framework, the role of CA is to define salient foreign language input which may assist L2 learners by raising their cross-language awareness. CA can now come out of the closet and cross-language Awareness can be practised in classrooms as a legitimate activity, for it will at the very least sensitise learners to the decisions to be made in their FL [foreign language] production, advising them when they can and can not resort with profit to the MT [mother tongue] (James 2005: 11). Translation into L2 is a perfect ‘pushed output’ task that requires stretching one’s linguistic resources. To produce a good translation, learners cannot
BATIA LAUFER and NANY GIRSAI
711
avoid problematic words or structures since they are predetermined by the source language. The CAT group in our study was submitted to pushed output, which may have contributed to the success in learning in general, and to good active recall, in particular. Active knowledge implies a better memory trace than passive knowledge and is more difficult to achieve. In the specific case of collocations, correct production can be considered a special accomplishment, as active collocational knowledge is not readily attained even by advanced learners. We ascribe the superior results of active knowledge to the L1–L2 translation task, which entailed recalling and using the target word forms. This task is more demanding than any other vocabulary task which requires recall of word meaning, recognition of word form, or recognition of word meaning (Laufer et al. 2004; Laufer and Goldstein 2004). The participants’ engagement in a cognitively demanding task may have enabled them to remember and later produce the correct lexical items and collocations. In terms of the involvement load hypothesis, the translation tasks had a higher involvement load than the tasks in the other two conditions. Let us look at the involvement load in each condition. In the MFI condition, one task required answering comprehension questions; the other was a general discussion of a topic related to the text. In both of them, learners could have used the target words, but they could also have used different vocabulary as well, particularly in the general discussion. (This was indeed the case reported by the teacher.) Hence, each task in this condition could be described as þ/ need, þ/ search, and þ/ evaluation. If a learner perceived a need to use a particular word, s/he would search for its meaning and make some decisions pertaining to its context. In this case, each of the three elements would receive a plus (þ). But the learner could also convey the intended meaning using different words, in which case the involvement load in the case words not used would be a zero (0). The first task in the FFI condition was a multiple choice exercise that required learners to choose the meaning of the target words from three alternatives. The involvement components included in it were ‘need’ (the word had to be focused on to complete the task) and ‘search’ (search for meaning). As there was no context in this task, there was no ‘evaluation’. The second task, which required filling in the correct target words in given sentences, included the element of evaluation as well, as a decision had to be made as to which word fitted which context. Hence the involvement load of the FFI condition was: task 1: þ need þ search evaluation, task 2: þ need þ search þ evaluation. In the CAT condition, the learners were asked to translate sentences with the target words, first from L2 to L1, and then from L1 to L2. The ‘need’ element was present as there was no escape from focusing on the target words. Translation into L1 required a search for meaning whereas translation into L2 required a search for form. Hence the element of ‘search’ was present
712
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
in both translation tasks. Usually, a sentence can be translated in more than one way. The final choice of the translation must have been made after an evaluation of several translation alternatives. In each option, the target word was evaluated against the other words surrounding it. Moreover, in L1–L2 translation, the entire L2 context was created by the learner. Hence, the element of ‘evaluation’ was moderate in the L2–L1 translation task and strong in the L1–L2 translation. The involvement load in the CAT condition was therefore: task 1: þ need þ search þ evaluation, task 2: þ need, þ search, þþ evaluation. The different involvement loads of the three conditions are reflected quite accurately in our results. The participants in MFI condition learnt hardly any vocabulary at all, the FFI condition fared much better, and the CAT condition yielded the highest scores. On the basis of our results, some teaching implications are evident, mainly for classes where learners share a common L1. In order to perform a contrastive analysis of selected L1 and L2 areas and subsequent translation, teachers should be familiar with learners’ L1 and with its linguistic structure. Since learners draw upon L1 conceptual knowledge when making assumptions about L2 words and connections between them, teachers should make them aware of interlingual differences which may result in errors. Since words and word combinations which embody such differences are more difficult to learn, teachers should devote more teaching and practice time to these lexical items than to vocabulary which is similar in L1. This can take the form of additional exposure, use in communicative tasks, or in form-focused activities of a contrastive or noncontrastive nature.
CONCLUSION James (1994) showed that both contrastive analysis and error analysis remained vital components of applied linguistics and language teaching. Neglecting the L1 would amount to ‘burying your head in the sand and hoping that effortless acquisition will take place in time’ (James 2005: 11). The evidence from the present research, together with evidence from grammar studies, suggests that there is indeed a place for contrastive analysis and translation activities in L2 teaching. This does not mean that we should abandon the communicative classroom and return to the ‘grammar– translation’ method, nor does it mean that we should teach the skill of translation at the expense of the ability to function in a foreign language. Meaningful communication has been the goal of communicative language teaching, but the best method for achieving this goal may not be identical to the goal itself. The research presented here suggests that second language learners may benefit from contrastive form-focused instruction in selected L2 areas through raising their awareness of interlingual difficulties, stretching their linguistics resources, and engaging in involving tasks.
BATIA LAUFER and NANY GIRSAI
713
NOTES 1 One reason for the relative paucity of contrastive FFI studies may be that researchers of a Second Language context did not find L1-based tasks relevant to the teaching of multilingual classes. However, contrastive FFI is of relevance in a Foreign Language context and is of theoretical interest to the Language Learning field. 2 The distinction between passive and active knowledge is not as simple as it may appear. Read (2000) points out that different people have made this distinction differently, which has created problems in comparing passive and active knowledge across studies. Here, we follow the distinctions of Nation (2001) and Laufer et al. (2004) and refer to the ability to supply the word form as active knowledge and to the ability to supply the word meaning as passive knowledge. 3 In the Israeli educational system, each class in high school is further subdivided into three proficiency groups: 3 points, 4 points, 5 points on the basis of an internal school test. 4 The second author was the teacher who carried out the study in the three groups of learners. 5 It was suggested by one reviewer that learners saw the target L1 meanings on the test of active recall and that may have aided them on the test of passive recall. We cannot know for sure whether there was a learning effect or not. The passive test was given after a 20 minute distracting task and the words
were in a different order. Furthermore, in all vocabulary studies we are familiar with, passive knowledge proved to be better than active knowledge. Most importantly, even though the possibility of a learning effect cannot be excluded, it does not invalidate the main findings of the study which pertain to the differences between different instructional tasks. These were investigated separately for each type of vocabulary knowledge: passive and active. 6 One reviewer suggested that the test instruments were close to the practice tasks of the CAT group and therefore might have favoured it. First, in the translation tasks, participants were required to translate sentences, while in the tests they were required to translate isolated words. Second, in the passive recall test learners had the option of providing English explanations for the target words. Since the target words were explained in English in the other two groups, the test requirement was not unfamiliar to them. Most important, however, is the argument that adolescent and adult learners do not need to be taught to link L2 words to L1 words. They do this instinctively, as explained in the Introduction. This is also supported by our data. Even though English explanations were allowed in the passive recall test, most of the answers of the non-CAT learners were translations into L1 even though no translation was practised in class.
REFERENCES Ammar, A. and P. M. Lightbown. 2005. ‘Teaching marked linguistics structures–more about the acquisition of relative clauses by Arab learners of English’ in A. Housen and M. Pierrard (eds): Investigations in Instructed Second Language
Acquisition. Berlin and New York: Mouton de Gruyter, pp. 167–98. Baddeley, A. 1997. Human Memory: Theory and Practice, rev edn. Hove: Psychology Press.
714
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
Bahns, J. 1993. ‘Lexical collocations: A contrastive view,’ English Language Teaching Journal 47: 56–63. Biskup, D. 1992. ‘L1 influence on learners’ rendering of English collocations: A Polish/German empirical study’ in P. J. L. Arnaud and H. Bejoint (eds): Vocabulary and Applied Linguistics. London: Macmillan, pp. 85–93. Blum, S. and E. A. Levenston. 1978. ‘Universals of lexical simplification,’ Language Learning 28: 399–415. Boers, F., M. Demecheleer, and J. Eyckmans. 2004. ‘Etymological elaboration as a strategy for learning idioms,’ in P. Bogaards and B. Laufer (eds): Vocabulary in a Second Language: Selection, Acquisition, and Testing. Amsterdam: Benjamins, pp. 53–78. Craik, F. I. M. and R. S. Lockhart. 1972. ‘Levels of processing: A framework for memory research,’ Journal of Verbal Learning and Verbal Behavior 11: 671–84. Dagut, M. 1977. ‘Incongruencies in lexical gridding—an application of contrastive semantic analysis to language teaching,’ International Review of Applied Linguistics 15: 221–9. DeKeyser, R. M. 1998. ‘Beyond focus on form: Cognitive perspective on learning and practical second language grammar’ in C. Doughty and J. Williams (eds): Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press, pp. 42–63. De la Fuente, M. J. 2002. ‘Negotiation and oral acquisition of L2 vocabulary. The roles of input and output in the receptive and productive acquisition of words,’ Studies in Second Language Acquisition 24: 81–112. Ellis, N. 1997. ‘Vocabulary acquisition: Word structure, collocation, word-class, and meaning’ in N. Schmitt and M. McCarthy (eds): Vocabulary: Description, Acquisition and Pedagogy. Cambridge: Cambridge University Press, pp. 122–39. Ellis, R. (ed.). 2001. Form-focused instruction and second language learning. A supplement to Language Learning 51: supplement 1. Ellis, R. and X. He. 1999. ‘The roles of modified input and output in the incidental acquisition of word meanings,’ Studies in Second Language Acquisition 21: 285–301. Ellis, R., Y. Tanaka, and A. Yamazaki. 1994. ‘Classroom interaction, comprehension and the acquisition of L2 word meanings,’ Language Learning 44: 449–91.
Gass, S. M. and L. Selinker (eds). 1983. Language Transfer in Language Learning. Issues in Second Language Research. Amsterdam: Benjamins. Granger, S. 1998. ‘Prefabricated patterns in advanced EFL writing: Collocations and lexical phrases’ in A. P. Cowie (ed.): Phraseology: Theory, Analysis and Applications. Oxford: Clarendon Press/Oxford University Press, pp. 145–60. Groot, P. J. M. 2000. ‘Computer assisted second language vocabulary acquisition,’ Language Learning and Technology 4: 60–81. http:// llt.msu.edu/vol4num1/groot/default.html. Hall, C. J. 2002. ‘The automatic cognate form assumption: Evidence for the parasitic model of vocabulary acquisition,’ International Review of Applied Linguistics 40: 69–87. Hill, M. M. and B. Laufer. 2003. ‘Type of task, time-on-task and electronic dictionaries in incidental vocabulary acquisition,’ International Review of Applied Linguistics 41: 87–106. Horst, M., T. Cobb, and H. Nicolae. 2005. ‘Expanding academic vocabulary with an interactive on-line database,’ Language Learning and Technology 9: 90–110. http://llt.msu.edu/ vol9num2/horst/default.html. Housen, A. and M. Pierrard (eds). 2005. Investigations in Instructed Second Language Acquisition. Berlin and New York: Mouton de Gruyter. Hulstijn, J. H. 1992. ‘Retention of inferred and given word meanings: Experiments in incidental vocabulary learning’ in P. J. L Arnaud and H. Bejoint (eds): Vocabulary and Applied Linguistics. London: Macmillan, pp. 113–25. Hulstijn, J. H. and B. Laufer. 2001. ‘Some empirical evidence for the Involvement Load Hypothesis in vocabulary acquisition,’ Language Learning 51: 539–58. James, C. 1994. ‘Don’t shoot my dodo: On the resilience of contrastive and error analysis,’ International Review of Applied Linguistics 32: 179–200. James, C. 2005 ‘Contrastive analysis and the language learner: A new lease of life?’ in D. J. Allerton, C. Tshichold, and J. Wieser (eds): Linguistics, Language Learning and Language Teaching: The International Cooper Series on English Language and Literature (ICSELL) 10. Seiten: Broschiert, pp. 1–20. Jiang, N. 2002. ‘Form–meaning mapping in vocabulary acquisition in a second language,’ Studies in Second Language Acquisition 24: 617–37. Jiang, N. 2004. ‘Semantic transfer and development in adult L2 vocabulary acquisition’ in
BATIA LAUFER and NANY GIRSAI
P. Bogaards and B. Laufer (eds): Vocabulary in a Second Language: Selection, Acquisition, and Testing. Amsterdam/Philadelphia: John Benjamins, pp. 101–26. Kitajima, R. 2001. ‘The effects of instructional conditions on students’ vocabulary retention,’ Foreign Language Annals 34: 470–82. Knight, S. M. 1994. ‘Dictionary use while reading: The effects on comprehension and vocabulary acquisition for students of different verbal abilities,’ Modern Language Journal 78: 285–99. Kupferberg, I. 1999. ‘The cognitive turn of contrastive analysis: Empirical evidence,’ Language Awareness 8: 210–222. Kupferberg, I. and E. Olshtain. 1996. ‘Explicit contrastive instruction facilitates the acquisition of L2 forms,’ Language Awareness 5: 149–65. Laforest, M. H. 1980. ‘Towards a typology of lexical errors,’ Anglistika 23: 1–30. Laufer, B. 1998. The development of passive and active vocabulary in a second language: Same or different?’ Applied Linguistics 12: 255–71. Laufer, B. 2005. ‘Focus on Form in second language vocabulary acquisition’ in S. H. FosterCohen, M. P. Garcia-Mayo, and J. Cenoz (eds): EUROSLA Yearbook 5. Amsterdam: John Benjamins, pp. 223–50. Laufer, B. 2006. ‘Comparing focus on Form and Focus on FormS in second language vocabulary learning,’ Canadian Modern Language Review 63: 149–66. Laufer, B. and N. Girsai. 2008. The use of native language for improving second language vocabulary’ in A. Stavans and I. Kupferberg (eds): Studies in Language and Language Education. Essays in Honor of Elite Olshtain. Jerusalem: The Hebrew University Magnes Press, pp. 261–75. Laufer, B. and Z. Goldstein. 2004. ‘Testing vocabulary knowledge: Size, strength, and computer adaptiveness,’ Language Learning 54: 469–523. Laufer, B. and J. Hulstijn. 2001. ‘Incidental vocabulary acquisition in a second language: The construct of task-induced involvement,’ Applied Linguistics 22: 1–26. Laufer, B., C. Elder, K. Hill, and P. Congdon. 2004. ‘Size and strength: Do we need both to measure vocabulary knowledge?’ Language Testing 21: 202–26. Lewis, M. (ed.). 2000. Teaching Collocation: Further Developments in the Lexical Approach. Hove: Language Teaching Publications.
715
Lightbown, P. M. and N. Spada. 1999. How Languages Are Learned, rev edn. Oxford: Oxford University Press. Long, M. 1991. ‘Focus on Form: A design feature in language teaching methodology’ in K. de Bot, R. Ginsberg, and C. Kramsch (eds): Foreign Language Research in Cross-cultural Perspective. Amsterdam: John Benjamins, pp. 39–52. Luppesku, S. and R. R. Day. 1993. ‘Reading, dictionaries and vocabulary learning,’ Language Learning 43: 263–287. Lyster, R. 1994. ‘The effect of functional-analytic teaching on aspects of French immersion learners’ sociolinguistic competence,’ Applied Linguistics 15: 263–87. Mondria, J.-A. and B. Wiersma. 2004. ‘Receptive, productive, and receptive þ productive L2 vocabulary learning: What difference does it make?’ in P. Bogaards and B. Laufer, (eds): Vocabulary in a Second Language: Selection, Acquisition and Testing. Amsterdam/Philadelphia: John Benjamins, pp. 79–100. Myint Su. 1971. The Analysis of Lexical Errors. M. Litt. Thesis, University of Edinburgh. Nation, I. S. P. 2001. Learning Vocabulary in Another Language. Cambridge: Cambridge University Press. Nesselhauf, N. 2003. ‘The use of collocations by advanced learners of English and some implications for teaching,’ Applied Linguistics 24: 223–42. Nesselhauf, N. 2005. Collocations in a Learner Corpus. Amsterdam: John Benjamins. Newton, J. 1993. ‘Task based instruction among adult learners of English and its role in second language development’, Unpublished PhD thesis, Victoria University, New Zealand. Norris, J. and L. Ortega. 2000. ‘Effectiveness of L2 instruction: A research synthesis and quantitative meta-analysis,’ Language Learning 50: 417–528. Odlin, T. 1989. Language Transfer. Cross-Linguistic Influence in Language Learning. Cambridge: Cambridge University Press. Paribakht, T. S. and M. Wesche. 1997. ‘Vocabulary enhancement activities and reading for meaning in second language vocabulary acquisition’ in J. Coady and T. Huckin (eds): Second Language Vocabulary Acquisition. A Rationale for Pedagogy. Cambridge: Cambridge University Press, pp. 174–200. Peters, E. N. 2006. ‘Vocabulary acquisition in a foreign language. The influence of task and text
716
FORM-FOCUSED INSTRUCTION IN L2 VOCABULARY LEARNING
variables’, Unpublished PhD thesis, University of Leuven. Prince, P. 1996. ‘Second language vocabulary learning: The role of context versus translation as a function of proficiency,’ The Modern Language Journal 80: 478–93. Qian, D. D. 1996. ‘ESL vocabulary acquisition: Contextualization and decontextualization,’ The Canadian Modern Language Review 53: 120–42. Read, J. 2000. Assessing Vocabulary. Cambridge: Cambridge University Press. Ringbom, H. 1982. ‘The influence of other languages on the vocabulary of foreign language learner’ in G. Nickel and D. Nehls (eds): Errors Analysis, Contrastive Analysis and Interlangauge. Heidelberg: Gross Verlag, pp. 85–96. Ringbom, H. 1983. ‘Borrowing and lexical transfer,’ Applied Linguistics 4: 207–12. Robinson, P. 1995. ‘Attention, memory and the ‘‘noticing’’ hypothesis,’ Language Learning 45: 283–331. Schmidt, R. 1990. ‘The role of consciousness in second language learning,’ Applied Linguistics 11: 17–45. Schmidt, R. 1994. ‘Deconstructing consciousness in search of useful definitions for applied linguistics’ in J. H. Hulstijn and R. Schmidt (eds): Consciousness in Second Language Learning. Special issue of AILA Review 11: 11–26. Schmidt, R. 2001. ‘Attention’ in P. Robinson (ed.): Cognition and Second Language Instruction. New York: Cambridge University Press, pp. 3–33. Sheen, R. 1996. ‘The advantage of exploiting contrastive analysis in teaching and learning a foreign language,’ International Review of Applied Linguistics 24: 183–97. Snellings, P., A. van Gelderen, and K. de Glopper. 2002. ‘Lexical retrieval: An aspect of fluent second language production that can be enhanced,’ Language Learning 52: 723–54.
Swain, M. 1985. ‘Communicative competence: Some roles of comprehensible input and comprehensible output in its development’ in S. Gass and C. Madden (eds): Input in Second Language Acquisition. New York: Newbury House, pp. 235–53. Swain, M. 1988. ‘Manipulating and complementing content teaching to maximize second language learning,’ TESL Canada Journal 6: 68–83. Swain, M. and S. Lapkin. 1995. ‘Problems in output and the cognitive processes they generate: A step towards second language learning,’ Applied Linguistics 16: 371–91. Watanabe, Y. 1997. ‘Input, intake and retention. Effects of increased processing on incidental learning of foreign language vocabulary,’ Studies in Second Language Acquisition 19: 287–307. Webb, S. 2005. ‘Receptive and productive vocabulary learning: The effects of reading and writing on word knowledge,’ Studies in Second Language Acquisition 27: 33–52. Webb, S. 2007. ‘Learning word pairs and glossed sentences: The effects of single context on vocabulary knowledge,’ Language Teaching Research 11: 63–81. Wolter, B. 2001. ‘Comparing the L1 and L2 mental lexicon: A depth of individual word knowledge model,’ Studies in Second Language Acquisition 23: 49–69. Wolter, B. 2006. ‘Lexical network structures and L2 vocabulary acquisition: The role of L1 lexical/ conceptual knowledge,’ Applied Linguistics 27: 741–7. Xiao, R. and McEnery, T. 2006. ‘Collocation, semantic prosody, and near synonymy: A crosslinguistic perspective,’ Applied Linguistics 27:103–29. Zimmerman, C. 1997. ‘Do reading and interactive vocabulary instruction make a difference? An empirical study,’ TESOL Quarterly 31: 121–40.
Applied Linguistics 29/4: 717–722 ß Oxford University Press 2008 doi:10.1093/applin/amn040 Advance Access published on 24 October 2008
FORUM
The Significance of First Language Development in Five to Nine Year Old Children for Second and Foreign Language Learning ROBERT VANDERPLANK University of Oxford, UK In this article, the language learning experiences and development of a child (the author’s daughter) between the ages of five and nine are drawn on to argue that we should re-focus our comparison of first and second language acquisition away from early L1 acquisition to the early schooling/middle childhood period. In addition to the transforming experiences of formal and informal education in and out of school, it is suggested that children also acquire essential attributes as part of their natural development such as the ‘mind’s eye’ and adult-like de-contextualized memory, which ensure that they may eventually become adult L1 users. The article concludes by advocating that we should re-read texts comparing first and second language acquisition and learning using the child in the early years of formal schooling as the basis for comparison with second language learning.
INTRODUCTION The essence of my argument is that the period from five to nine years of age, often referred to as middle childhood, provides a more relevant and useful model for foreign language learning by adolescents and adults than early child language acquisition models. My observations are based on my daughter, Pippa, now nine years old. In four years, from the five year old with charming but impoverished, formulaic and error-laden child-speech, she has developed through natural development plus a great deal of effort, application and support into someone who is not only linguistically socialized and literate but also, and more important for second language acquisition research, someone with adult-like memory and recall, with inner speech and a mind’s eye. It may appear that in focusing on this period, when full-time schooling, socialization and becoming literate take over much of a child’s life, I am arguing a wholly environmental, nurturing case. As will be suggested, there are changes in children during this period which may be regarded as developmental and which are at least as important for them in becoming functioning adult native speakers as any developmental linguistic milestones which have gone before.
718 FORUM
DEVELOPING ADULT-LIKE MEMORY First, one key attribute Pippa did not have prior to the age of five was the sort of memory which you and I have. Her memory was still largely locative and contextualized (recall of people, activities, events triggered by places), in need of considerable scaffolding and very limited in its capacity to retain much information in any way which required deliberate conscious recall, as with most children at this age (Wood 1998: 70). But surely children can learn songs and are able to repeat stories back word-by-word? Of course, they do have a memory for these things but it is not the same memory which we adults have, the ability to memorize consciously, both verbally and visually, then retain and recall in a flexible way (Wood 1998). Ask children under five to rehearse what they are going to say to grandma on the telephone and most will find this a very difficult exercise. They can happily speak spontaneously but rehearsal and recall are usually beyond them. The literature on child memory suggests that while there are still issues as to whether the development of children’s memories is largely a matter of its being fostered through social interaction and schooling (Wood 1998: 80–1) or whether it is a matter of mental development (Case 1985, 1992; Halford 1992), it is clear that there is significant agreement about the qualitative changes in the speed and capacity of children’s memory from the age of five—changes which are crucial to the development of thinking in language and learning through language.
AN INNER VOICE AND MIND’S EYE Another crucial milestone of language development which develops during the critical five to nine years period is the ‘inner voice’ or ‘inner speech’ (Vygotsky 1978, 1986). Pippa’s ‘inner voice’ developed during the summer following her sixth birthday when she started reading silently. Occasionally she would sub-vocalize a new or difficult word or phrase, but most of the time, the reading of Pippi Longstocking was silent and inside her. She appeared to be able to store written information or hold words long enough (or perhaps long enough strings of words) in memory to process text. Once you have an inner voice or inner speech, you have an invaluable learning tool, one which allows you to use language as a tool for thinking (Olson and Astington 1990; Olson 1996; Wood 1998). At the age of five, Pippa also lacked the crucial ‘mind’s eye’ for language (the mind as a means of imagining scenes, views, language, etc.) and some of her friends even at eight years of age still do not appear to have it (or, if they do, it is relatively unused, unconscious or undeveloped). There are abundant references to the term ‘the mind’s eye’ in the literature on child development (e.g. Lerner 2002) and first language memory (e.g. Bower 1972; Kosslyn 1988), especially with regard to children with language impairment, but there is little or nothing in the second language literature. I had asked her if she could see something written or depicted in her head occasionally over two
FORUM
719
years and she was not able to do it. At seven plus (more accurately, 7;4), while testing her spelling, I realised that Pippa had reached this most crucial of developmental milestones. The mind’s eye allows you to develop your memory and recall exponentially. In the mind’s eye, a child has a reasonably reliable means of recall and can begin learning through language and visualization in earnest. Unlike children in whom it has not yet appeared, children with the mind’s eye do not need to learn words and strings of words in strict sequences, as in songs and nursery rhymes. Words and phrases in lists can be recalled in random order. English spelling can be mastered with relative ease. It is no accident that it appears at the age when serious schooling is under way.
MAKING CONNECTIONS Observations on first language acquisition past the age of five are hardly new and there is a rich literature to be tapped for insights into second language learning. For example, within the developmental psychology framework, Lloyd et al. (1998) in their study of how children miss or understand ambiguity at different ages highlight the phases which five to eleven year old children go through in developing their speaking and listening skills. Lee et al.’s (2001) study of ‘say’ and ‘mean’ in three to seven year olds reminds us that the concept of wording as distinct from meaning appears not to be fully developed until about age six or seven. In addition, there is a substantial body of research which has looked at the development of language and culturally situated behaviour in different societies at different ages within the framework of language socialization (e.g. Ochs and Schieffelin 1984; Schieffelin and Ochs 1986; Harkness 1990; Cook 1999) and another body of research within the framework of literacy studies, which examines the close connections between language development in and out of school, socialization in family and school contexts, and specific functional skills (Taylor 1983; Maybin 2006, 2007). What emerges from these research areas of particular relevance to language learning, both L1 and L2, is that there is an enormous amount of language development during this period, closely linked to the development of the self and usually linked to schooling. From all three perspectives, language is a powerful resource which is not only central in developing social interaction during middle childhood but also acts as a toolkit for developing the mind during this period. In view of such rich research resources which stress the developing and multi-faceted nature of the child’s language during middle childhood, it seems surprising that so few links have been made between middle childhood language development and second language learning. As they steadily develop the knowledge, skills, and attributes described above, children are becoming more like adults, equipped with the cognitive and linguistic tools for undertaking large learning tasks, reflecting on their knowledge and experience.
720 FORUM
HOW FOREIGN LANGUAGES ARE LEARNED: A NEW PERSPECTIVE Where do the above observations and insights take us? One starting point would be to re-read Chapter 2 of Lightbown and Spada’s (2006) How Languages are Learned from the perspective of an L1 learner in middle childhood. Chapter 2, ‘Explaining second language learning’ describes a number of theories and hypotheses regarding first and second language learning. However, the comparisons are only between early childhood first language learning and later (from school age onwards) second language learning. Missing is any discussion about first language development during the crucial early school years and its relevance to second and foreign language learning. As you read each section on different theories and hypotheses, consider the novice first language learner in the primary school rather than the toddler. From this perspective, Schwartz’s (1993) view that instruction and feedback change only the superficial appearance of language performance and do not really affect this underlying systematic knowledge of the new language becomes an absurd proposition. In the primary school context, through instruction and feedback, teachers transform young learners’ knowledge of their mother tongue and how they use it on a daily basis, as do parents and peers. While comprehensible input and Krashen’s other hypotheses have long been useful metaphors for the early stage of naturalistic language learning, whether first or second, they do not take us to the ‘learning user’ which a six year old is. In the cognitivist/developmental section of the chapter, attention is focused on how learners build up knowledge which can eventually be called on automatically for speaking and understanding. Is a six year old a proficient speaker, someone who can choose words, pronounce them and string them together with appropriate grammatical markers? Many of the interactional encounters that the adult meets in the course of a day would be beyond the child even in linguistic terms, let alone in functional terms. At least this model allows for linguistic development from middle childhood onwards in a way which Universal Grammar and Krashen’s hypotheses do not. So, too, do the connectionist and sociocultural models which Lightbown and Spada describe, the latter having obvious relevance to my argument. In the five to nine period, children are in the process of gaining mastery over their language, learning through social interaction, developing language skills with explicit support from teachers, parents, and others, practising these skills a great deal to the point where they become internalized and available for instant use. What are the implications for applied linguistics in shifting the comparison with second language learning from early childhood to middle childhood, or extending the comparison to encompass a much broader age range which includes all the developmental attributes of middle childhood? Just as five year old children bring their language to school and develop it, so adults
FORUM
721
bring at least one language to their foreign language learning and transform it. Some of the attributes of the adult or adolescent learner outlined by Lightbown and Spada (2006: 29–31) are precisely those which are developing in the L1 learner as well as in the young L2 learner. Our models of second or foreign language learning would need to include more explicit teaching and learning, more emphasis on memory, conscious effort, rehearsal and reflection, more emphasis on reading and writing, in addition to oral communication and interaction. In reading, for example, there are comparisons to be made between reading ages and ability in children and equivalent reading ability in older language learners. It may be that reading in L1 and L2 is the site where conflicting models may be reconciled since those advocating the primacy of acquisition over learning in a foreign language must explain how extensive reading, which requires such a dynamic, reflective, and learned skill as literacy is to be regarded as an ideal source of comprehensible input (Mason and Krashen 1997; Ying Lao and Krashen 2000). My own informal research in this area suggests that L2 learners find it quite helpful to be informed that they have a reading ability equivalent to an eight or nine year old (in many languages, books for children are helpfully graded in this way). First year university students entering a British modern language degree course with good grades tend to have a reading age of about twelve in the foreign language—something which comes as a shock when you are being asked to read Camus, Balzac, Calvino, or Bo¨ll. From the second language perspective, an eight or nine year old with average reading ability would struggle to tick the B2 boxes on the Common European Framework scale for reading (and listening) proficiency. Most nine year old readers have a very long way to go to reach assumed adult reading ability. In addition to re-reading Lightbown and Spada (2006), it would also be worth re-reading the seminal article by Firth and Wagner (1997) and the responses to it with young school-age children in mind. The importance of the five to nine period is that language development is not just a matter of nurture, although schooling obviously dominates most children’s lives. During this period, I have suggested that there are also critical developmental milestones which are recognised in the L1 literature but are missing from the L2 literature. Achieving these milestones enables children to become adult language users later on. Without them, our children would be functioning at the level of the impoverished models of language learning which have tended to dominate language teaching and learning for some thirty years.
REFERENCES Bower, G. H. 1972. ‘Mental imagery and associative learning’ in L. Gregg (ed.): Cognition in Learning and Memory. New York: Wiley, pp. 51–88. Case, R. 1985. Intellectual Development: Birth to Adulthood. New York: Academic Press.
Case, R. 1992. The Mind’s Staircase: Exploring the Conceptual Underpinnings of Children’s Thought and Knowledge. Hillsdale: Erlbaum. Cook, H. M. 1999. ‘Language socialization in Japanese elementary schools: Attentive
722 FORUM
listening and reaction turns,’ Journal of Pragmatics 31: 1443–65. Firth, A. and J. Wagner. 1997. ‘On discourse, communication, and (some) fundamental concepts in second language acquisition research,’ Modern Language Journal 81/3: 285–300. Halford, G. S. 1992. Children’s Understanding: The Development of Mental Models. Hillsdale, NJ: Lawrence Erlbaum. Harkness, S. 1990. ‘A cultural model for the acquisition of language: Implications for the innateness debate,’ Developmental Psychobiology 23/7: 727–40. Kosslyn, S. M. 1988. ‘Imagery in learning’ in M. S. Gazzaniga (ed.): Perspectives in Memory Research. Harvard: MIT Press. Lee, E., N. Torrance, and D. Olson. 2001. ‘Young children and the say/mean distinction: Verbatim and paraphrase recognition in narrative and nursery rhyme contexts,’ Journal of Child Language 28: 531–43. Lerner, R. M. (ed.) 2002. Concepts and Theories of Human Development 3rd edn. Mahwah, NJ: Erlbaum. Lightbown, P. M. and N. Spada. 2006. How Languages are Learned 3rd edn. Oxford: Oxford University Press. Lloyd, P., S. Mann, and I. Peters. 1998. ‘The growth of speaker and listener skills from five to eleven years,’ First Language 18: 81–103. Mason, B. and S. Krashen. 1997. ‘Extensive reading in English as a foreign language,’ System 25: 91–102. Maybin, J. 2006. Children’s Voices: Talk, Knowledge and Identity. Basingstoke, UK: Palgrave Macmillan.
Maybin, J. 2007. ‘Literacy under and over the desk: Oppositions and heterogeneity,’ Language and Education 21/6: 515–30. Ochs, E. and B. Schieffelin. 1984. ‘Language acquisition and socialization: Three developmental stories and their implications’ in R. A. Shweder and R. A. LeVine (eds): Culture Theory: Essays on Mind, Self and Emotion. Cambridge: Cambridge University Press. Olson, D. R. 1996. ‘Toward a psychology of literacy: On the relations between speech and writing,’ Cognition 60: 83–104. Olson, D. and J. Astington. 1990. ‘Talking about text: how literacy contributes to thought,’ Journal of Pragmatics 14: 705–21. Schieffelin, B. and E. Ochs. 1986. ‘Language socialization’ in B. J. Siegel, A. R. Beak, and S. A. Tyler (eds): Annual Review of Anthropology 15: 163–91. Schwartz, B. 1993. ‘On explicit and negative data effecting and affecting competence and linguistic behavior,’ Studies in Second Language Acquisition 15: 147–63. Taylor, D. 1983. Family Literacy: Young Children Learning to Read and Write. Exeter, NH: Heinemann. Vygotsky, L. S. 1978. Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press. Vygotsky, L. S. 1986. Thought and Language. Cambridge, MA: MIT Press. Wood, D. 1998. How Children Think and Learn 2nd edn. Oxford: Blackwell. Ying Lao, C. and S. Krashen. 2000. ‘The impact of popular literature study on literacy development in EFL: more evidence for the power of reading,’ System 28/2: 261–70.
Applied Linguistics 29/4: 723–736
ß Oxford University Press 2008
REVIEWS H-Dirksen Bauman, Jennifer Nelson, and Heidi Rose: SIGNING THE BODY POETIC. ESSAYS ON AMERICAN SIGN LANGUAGE LITERATURE. University of California Press, 2006. Signing the Body Poetic offers a rich compilation of documents, analyses, and references. Both people involved in the ‘sign’ field and/or concerned with d/Deafness would use this volume either as a direct source of poetic material, or as an indirect access to references on videos, literary theories, accounts, and reports about cultural events. The volume provides complementary perspectives: it adds description on literate events and poetic corpora. At the same time, it puts the light on the epistemic dimensions. As the writers belong to both hearing and deaf communities, this allows for multiple points of view. Finally, having language as its matter, the volume is cross-disciplinary, as it connects with literary studies, linguistics, philosophy, arts, and semiotics. The title of the volume invites a focus on the body as a crucial element when considering Literature in American Sign Language (ASL). From the body, two main questions ensue, one about the formal aspects of the visuo-gestural modality and another about meaning. The volume’s aim is twofold: (i) it describes poetry as it is embodied and shows how poetics should take this ‘embodiment’ into account; (ii) it underlines the political dimension of the signed literature. Although the authors have organized the collection around three separate themes (‘Framing the ASL literature’, ‘The embodied text’ and ‘The political text’), it is clear that the formal and political dimensions interconnect. As for the primary texts, the reader will find, in the book and DVD, clips with or without subtitles, transcriptions of analysed performances, and pictures illustrating performances. The time has come to view for the first time (in a single DVD) well-known performances. Indeed, performances such as the ‘BOAT BOAT’ by Kannapell, or ‘Snowflake’ by Valli, were described or even transcribed through glosses and pictures in previous studies, but this cannot be compared with direct access to recorded performances. How should one express one’s feeling when watching the original version of ‘Poetry’ by Flying Words Project, or the ‘Gallaudet Protest’ by Eastman, performed in their authentic context? The authors give us the opportunity to come close to ‘attending’ the performance by including the audience’s behaviour, including breathing sounds! As for the issues presented here, it may be that, as a linguist, you feel hungry for more explicit description of either morpho-syntactic or phonological structure (as on p. 141 when Rose mentions ‘specific rhythms and spatial prosody’ without telling more, or on p. 151 when Wolter and Cook start addressing the issue of eye contact in the topic–comment format), and it would be interesting to complete the linguistic approach with Sutton–Spence’s recent work on British
724 REVIEWS
Sign Language poetry for instance. In addition, the reader will note a strong philosophic dimension (with many references to Derrida, Deleuze a.o.) and she will note that the interdisciplinary perspective announced (p. 13) is respected. One of the main benefits and a novelty of this volume consists of its bilingual bimodal medium. The clips on the DVD should not be seen just as an illustration of the written chapters, nor should they be viewed after the reading part. Early on in the Introduction, the editors invite the reader to consider the signed version of their introduction (it is more than a translation of the written form); and from the very start of the book, the editors remind the reader that the best way to grasp ASL is to watch poetry in it. Then, every chapter invites the reader to go back and forth: reading and watching. The user’s guide secures optimum use of this double material, but would have been spent better at the very beginning of the book. The reader may feel a bit confused with the proliferation of contributors and references or, on the contrary, the reader may appreciate the profusion of information sources. In both cases, the Afterword by Padden builds an interesting bridge between the literate events and the analyses proposed in the 10 chapters. The volume and DVD consist of original mixed contributions by deaf and hearing people, poets, teachers, and/ or researchers (all three in one for some of them!) and the forms they take on are also original and ‘mixed’: this is not only true for the performances that you can get in their signed forms but also for some of the analyses, either echoing the written form or complementary to it. Hence, this collection underlines the comparison between various types of artistic composition: graphic, cinematic, dramatic, vocal, or gestural. This is an opportunity to point out the differences and analogies between these means of expression and art forms. It should not come as a surprise that Gallaudet is the ideal melting pot for the majority of the performances and issues presented here. As the university embodies the Deaf pride (see the clip of the ‘Gallaudet protest’ as a piece of evidence), the ‘Gallaudet people’ cannot be accused of ‘hearing thinking’. Hence, they legitimately address such questions as influence and interconnections between the hearing and the deaf cultures, the role of the voice, etc. Note, for instance, in Chapter 2, that Bahan, as a deaf poet and linguist, introduces his overview of deaf literary patterns by considering that this is ‘something very much like an oral tradition’ (p. 21) and even that ‘it exists along [in the Deaf community] an oral–literate continuum’ (p. 22). Similarly, Davidson, in Chapter 11, addresses the issue of the ‘Hearing things’ (the speech in deaf performances). It is all the more surprising then to fall on a remark like ‘the trend of more hearing people learning to sign [. . .] might [. . .] even affect the purity of ASL itself’ (Krentz, p. 67), as if ASL was not a living language, subject to variation and interference with its linguistic environment, like any language. My main, but slight, reserve is about what I would call ‘Gallaudetocentrism’. The subtitle ‘ASL Literature’ clearly shows that the editors do not pretend to be embracing the whole of d/Deaf culture around the world, so it would be unfair to criticize them for presenting only ASL poetry (especially as they also raise the question of future directions, p. 15). Nevertheless, I note
REVIEWS 725
a few points where some contributors seem to omit if not ignore the rest of the deaf community, or the rest of the world (at least the non-Western one). ASL poetry and ASL poets are also presented as special, even unique, although the authors could have considered these characteristics in comparison with European deaf poetry (in order not to focus on ASL) or with oral-transmission cultures (in order not to focus on gestural versus vocal channel). The unicity claim is all the more obvious when poetics and literary theory are criticized as ‘phono-centric’. I would suggest that Western (hearing) written poetics is just as challenged when facing the oral literary traditions of Asia and Africa. There is also the editors’ suggestion to ‘break away from the vocabulary of the spoken– written form literature’ (p. 12): is this not true for all face-to-face literature, whether signed or oral? When Bahan proposes that ‘some elements of vocal songs are transposed into the signed modality, such as fluidity of words/signs and the rhythm’ (p. 34), I wonder why he is considering fluidity and rhythm as vocal rather than gestural. As a further illustration, see for instance Cook signing, in clip 5.5, that Bragg ‘has created’ the ‘visual vernacular’ technique, underlining ‘the ability to shift point of view (p. 151)’: would it not be right to think that this is a common way of playing with signs, commonly used in the Deaf folklore? It seems to me that the real creation is the way this process, through its analysis, is considered a resource of a poetic register. In another chapter, some poetic patterns are called ‘emerging’, ‘new’ or ‘original’, when they actually receive exposure. On the contrary, Krentz considers that the ‘one, two, onetwo-three’ routines [have] largely disappeared’, while one finds this kind of rhythmic patterns nowadays in signed nursery rhymes and in chants (either to support one’s team or to protest). In both cases, it seems that the authors mistake the existence of a form for its recognition and distribution. My critical comments originate in the abundance and richness of the volume: as it offers a lot, one is attending to everything! ‘Ethnocentrism’ also has a political dimension: the contributors are committed to the recognition of their cultural identity through artistic expression and performance analysis. The struggle for recognition is presented as never-ending and it would appear that this is a strong subjectivity that cannot be by-passed easily. Still, the volume is a noticeable step forward for research in deaf culture, sign languages, and poetics. Hence, I would recommend any deaf studies department to procure this book and DVD for their students. The time has also come to support the same kind of collections for other sign languages, and d/Deaf communities (as the European Cultural Heritage Online program indeed proposes, by collecting poetry and other performances, alongside those of endangered languages, and by sharing annotated corpora and analyses on the Web). Reviewed by Marion Blondel CNRS-Universite´ de Paris 8, France doi:10.1093/applin/amn041 Advance Access published on 24 October 2008
726 REVIEWS
Ivana Markova´, Per Linell, Miche`le Grossen, and Anne Salazar Orvig: DIALOGUE IN FOCUS GROUPS: EXPLORING SOCIALLY SHARED KNOWLEDGE. Equinox, 2007. Focus groups are a technique for eliciting research data by holding a series of moderated discussions with small numbers of people, perhaps six or eight, which are often recorded and transcribed for analysis. Robert Merton was one of the originators of the technique with the use of ‘focused interviews’ for the evaluation of propaganda in the 1940s. He left them behind when he finished his war work and went on to other things, but forty years later he commented on the huge industry that his work had helped start. In an aside he notes that focus groups ‘are not, of course, groups in the sociological sense of having a common identity or continuing unit, shared norms and goals’ (Merton 1987: 555). One way to read Dialogue in Focus Groups is as a response to Merton. The authors argue that focus groups can usefully be seen as groups, however temporary, contingent, heterogeneous, and in some cases awkward, and that this approach to them has implications both for social research methodology and for our understanding of language use and of social knowledge. For some researchers, the fact that focus groups are groups is a source of bias, distorting what should be purely individual responses, but for the authors of this book, the fact that each person in the group accommodates others is not a problem, but an opportunity. The book is not a guide to the practice of focus group research, though the authors provide a good list of such guides (p. 222). The focus in the title is on the first word: dialogue. As they point out, ‘it seems that everyone is in dialogue’ (p. 7), nation and nation, governments and citizens, corporations and stakeholders. But they mean the word in a more profound sense than just the exchange of words or the exchange of views; they mean, among other things, the way all utterances are packaged for others, in terms of shared knowledge, and in terms of a dialogue each speaker has with him or herself. It is routine now to cite Bakhtin and Vygotsky in social research, but I don’t think I had quite grasped what it means in research practice to follow through their lines of thought. One of the appeals of the book is its diversity: four researchers, from three disciplines (sociology, social psychology, and linguistics), working on six studies that were done in five countries and (I think) five languages (with the examples translated into English, which is not an easy task for these data). The book is not an edited collection, but an attempt at a synthesis of the rather different styles, interests, and aims of these researchers in a single work. I thought there were some places in which the joins showed, for instance in the overlap of Chapters 6 and 7, which in my reading seemed to apply similar terms to different data with different emphases. It might have helped to have the authors of each chapter take their examples
REVIEWS 727
from all the sets of data, so we could see more clearly the conceptual development from one chapter to the other. For the most part, chapters by the different researchers fit together and illuminate one another, for instance in the way the detailed discussion of socially shared knowledge in Chapter 1 connected to both Chapters 6 and 7. The book lists its key concepts (p. 203) as: (i) focus groups as a form of interaction; (ii) heterogeneous subjects, with conflicting roles and tensions between various possible stances and voices; (iii) the circulation of ideas; and (iv) socially shared knowledge. Others have noted that focus groups involve conversational interaction (Puchta and Potter 2003; Myers 2004), but there are new insights here in the analysis of focus groups as a communicative activity type. The authors stress the importance of sociability, a kind of interaction for its own sake, in contrast to the kinds of instrumental motivations that are often the basis for genre analysis. The heterogeneity of the subject is of course a touchstone of poststructuralist social science. The new contribution here is in the analysis of how aspects of the subject emerge in the data, for instance in the use of ‘virtual voices’, hypothetical reported speech (e.g. ‘I don’t go to a therapist and say ‘‘look usually every Friday I kill someone’’ ’ (p. 124)). Circulation of ideas refers to the ways participants frame utterances rhetorically, both drawing on and developing common arguments and themes; examples include we/they dichotomies, extreme case formulations, and irony (p. 99). That knowledge is socially shared, not contained within individuals, is another touchstone of poststructuralist approaches. As the authors note, following Schu¨tz (p. 29), knowledge can be shared in several ways: in the assumption of reciprocity (you know what I know); in the reliance on various kinds of common sense (including topoi and familiar arguments); and in the social distribution of knowledge, so that it is assumed that some people know about one thing, while others will know about other things (and the implication is that we are all entitled to be ignorant of some things). Again the new contribution, for me, is in the analysis, tracing themata (in this case broadly held implicit beliefs defined in terms of oppositions) in collaboratively produced utterances (p. 183). One implication of these studies for methodology is to break down the opposition between traditional content analysis and discourse analysis of research transcripts. It is argued that ‘contents of talk have dynamic structures and the themes of talk show progression and changes’ (p. 202). Again, the point has often been made, but the practical implications for analysis have not always been taken up. The implications are that the topics and themes coded in a social research project cannot be taken out of their context in the talk (as many social researchers do), and that the transcript of one session is never enough (as discourse analysts may want it to be), because the talkers presuppose shared knowledge of activity types and cultural themes. If Dialogue in Focus Groups were just another methodological guide, it would be useful to focus group researchers and not to other applied linguists.
728 REVIEWS
Instead, it is a much deeper, and more important, inquiry into how we analyse and understand talk that has been conducted for social research. Reviewed by Greg Myers Lancaster University, UK doi:10.1093/applin/amn042
Advance Access published on 24 October 2008
REFERENCES Merton, R. K. 1987. ‘The focused interview and focus groups—continuities and discontinuities.’ Public Opinion Quarterly 51(4): 550–66. Myers, G. 2004. Matters of Opinion: Talking about Public Issues. Cambridge: Cambridge University Press.
Puchta, C. and J. Potter. 2003. Focus Group Practice. London: Sage.
Simon Gieve and Ines Miller: UNDERSTANDING THE LANGUAGE CLASSROOM. Palgrave, 2005. Understanding the Language Classroom is a collection of twelve papers which respond to Dick Allwright’s six directions for charting developments in classroom research over the last five decades, while suggesting a number of improvements (Allwright 2006). The book opens with a foreword by Bailey who acknowledges Allwright’s influential contribution in broadening our understanding of the scope of research in the language classroom. Her succinct and excellent synopsis illustrates the significant and complex developments which have marked the field. The introduction by Gieve and Miller provides important background information to enable the reader to follow the rationale behind Allwright’s six ‘promising’ directions. It explains that Allwright’s own understanding of language development in the classroom stems from his earlier work that ‘learners don’t simply learn what teachers teach, but that learning arises out of the learning opportunities’ (p. 1) which are created through the interactive work of all those present in the classroom. From this perspective, it is not helpful to try to make sense of what goes on in the classroom by looking for the ‘best method’, the ‘perfect task’ or the ‘ideal condition’. Classroom processes are the result of participants’ interactions which are rooted in the (inter-)subjective processes which underlie their collective work and cannot therefore be accounted for by technical efficiency and general solutions to local situations. This is echoed in Allwright’s six directions, as stated in Chapter 1: 1 From prescription to description to understanding moves from the rejection of ‘method’ as a single and prescriptive influential variable of the 1960s
728 REVIEWS
Instead, it is a much deeper, and more important, inquiry into how we analyse and understand talk that has been conducted for social research. Reviewed by Greg Myers Lancaster University, UK doi:10.1093/applin/amn042
Advance Access published on 24 October 2008
REFERENCES Merton, R. K. 1987. ‘The focused interview and focus groups—continuities and discontinuities.’ Public Opinion Quarterly 51(4): 550–66. Myers, G. 2004. Matters of Opinion: Talking about Public Issues. Cambridge: Cambridge University Press.
Puchta, C. and J. Potter. 2003. Focus Group Practice. London: Sage.
Simon Gieve and Ines Miller: UNDERSTANDING THE LANGUAGE CLASSROOM. Palgrave, 2005. Understanding the Language Classroom is a collection of twelve papers which respond to Dick Allwright’s six directions for charting developments in classroom research over the last five decades, while suggesting a number of improvements (Allwright 2006). The book opens with a foreword by Bailey who acknowledges Allwright’s influential contribution in broadening our understanding of the scope of research in the language classroom. Her succinct and excellent synopsis illustrates the significant and complex developments which have marked the field. The introduction by Gieve and Miller provides important background information to enable the reader to follow the rationale behind Allwright’s six ‘promising’ directions. It explains that Allwright’s own understanding of language development in the classroom stems from his earlier work that ‘learners don’t simply learn what teachers teach, but that learning arises out of the learning opportunities’ (p. 1) which are created through the interactive work of all those present in the classroom. From this perspective, it is not helpful to try to make sense of what goes on in the classroom by looking for the ‘best method’, the ‘perfect task’ or the ‘ideal condition’. Classroom processes are the result of participants’ interactions which are rooted in the (inter-)subjective processes which underlie their collective work and cannot therefore be accounted for by technical efficiency and general solutions to local situations. This is echoed in Allwright’s six directions, as stated in Chapter 1: 1 From prescription to description to understanding moves from the rejection of ‘method’ as a single and prescriptive influential variable of the 1960s
REVIEWS 729
2
3
4
5
6
and early 1970s to the descriptive processes of the 1980s revealing lessons as ‘socially constructed events’ and thus highlighting the importance of the learners’ roles in this process. From simplicity to complexity realises that these descriptive processes have uncovered a constellation of complex variables which makes it futile to search for the ideal method that suits all teachers and all learners in all classrooms. From commonality to idiosyncrasy accepts that learner idiosyncracy is a phenomenon that is far more representative of the classroom situation than is commonality. From precision to scattergun recognizes that, rather than predicting precisely what will be learned by manipulating psycholinguistically motivated task features, the creation of a co-constructed understanding of the local situation must be encouraged to enable the emergence of a wide range of productive learning opportunities from which all can benefit. From teaching and learning as ‘work’ to teaching and learning as ‘life’ proposes that learning opportunities do not arise from efficiency in the work but from a collegial atmosphere conducive to a better quality of life. From academics to practitioners as the knowledge-makers in the field is about the realisation that understanding the local situation is an essential condition for any attempt at changing an irreducibly complex situation; it has led Allwright to propose ‘Exploratory Practice’ as a more inclusive and viable alternative to researching the language classroom.
As the editors explain, the book is not about Allwright’s work but about how it is echoed in the work of so many others. The contributors’ responses fall into two main themes, albeit not so coherently: Chapters 2–8 relate to what goes on in the classroom and Chapters 9–12 to language teacher education and development. In Chapter 2, Gieve and Miller concentrate on the concept of quality of life, arguing for a notion of ‘classroom awareness’ which grows out of the participants’ collegial efforts to develop a mutual understanding of their context in order to create an environment conducive to learning. Next, Holliday denounces native-speakerism as a central issue in TESOL and exposes the simplicity with which the profession has appropriated the custody of the English language, prescribed its teaching methodology, and ignored its global ownership. Chapter 4 by Wright contrasts the view of classroom management centred on organization and control of language groups with one where opportunity for development and ‘quality of experience’ lie at the heart of intertwined social, cultural, psychological and institutional forces. Wood in Chapter 5 focuses on understanding how the negotiation of decision-making takes place and what aspects might constrain its achievement. He concludes that what is being negotiated between teachers and learners is not techniques but beliefs, roles, and responsibilities. Next, Coleman uses a Darwinian metaphor arguing that external and prescriptive proposals for improving
730 REVIEWS
pedagogical efficiency inevitably fail to consider the determining pressures which, uncontrollably, shape the achievement of the local adaptation. In Chapter 7, Tseng and Ivanic find common ground with Allwright’s directions and build their conceptual framework for adult literacy education on the integration between teaching, learning, and ‘life’ as well as the recognition of adult learners’ idiosyncracies. Chapter 8 by Tarone explores why the same successful language immersion experiment which produced excellent speakers of French at lower grade levels failed to persuade those at upper grade levels to speak the school language. Tarone detects a mismatch between the classroom speech community, academic in nature, and the students’ actual social roles and social needs outside the classroom. In Chapter 9, Fanselow and Barnard propose their own interpretation of what might constitute ‘Exploratory Practice’ as a tool to be used by teachers to develop a better understanding of their professional activity. Next, Breen identifies four pressures which challenge the ELT practitioners’ professional identities: the knowledge that teachers are supposed to have; the pressures for change in the way they teach; the preoccupation with ‘performativity’, and insecurity in their working conditions. Breen subsequently proposes seven desirable features for teacher programme development grounded in localized communities of practice. These, at the same time, deal with global issues which impact on teachers. Celani’s own professional context in an inservice English teacher education programme in Chapter 11 finds resonance with Allwright’s directions as opposed to the many other Brazilian teacher education courses where the prevailing assumption about technical efficiency, myth of the best method, and simplistic problem-solving seem to be the norm. Finally, Freeman in Chapter 12 identifies three primary obstacles to teachers’ professional development: the tendency to believe that what happens in the classroom is between individuals not collectivities; the focus on the here-andnow ignoring historical processes, as well as the belief that the actions of an individual teacher cause individual learning. The contributors have all found resonance with the broad applicability of Allwright’s thinking. Their conclusions rise against global and top-down prescriptive solutions that are universally applied. They also coherently converge towards the acknowledgement that language lessons are locally negotiated and are the result of a complex and idiosyncratic sociolinguistic dynamic. It is precisely this complexity and idiosyncracy that have been overlooked by some task-based syllabi research which recommends to teachers particular tasks on the sole basis that they generate, for instance, a high number of conversational adjustments. Research of this kind has ignored the qualitative aspects of the quantitative use of conversational modifications and hopes that predictability is not a perennial problem but just a temporary technical one that some imaginative task designers will resolve with just a little more effort. By way of contrast, the explicit message that Allwright sends out and is
REVIEWS 731
reflected throughout the contributors’ responses is that language learning in the classroom is not a simple matter and that its inherent complexity is an inescapable reality that we must embrace. It is worth noting here that LarsenFreeman’s work on chaos/complexity theory does address complexity issues, even though it is related to second language acquisition rather than language teaching and learning in the classroom. Surprisingly, her work has only received a brief mention in Chapters 7 and 12. Nevertheless, the book remains an outstanding contribution to sociocultural perspectives of language teaching and learning stressing that the countable and observable elements of classroom life are only the tip of the iceberg. In concert, the contributors see that the human qualities of the classroom participants are at least as important, or more, than their purely cognitive functioning. They have succeeded in shaking the much taken-for-granted belief that efficiency in work and technical solution would be the way forward. In this respect, Allwright has maintained his well established way of questioning our deep seated assumptions about teaching and learning processes. The reader is reminded that he did so nearly 30 years ago through ‘Igor’ (Allwright 1980) when he warned us against equating ‘high input generators’ with ‘efficient and proficient speakers’. In the midst of our sometimes desperate efforts to relate interaction to language progress, Allwright cautioned us to recognize the complexity of classroom language learning and redirected our attention to the effect of input generation on those learners who ‘silently’ attend to other learners’ involvement. In other words, as early as the 1970s, directness, simplicity, and commonality had little room in Allwright’s consideration of classroom life. The book is impeccably edited. The clarity in its organization, its wide-ranging topics, its thought-provoking ideas as well as the reference lists spanning several decades of research will make it compelling reading for academics and postgraduate students in applied linguistics for many years to come. Reviewed by Assia Slimani-Rolls Regents’ College, London doi:10.1093/applin/amn043 Advance Access published on 24 October 2008
REFERENCES Allwright, R. L. 1980. ‘Turns, topics and tasks: Patterns of participation in language learning and teaching’ in Larsen-Freeman (ed.): Discourse Analysis in Second Language Research. Rowley, MA: Newsbury House, pp. 165–87.
Allwright, D. 2006. ‘Six promising directions in applied linguistics’ in S. Gieve and I. K. Miller (eds): Understanding the Language Classroom. Palgrave Macmillan, pp. 11–17.
732 REVIEWS
Sandra Mollin: EURO-ENGLISH: ASSESSING VARIETY STATUS. Gunter Narr Verlag, 2006. Jennifer Jenkins: ENGLISH AS A LINGUA FRANCA: ATTITUDE AND IDENTITY. Oxford University Press, 2007. These two books address core issues in the debate on English as a Lingua Franca and come to very different conclusions. They are both good summaries of the position held by applied linguists who favour English as a Lingua Franca [ELF] as an emerging variety of English (Seidlhofer 2004; Jenkins 2006) and those who find it a mere ‘Yeti of English varieties: everyone has heard of it, but no one has ever seen it’ (Mollin 2006: 1). The debate concerning the nature of ELF has been lively but often lacking in hard data on what users actually say and do with the language. On the one hand, corpus linguistics has drawn sweeping pedagogic conclusions from one-sided data based on ‘native-speaker’ uses of English (Carter and McCarthy 1995). The insights from native-speaker corpora have been applied, across the board, to L1- and L2-users of English, not always taking into account the diversity of modern uses of English around the world and practical classroom implications of ‘native-like’ authenticity (McCarthy and Carter 1995; Cook 1998). On the other hand, the school of thought which has emerged around the concept of ‘English as a Lingua Franca’ seems to be making the same mistake as the corpus linguists in sliding, uncritically, from description to prescription and, so far at least, on the basis of very little description. While the native-driven corpus linguist would seem to offer the L2-user ‘too much reality’ (including informal, culture-bound idiomaticity—Cook 1998; Prodromou 2008), ELF scholars, with their ‘reduced’ common core, do not offer enough reality: they simplify rather than amplify the options open to the L2-user. Proponents of ELF have identified what they see as ‘the reality of the spread of ELF’ (Jenkins 2007: 148) but this reality has not yet materialized in terms of frequent, systematic lexico-grammatical and pragmatic features. It is within this contentious framework that Mollin decided to collect her own corpus of L2 speech in international contexts, with a focus on European uses of English or Euro-English. Sandra Mollin’s book is one of the first attempts at beginning to fill the gap between the reconceptualization of ‘non-native speaker’ uses of English as a putative new variety and empirical evidence for the existence of such a variety. The two books under review take very different views of whether EuroEnglish is a variety of English, but they overlap in one area: their interest in the attitudes of users towards the validity of ELF. The main difference between the two books in terms of data is that Mollin goes out in search of the realworld existence of ELF forms and functions whereas Jenkins takes them for granted (Jenkins 2007: 238). Jenkins’ main focus is on the attitudes and beliefs of teachers towards ELF and how these are shaped by questions of identity. Whereas Mollin is interested in finding out whether Euro-English exists at all, Jenkins wants to know how teachers feel about it, and given the results of her
REVIEWS 733
research, to explain why there is relatively little acceptance of ELF amongst those on behalf of whom it is put forward as an alternative to native-speaker dominated Standard English. For Jenkins, this overwhelming resistance to the concept of ELF, by L1- and L2-users alike, is where beliefs about identity come in: the native-speaker ideology that sees native English as superior to other varieties has been imposed in many cases on the lingua franca user, contrary to their own interests (Jenkins 2007: 185–7, 230). Mollin investigates the empirical validity of ELF as a distinct and sui generis variety of English. The doctoral research on which this book is based takes Seidlhofer’s (2004) list of English as a Lingua Franca core features (e.g. simple present 3rd person -s omitted; omission of article; the interchangeability of relative pronouns who and which, etc.) and through a corpus of 400,000 words of spoken and email uses in Europe, attempts to establish whether these features are frequent enough and accepted enough by users to constitute an emerging Euro-English. In Chapters 1–3, Mollin outlines the debate about Euro-English within the framework of the historical emergence of New Englishes and the criteria for ESL status in terms of form, function and attitude. ELF-proponents frequently bracket ELF with well-established ESL varieties of English or World Englishes (Jenkins 2006: 157; 2007: 14–15, 205). Jenkins’ arguments in favour of ELF as a manifestation of the broader World Englishes movement often depend on predictions about future developments rather than existing facts: for example, she says ‘it seems entirely reasonable that there will ultimately be ELF literature’. This ‘ELF literature’ will be written in an English which will have been captured in ELF corpora; the assumption is that this variety of English will ‘make use of frequent and systematic’ linguistic features (Jenkins 2007: 14). The evidence for this ELF variety is yet to come: ‘it is critical to the future of ELF for its corpus linguists to provide detailed descriptions as quickly as possible’ (Jenkins 2007: 15). Meanwhile, Jenkins asserts that ELF sits ‘more comfortably within the World Englishes framework’ than in monolithic native-driven models that leave little scope for non-native speaker variation and innovation (Jenkins 2007: 17). How comfortably does ELF sit within a World Englishes framework? Mollin’s examination of specific criteria for ESL-status is particularly helpful in assessing the degree to which ELF has achieved sui generis varietal status. Mollin, in fact, picks up on Jenkins’ earlier prediction that Euro-English will follow developmental stages similar to that of codified Asian Englishes (Jenkins 2003: 38). She focuses on the criteria of nativization, institutionalization (the acceptance of a new and independent norm) and an expansion of the functions of English in the community (Mollin 2006: 44). These criteria involve extensive bilingualism, the use of English in education, administration, the media and creative writing. In ESL or World Englishes, Mollin argues, we find an extended range of styles and registers, a distinctive lexicon, phonology, syntax and discourse style. All of these characteristics must be ‘communal’ and systematic. The speakers of established nativized varieties accept the variety as a model; the model has begun to be codified as an officially recognized variety.
734 REVIEWS
The chapters that follow attempt to identify, through an examination of the data, the degree to which ‘Euro-English’ fulfils at least some of these criteria. Chapter 4 describes the functional uses of Euro-English and concludes that ‘Europe is a special case that is not easy to grasp with the instrument of the distinction between ESL and EFL’. Mollin’s is a refreshingly nuanced approach to the traditional tripartite distinction (ENL, EFL, ESL) which, Mollin argues, has been overtaken ‘by the realities of a globalized world: a distinction between settings with, on the one hand, typical postcolonial characteristics and, on the other, settings with no historical connection to the English-speaking world may no longer hold’ (Mollin 2006: 86). Thus, Euro-English lies somewhere between ELF and ESL, in terms of its functions. In Chapters 5 and 6, Mollin reports on the insights afforded by her Euro-English corpus. She presents evidence, or the lack of it, for the ‘formal independence’ of Euro-English and the attitudes of its users towards a putative Euro-variety. Her conclusion revisits the question with which her book began: is the label ‘Euro-English’ a legitimate description of an existing trend or is it a mere illusion? Mollin’s answer is that Euro-English is not a variety of English (p. 201) and ‘quite simply, Europeans do not want Euro-English’ (p. 200). Ironically enough, Jenkins’ book reports on research which confirms the unpopularity of ELF amongst both L1-users and L2-users of English, but her explanation for why this is so differs in important respects from Mollin. In Chapters 1 and 2 Jenkins attempts to put the record straight regarding the real nature of ELF, rebutting a host of critics of ELF and placing these doubters within a ‘standard language ideology’, which, she claims, prioritizes ‘nativespeaker’ models of English over new Englishes. Jenkins includes ELF as one offshoot of the New Englishes movement. In Chapters 3 and 4, the author reviews previous research on the impact of attitudes towards language varieties in order to expose the prejudice she detects on the part of applied linguists towards ELF varieties (p. 66). The section on previous research into attitudes towards ELF is criticized for not probing deeply enough into what Jenkins considers to be a ‘deep-seated’ bias in such research (p. 105). What Jenkins finds lacking in attitudinal research is a qualitative exploration of why teachers express scepticism towards ELF. In Chapters 5 and 6 Jenkins undertakes such an exploration in order to demonstrate empirically ‘how the prevailing native speaker standard ideology in ELT and applied linguistics may be affecting NNS teachers’ attitudes, beliefs and feelings about their own and other NNSs’ English accents’ (p. 65). Chapter 5 is a remorseless critique of the numerous dissenters from the official ELF position, all of whom Jenkins accuses of ‘misrepresenting’ (pp. 115, 125), ‘misunderstanding’ (pp. 112, 117) and ‘misconceiving’ (pp. 127, 133) the real nature of ELF. The second part of Chapter 5 provides an analysis of the largely anti-ELF attitudes expressed by participants in MA and other workshops conducted by the author. As in the case of dissenting applied linguists discussed earlier in the same chapter, Jenkins explains negative attitudes towards ELF on the part of MA students and teachers as the result of ‘confusion’ (p. 133) and
REVIEWS 735
‘ignorance’ (p. 138). Chapter 6 reports on the results of Jenkins’ questionnaire to teachers, which was designed to elicit attitudes towards ELF accents. The focus is on ESL accents rather than ELF accents, in spite of the book’s title. Nowhere in the book does Jenkins examine examples of what she sees as ELF as opposed to ESL/EFL accents and nowhere do we find data on the lexis, grammar, discourse and pragmatics of an albeit putative ELF. As far as accents are concerned, Jenkins finds the teachers to be ‘ill-informed’ (p. 148) and ‘biased’ and ‘prejudiced’ against ELF accents (pp. 156, 181) and holding the ‘same kind of prescriptive views of correctness’ as the general population (p. 148). Another reason, says Jenkins, why ‘non-native speakers’ hold these attitudes is as a result of ‘insecurity’ (pp. 162, 188). Moreover, argues Jenkins, non-native teachers have taken on board native-speaker ideology: ‘we must consider the possibility that the NNS teachers who took part in the study (along with most of their peer group) have, at least to some extent, been so strongly influenced (‘‘brainwashed’’ even) by the prevailing standard NS English ideology’ (p. 187). The final explanation Jenkins offers for the reluctance of her respondents to welcome ELF is that such scepticism relates to issues of identity, and specifically the clash between how the teachers see themselves as successful English teachers and as members of their own L1 groups. These issues are examined in detail in Chapter 7. In this chapter, Jenkins looks at the question of identity and ELF in greater detail, drawing on the work of Joseph (2004) and sociocultural views of language. Jenkins summarizes sociocultural and post-structural work on language and identity. This framework highlights the complexity of identity in ELF communities (p. 204). Again, Jenkins equates ESL identity formation with ELF: ‘precisely the same obtains for ELF varieties’ (p. 205). We are now approaching the end of the book and Jenkins has referred repeatedly to ‘ELF varieties’ (p. 205) but in 300 pages and eight years after Jenkins’ pioneering work on phonology in an ELF context (Jenkins 2000) we have still not been given examples of the frequent and systematic lexico-grammatical or functional features of ELF varieties. In this book, Jenkins does not even reproduce the short list of ‘unproblematic’ grammatical items put forward by Seidlhofer (2004) and quoted in her own earlier papers (e.g. Jenkins 2006). In her conclusion to the book, Jenkins insists that, although it is not yet possible to teach ELF (p. 238), this is no reason for not changing our mindset, until it is possible to do so. She argues against the sceptics that it is ‘manifestly clear to all who communicate internationally in English that ELF exists’ (p. 238). In her book, Jenkins describes what she sees as the obstacles that hamper the acceptance of ELF: gatekeeping practices by the ENL establishment, ENL ideology, and so on, but she is confident that, in future, the teaching of ELF will be possible; once the data have been collected and codification of the new variety achieved (p. 238). Jenkins’s argument would be immeasurably more convincing if at the heart of it there were not a huge absence, and that is the absence of evidence that ELF exists at all as an independent norm-generating variety of English. Instead, we get a re-run of points made about a lingua franca phonological
736 REVIEWS
core from Jenkins (2000), which itself is constructed on a narrow database and limited in application. Jenkins, unwittingly, enters into a dialogue with sceptics like Mollin and argues forcefully for the reality of ELF. Her argument is undermined by the lack of reference to the lexico-grammatical nature of ELF, which Mollin at least does try to identify via empirical analysis. There is much work yet to be done on discourse and pragmatics in settings where English is used as an international lingua franca. Mollin’s book has made a start in researching the area of lexico-grammar, as Jenkins had done in her earlier book on phonology. One looks forward to more fruits from the Vienna VOICE corpus of ELF spoken English. In the meantime, it is a useful exercise for those interested in varieties of English and the ELF debate to read the books by Mollin and Jenkins side by side. Reviewed by Luke Prodromou University of Nottingham, UK doi:10.1093/applin/amn044 Advance Access published on 24 October 2008
REFERENCES Carter, R. and M. McCarthy. 1995. ‘Grammar and the spoken language,’ Applied Linguistics 16/2: 141–58. Cook, G. 1998. ‘The uses of reality: A reply to Ronald Carter,’ ELT Journal 52/1: 57–63. Jenkins, J. 2000. The Phonology of English as an International Language. Oxford: Oxford University Press. Jenkins, J. 2003. World Englishes: A Resource Book for Students of English. London: Routledge. Jenkins, J. 2006. ‘Current perspectives on teaching World Englishes and English as a Lingua Franca,’ Tesol Quarterly 40/1: 157–81.
Joseph, J. 2004. Language and Identity. London: Macmillan. McCarthy, M. and R. Carter. 1995. ‘Spoken grammar: What is it and how can we teach it,’ ELT Journal 49/3: 207–18. Prodromou, L. 2008. English as a Lingua Franca: A Corpus-based Analysis. London: Continuum. Seidlhofer, B. 2004. ‘Research perspectives on teaching English as a lingua franca,’ Annual Review of Applied Linguistics 24: 200–39.
